Arachne DOS Coders1 Message Board ,----------------------------------------------------------------------------- | | DOS Coders1 Message Board | | [ For the Operational discussions of the Arachne C Coders ] | ---------------------------------------------------------------------------- | | | ============================================================================ | | Thu, Jan 01, 2004 - 6:39:05 am | | glennmcc> | A hint ? | | Sure..... no problem. | | Set it up as described above..... | _______________________________________________________________________ | | Try doing that same experiment but this time while using command.com | | | and a copy of mime.cfg with all of the [nnn] mem specs in-place......| | ------------------------------------------------------------------------ | | Now, leave it that way. ;-) | | | ============================================================================ | | Thu, Jan 01, 2004 - 6:53:47 am | | glennmcc> | Happy New Year !!!! :))) | | ============================================================================ | | Thu, Jan 01, 2004 - 9:37:44 am | | ray> | You tease! | | So there is no way to use 4dos with the [nnn]'s after all :-( | If I have to choose, I'll stick with 4dos inspite of the problem. | I *hate* command.com the way you hate Windows. Without 4dos, I would | have gone to Linux ages ago. | | Hmmm ... unless I can get Arachne to launch from 4dos, run under | command.com, shell to 4dos, and quit back to 4dos ... it might be | possible. | | Anyway I'm still stumped as to why Arachne has this problem with | 4dos, as I said I've never seen anything comparable. | | Happy New Year! Š| | | ============================================================================ | | Thu, Jan 01, 2004 - 9:42:54 am | | glennmcc> | Arachne does _not_ have a problem. | | The problem is 4dos itself. | | And the folks who wrote 4dos know that it has problems. | | Why else would include 4dostest.exe ??? | | | ============================================================================ | | Thu, Jan 01, 2004 - 12:47:48 pm | | ray> | ... not pointing fingers either way: 4dos + Arachne = problem, that's | all I know at present. Since the onus is on 4dos to maintain | compatablily, it would seem logical to 'blame' it for the problem. | OTOH, (as I said) this is the one and only compatablity problem I've | ever had with it, so I'm mighty curious as to why. From what I can | determine, it seems that Arachne is never satisfied that '[nnn]' | KB of memory is available even when it is (I'm "160+"). Possibly, there | is some minor tweek in Arachne that could deal with this, or maybe not. | Anyway, the problem is a very minor one -- really just a question | of programmers curiosity. | | 4dostest.exe eh? ... Haven't come across it ... I'll have a look-see. | | | | ============================================================================ | | Thu, Jan 01, 2004 - 7:42:46 pm | | ray> | Glenn, | | I googled '4dostest.exe' and found nothing; and it' isn't part | of the 4dos distribution package. Can you tell me any more about | it, or where I can get it? | | ============================================================================ | | Fri, Jan 02, 2004 - 3:05:04 am | | Joe> | Interesting problem. If "Arachne is never satisfied that '[nnn]' KB | of memory is available", then why would you need to _remove_ the | '[nnn]' stuff?! If that was the exact problem, you would simply get Š| the same number of beeps, with or without the '[nnn]' stuff. So the | problem cannot be that "Arachne is never satisfied ...", it must be | what happens after Arachne _is_ satisfied ... | | | | ============================================================================ | | Fri, Jan 02, 2004 - 4:19:29 pm | | ray> | Logic ... the ultimate weapon ... | | Good thinking Joe, but there might be more to it than that (might!). | I've managed to get Arachne to swallow '[nnn]' strings in front of | functional mime.cfg lines, with the result being just as you predict -- | the same number of beeps as before; but there are other lines that | won't swallow the '[nnn]' at all -- she gives the usual error | messages but don't do nothin else. It's the sort of thing | that only Glenn can deal with, but I don't think he's sympathetic to | the needs of a 4doser like me. I'll get to the bottom of it one day, | in the mean time it's just a very small anoyance. | | ============================================================================ | | Fri, Jan 02, 2004 - 7:49:26 pm | | Joe> | I think the problem is "comspec" - make sure this env. variable points | to '4dos.com' ... my experiment with 4dos 5.5.2 showed that this did | not modify/set "comspec" automatically. Cheers. | | ============================================================================ | | Sat, Jan 03, 2004 - 11:13:56 am | | ray> | joe, | | No, it's not 'comspec': 4dos 6.02 sets that fine; also there is | no problem shelling out (ALT-E takes me to a 4dos shell with no | problems) ... it's something much more subtle. | | | ============================================================================ | | Sat, Jan 03, 2004 - 4:08:02 pm | | glennmcc> | Yep, yopu are right. | | I worked on it for several hours last evening and I'm working on it | again this evening. | | I think that it might be a problem with environment space. Š| | Not what we set in our shell line in config.sys but with what 4dos uses | by default when called from within another program. | | I think that if I can figure-out how to get Arachne to assign more | environment space when calling 4dos.com from inside of core.exe...... | we will have the 'fix' we need. | | Shelling to dos with Alt+E is no problem becuase we assign the | environment space via this line in arachne\system\dgi\dosshell.bat | | %COMSPEC% /E:512 | | Now to find out hopw to do that when we are staying inside of | core.exe and not 'shelling-out' | | ============================================================================ | | Sat, Jan 03, 2004 - 4:10:08 pm | | glennmcc> | BTW, | I am doing my testing with 4Dos v6.01 | | ============================================================================ | | Sun, Jan 04, 2004 - 12:48:40 pm | | ray> | Glenn, | | ... Now that's a HOT lead; I haven't looked into it yet, but while | I was studying the 4dos docs (in the troubleshooting section I think), | there was some mention of just that -- programs having trouble | calling the command processor due to ( ... what was it ? ... ) the | environment being too large? | | BTW, following a tip I got in the | dusty archives I tried 4dos with '1.6b1' and found that I *could* | get to the inbox w. 4dos (6.02) and the '[nnn]'s' intact in 'mime.cfg', | but could not read any messages. That seems to sugest that | 4dos and Arachne will read '[nnn]'s' in *principal* even if not | in *practice* ... back to the lab. | | | | ============================================================================ | | Sun, Jan 04, 2004 - 2:54:59 pm | | ray> | Glenn, | | The results of some tests with environment size on mime.cfg: | Š| ;Arachne shows "157+", Evironment is 512, 381 free. | | ;No. No beep, memory is available, so Arachne tries internal execution but f | ;file/inbox.dgi >HTM|[147]$einsight.exe -i $M*.cnm -cache=$t>$2 | | ;Yes, but beeps. Requested mem is unavailable and Arachne uses external exec | ;file/inbox.dgi >HTM|[148]$einsight.exe -i $M*.cnm -cache=$t>$2 | | ;Arachne shows "160+", Evironment is 1024, 607 free. | | ;No. No beep, memory is available, so Arachne tries internal execution but f | ;file/inbox.dgi >HTM|[150]$einsight.exe -i $M*.cnm -cache=$t>$2 | | ;Yes, but beeps. Requested mem is unavailable and Arachne uses external exec | ;file/inbox.dgi >HTM|[151]$einsight.exe -i $M*.cnm -cache=$t>$2 | | Conclusion: A smaller environment seems to make no difference. 4dos + | Arachne can process the '[nnn]' strings OK, but any attempt at | internal execution fails even though plenty of memory is available. | | | ============================================================================ | | Sun, Jan 04, 2004 - 3:10:31 pm | | glennmcc> | Fantastic testing. | | Now we're getting somewhere. | | Were all of those tests with Arachne v1.6b1 ? | | | ============================================================================ | | Sun, Jan 04, 2004 - 8:16:16 pm | | Joe> | Arachne 1.6b1??? Well, I was wondering why Ray had so much | free memory in Arachne. I was suspecting he had a "radical" | memory configuration that was causing him side-effects. My | own brief tests with 4dos 5.5.2 worked fine, I just had to | set "comspec", that's all. | | | ============================================================================ | | Sun, Jan 04, 2004 - 10:18:03 pm | | ray> | Glenn, Joe, | | The above was under 1.73GPL (should have reported that ...) I'll | try the same strings under 1.6b1 tommorow. | Š| Hey Joe, the above '160+' is with your lsppp ... sorry not *your* | lsppp package but loaded high, that's why I've got so much memory :-) | And you were quite right that my earlier conclusion about Arachne | "thinking there isn't enough memory when there is" was wrong; it seems | the opposite is true -- if she thinks there *is* enough memory, that's | when the problem occurs. | | | ============================================================================ | | Mon, Jan 05, 2004 - 5:13:22 pm | | ray> | Hey gregy ... you haven't frozen to death on us have you? It's even | cold here in Vancouver; -12C is almost unheard of. | | ============================================================================ | | Mon, Jan 05, 2004 - 5:20:05 pm | URL | | glennmcc> | ray, | | It's even a more complex problem than we first thought. :( | | Try the tests under URL. | | | ============================================================================ | | Mon, Jan 05, 2004 - 5:29:17 pm | | glennmcc> | With 4dos.com, | | load error | the word "testing" is shown but no beep is heard | screen blinks, no beep | load error | "testing", no beep | screen blinks, beep sounds | | | With command.com, all of them work perfectly. | | "testing" is shown, beep is heard | "testing", beep | screen blinks, beep | "testing", beep | "testing", beep | screen blinks, beep | | ============================================================================ | Š| Mon, Jan 05, 2004 - 8:41:56 pm | | ray> | Glenn, | | What do you mean "under URL" ? | | Anyway, I tried the previous tinkering with 1.6b1 and: | | ;135+, goes to inbox w.o. beep! ... but can't view any msg. | ; after dialing, 66+ and all is OK, but with beeps! ... Arachne is | ;ignoring the absence of the [nnn] string! When memory is high, she | ;tries 'internal' and fails, when memory is low, she shells to DOS | ; and suceeds. | ;file/inbox.dgi >HTM|$einsight.exe -i $M*.cnm -cache=$t>$2 | | ;All identical to above. | ;file/inbox.dgi >HTM|[130]$einsight.exe -i $M*.cnm -cache=$t>$2 | ;All identical to above. | ;file/inbox.dgi >HTM|[136]$einsight.exe -i $M*.cnm -cache=$t>$2 | | ... So much for the notion that there is no problem with 1.6b1. | But there is one big difference, and that is that at least you can | get to the inbox w.o. any beep (even if you can't read any messages). | | My suspision is that those folks who report 'no problem' with | 4dos and whatever version of Arachne are infact running in a low memory | situation, such that Arachne never tries to pull the 'run internaly' | trick ... she is just shelling to DOS and running things that way. | | Joe, can you confirm or refute this? The trick would be to put a | 'beep' into your 'arachne.bat' and see if you get beeps while | acessing your inbox. If you do get beeps, were no farther ahead, if | you don't get beeps, then we've got to figure out why, but we're on | the scent. | | ============================================================================ | | Mon, Jan 05, 2004 - 9:09:26 pm | | glennmcc> | ray, | | I just read your post on the Yehaw group concerning what you are | 'thowing out' of the SRC code files. | | Please, stop for a moment and ponder this.......... | | I have been working with this code for almost 3 years. | | The number of improvements, bug fixes, feature additions, etc...etc... | which I have made is too numerous to put an exact number onto right off | the top of my head. | | Q) How much of the code that I started with did I delete ? Š| | A) None of it. | That's right... absolutely no deletions were ever done. | | Every change that I made is via commenting-out of the old section and | adding my section. | | Every one of my changes is fully explained to the best of my ability. | (right down to the simple little comments sying what a new variable name | is being used for) | | --- such-as this section in misc.c --- | //!!glennmcc: begin Apr 23, 2002 | //needed for $K | char *cfgkw="_"; | int kcnt=0; | //!!glennmcc: end | ________________________________________________ | | | Q) Why would I keep all of the old stuff that is no longer being used. | (some it has not been used since v1.10) | | A) Simple, all of it _is_ still used. | Not in the executable itself but rather by the mind of the person | looking at the code. | | Q) Why would a programmer need to see how it | used to be done, but no longer is being done ? | | A) For the same reason that the old saying came-to-be. | | If we do not learn from history... we are destined to repeat the same mistak | By seeing how a particular section of code was before and now is... | This helps a programmer to better understand why the change was made. | | It also helps them to find even better ways of doing it. | | Believe me when I tell you that had it not been for some of those old | sections of commented-out code.... v1.71;UE01 never would have come-to-be. | And of-coarse... neither would v1.73;GPL | (nor the one I am using right now...... v1.75;GPL) | | Again, please believe me when I say that: | | EVERY LITTLE TINY BIT OF PREVIOUS CODE IS STILL NEEDED. | PLEASE, DO NOT DELETE SO-MUCH-AS A SINGLE CHARACTER. | | (capitalized for emphisis... not meant to be taken as shouting) | | | ============================================================================ | | Mon, Jan 05, 2004 - 9:10:47 pm | URL Š| | glennmcc> | What do you mean "under URL" ? | | The URL above my nickname on that post. (and now on this one) | | http://www.cisnet.com/glennmcc/my-stuff/4dostest.htm | | | ============================================================================ | | Mon, Jan 05, 2004 - 9:51:42 pm | | ray> | Glenn, | | Yes, now I understand ... will get to it tommorow. | | And thanks for your input vis. cleanning; I very greatly want as much | consensus as possible on that subject, and I don't consider my | own oppinion to be very important, but I will argue a point for the | sake of understanding. Anyway, I'm not sure if you are aware of | some of the backgound; we've been discussing it for quite some time, | sepecialy Michal and I. I was only talking about deleting truely | trivial 'ID' comments like " xxxx //mp" or "xxx //haro" ... | I wouldn't delete any other comments. As far as code goes, I | agree with what you say. I have only deleted very large blocks of | commented out code for compactness sake, and these, if needed | can be re-inserted without much trouble as required from the originals. | When I started this I was without any guidance at all, and now that | I have some feedback I'm quite prepared to go thru the whole thing | again to make whatever adjustments in content or style are | deemed necessary. Needless to say, your oppinions will carry the | greatest weight. | | | ============================================================================ | | Mon, Jan 05, 2004 - 10:29:31 pm | | glennmcc> | re: | I was only talking about deleting truely trivial 'ID' comments like | " xxxx //mp" or "xxx //haro" ... | | How does one define 'trivial' ? | | | IMO, | There is no such thing as 'trivial'. | Not even "//mp" or "//haro" | | In the lower right corner of "Nude descending a stair" | | Is "Duchamp" trivial ? Š| | As for..... | | "I have only deleted very large blocks of | commented out code for compactness sake, and these, if needed | can be re-inserted without much trouble as required from the originals." | | How will future developers know that those large blocks even exist in | the originals ? | | Are you planning to package the SRC with duplicate files ??? | | sendmail.c | sendmail.old | misc.c | misc.old | | etc... etc..etc... | | Or perhaps in 2 seperate zip files ? | | asrc-new.zip | asrc-old.zip | | If so... there goes the 'compactness' right-out-the-window. | | | ============================================================================ | | Tue, Jan 06, 2004 - 7:47:06 am | | Joe> | Hi Ray, | | I put peeps in the other day, when I tried 4dos 5.5.2, everything worked | as expected. Beeps when memory was insufficient, etc. I cannot check | mail stuff because I don't have a POP3 account . | | ============================================================================ | | Tue, Jan 06, 2004 - 7:50:45 am | | Joe> | Hi Glenn, | | Speaking of code history, what is the earliest source code that can | be made available? I'm thinking 1.69 and/or 1.70r3 would be useful. | For instance, I'd like to know at which version IE_MAXLEN was increased | to 600 (it's now 800) and what it was prior to this. | | ============================================================================ | | Tue, Jan 06, 2004 - 8:51:22 am | | glennmcc> Š| The earliest one I have is v1.64 | | It is zipped right now. | | I'll have a look at it later today and get back to you with the answer. | | | ============================================================================ | | Tue, Jan 06, 2004 - 9:02:11 am | | glennmcc> | Well..... it's later today. | | | ---------- l:\asrc164\ie.h | [7]#define IE_MAXLEN 512 | | ---------- l:\asrc166b\ie.h | [7]#define IE_MAXLEN 512 | | ---------- l:\asrc167\ie.h | [7]#define IE_MAXLEN 512 | | ---------- l:\asrc169\ie.h | [7]#define IE_MAXLEN 600 //to handle longer cookies | | ---------- l:\asrc170\ie.h | [7]#define IE_MAXLEN 600 //to handle longer cookies | | ---------- l:\asrc171\ie.h | [8]#define IE_MAXLEN 800 //to handle longer cookies | [9]//#define IE_MAXLEN 600 //to handle longer cookies | | ---------- l:\arachne\asrc\asrc173.gpl\ie.h | [8]#define IE_MAXLEN 800 //to handle longer cookies | [9]//#define IE_MAXLEN 600 //to handle longer cookies | | | ============================================================================ | | Tue, Jan 06, 2004 - 9:46:23 am | | Joe> | Thanks for that Glenn, | | I'm thinking of shrinking IE_MAXLEN again (excluding cookies, AFAICT | the minimum is now about 740, as required for the Cache Index) ... | | Oh, I just remembered something else - is the Miniterm source | available (I'm guessing this is supposed to be GPL mow too). | | ============================================================================ | | Tue, Jan 06, 2004 - 5:55:49 pm Š| | glennmcc> | AFAIK, | Michael has never made the SRC for miniterm available. | | | | ============================================================================ | | Tue, Jan 06, 2004 - 6:17:21 pm | | ray> | Joe, | | So she beeps when memory is insuffisient, and she does not | beep when memory is sufficient (but she works) ? ... If that's so | then at least some versions of Arachne can work with some versions | of 4dos. | | ============================================================================ | | Tue, Jan 06, 2004 - 6:41:02 pm | | glennmcc> | "Nude descending a stair" is the title of a painting | by an artist named Duchamp. | | http://faculty.dwc.edu/wellman/conceptual/nude_no2.jpg | | ============================================================================ | | Tue, Jan 06, 2004 - 8:14:36 pm | | ray> | Hmmm ... she's thought provoking. Superb color disipline. It will | be years yet before I really know how to judge cubist and latter | art, but I'd guess that she's a masterpiece. | | BTW, Glenn, can you send me (or upload) a copy of 'beep.com'? For | consistency, I'd like to use it testing those .dgi's you posted. | | | | ============================================================================ | | Tue, Jan 06, 2004 - 8:37:50 pm | URL | | glennmcc> | Need new glasses... do we ???? | | ============================================================================ | | Tue, Jan 06, 2004 - 8:55:22 pm | Š| ray> | Ooops, not new glasses, new way of thinking. The sad fact of the matter | is that I'm still getting used to the 'click on link' way of | making things happen :-( | | ... and while you're explaining the obvious ... how do I run those | 'made up' .dgi's ? As you no doubt noticed I had to 'steal' an | existing .dgi to run my tests; your way looks like it will be a whole | lot more elegant. | | ============================================================================ | | Tue, Jan 06, 2004 - 9:58:01 pm | | glennmcc> | How do you run them ??? | | You said it yourself......... | | .... 'click on link' way of making things happen.... | | http://www.cisnet.com/glennmcc/my-stuff/4dostest.htm | (contians 'clickable' links to each of them) | | | | | ============================================================================ | | Wed, Jan 07, 2004 - 7:25:49 am | | Joe> | Ray, | | About the beeps : yes. BTW, my beep was something like "echo ^G", | but with a real Ctrl-G character (in Borland editors, probably | borrowed from WordStar, I expect, you can prefix a control character | you want with ^P and the editor inserts the control character into | the text, instead of trying to interpret it as a command/sequence). | | I still have doubts about your memory configuration being too | "radical". Core.exe takes about 492,000 standard or about 16,000 | less if compiled for 386+387, so adding 160,000 gives a total of | 636,000 to 652,000 which implies as little as 3,000 bytes for | all other conventional memory usage. Just a hunch, but does 4dos | work properly for this stuff if you use a slightly more conservative | memory configuration? | | | ============================================================================ | | Wed, Jan 07, 2004 - 8:38:28 am | | Joe> | Thanks Glenn, Š| | I'll ask MP about Miniterm ... | | | ============================================================================ | | Wed, Jan 07, 2004 - 3:35:05 pm | | ray> | Glenn, | | AAAAA, now I see ... I had saved that page as text, just looking for | them 'mime' strings. Gotta learn something about HTML. | | | ============================================================================ | | Wed, Jan 07, 2004 - 4:58:02 pm | | ray> | Glenn, | | I started to look at your test package, but that 'beep.com' | crashes my system! | | ============================================================================ | | Wed, Jan 07, 2004 - 7:54:31 pm | URL | | glennmcc> | That's because I screwed it up when I uploaded it the forst time. :(( | | The .HTM itself also had incorrectly formatted 'clickable links'. | | I have fixed both of the screw-ups and uploaded the fixed files. | | Sorry about that. | | ============================================================================ | | Wed, Jan 07, 2004 - 8:12:43 pm | | glennmcc> | Just back from reading your posts on Yehaw. | | Modules using memory below 1 MB: | Name Total Conventional Upper Memory | -------- ---------------- ---------------- ---------------- | SYSTEM 33,760 (33K) 10,848 (11K) 22,912 (22K) | HIMEM 1,168 (1K) 1,168 (1K) 0 (0K) | EMM386 4,320 (4K) 4,320 (4K) 0 (0K) | 4DOS 7,392 (7K) 432 (0K) 6,960 (7K) | NANSI 3,184 (3K) 0 (0K) 3,184 (3K) | RAMDRIVE 1,456 (1K) 0 (0K) 1,456 (1K) Š| IFSHLP 2,864 (3K) 0 (0K) 2,864 (3K) | SMARTDRV 35,056 (34K) 0 (0K) 35,056 (34K) | 320-180C 5,296 (5K) 0 (0K) 5,296 (5K) | KSTACK 1,072 (1K) 0 (0K) 1,072 (1K) | CTMOUSE 3,360 (3K) 0 (0K) 3,360 (3K) | NOVAFONT 5,264 (5K) 0 (0K) 5,264 (5K) | Free 708,896 (692K) 638,000 (623K) 70,896 (69K) | | Memory Summary: | Type of Memory Total Used Free | ---------------- ----------- ----------- ----------- | Conventional 655,360 17,360 638,000 | Upper 158,320 87,424 70,896 | Reserved 0 0 0 | Extended (XMS) 132,945,296 12,872,080 120,073,216 | ---------------- ----------- ----------- ----------- | Total memory 133,758,976 12,976,864 120,782,112 | Total under 1 MB 813,680 104,784 708,896 | Largest executable program size 637,984 (623K) | Largest free upper memory block 70,688 (69K) | MS-DOS is resident in the high memory area. | | | ============================================================================ | | Wed, Jan 07, 2004 - 8:13:17 pm | | glennmcc> | Those memory stats say it all.... | | Ferchrissakes man. | | Not only are you using 4dos.com.... | | But you are booting to MShitDos v7.x that ships with WinShit9x !!!! | | There's no help for you. :( | | | ============================================================================ | | Wed, Jan 07, 2004 - 8:19:09 pm | | glennmcc> | That HDD is probably formated to FAT32 also. | | And if you are also trying to do all of this stuff from within a | 'dosbox' while still inside of W9x....... | | No wonder you are having all kinds of problems. | | My advice... | (take it for what it's worth or ignore it.... you choose) | | Dump WinShit by reformating the HDD to FAT16 Š| | Install OpenDos v7.01 | | Do not use 4dos | | Run any program you'd like.... problem-free. :))) | | | | ============================================================================ | | Wed, Jan 07, 2004 - 8:58:38 pm | | ray> | Glenn, | | Ooops, The information below was acumulated before reading | your post above, so you might not be interested anymore. Anyway | I'll respond to your remarks above first: | | Yes, one of my partitions has windows on it and is formated FAT32. | I consider FAT32 to be a natural progression from FAT16, just as FAT16 | progressed from FAT12. I don't like Windows neither, but there are | things I *have* to use it for, so I usualy boot to my FAT32 partition. | | I have found that almost all of my software (including Arachne) | works just fine on it (although I lament bitterly having to use | Dumbdrive instead of Ncache2) M$DOS7.10 seems benign enough to me that | I don't get a knot in my stomach using it. No, I don't try to | run DOS in a bleading window box; I'd be worthy of your reproach if | I did! | | The 4dos + Arachne problem seems identical in my DRDOS partition. And | this isn't 'my' problem; the 4dos + arachne problem has been around | for a long time, and on many different systems. As for 4dos, as I said, | this is the only incompatablity I've come across. I'm not the only | DOS user who is absolutely commited to 4dos, and it remains my opinion | that Arachne is considerably hobbled by her this problem -- it makes | her look bad *regardless* of who's 'fault' it is. | | Anyway, if your still interested: | | I found another 'beep.com' that works with my system (and only 32 bytes). | I found that the .dgi's 'beep' worked, except when I turned on the 'beep' | inside of 'arachne.bat'; even when I had the frequency inside of 'arachne.ba | set quite different from that of 'beep.com', only the 'arachne.bat' | beep was heard. So I changed my 'shell' test to a 'pause' inside | 'arachne.bat'. | | These were my results (DOS 7.10, 4dos 6.02, 160+) using the strings | that you posted: | | No shell, no sound, no 'testing', page must be reloaded to reactivate mouse. | file/beepechonoshell.dgi >txt|[130]beep.com \n echo testing>$2 | file/beepechonoshellnoext.dgi >txt|[130]beep \n echo testing>$2 Š| | No shell, no sound, 'testing' shows OK. | file/beepechoshell.dgi >txt|beep.com \n echo testing>$2 | file/beepechoshellnoext.dgi >txt|beep \n echo testing>$2 | | Shell, Beeps OK, no 'testing' (obviously). However the 'beep' came | *before* the pause, in other words 'beep.com' was executed internaly, | *then* the shell was done! | file/beepnoshell.dgi |[130]beep.com | file/beepnoshellnoext.dgi |[130]beep | | Conclusions: | | The '.com' extension makes no difference (a bummer if it did). | These results seem difficult to reconsile with my preious tests, which seeme | to show that Arachne could only execute a .dgi if the [nnn] string | was set artificialy high, thus forcing her to shell out. The first | pair of tests above seem to accord with that conclusion, but the second | and third pair seem to display the exact opposite result: In the second | pair she does not shell, and in the third pair, she does, but the beep | comes *before* the shell! | | More work on this tommorow. | | One more observation: Durring the first pair of tests above, I got this | msg: "Duplicate redirection ">nul" ... from $roura$.bat. I don't know | what it means, but I'll bet it's important. | | | | | ============================================================================ | | Wed, Jan 07, 2004 - 9:17:21 pm | | glennmcc> | Sorry if I stepped on your toes with my comments. | | I did mean to. | | | Ah HA !!! | | Yes, that means something. | | This...... msg: "Duplicate redirection ">nul" ... from $roura$.bat. | | suggests to me that in some way we have yet to determine, 4dos.com is not | interpreting the command line in the same way as does command.com | | Now... let's see if we can 'capture' the sequence of events. | | This does warrant further investigation. | | Š| ============================================================================ | | Wed, Jan 07, 2004 - 9:18:16 pm | | glennmcc> | Damn typos.... that should have been | "I did _not_ mean to." | | ============================================================================ | | Thu, Jan 08, 2004 - 12:42:49 pm | | gregy> | Hi....belated Happy New Year....!!!!....I've been offline for 5-6 days | ....I'll try to catch up and reply to messages later today... | | | | ============================================================================ | | Thu, Jan 08, 2004 - 5:11:51 pm | | ray> | Gregy, welcome back! | | Glenn, | | Sure, and thanks. I make it a policy never to take offence, although | 'prior restraint' is something I do apreciate. Glad to hear the | problem still interests you. | | | | ============================================================================ | | Thu, Jan 08, 2004 - 8:11:26 pm | | ray> | Glenn, | | I had a whole load of tests to show you, but I managed to overwrite | the whole thing :-< ... anyway here is something unexpected: | | (Now that I can see how your test file works I modified it | slightly: "test1.dgi" etc. etc. but otherwise the same). | | ;MIME.TST | | "Arachne load error" on 'press any key' | ;file/test1.dgi >TXT|copy /b _4prt.txt $s>NUL \n del _4prt.txt \n if exist $ | has been saved to $s>$2 | | Yes! No Shell !!!, copy OK, Msg is html! | file/test1.dgi >TXT|copy /b $earachne.bat r:\>NUL \n echo file has been save | Š| Shell !!! Copy OK, No msg. | file/test2.dgi |copy /b arachne.bat r:\>NUL \n echo file has been saved! | | The 'duplicate redirection ">nul" msg. is cured by using "\n" as | the command separator, not "|" ... finding out why should proove | interesting. | | Note that in the second test above, the copy and echoing of the | message happens *without* shelling !, and with no problems. | | The third test seems to show that without the ">TXT", not only | does the message not echo, but Arachne *will* shell ! This | all seems bisare to me, but I checked it four times, there is no | mistake. | | What I lost in the stuff I overwrote showed things working, but | the message echoing in text mode (not html), but I can't recall | what the hell I did to get that result. | | ============================================================================ | | Thu, Jan 08, 2004 - 9:28:49 pm | | glennmcc> | Here's one problem in your test lines. | | file/test2.dgi >TXT|copy /b arachne.bat r:\>NUL \n echo file has been saved! | | file/test2.dgi >TXT|copy /b arachne.bat r:\\>NUL \n echo file has been saved | | Dir specs require \\ in mime.cfg lines. | | | | ============================================================================ | | Thu, Jan 08, 2004 - 9:30:03 pm | | glennmcc> | I'll try to explain the best I can how this mime.cfg line works. | | file/test2.dgi >TXT|[100]copy /b arachne.bat r:\\>NUL \n echo file has been | | >TXT is used 'internally' by core.exe to plce the correct file extension | on the final resulting file in the cache dir. (>TXT does not get 'fed' | to the command interpretor on the command line) | | | is also used 'internally' by core.exe to indicate where the beginning | of the command line processing is to be. | | [nnn] again is used 'internally' to tell core.exe not to shell out if | there is more than nnn kb of memory currently free. | | Now, this/these command line(s) is/are formed. | Š| copy /b arachne.bat r:\>NUL | echo file has been saved! | | The last part... >$2 ... is again an 'internal command' to core.exe | telling it to now view the resulting file in our cache dir. | | OK, | Is that just about as clear as mud ???? | | ============================================================================ | | Thu, Jan 08, 2004 - 9:35:51 pm | | glennmcc> | Oh, I forgot to explain why \\ is needed for dir specs. | | core.exe 'strips' the first \ and then uses the next character to | determine what to do next. | | \n becomes n which is then interpreted as 'new line' | | c:\\download\\ becomes c:\download\ to indicate a dir spec | | | | ============================================================================ | | Thu, Jan 08, 2004 - 9:44:46 pm | | glennmcc> | Also notice that my example will not work. | | At the bottom of mime.cfg we find................ | | ; \n ... end of line (note: this won't work for "|[nn]xx" style commands!) | | So all those 'multi line' examples which include [nnn] will 'fail'. | | | | ============================================================================ | | Thu, Jan 08, 2004 - 9:54:14 pm | | glennmcc> | I just checked to see why they worked (seemed to work) for me before. | | I had my CD-Rom drivers loaded. | | So I only had 80kb of free memory. | | \n works with [nnn] when not enough memory is availabe to stay inside | instead of shelling. | | When memory is low... we shell and \n works. Š| | When memory is high enough to stay inside..... \n fails. | | | | ============================================================================ | | Thu, Jan 08, 2004 - 9:57:49 pm | | glennmcc> | BTW, | | Removeing [nnn] 'forces' a shell just as does @ but simply does not | switch to text mode. | | Our screen will blink memenarily as it is re-rendered but we don't go to | a black screen with our command line showing as we would if we used this | line in mime.cfg | | file/test2.dgi >TXT|@copy /b arachne.bat r:\\>NUL \n echo file has been save | | | ============================================================================ | | Fri, Jan 09, 2004 - 5:18:58 pm | | ray> | Glenn, | | Sometimes things get murkier before they get clearer, eh? | | Much to say but for now the bad news: In 'utilities' I tried 'cache | index' (for the first time) and got 'incompatible version' :( ... | so everything I've reported is possibly compromised by that fact. | I rmember that you did something that triggered that but I can't | remember what ... something about buffer being enlarged? ... can | you quote me the affected line in wwwman so that I can recompile? | | Yeah, wasn't it URLBUF or something like that .... | | | ============================================================================ | | Fri, Jan 09, 2004 - 5:21:36 pm | | ray> | OH, one more thing for now. Note that one of my tests above, although | it had no [nnn] did *not* shell out! That's what I found so | interesting, it sounds like it should be impossible. | | | ============================================================================ | | Fri, Jan 09, 2004 - 8:55:25 pm | URL Š| | glennmcc> | No need to recompile wwwman.exe | | I already did it. | | Link is to new3wman.zip which contains the 'fixed' .EXE | and the updated SRC code. | | ============================================================================ | | Fri, Jan 09, 2004 - 9:01:40 pm | | glennmcc> | That sounds like it's impossible because it IS impossible. | | | [nnn] == do not shell unless nnn kb is not available. | (stay in graphics mode in either case, do not switch to text mode) | | No [nnn] spec == shell to Dos but stay in graphics mode, do not switch to te | | @ == shell and switch from graphics mode to text mode. | | And that is the only way possible for it to work. | | | ============================================================================ | | Fri, Jan 09, 2004 - 9:03:40 pm | URL | | glennmcc> | Oops, bad link. :( | | Try this one. | | ============================================================================ | | Fri, Jan 09, 2004 - 9:05:03 pm | URL | | glennmcc> | Oops again. :( | | The SRC is not in that .ZIP | | If you need it let me know and I'll upload it. | | ============================================================================ | | Fri, Jan 09, 2004 - 9:42:02 pm | | ray> | For uptodateness I'd like the SRC, but I thought it was just one Š| line OTOH, just as easy to download the whole thing again. | | But I gotta tell ya Glenn, the above 'imposibility' did happen. I tried | with both a beep and a pause in 'arachne.bat', and there was no | mistake: the string above did *not* shell (i.e. the batch file was | not called). I couldn't believe it neither but I ran it four times. | Maybe this is a clue to the problem ... some mistake parsing | the lines. I'll run those strings again, but ... | | | ============================================================================ | | Fri, Jan 09, 2004 - 9:42:15 pm | | gregy> | Yes...please....upload and link to it.......(me, always with the | inconvenient request).... | hey...Ray........I've a request for you coder-guys... | | | ============================================================================ | | Fri, Jan 09, 2004 - 9:48:08 pm | | gregy> | Ray...Joe...Michal...et al you coder-cleaners....REQUEST: | ... | While you're cleaning each file...you get SOME idea of what the | overall function/purpose of the file is....such as that file Michal | says has all the function-calls set out in COMMENTs.... | ... | So...while your cleaning...or just after you finish a module...would | you jot down notes for yourself as to what that module is for, etc... | ...and then type those notes into a plain text file (i.e. MAIN.C | would have MAIN.TXT as its corresponding notes file) and upload it | here, or email attach to me...??? | ... | We are going to have to do this, sooner-or-later....and it would be | wise to get an informal start on it, right now... | ... | | | | ============================================================================ | | Fri, Jan 09, 2004 - 9:58:01 pm | | gregy> Š| Also....I now have a complete site for this ArachneDevelopment | Community.....175Mbytes storage, 1Gbytes traffic, etc...plus a bunch | of other bells and whistles.....it's a "paid-up" site...already in | existance....and I just need to clean some things up and set up some | file and message board areas.... | ... | FWIW....I am still looking long-term at getting a SourceForge site | set up for CORE.EXE...INSIGHT.EXE...WWWMAN.EXE....because the CVS | will come in real handy there (if not real crucial, there)...but.. | ...NOTE...I think we cannot host the whole ArachnePackage there... | ...because....Arachne is NOT just CORE, Insight, WWWMAN...it is a | WHOLE bunch of other linked-in little packages....many of which are | not ours...belong to other authors...or...they are small utilities | (such as APM.EXE) which do NOT require CVS, etc...so...we need a | central HOST site, linking to the other subsidiary sites (such as | these message board sites....TheChicagoBear's and Bastiaan's download | sites.....whereever.).... | ... | For another example....Joe...your suggestion to replace PKUNZIP with | the free InfoZIP (or something such) is not really necessary...we can | give users the option....but...PKUNZIP is free for non-commercial | use...and.....we are NEVER going to be a "commercial use" in this | ArachneGPL community...so...why should we go to the trouble of putting | another unZIP in the package...?? | ... | Have I stated this clearly....???....or have I just confuzzeled all | of you...??? | | | ============================================================================ | | Fri, Jan 09, 2004 - 10:31:16 pm | | ray> | OHHH NOOooo .... | | I love Arachne, but she can be very cruel. I was retesting the | 'imposibility', the echo string I had been using was "file has been | saved!", I changed the string to "this is test1.dgi", ran it and, | nothing changed (still "file has been saved!") I searched my entire | computer, eventualy, for that string and it does not exist on my | computer! After much gnashing of teeth I realized that to get things | to work properly the file you sent had to have the HTML edited to | call "file://test1.dgi" not "file:test1.dgi". After the change, | you said it, ... it beeps :-( and shows the correct message. So I'll | eat my hat if you want. But God knows where that "file has been saved!" | string was comming from, and why the "file:test1.dgi" string | 'sort of' worked -- just enough to throw me off the scent. Anyway, you | were right the 'imposiblity' is impossible. | | | | ============================================================================ | | Fri, Jan 09, 2004 - 10:36:33 pm Š| | ray> | gregy, | | RE taking notes: I'm sort of doing that. As I've gone thru each file | I've noted the #includes, and made notes about what each file seems | to be doing (mostly just quoting Michael's own notes). I'm hoping it | will be expanded eventualy to give a full explanation of the | geography of the project. | | | ============================================================================ | | Fri, Jan 09, 2004 - 10:55:18 pm | | glennmcc> | ray, | | See ? | | I told you so. | | Couldn't resist... had to say it. | | ============================================================================ | | Fri, Jan 09, 2004 - 11:00:28 pm | URL | | glennmcc> | gregy, (and ray as-well, of-coarse), | | Just now uploded new3wsrc.zip | | | ============================================================================ | | Fri, Jan 09, 2004 - 11:03:01 pm | | glennmcc> | Oh and ray, | | Hey, look at it this way... | | | You just leaned another good 'rule' to try and follow. | | file://filname == good | | file:filename == bad | | | | ============================================================================ | Š| Sat, Jan 10, 2004 - 1:12:15 am | | gregy> | Glenn....thanks...got it... | ... | re:....file://etc | ... | Yes...isn't that the STANDARD for major protocols...???... | ... | seems like that comes up every time...... | ... | hmmmmm...maybe EXCEPT with that cursed javasjunk.... | .. | BTW...I just gave Christof the link to this board..... | | | | ============================================================================ | | Sat, Jan 10, 2004 - 9:56:14 am | | ray> | Glenn, | | Yup, lesson learned ... but "file:test1.dgi" did call something! It | half executed the 'test1.dgi' line in 'mime.cfg', but came up with it's | own echo string -- I'd still like to know what the hell was going on | there. | | ============================================================================ | | Sat, Jan 10, 2004 - 12:15:38 pm | | ray> | Glenn, | | Grabbed your sources; IINM, the only change needed to fix the | 'incompatible version' msg was changing "URLSIZE" from 256 to 512 ? | I recompiled with that change (wwwman.exe = 26448 bytes), and all | seems well, thanks. | | | | ============================================================================ | | Sat, Jan 10, 2004 - 12:18:19 pm | | ray> | Gregy, | | I pending feedback from the others, I at least would like it if this | window (hate to even *speak* that word) ... was a little bigger, it | would make writing a message a little easier IMHO if we could see more | than 6 lines. | Š| ============================================================================ | | Sat, Jan 10, 2004 - 5:33:15 pm | | glennmcc> | What compiler did you use ? | | My copy of wwwman.exe (compiled with Borland v3.1), is 28,040 bytes. | | ============================================================================ | | Sat, Jan 10, 2004 - 5:36:27 pm | | glennmcc> | As to the 6 lines.... | | F4 to edit the SRC of this page. | | Change the rows from 6 to 16 (or what ever you want) in this section. | | | | F2 to save. | | It will of-coarse be back to 6 after posting you message. | | So, F4 again. ;-) | | | | ============================================================================ | | Sat, Jan 10, 2004 - 6:07:20 pm | | ray> | Yikes, I was afraid of this: different folks compiling slightly | different sources with slightly different compiler settings ... it | could get out of hand quickly. That's one of the reasons I wanted | to develop a 'standard code base' so we'd each be refering to the | same thing. | | Anyway, I compiled my cleaned sources, which were identical in | compile, to the ones from you that included MHT's fix, and URLSIZE | was increased to 512. The compile was from the command line using | bcc.exe. The command string was: get back to you, got to retrive it | ... | | ============================================================================ | | Sat, Jan 10, 2004 - 6:14:23 pm | | ray> | ... | | bcc -Ic:\apps\tc\include -Lc:\apps\tc\lib wwwman.cpp Š| | The standard command line compile is always tighter than from the IDE, | since the defaults are for minimum size. | | | ============================================================================ | | Sat, Jan 10, 2004 - 8:27:52 pm | | glennmcc> | The 'standard code base' was D/Led from my site. | | As to all of us using the same compiler settings. | | My way is the only right way to do it. | | For compiling core.exe | 1) cd to the dir containing the SRC | | 2) run bc.exe | (since nokey.prj is the only .prj file in that dir... | it will be opened automatically) | | 3) Alt+C | | 4) choose to build all | | 5) Alt+X to exit BC | | 6) run nokeyexe.bat | (after editing it to point to your 'testing dir' structure | | 7) cd to your testing dir | | 8) run arachmne.bat which will now use your new compile of core.exe | that was copied to that dir by nokeyexe.bat | | Now for wwwman.exe | | 1) cd to the dior contianing the SRC | | 2) run bc.exe | | 3) open wwwman.cpp | | 4) Alt+C | | 5) choose to build all | | 6) Alt+X | | 7) copy the new wwwman.exe to that same 'testing dir' | | 8) cd to that dir | Š| 9) run arachne.bat to use your new compile of wwwman.exe | | Do it 'my way' ... and you will have no problems at-all. | | | ============================================================================ | | Sat, Jan 10, 2004 - 8:39:36 pm | | glennmcc> | Please forgive me for being blunt..... | But are you now re-thinking your position on this guarantee that you made ? | | --------- | Mon, Dec 08, 2003 - 4:02:24 pm | Ray> | Gregy, yeah, don't worry about that -- no code will be deleted at this | point. After the first clean .exe's will be guaranteed to be identical. | --------- | | Do you feel that this is no longer feasable ? | | Please don't get me wrong here. | | I'm not trying to be an ass about this. | | I just want to get it clear where we all stand on our various ideas. | | | ============================================================================ | | Sat, Jan 10, 2004 - 8:43:11 pm | | gregy> | Ray....yes....Glenn's "fix" for more-than-6 lines is the way to go... | ....since we are more-or-less standardized on one MessageBoardSource | ...and we don't want to set it for more... | ... | However....I have an alternative.... | ... | I compose messages, especially lengthy and complicated ones, offline, | using a very simple small editor that came with my first DRDOS version | 3.35..... | ... | I compose/type the whole message, then edit it until I am satisfied | with it and it has mostly no errors. | ... | Then I break it into 10-15 line chunks saved to files named 1, 2, 3, | 4, etc... | ... | Then I come here....enter my Nick: .....put my cursor into this message | box ....F3 and enter 1 then "Send/Refresh"....then put the cursor back | in the message box, F3 and enter 2 then "Send/Refresh"....then...etc | for each sequential chunk of my originally composed message... | ... Š| The advantages are that I'm composing off line, and just copying in | when I get here to the board (or any other board, also)...I ALWAYS | do this when sending an email to the editor on some website, etc.... | ... | On message boards...I find it is the only way to go... | | | ============================================================================ | | Sat, Jan 10, 2004 - 8:44:11 pm | | gregy> | Glenn...hi... | | | | ============================================================================ | | Sat, Jan 10, 2004 - 8:54:39 pm | | gregy> | Ray, et al...did you guys note Christof's latest (9 Jan 04) posting | of aratrans.zip on his website....???? | ... | It should be still there.... | ... | | | ============================================================================ | | Sat, Jan 10, 2004 - 9:06:32 pm | | ray> | Hmmm? I'm missing something here, I think. (BTW, I don't like rudness, | but IMHO bluntness isn't rudness, just the opposite, since you know | where you stand. I am the world's least 'touchy' person). | | What did I say? | I was just mentioning, that as folks work on various modules, we | could easily get into a state where no one knows who's got what. | Already I've got 3 wwwman's (four, if you include the cleaned | version): pre MHT's fix, post MHT's fix, and the URLSIZE change. | Somehow I was running with the wrong one, it could be a harbinger of | trouble to come. Unless a protocol is developed for this sort of thing. | | As far as my cleaning goes, it's on hold pending something like a | consensus on wether or not I'm wasting my time; I have not recieved | a desision on this point. If it is to be used, then *of course* the | compile will be identical (as you quoted me above), the difficulty | lies in the 0.01% of comments I've modified, and in the large | deleted blocks (which can be easily returned if needed). If the | cleaned code is to be used, then the up to date changes need to | be incorporated, no? That was all I was getting at. | | Oh, one more thing: I disagree with you that your compile is the only Š| way to do it. In my oppinion the final distribution of any .exe should | always be a command line compile, since they are always smaller, and | are identical in function ... but whatever the desision is, we need | to be clear on it, no? | | ============================================================================ | | Sat, Jan 10, 2004 - 9:09:09 pm | | ray> | gregy, | | yeah, that's the way to do it: compose with an editor, then import. | | | ============================================================================ | | Sat, Jan 10, 2004 - 9:43:07 pm | | glennmcc> | If you'll notice in that post of mine.... there are lots of gs | | You may be the world's least touchy person. | | However, I hold the title of being the world's biggest smartass. ;-) | | And I defend that title whenever I get the chance. | | ============================================================================ | | Sat, Jan 10, 2004 - 9:49:15 pm | | glennmcc> | IMO, | | file size don't mean shit now-a-days. | | After all... I a have almost 4 gigabytes of free space on my HDD. | | What do I care if a particular file is 6% larger in size? | | If every file on this drive were 6% larger, I would have 3.9 gigs free | instead of 4 gigs. | | ============================================================================ | | Sat, Jan 10, 2004 - 10:23:51 pm | | gregy> | OKAY...!!!....ALRIGHT, ALREADY.......... | ... | glennmcc MAY be the "worlds bigest smartass".....but he has to | contend that title against me....!!!!...... | ... | Ray.... Š| As I said to you at the VERY beginning...(and apparently Joe said | the same thing)......do NOT get frustrated...!!!.....PLEASE do NOT | get hurt feelings...!!!! If we disagree on something....it is only | on that something... | ... | NOW.... | When I set this "sort of organized effort" up....I laid out the | directory structure that we were going to use.....and I asked all of | you if it suited you.... | ... | You all expressed NO negative feedback...so...we ARE going with what | we have, right now....UNLESS...!!!!...UNLESS....!!! you NOW give me | some criticism and direction on how to "standardize" our efforts' | organization..... | ... | Because.....the directory structure I set up here was designed to | avoid and eliminate the circular discussions we have been getting into | both here and on your ArachneDevelopmentYahooGroup List.... | ... | I will reiterate the structure I have set up...and WHY it is designed | to avoid the spinning we are running into now.... | ...NOTE - it may not actually do this....but....tell me NOW how to | improve it so that it will more nearly do this. | | (see next post) | | | | ============================================================================ | | Sat, Jan 10, 2004 - 10:31:50 pm | | gregy> | The directory/subdirectory structure I have set up here is: | | /~gfeig/arachne/CORE (INSIGHT, WWWMAN)/ | | | /~gfeig/arachne/CORE/BASE0 (BASE1, BASE2, BASE3)/JOE (MICHAL, RAY)/ | | You can look at: | | http://www.hi-line.net/~gfeig/arachne/ | | and see the whole directory structure....I have not "ACCESS DENIED" | this directory structure, so my ISP's servers will automagically give | you a clickable directory structure that you can follow down thru and | check out... | | There is a CLEAR reason for this structure, which I explained (probably | not very clearly....I apologize)....and this reason is to keep us | organized, and keep us working step-by-step ....not trying to jump to | the ultimate goal all at once. | | (see next post) Š| | | | | ============================================================================ | | Sat, Jan 10, 2004 - 10:41:16 pm | | gregy> | BASE0 - the code we inherited from Michal, et al - this code is NOT | posted here, because I have a limited site space, and I don't want | to put anything here that isn't crucial for this INITIAL stage of our | efforts. | | BASE1 - the FIRST results of our cleaning/sprucing up of the legacy | code. This should be where we are NOW at. | | BASE2 - the SECOMD stage of cleaning/sprucing up...still of the | legacy code....should NOT include any new functions or optimizations... | ...strictly the legacy code.....if you guys are working on enhancements | to the code....FINE...but...it doesn't belong here....save it for when | we can use it.. | | BASE3 - the FINAL stage of cleanup of the legacy code (I think we can | do it with only three "bases"....maybe we will only need two "bases" | or maybe four "bases"...but we can delete/add if necessary)... | | NEXT.....(see next post) we will take our finalized cleaned-up code... | ...add all the new enhancements that you guys have been making as | we go along thru this cleanup.....and start issuing new versions.. | ... | | Ara175r1 - this will be the next version which will be built on our | cleaned-up BASE3 legacy code, PLUS the new enhancements you guys have | been coding. | | ARA175r2 - as many of these until we get it enhanced/advanced to a | major improvement. | | ARA177r1 - the next version | | ....etc - just keep on improving it and issuing new versions... | | (see my next post) | | | | ============================================================================ Š| | Sat, Jan 10, 2004 - 10:55:02 pm | | gregy> | Now come back to my, guys...because this IS the project structure, as | expressed in this directory structure....and it was designed to | promote that and try to keep us from getting sidetracked.... | ... | For sure....Ray's deletions may (probably) have to be added back, at | least partly...BUT WE DON'T HAVE TO CROSS THAT BRIDGE right now...in | the BASE1 effort... | For now....let's just get the code cleaned....and get some kinda of | explanation file for each module (at least an empty file for future | use, if nothing else)...then we'll get on to the next, BASE2, stage. | .... | I have procured a 175Mbyte site for general use of the Arachne | Community - it is already basically set up, and Jake and I just need | to clean it up somewhat. | ... | We will keep your working directories here (and I will make one for | Christof and one for glennmcc)....but.....I will set up the SAME | directory structure over on the main Community site...so that we have | duplicate mirrors in both places. | ... | SO....I NEED your criticism/approval NOW...because you don't want me | setting up what is totally unworkable..... | | I have been "winging it" with just my prior experience to go on.... | ...and if I've done it cock-eyed....now is the time to straighten it | up....BEFORE we start moving to our main Community site...and start | working on the BASE2 stage.. | ... | ...so.......here I sit.........waiting for your deluge.... | | | ============================================================================ | | Sat, Jan 10, 2004 - 10:59:23 pm | | gregy> | FWIW....my prior experience was to setup/manage(with two other senior | NCOs) the whole Brigade Data Center when my regiment upgraded to a | brigade....and, also supervise the Programmers' Section.....but... | ...but...that whole thing NEVER included setting up and administering | a whole GPL project.... | ... | ...so...my toes may be a little wet....but I don't know the depth of | the water, yet.....and can't even tell it's temperature, yet.... | ... | | | | ============================================================================ | | Sat, Jan 10, 2004 - 11:04:18 pm Š| | gregy> | Joe....one of the reasons why I needed to run this stuff back past | everyone is because I see you're uploading files to your YehGroup | filearea...and I need to d/l them to stash here....but...I need to | know that they are "BASE1 legacy" and do not include any new stuff | or enhancements/fixes... | ... | ...so...please inform me of the status of those files...I joined your | group....but...only to monitor...and grab files to store here... | ...so....come back to me, here, will you....??? | | | | ============================================================================ | | Sat, Jan 10, 2004 - 11:09:54 pm | | gregy> | Oh...BTW....Joe...did I ever send you the Username:Password combo for | your directories, here....????.....or did we decide to work on using | ArachneFTP for you after you came back from vacation....???? | ... | I remember you said you were behind routers, etc, and needed to try to | use Arachne since it was "passive??" FTP upload, or something....so.. | ... | Michal uses ArachneFTP to put stuff in his directories...so...we should | be able to get it to work for you, too.. | | | ============================================================================ | | Sun, Jan 11, 2004 - 4:24:31 am | | Joe> | Ray, Gregy, | | The BaseCode (IIRC;-) directory at Yahoo Groups contains a full | set of Arachne 1.73-GPL sources, including Glenn's Wwwman (from | thechicagobear; hi Ray, were you looking for something?;-). This | is basically what Gregy describes as Base0. The Tools directory | contains proprietry stuff needed to build/support Arachne, which | may or may not have sources available. Finally, the Joe directory | contains stuff I've enhanced and the Ray directory contains the | stuff Ray's "cleaned up". I trust that clears things up. | | To download any of this stuff, first sign-in to Yahoo, then you should | be able to access it via the ArachneDevelopment web interface or like: | http://groups.yahoo.com/group/ArachneDevelopment/Files/Ray/html.c ... | | | | ============================================================================ | | Sun, Jan 11, 2004 - 4:30:38 am Š| | Joe> | Gregy, | | The problem(s) with PKUNZIP is that recent versions or Arachne are | dependent on this non-GPL software (see Mime.cfg), whereas InfoZip | is GPL. Furthermore, PKUNZIP is no longer free for non-commercial | use! Therefore we should not be making Arachne dependent on PKUNZIP. | If someone wants to add this dependency to Mime.cfg, fine, but it | should not be this way as standard. | | | | ============================================================================ | | Sun, Jan 11, 2004 - 4:41:43 am | | Joe> | Ray, Gregy, et al. | | Actually, there's one step I forgot to mention about specifying a | file URL for downloading from ArachneDevelopment. It seems (at least | with Arachne) that you have to navigate to the file's directory | first. This may be a "referrer" thing. From there, you can either | click on the file's link or, failing that, just add it's name to | Arachne's "URL" line. | | | ============================================================================ | | Sun, Jan 11, 2004 - 4:54:50 am | | Joe> | Gregy, | | No, you haven't sent me Username:Password stuff yet. IIRC, you were | still sorting out how to do this stuff using Arachne. Remember, not | only am I normally behind a firewall that excludes any Internet | access except via IE (grrr!!!), but I don't have /have never used | a "real" FTP client. So, after you've sent me the info, I will try | using Arachne for this. OTOH, HTML uploading is very easy for me, | I can even do this with Lynx in DOS (with a few limitations), so | Yahoo Groups a far easier option for me. | | Also, since you have stated that you don't want us to upload enhanced | code, there's not much to upload from me (or Glenn, I expect). | | | ============================================================================ | | Sun, Jan 11, 2004 - 12:18:45 pm | | ray> | Wow, busy! ... I might have missed something. | Š| Anyway, gregy, I think that at least you and I are on almost the same | wavelenght vis. the need for 'CORE1'. This is your department, so I'm | not quibeling, but I'd consider CORE1 to be cleaned up, translated, | proof read, and aproved by all. At this point we're (what I'd call) | 'CORE O.1' i.e. cleaned (mostly), but almost no translation, only one | or two files proofread, and still no aproval or consensus (note Glenn's | remarks from yesterday). What do you guys want me to do? | | | | ============================================================================ | | Sun, Jan 11, 2004 - 2:55:01 pm | | glennmcc> | | | Do it my way | | | | ============================================================================ | | Sun, Jan 11, 2004 - 2:59:07 pm | | glennmcc> | Gregy, | | Joe is correct. | | I would have nothing to upload at this point in the game. | | You already downloaded from my site back in November, | everything I have that is 'unenhanced'. | | If I understand correctly this where it will go. | | /core/base0/ | | | ============================================================================ | | Sun, Jan 11, 2004 - 9:07:51 pm | | gregy> | Glenn...yes....BASE0 is the unenhanced "legacy" code....I do not have | it uploaded here, as I said, because of space restrictions....but..on | the new site...it will be there... | ... | Right now...all the "first cleaning pass" code will be in the BASE1 | directories... | ... | The "second cleaning pass" code will be in the BASE2 directories. | .. | The (if needed) "third cleaning pass" code will be in the BASE3 Š| directories... | ... | | | | ============================================================================ | | Sun, Jan 11, 2004 - 9:12:54 pm | | gregy> | Glenn...yes....that is what I figured....you are not doing much, if | any, code cleaning. You are mostly working on further enhancements... | ...BUT... | ....as you surmised....those enhancements do NOT now enter into this | preliminary BASE0 step of the project.... | ... | Actually....this project is ONLY about cleaning....not about enhancing | .... | I mean....we have to get us a clean code-base FIRST....and that is the | object of this Project... | ... | Of course....the Ultimate Goal...is to start enhancing the Arachne | Package...which includes some changes to CORE, INSIGHT, WWWMAN...(some | of which are already begun or finished).... | ... | ...BUT...right this first step...it's just a cleaning operation...which | also includes the translation from Christof.. | ... | | | ============================================================================ | | Sun, Jan 11, 2004 - 9:19:11 pm | | gregy> | Joe....please give me a URL for the varification of that, "PKUNZIP is | no longer available for non-commercial use."... | ... | I can find nothing anywhere that says that.... | ... | If this is indeed so....then...we cannot continue to issue the Arachne | Package with default PKUNZIP.....but....as far as I can see...it's | still (with every indication of "forever") available for non-commercial | use..... | ... | NOTE - this does NOT mean their whole package of ZIP tools is available | ...but....so far as I can see....they're behaving like Adobe....the | "reader" is free..... | ... | | | ============================================================================ | | Sun, Jan 11, 2004 - 9:23:15 pm | Š| gregy> | Ray...does the "Ray" directory on Yahoo contain more stuff than your | directory, here...??? | ... | I have only had that one email attach from you...have you cleaned up | more than that....??? | .... | In any event....I can easily get the stuff from Yahoo....I owned 5 | groups way back when it was still OneList....then eGroups....then | Yehaw bought it...and....the rest is history.....:((( | ...I only own 1 1/2 groups, there, now....although I'm a member of | 16 groups... | | | | ============================================================================ | | Sun, Jan 11, 2004 - 9:25:57 pm | | gregy> | Joe...you are correct....your enhanced stuff doesn't yet enter in to | this project........from my above descriptions, you can see | that it will only start entering in, when we get ready to REALLY | issue the next version release... | ... | ....and....from the amount of work being done right now....that next | version outta have some VERYYYYYYYYY interesting enhancements.... | | | ============================================================================ | | Sun, Jan 11, 2004 - 9:38:42 pm | | gregy> | Ray....I think maybe you are thinking TOO perfectionist...... | ... | What I mean is....this "first cleaning pass" is just that...a "first" | ....and it will end up as BASE1...... | ... | Then, with further proof reading and cleaning...we will arrive at | BASE2.... | ... | Then...a final "third cleaning pass" should result in a fully cleaned | and commented and explained code base as BASE3... | ... | So....you really ARE working on BASE1, right now....try not to think | of it as resulting in finalized code.... | ... | WHY....???? | .. | ...because I see us starting to get pickey-pickey right now....when | we're not even TO that stage yet.... | ... | All we need, so far, is general agreement on some simple conventions | ...and I've seen you guys work those out.....and now you're progressing Š| with the cleaning.. | ... | The more "correct and pickey" cleaning can take place in the next pass. | ... | The "final polishing" cleaning can take place in the third pass... | ... | Otherwise we start "backing and filling" and spinning in circles... | trying to get perfection, before we even have preliminarily cleaned | code... | ... | Oh, Ray....at this stage....I guess I have to go along with Glenn and | Joe...about removing things from the code.... | ... | I agree with you, that some things don't need to be in there...and | should be excised.....but...this "first cleaning pass" isn't the place | where we want to loose anything.... | ... | Maybe not even until we get one version issued, down the road.....and | maybe not even then.... | ... | But...someday....we are going to need to cut unused/excess stuff out | ...but...today isn't the day... | ...AND.... | ..today isn't really when we need to argue about this, either.... | .. | Let's try to remember that the ONLY thing we are after, right now, is | preliminarily clean code, translated, and a bit more organized...and | commented/explained as much as possible... | ... | | | | ============================================================================ | | Sun, Jan 11, 2004 - 9:44:08 pm | | gregy> | Oh...and guys....let's PLEASE try to remember this, too....it's been | less than 2 1/2 half months since we received this code. And (to me) | it sometimes looks like you guys are "thinking development" and getting | antsy about "no progress"....and....and... | ... | ...but....Ray is doing most of the cleaning....and it is only within | the past month, that he has had any REAL feedback from ProgrammerGroup | members about how he should proceed... | ... | I gotta tell you guys....Ray don't DO this stuff for a living....!!!! | ...he's an upper-level, over-payed, Canadian Gov bureaucrat....and | this is NOT is usual cup of tea.....>>>>>>> | ... | | | | ============================================================================ | Š| Sun, Jan 11, 2004 - 9:50:40 pm | | gregy> | Michal, Joe, Glenn, Ray.....I asked a favor a few posts ago....about | maybe making notes as you go thru a module/file...then making a | FILE.TXT comment file to go with that file.... | ... | Glenn maybe understands more about this code than any other...so... | would he be willing to make a concentrated effort to dash off a very | brief file for each of the code-files....??? | .... | Anyone else reading/cleaning the code would be advancing our effort if | they would jot down their thoughts, also....even it those jottings | turn out to be wrong.....at least that will advance the "explaining" | part of this project.. | ....????????????????????.... | | | ============================================================================ | | Sun, Jan 11, 2004 - 10:03:57 pm | | gregy> | Oh...and Ray.....what you describe as CORE 0.1, "cleaned (mostly) | but without translation, etc".....when we get Christof's translations | inserted....THAT IS our BASE1 code base.... | ... | Then we go on to proofreading, etc, etc....which will result in our | BASE2..... | ... | So...you are well along, now...all we need it Christof's further | translations.....and he told me he would try to finish within two | weeks.....and....since that ain't Christof's cup of tea, either... | we should be patient with him and consider that if he gets it to | us within a month...we are VERYYYY fortunate..... | ... | AGAIN....we're ALL antsy...but....things are going pretty fast...I | mean...I'm even managing to finally get my "new" P133 box more-or-less | finished..... | ... | So....probably by the end of February we will have our BASE1....and | will have some work done on our BASE2 (Joe and others have already | proofread some of Ray's cleaned-code...so part of BASE2 is already | being done)... | ... | I think we can reasonably expect that, at our current steady pace, | we should have our BASE3 by May.....then some polishing/quality-checking | ...decisions about what to include (i.e. what's ready for "big time") | ..we should be ready to release our first "official" new GPL version | package right around August/September..... | ... | ...and...with the enhancements that are already being worked on, worked | out, that new version should be a REAL enhanced eyepopper..... | ... | Now, that's just my sober appraisal of our situation....and...of Š| course.....since I've never been wrong before, this will turn out to | be the first time........ | | | ============================================================================ | | Sun, Jan 11, 2004 - 11:07:10 pm | | ray> | | Gregy, | | Ray...does the "Ray" directory on Yahoo contain more stuff than your | directory, here...??? | | Your an inspiring man Gregy (an NCO, eh? -- Sargent Feig maybe ?? -- anyway, | no one knows the need for coordination better than the army! so you-da-man). | I was at 6 files left when I stoped work. I'll start again. I had hoped | for Hans's help, but haven't heard back from him. In any case, I'll send | you everyting when I'm done. | | What I mean is....this "first cleaning pass" is just that...a "first" | ....and it will end up as BASE1...... | ... | Then, with further proof reading and cleaning...we will arrive at | BASE2.... | ... | Then...a final "third cleaning pass" should result in a fully cleaned | and commented and explained code base as BASE3... | ... | | OK. I liked the idea (as with software) of saving the 'whole number' | releases for major steps taken, thus base1 would be the *finished* | sources for further development; but never mind, you've got a handle | on this, so whatever works for you works for me to. | | Otherwise we start "backing and filling" and spinning in circles... | trying to get perfection, before we even have preliminarily cleaned | code... | ... | | Yes! That's one of the reasons I've been maybe a little to worried about | this, it's hard to change the tune 90% of the way thru the song ... | | Oh, Ray....at this stage....I guess I have to go along with Glenn and | Joe...about removing things from the code.... | ... | | Agreed -- "loooooonnnngggg cookies" will NOT be sortend to "loooonnng cookie | even if I have to wrap the line. | | ...but....Ray is doing most of the cleaning....and it is only within | the past month, that he has had any REAL feedback from ProgrammerGroup | members about how he should proceed... | Š| Well Thanks! I apreciate the note of understanding very very much. | | I gotta tell you guys....Ray don't DO this stuff for a living....!!!! | ...he's an upper-level, over-payed, Canadian Gov bureaucrat....and | this is NOT is usual cup of tea.....>>>>>>> | | Wadayamean 'upper-level'? You have to be French for that! ('over-payed', | now that I'll buy) | | So...you are well along, now...all we need it Christof's further | translations.....and he told me he would try to finish within two | weeks.....and....since that ain't Christof's cup of tea, either... | we should be patient with him and consider that if he gets it to | us within a month...we are VERYYYY fortunate..... | | There is a real problem with the translation. I've asked Christof to | work inside the cleaned files, but he won't do it. If all his work | is in the old sources, then the translations will have to be copied | one at a time over to the cleaned files, and that will make things take | one hell of a lot longer. I asked Michal to take this issue up with | Christof, but I've heard nothing back from either of them. | | | ============================================================================ | | Mon, Jan 12, 2004 - 12:24:31 am | | gregy> | ahhh....I see....well...yes....translating in the legacy files, then | having to transfer the translations to the cleaned files adds another | step....but...if Christof is doing it that way.....I guess we'll just | have to accept his work.....I mean...he he he....unless we can hire | another Czech-speaking Arachne-familiar User as additional Translator | ..... | | | ============================================================================ | | Mon, Jan 12, 2004 - 12:35:12 am | | gregy> | he he...."over-payed" was facitious...I know that you're not getting | a BIIGGGGGGGG wage.....gov'mt bureaucrat or not..... | ... | hmmmm...in the Marines they call us "Gunney"...in the Army they call | us "Platoon Daddy".....it's the E7, Gunnery Sergeant/Platoon Sergeant | rank...which is the most senior NCO on the "gruntline"...the actual | fighting units, whether infantry or armor......everyone of higher rank | is "HQ"...which THEY think means "headquarters"....but we members of | the "gruntliner" ranks know means "hindquarters"......and I've had | many a reararea NCO/Officer brace me about using that term to refer to | them....and I've backed ALL of them down (including, at least, two | General Officers)....but giving a wolfish grin and saying, "But, of | COURSE,...you guys are beHIND us....so.....you're HINDquarters....got | it....????"......whereopon they realized they were dealing with a crazy Š| and they left........ | It really doesn't matter what those kinda guys think....they'll leave | you along....'cause they realize that it's you and your "kids" who'll | do the actual fighting/dying.....and....as long as you do a competent | job of that......hey...you can't beat success with a stick.... | | | ============================================================================ | | Mon, Jan 12, 2004 - 12:39:55 am | | gregy> | ahhh...now I see.....BASE1, BASE2, BASE3.....you thought I was | referring to a "version" number......heck no...!!!.....we won't even | be able to think of "versions"....until we get BASE1-3 cleaned and | finalized....THEN we'll start putting the package for the next "Real | Official" version release together (Glenn's got a temporary, personal | version 1.75...but he's really in test/development mode......) | | | | ============================================================================ | | Mon, Jan 12, 2004 - 12:41:41 am | | gregy> | and...BTW....Christof uploaded more translations to his website on | 9 Jan....did you get it.....??? | .... | | | ============================================================================ | | Mon, Jan 12, 2004 - 1:08:04 am | | Joe> | About PKUNZIP, first check http://nct.digitalriver.com/fulfill/0018.4 | where you will see that PKZIP/PKUNZIP/etc. are no longer available as | shareware, you must buy them. Then read the current (sorta;-) doc files | and you will find no mention of free use for PKUNZIP (free use is only | mentioned for PKSFANSI, which ain't much use as a PKUNZIP substitute, | forgetting the fact that they no longer bundle PKSFANSI anyway. Finally, kee | doc files until you find the following : | | "Use of PKWARE programs for personal use is covered by your purchase | of a single license for PKZIP. A Distribution License is required by | a corporation, company, educational institution or governmental | agency to use PKWARE programs to facilitate the distribution of | software to outside parties." | | There is _no_ differentiation made between PKZIP and PKUNZIP licensing, | if you type "pkunzip -l" or "pkzip -l" you get pretty much the same | message, at least for version(s) 2.50. | | Heck, you can't even distribute a self-extracting (PK) ZIP file anymore, Š| unless you buy a special (no doubt expensive) license to do so! | | | ============================================================================ | | Mon, Jan 12, 2004 - 1:14:30 am | | Joe> | Gregy, | | No need to download files from the "Ray" directory at Yahoo Groups, | except for 2w3mansr, they were downloaded from your site anyway. | | You might download one file from the "Joe" directory (inetstr.c | or similar) though, that's one I spruced up a bit and enhanced the comments. | | | ============================================================================ | | Mon, Jan 12, 2004 - 1:18:16 am | | Joe> | Ray, Gregy, | | Please note that we are being "picky" at this "early" stage with | Ray's changes because this means he can avoid the same problems | with the subsequent code he works on. If we wait for Base1 to be | completed, then there'll be much more involved in getting to | Base2 ... | | ============================================================================ | | Mon, Jan 12, 2004 - 4:05:01 pm | | ray> | Gregy, | | Hmmm, a Platoon Sargeant (what? 4 or 5 stripes?) ... and such a | nice guy! The movies lead us to believe that all sargents behave | all the time like woonded wolverines with toothaches ;-) | | Joe, Gregy, | | Thanks for the support Gregy, it helps, but I know were Joe is comming | from and he's right to want things as tight as possible as soon | as possible. Joe, to save time would you just email me your | inetstr,c? ... I'd like to see what you've done. | | Gregy, | | About Christof, you said it: it will take maybe 20 - 50 hours to | move all the comments 'by hand' ... I was hoping that you might talk | him out of doing it that way, but ... | | Hey where is Michal? Š| | ============================================================================ | | Mon, Jan 12, 2004 - 4:33:33 pm | | glennmcc> | Joe, | | Pkzip and Pkunzip v2.04g | | | --- result of pkzip.exe -L ---- | | PKZIP (R) FAST! Create/Update Utility Version 2.04g 02-01-93 | Copr. 1989-1993 PKWARE Inc. All Rights Reserved. Shareware Version | PKZIP Reg. U.S. Pat. and Tm. Off. Patent No. 5,051,745 | | You are free to use PKZIP for private non-commercial use IF: | NO FEE IS CHARGED FOR USE, COPYING OR DISTRIBUTION. | IT IS NOT MODIFIED IN ANY WAY. | | PKWARE disclaims all warranties as to this software, whether express or | implied, including without limitation any implied warranties of merchant- | ability, fitness for a particular purpose, functionality or data integrity | or protection. See documentation for further disclaimers and information. | | For use on multiple machines or to use to distribute your information | or software, a separate license is required. See the file LICENSE.DOC | for further information. | | If you use PKZIP on a regular basis you are strongly encouraged to register | it. With a full registration of US $47.00 you will receive a comprehensive | printed manual, the latest version of PKZIP, PKUNZIP & PKSFX software, and | when available, the next version of the software. See the documentation for | commercial and distribution licensing info. Send check or money order to: | PKWARE, Inc. | 9025 N. Deerwood Drive | Brown Deer, WI 53223 | | __________________________________ | | | --- result of pkunzip.exe -L ----- | | PKUNZIP (R) FAST! Extract Utility Version 2.04g 02-01-93 | Copr. 1989-1993 PKWARE Inc. All Rights Reserved. Shareware Version | PKUNZIP Reg. U.S. Pat. and Tm. Off. | | You are free to use PKUNZIP for private non-commercial use IF: | NO FEE IS CHARGED FOR USE, COPYING OR DISTRIBUTION. | IT IS NOT MODIFIED IN ANY WAY. | | PKWARE disclaims all warranties as to this software, whether express or | implied, including without limitation any implied warranties of merchant- | ability, fitness for a particular purpose, functionality or data integrity Š| or protection. See documentation for further disclaimers and information. | | For use on multiple machines or to use to distribute your information | or software, a separate license is required. See the file LICENSE.DOC | for further information. | | If you use PKUNZIP on a regular basis you are strongly encouraged to registe | it. With a full registration of US $47.00 you will receive a comprehensive | printed manual, the latest version of PKZIP, PKUNZIP & PKSFX software, and | when available, the next version of the software. See the documentation for | commercial and distribution licensing info. Send check or money order to: | PKWARE, Inc. | 9025 N. Deerwood Drive | Brown Deer, WI 53223 | ___________________________________ | | | | | ============================================================================ | | Mon, Jan 12, 2004 - 4:36:42 pm | | glennmcc> | And now v2.50 | | AFAIK, v2.50 is the last version for DOS | | --- result of pkzip.exe -L ---- | | PKZIP (R) FAST! Create/Update Utility Version 2.50 03-01-1999 | Copr. 1989-1999 PKWARE Inc. All Rights Reserved. Shareware Version | PKZIP Reg. U.S. Pat. and Tm. Off. Patent No. 5,051,745 | | | PKWARE disclaims all warranties as to this software, whether express or | implied, including without limitation any implied warranties of merchant- | ability, fitness for a particular purpose, functionality or data integrity | or protection. See documentation for further disclaimers and information. | | For use on multiple machines or to use to distribute your information | or software, a separate license is required. See the file LICENSE.TXT | for further information. | | | If you use PKZIP on a regular basis you are strongly encouraged to register | it. With a full registration you will receive a comprehensive printed manual | the latest version of PKZIP, PKUNZIP & PKSFX software, and discounts on | future upgrades of PKZIP for DOS and other platforms. See the documentation | for ordering and commercial & distribution licensing information. | PKWARE, Inc. Internet: www.pkware.com Phone: 414-354-8699 | 9025 N. Deerwood Drive Sales: sales@pkware.com Fax: 414-354-8559 | Brown Deer, WI 53223 Tech supp.: support@pkware.com BBS: 414-354-8670 | _______________________________ | Š| --- result of pkunzip.exe -L ---- | | PKUNZIP (R) FAST! Extract Utility Version 2.50 03-01-1999 | Copr. 1989-1999 PKWARE Inc. All Rights Reserved. Shareware Version | PKUNZIP Reg. U.S. Pat. and Tm. Off. | | | PKWARE disclaims all warranties as to this software, whether express or | implied, including without limitation any implied warranties of merchant- | ability, fitness for a particular purpose, functionality or data integrity | or protection. See documentation for further disclaimers and information. | | For use on multiple machines or to use to distribute your information | or software, a separate license is required. See the file LICENSE.TXT | for further information. | | | If you use PKUNZIP on a regular basis you are strongly encouraged to registe | it. With a full registration you will receive a comprehensive printed manual | the latest version of PKZIP, PKUNZIP & PKSFX software, and discounts on | future upgrades of PKZIP for DOS and other platforms. See the documentation | for ordering and commercial & distribution licensing information. | PKWARE, Inc. Internet: www.pkware.com Phone: 414-354-8699 | 9025 N. Deerwood Drive Sales: sales@pkware.com Fax: 414-354-8559 | Brown Deer, WI 53223 Tech supp.: support@pkware.com BBS: 414-354-8670 | | _______________________________ | | | | ============================================================================ | | Mon, Jan 12, 2004 - 4:44:22 pm | | glennmcc> | While it IS true that v2.04g is no longer offfered for D/L directly from | the Pkware site itself. | | It is still available from many sites all over the web. | | Googling for "pk204g.exe" yields 218 hits. | | ============================================================================ | | Mon, Jan 12, 2004 - 4:51:15 pm | URL | | glennmcc> | | --- readme.txt included with v2.50 (DOS) --- | --------------------------------------------------------------------- | PKZIP (R) FAST! Create/Update Utility Version 2.50 03-01-1999 | Copyright 1989-1999 PKWARE Inc. All Rights Reserved | PKZIP Reg. U.S. Pat. and Tm. Off. Patent No. 5,051,745 | --------------------------------------------------------------------- Š| | This SHAREWARE version of PKZIP 2.50 for DOS is intended for use on | DOS 2.0 or later running on an x86/Pentium processor. If you plan to | distribute the Shareware version of PKZIP 2.50 for DOS to friends, | associates, or to an online service, please distribute the file | PK250DOS.EXE rather than individual files. | | Installation | -------------- | To install PKZIP 2.50 for DOS: | | Run the PKZIP Shareware self-extracting setup program by typing | the following and pressing ENTER: | | a:pk250dos.exe -d c:\pkware | | In this example, the PKZIP for DOS file set will be installed from | the a: drive into the c:\pkware directory. | | If you need further assistance with installing or using this | software, please contact our Technical Support Department as follows: | | Phone: 414-354-8699 | Fax: 414-354-8559 | Email: support@pkware.com | Web: http://www.pkware.com | CompuServe: Go PKWARE | PKWARE BBS: 414-354-8670 | | Address: 9025 N. Deerwood Drive | Brown Deer, WI 53223 USA | | Files included with this release | ---------------------------------- | | PK250DOS.EXE | -------------- | This is the installation file for PKZIP 2.50 for DOS. | | The following files will be installed on your system: | | ADDENDUM.TXT A file describing the new functionality in this | release. | | HINTS.TXT Document with answers to frequently asked questions | about PKZIP for DOS. | | LICENSE.TXT License information for using PKZIP 2.50 for DOS. | | MANUAL.TXT ASCII version of the PKZIP for DOS User Guide. | | OMBUDSMN.ASP Association of Shareware Professionals statement. | | ORDER.TXT A list of distributors and an order form for | purchasing PKWARE products. Š| | PKUNZIP.EXE PKUNZIP 2.50 for DOS. | | PKUNZJR.COM PKUNZIP 2.50 for DOS Junior. | | PKZIP.EXE PKZIP 2.50 for DOS. | | PKZIPFIX.EXE PKZIPFIX 2.50 for DOS. | | README.TXT Information about PKZIP 2.50 for DOS (this file). | | WHATSNEW.TXT A file describing the new features in this release. | | ZIP2EXE.EXE ZIP2EXE (PKSFX) 2.50 for DOS. | | Windows 9x/NT Compatibility | ---------------------------- | While PKZIP 2.50 for DOS is generally compatible with the Windows NT | operating system, some functionality (e.g. Long File Names; NTFS file | system) is NOT currently supported. Additionally, PKUNZJR and PKSFXJR | do NOT presently provide support for Long File Names under the | Windows 9x and Windows NT operating systems. | | ____________________________________________ | | I wonder what we get with a google for "pk250dos.exe" | | | | ============================================================================ | | Mon, Jan 12, 2004 - 4:59:33 pm | URL | | glennmcc> | 631 hits | | And 205 with | http://www.google.com/search?q="pk250dos.exe" "www.pkware.com" | | __________ | | In your opinion, | Is it a problem having these lines in mime.cfg ? | | file/bmp2zbm.dgi >TXT|pkzip -m $s _4prt.bmp|del _4prt.bmp \n if exist $s e | been converted to $s>$2 | file/.zzt >TXT|pkunzip >nul $1 -@$2 | file/.zzh >TXT|pkunzip >nul $1 -@$2 | file/.zza >TXT|pkunzip >nul $1 -@$2 | file/.zbm >BMP|[150]md $a\\zbm \n pkunzip -o $1 $azbm \n copy /b $azbm\\*.bm | $azbm\\*.bmp | image/x-zbitmap ZBM>BMP|[150]md $a\\zbm \n pkunzip -o $1 $azbm \n copy /b $a | \n del $azbm\\*.bmp | image/zbm ZBM>BMP|[150]md $a\\zbm \n pkunzip -o $1 $azbm \n copy /b $a Š| \n del $azbm\\*.bmp | application/zip ZIP>TXT|[300]PKUNZIP -v $1>$2 | application/x-zip ZIP>TXT|[300]PKUNZIP -v $1>$2 | application/x-zipped ZIP>TXT|[300]PKUNZIP -v $1>$2 | file/saveaszbm.dgi |pkzip -ex $s $l\n ren $s *.zbm | file/savexport.dgi |pkzip -m $w\\_4prt.zbm $e\\_4prt.bmp | file/.zip >TXT|[300]PKUNZIP -v $1>$2 | | | | | ============================================================================ | | Mon, Jan 12, 2004 - 5:07:34 pm | | glennmcc> | Since you have suggested the use of the freeware zip and unzip..... | | Is this what you would prefer to see ? | (the letters 'pk' are now removed) | | | file/bmp2zbm.dgi >TXT|zip -m $s _4prt.bmp|del _4prt.bmp \n if exist $s | echo file has been converted to $s>$2 | file/.zzt >TXT|unzip >nul $1 -@$2 | file/.zzh >TXT|unzip >nul $1 -@$2 | file/.zza >TXT|unzip >nul $1 -@$2 | file/.zbm >BMP|[150]md $a\\zbm \n unzip -o $1 $azbm \n copy /b | $azbm\\*.bmp $2 \n del $azbm\\*.bmp | image/x-zbitmap ZBM>BMP|[150]md $a\\zbm \n unzip -o $1 $azbm \n copy /b | $azbm\\*.bmp $2 \n del $azbm\\*.bmp | image/zbm ZBM>BMP|[150]md $a\\zbm \n unzip -o $1 $azbm \n copy /b | $azbm\\*.bmp $2 \n del $azbm\\*.bmp | application/zip ZIP>TXT|[300]UNZIP -v $1>$2 | application/x-zip ZIP>TXT|[300]UNZIP -v $1>$2 | application/x-zipped ZIP>TXT|[300]UNZIP -v $1>$2 | file/saveaszbm.dgi |zip -ex $s $l\n ren $s *.zbm | file/savexport.dgi |zip -m $w\\_4prt.zbm $e\\_4prt.bmp | file/.zip >TXT|[300]UNZIP -v $1>$2 | | | Now we would not be "stepping on anyones' toes" by reccomending the use | of a shareware program. | | We would instead be suggesting the use of "freeware". | | | ============================================================================ | | Mon, Jan 12, 2004 - 5:08:59 pm | | glennmcc> | Or would you also like to see both zip and unzip included in the package ? | | Š| | | ============================================================================ | | Mon, Jan 12, 2004 - 7:02:30 pm | | gregy> | Joe, glennmcc.....for the time being...we seem to be covered for free | distribution of the 2.04g with Arachne... | ... | ...BUT.....I'll check further....and....since there appears to be at | least the possibility of a problem arising at some time in the future, | ...we had better start NOW thinking about a good replacement for it.. | ... | Since it is NOT a present urgent item....let's figure on THINKING | about what free zip/unzip utils we can get to replace pk.... | ... | | | ============================================================================ | | Mon, Jan 12, 2004 - 7:25:18 pm | | glennmcc> | But the point I'm trying to make is.... | | | We are not and never have distributed ANYTHING from Pkware. | | All we have is 10 refferences to pkunzip and 3 to pkzip inside of mime.cfg | | Nowhere in our package have we done ANYTHING that could be miscostrued | as distribution of ANYTHING. | | All we have is the word "pkunzip" in 10 places | and the word "pkzip" in 3 places. | | I don't see the problem. | | Could someone please explain it to me ??? | | ============================================================================ | | Mon, Jan 12, 2004 - 7:30:51 pm | | glennmcc> | All of that not withstanding.... | | Does this not make it quite clear that Pkware HAS given | permission to distribute the file pk250dos.exe ??? | | .............If you plan to | distribute the Shareware version of PKZIP 2.50 for DOS to friends, | associates, or to an online service, please distribute the file | PK250DOS.EXE rather than individual files........... Š| | | | | ============================================================================ | | Mon, Jan 12, 2004 - 7:42:15 pm | | gregy> | yes, glennmcc...that is true...all of it......(however, don't we also | include PKUNZIP somewhere in the package...???....or...where does | it normally come from...?).....I don't remember initially d/ling it | to use with my original Arachne..... | ...however.....I've d/led it numerous times..... | ... | I went and checked digitalriver....the www.pkware.com links lead there | from their "FREE Download" links... | ... | HOWEVER.....it's still around on simtelnet...I just went and checked | one of their mirrors... | ... | SO....for now....let's just leave things as they are.....and we can | decide that when/if we need to in the future... | .. | For right now....it isn't a crucial/urgent item.....noone is suing | us..... | ... | | | ============================================================================ | | Mon, Jan 12, 2004 - 7:44:18 pm | | gregy> | ...and....from the ARCERS/, ARCUTILS/, COMPRESS/and EXECOMP/ | directories on simtelnet.....I'd say we have a WIDDDDDEEE range of | alternatives to chose from, if/when we have to... | .... | ...just....we don't have to decide now.... | | | ============================================================================ | | Mon, Jan 12, 2004 - 7:47:37 pm | | gregy> | Ray, Joe....okay...picky it is......what I want us to try to | avoid, as much as possible, though, is "show-stopping" bouts of | "infinite recursion" back-and-forth thru this legacy code...trying | to come up with "the final BASE" with only one pass....... | ... | ....as long as cleaning isn't being halted for "go back for | perfection this first time"....and you all are comfortable with how | you're proceeding....then.....I am, too........ | Š| | | ============================================================================ | | Mon, Jan 12, 2004 - 7:55:28 pm | | gregy> | Joe..... I'll go get your inetstr.c from your "JOE" directory...I | mean, of course, it goes in your /CORE/BASE1/cleaned up directory.... | heck...!!!!....if you wanna cleanup ALL the source, we'll be glad to | have it.....!!!! | ... | ... | | ... | RAY....I don't think I want to push Christof any more than I already | have.....he was pretty reluctant to get involved in the ProgrammerGroup, | originally....I fast-talked and he got a good idea that he could be | useful translating...so.....he jumped it, there.....but....I'm | inclined to NOT twist his arm any further... | ... | We'll just have to manually move the trans from the legacy to the | new files... | ... | | | ============================================================================ | | Mon, Jan 12, 2004 - 7:58:39 pm | | gregy> | I will commit to help with that manual moving...starting 15 Feb 04... | ... | I can do it in my OS/2...with two side-by-side windows and CPfrom | one to the other.... | ... | ...as long as I use a straight Programming Editor (and OS/2 has | several real good ones) and not a wordprocessor... | ...then....everything should end up okay.....and it may go pretty | fast, once I get into it... | ... | | | ============================================================================ | | Mon, Jan 12, 2004 - 8:02:31 pm | | gregy> | Meantime....how about you guys' EXPLAIN.TXT files for each module..?? | ... | As I do the trans-move...I will also get some idea about each file, | so...I'll put notes in a third window...and just already have it when | I get done.. | .. | Š| | | ============================================================================ | | Mon, Jan 12, 2004 - 8:42:29 pm | | ray> | Gregy, | | Yeah, I see where you're comming from vis. Christof (but it would | have been nice). Anyway if you lend a hand, that would be *very* nice; | I don't know if I could work up the energy by myself. I figured about | 200 hours to clean core.exe, and that's about what it's going to be; | after that I'd like a rest. | | ============================================================================ | | Tue, Jan 13, 2004 - 12:32:35 am | | Joe> | OK, let me be as clear as possible about this PKWARE issue. | | 1. PKZIP, PKUNZIP *used* to be available under a shareware license. Past | tense! If you try to register it now, you are liable to get a response | from PKWARE to say, "Sorry, you'll have to send us $$$ more to buy this | software, as we no longer think the shareware payment is adequate". | | 2. If older versions of PKUNZIP allowed for free non-commercial use, then | fine, we could probably bundle this with Arachne and the user could get | away without attempting a shareware registration, but what about PKZIP, | which Mime.cfg also now requires? We can't get away with that trick there. | | 3. Arachne should work "out of the box", without requiring users to | purchase some third-party utility or simply assuming all potential users | already have it. | | 4. We don't need this PKWARE stuff. The Info-Zip equivalents are free, | open source alternatives which we *can* bundle with Arachne. No more | "let's just assume everyone has the PKWARE stuff". Arachne should not | be broken for any user for lack of this _commercial_ software. | | We have no excuses. Let's do this properly, we have no reason not to. | Heck, the Info-Zip stuff is even BSD licensed now, so it's even more | "free" than it used to be under the GP License. | | ============================================================================ | | Tue, Jan 13, 2004 - 12:36:45 am | | Joe> | Ray, | | I'll email you tomorrow about/with "inetstr.c" ... | | Š| ============================================================================ | | Tue, Jan 13, 2004 - 12:45:29 am | | Joe> | Glenn, | | Not to labour the point, but the PKWARE isssue is not about whether | or not we distribute anything from PKWARE with Arachne, we *are* | permitted to do so if we wish. We can distribute the shareware versions | of any software ad infinitum, unless there is a "sunset clause" included | in the original distribution. The problem is that we are then requiring | users to purchase this software after a "trial period", which IIRC is | 30 days standard for shareware. Why do we owe such a favour to PKWARE? | Would we do the same for MonoSoft? Of course not! | | ============================================================================ | | Tue, Jan 13, 2004 - 3:10:39 pm | | ray> | Guys, | | I haven't been following too closely on the 'zip' question, so pardon | an off-hand remark, but it seems to me that: | | - it's not a very big deal either way. | - gregy's the admin. so he should have the final say (he out ranks us | ;-) | - ... but Joe is probably right ... we should use the 'openest' software | we can. | | | | I see the future of DOS as robust and growing, not hiding in 486's | waiting to go extinct when the last old box it chucked. We should | take a lesson from our 'big brother' Linux, and think in terms, not | just of software packages, but of *distributions* of whole suites | of programs, all integrated and working together -- in that spirit, | why not distribute Arachne with info-zip and with a whole bunch of | other open source stuff, all pre-tested; that's the future as | I see it. | | | | ============================================================================ | | Tue, Jan 13, 2004 - 4:29:49 pm | | gregy> | Joe.....I agree with everything you say.....so...we don't have any | argument....since I noticed you didn't class it as something we have | to do today..... | ... | As glennmcc as pointed out....the changes in MIME.CFG are already Š| made....(he posted his example, here)....and we can easily put the | changes in the ArachnePackage for the next release..... | ...however....I'm not real familiar with the zip you are suggesting.. | ..so...will you post the URLs here (in the text box, please, not in | the URL box) so we can go look at and get it and start experimenting | with it to see how it works, etc...?? | .... | | | | ============================================================================ | | Tue, Jan 13, 2004 - 4:34:33 pm | | gregy> | Ray...I learned long ago...working as technician, M & O Supervisor, | MÒManager, Engineer, Managing Engineer, for large international | corporations....that the admin/boss is NOT the "boss"...but is the | ultimate gofer......so....that sets THAT straight... | ..... | NOTE - I don't outrank anyone.....FYI...PlatoonSergeants (Daddies and | Gunnies) don't "boss" their troops...they lead their troops...and.. | get shot first....while their troops take care of the job to be done. | ... | ....so...THAT clears THAT up, also....... | ... | | | ============================================================================ | | Tue, Jan 13, 2004 - 4:43:19 pm | | gregy> | Ray.........complete packages.....well...Arachne has always | been made available with a whole bunch of proggies/utilities....that's | a "suite" if you will..... | .. | One of the frustrating problems with Arachne is that so many users, | even older users (my self included) do NOT know or use what is there | ...stuff that is REALLY useful and even advanced..... | ... | For example.....our PDF2TXT and PDF2HTM (thanks to Joerg) .APMs | let us view, edit ANY PDF....and Joerg is constantly updating his | PDF2HTM to cover new PDF versions.... | ... | That's only one example...there are countless others.... | ... | BUT...some users want this stuff "compiled-in" into the CORE, etc.. | ...when the stuff is (and should be) offered as installable addons.. | .... | I think part of this problem is PR....but....maybe not...remember | that DOS versions of then popular "Office" suites didn't keep their | user base.... | ... | But....you are right that we need to offer lots of functionality with Š| Arachne....and we do....and should continue to...and enhance what we | have.. | | | ============================================================================ | | Tue, Jan 13, 2004 - 5:26:17 pm | | ray> | Gregy, | | Hey, no need to explain about the 'leader' thing, don't take it | too seriously, I plan to kid you hard about it. But I have a very small | ego about this kind of thing. I myself am a natural leader, and have | demonstrated it (not here!) many times, but I'm not insecure, I very | much enjoy having someone else take charge when it's appropriate. | My view of authority is that as long as the admin. is moving in the | right direction, I'm more than happy to 'take orders' from them. | Structured groups get things done faster, no? | | | | | ============================================================================ | | Tue, Jan 13, 2004 - 6:42:54 pm | | glennmcc> | Joe...... | | Pkzip and Pkunzip v2.04g | --- result of pkzip.exe -L ---- | PKZIP (R) FAST! Create/Update Utility Version 2.04g 02-01-93 | Copr. 1989-1993 PKWARE Inc. All Rights Reserved. Shareware Version | PKZIP Reg. U.S. Pat. and Tm. Off. Patent No. 5,051,745 | | You are free to use PKZIP for private non-commercial use IF: | NO FEE IS CHARGED FOR USE, COPYING OR DISTRIBUTION. | IT IS NOT MODIFIED IN ANY WAY. | | | ============================================================================ | | Tue, Jan 13, 2004 - 6:45:05 pm | | glennmcc> | | Plain and simple...... | | We CAN bundle pk204g.exe along with Arachne. | | The same is true of pk250dos.exe | | | Š| ============================================================================ | | Tue, Jan 13, 2004 - 6:51:49 pm | | glennmcc> | There is no 'trial period' with v2.04g | | You are free to use PKZIP for private non-commercial use IF: | NO FEE IS CHARGED FOR USE, COPYING OR DISTRIBUTION. | IT IS NOT MODIFIED IN ANY WAY. | | We are no more 'requireing' a user to register it than Michael was | 'requireing' users to register Arachne. | | All of this not withstanding...... | | I agree with your position that everything included in our now 'opensrc' | package should also be 'opensrc'. | | Therefore..... | | Let's do it. | | Let's remove 'pk' from mime.cfg and let's include infozip in the package | right along with Arachne. | | | | ============================================================================ | | Tue, Jan 13, 2004 - 7:02:14 pm | | glennmcc> | I do not ythink that we should include the entire infozip package but | rather just zip.exe and unzip.exe | | The entire package is a 1.8mb zip file in itself. | Which makes the infozip package bigger than the Arachne package. | | ============================================================================ | | Tue, Jan 13, 2004 - 10:25:54 pm | | gregy> | Well...where'd you find it.....??? | ... | Or are you gonna make me hafta go google some more to find it...??? | ... | .... | And....I agree...just the zip and unzip....that's all we've been | including from PK in the ArachnePackage...isn't it...??? | | | ============================================================================ | Š| Tue, Jan 13, 2004 - 10:27:39 pm | | gregy> | And...we can put a link to the whole shebang in the ArachnePackageHelp, | or something.....or I can stash the whole thing on the AVA site...but | I'd rather give an offsite link for something like that....probably | the main InfoZip download link.... | ... | | | | ============================================================================ | | Tue, Jan 13, 2004 - 10:28:55 pm | | gregy> | And...anyone who wants to can modify your aunzip thingee to use either | Info or PK.... | ...leave it up to them... | ... | | ============================================================================ | | Tue, Jan 13, 2004 - 10:33:30 pm | | gregy> | Ray....I ain't being "touchy" about that leader-thing...I'm just poking | you back..... | ...BUT... | I DO want everyone to understand EMPHATICALLY...that I ain't....."The | Boss"...that is..... | ... | He..he..he....I am NOT a natural leader..........I'm naturally | lazy....I'd rather be lead....so....I've had to train myself to get | down and lead.... | ...the military thing helped me do that... | ... | but...natural, I'm not..........if you look close at any milling | group....you'll spot me hovering in the back...waiting for someone | to step up and lead... | ... | Then...if no one does....and someone's gotta....well....I can groan | and shamble forward....and then, usually...I can fumble thru and get | the job done... | ... | | | ============================================================================ | | Wed, Jan 14, 2004 - 12:10:14 am | | Joe> | Glenn, | | Thanks for pointing out that non-commercial free use applies to PKZIP Š| 2.04g, not just PKUNZIP 2.04g (I have the registered version of 2.50, | so I know that free use is not an option for _that_ version, neither | is a (low-cost) shareware payment, now). This means bundling 2.04g is | actually an option, although as everyone now agrees, not the best option. | | | ============================================================================ | | Wed, Jan 14, 2004 - 12:28:06 am | | Joe> | Gregy, Glenn, | | InfoZip's home page is : http://www.info-zip.org/ | | I'd suggest using the 16 bit version of "unzip" and the 16 bit "sm" | (small memory) version of "zip". The combined size of these two | executables is about 100k, which is about 20% more than the PKWARE | equivalents. | | Note that the command line options for the InfoZip stuff are a bit | different to the PKWARE stuff, but since this is internal to Mime.Cfg, | nobody's going to notice. | | You will find the InfoZip stuff works much better than the PKWARE stuff | in a number of respects. It can extract from a Windoze32 EXE file with | embedded ZIP archive, whereas PKUNZIP gets confused, and is *much* | faster scanning large archives. It can extract LFN cr*p in a compatible | manner under non-LFN DOS, whereas PKUNZIP 2.04g and 2.50 can really | mess things up. Just try installing a bunch of DJGPP stuff and you'll | see what I mean. Of course, for full LFN support (for newer DOSes with | LFN API installed), you'd need to use the 32 bit versions of the InfoZip | utilities, but this shouldn't be relevant to Arachne's needs. | | ============================================================================ | | Wed, Jan 14, 2004 - 1:15:01 am | | gregy> | Joe, Glenn, | | I went and d/led all the files in: | | ftp://ftp.uu.net:21/pub/archiving/zip/MSDOS/ | | This is the directory listing for that directory, which is where | the InfoZip links you to: | | Please read the file /info/README.ftp | it was last modified on Fri Jun 29 00:54:02 2001 - 929 days ago | Please read the file /info/README | it was last modified on Fri Jun 29 00:54:02 2001 - 929 days ago | =============================================================== | | total 1764 Š| -r--r--r-- 1 21 1601 Jun 28 2001 README | -r--r--r-- 1 21 6693 Jun 28 2001 README.DOS | -rw-r--r-- 1 21 119146 Jun 28 2001 gzip124.exe | -rw-r--r-- 1 21 25022 Jun 28 2001 ship16bnd.zip | -r--r--r-- 1 21 162418 Jun 28 2001 unz542x.exe | -r--r--r-- 1 21 232972 Jun 28 2001 unz542x3.exe | -r--r--r-- 1 21 308972 Jun 28 2001 zip23x.zip | | =============================================================== | (7 items listed) | | If this is NOT what you had in mind, please tell me...so I can | go get the correct files..... | | ============================================================================ | | Wed, Jan 14, 2004 - 3:30:24 pm | | ray> | Gregy, | | Yeah, were cool ... Boss . | | I love your attitude to 'leadership', have you read the old Roman | story of Cincinatus? | | | ============================================================================ | | Wed, Jan 14, 2004 - 5:10:10 pm | URL | | glennmcc> | Here's the FTP site you want for the latest stuff. | | ============================================================================ | | Thu, Jan 15, 2004 - 12:06:28 am | | Joe> | Hi Greggy, | | Yep, that's the stuff except ... it's outdated. I guess you selected | the US download mirror as the closest, which is not unreasonable. What | is unreasonable is that "uu.net" haven't kept up-to-date. I guess this | partly explains it (from the home page) : | | "Our primary web site is hosted by our very own Hunter Goatley. Our primary | hosted by the fine folks at LEO, the Technical University of Munich's huge L | Everything Online archive. Secondary distribution sites are hosted by the Co | TeX Archive Network, the UK Mirror Service, UUNET Technologies, Swiss SunSIT | (see below)." | | Anyway, use Glenn's link, unz550x and zip23x have what we need ... | Š| ============================================================================ | | Thu, Jan 15, 2004 - 5:59:05 pm | | ray> | OK, folks, I've finnished my first pass thru the code. In the next | day or two I'll post on yahoo a detailed description of everything | I've done, and I'm hoping that all will read it and comment, or | "forever hold their peace". We've got to nail down questions of | formating, etc. Just now I was talking to Joe, and he said we'd | agreed on two space indentation, but I thought it was 4 spaces! ... | let's get this stuff all figured out before I send it all to Gregy. | | | | | ============================================================================ | | Thu, Jan 15, 2004 - 8:30:42 pm | | glennmcc> | You're both wrong. :( | | It's 3 spaces. | | And absolutely.... NEVER any tabs. | | ============================================================================ | | Thu, Jan 15, 2004 - 9:58:45 pm | | ray> | Well at least Glenn is weighing in before the fact :-). I can see | I'm whipped on the tabs issue ;-( Michal says that 4 spaces is | standard, and that's what I've planned my line wraps around. Me I like | 5 spaces. Joe wants 2! I can live with 3 if Joe and Michal agree. | BTW, I'll always keep a 'tabs' verion anyway, and it is simple to | convert tabs to whatever is wanted with spaces, so there's no | crossing of any Rubicon; we can change our oppinion any time :-) | | While were at it, what about "if (xxx)" vs. "if(xxx)" ... I very | slightly like the latter, Michal prefers the former. Same question | with "for (xxx), while (xxx), switch (xxx). | | | ============================================================================ | | Thu, Jan 15, 2004 - 10:48:56 pm | | glennmcc> | I'm with Michal, I prefer the space in between. | | If (xxx) | | Š| ============================================================================ | | Thu, Jan 15, 2004 - 10:49:35 pm | | glennmcc> | And I think Michal is right.... 4 is better. | | ============================================================================ | | Thu, Jan 15, 2004 - 10:54:57 pm | | glennmcc> | I also prefer to start the next eg of code which will be executed on a | true result of the if on the line following that if | | if (xxx) | { | do this; | and this; | and this; | } | else | { | do that; | and that; | and that; | } | while (xxx) | { | do this; | and that; | } | | | ============================================================================ | | Thu, Jan 15, 2004 - 10:58:33 pm | | glennmcc> | Unless it's just a single line of code. In that case...... | | if (xxx) do this; | else do that; | while (xxx) do this; | | | | ============================================================================ | | Fri, Jan 16, 2004 - 7:21:03 am | | phantom> | Ray, | | About indentation, I said I prefer 3 spaces for general use, 2 spaces Š| for complex code with much nesting, eg. Arachne. | | As I recall, Michal said something similar, except his numbers were | 4 and 2 respectively. | | If, while : I slightly prefer the space in between. | | I never prefer "in-line" single-line 'if', 'while' statements. First you | have the 'if' statement or whatever, then you have the statement that | gets executed depending on the result of the 'if' statement. That's two | statements, so is more readable as two lines. | | Braces, I tend to prefer them non-indented, just the enclosed statements | indented. | | | ============================================================================ | | Fri, Jan 16, 2004 - 3:36:00 pm | | ray> | | | I'm with Michal, I prefer the space in between. | | | If, while : I slightly prefer the space in between. | | OK, done. | | | And I think Michal is right.... 4 is better. | | OK, its 4 spaces unless something unexpected happens. | | | I also prefer to start the next eg of code which will be executed on a | | true result of the if on the line following that if | | | | if (xxx) | | { | | do this; | | and this; | | and this; | | } | | else | | { | | do that; | | and that; | | and that; | | } | | while (xxx) | | { | | do this; | | and that; | | } | | | | glennmcc> | | Unless it's just a single line of code. In that case...... Š| | | | if (xxx) do this; | | else do that; | | while (xxx) do this; | | | | | | I never prefer "in-line" single-line 'if', 'while' statements. First | | you | | have the 'if' statement or whatever, then you have the statement that | | gets executed depending on the result of the 'if' statement. That's | | two | | statements, so is more readable as two lines. | | | | Braces, I tend to prefer them non-indented, just the enclosed | | statements | | indented. | | | | phantom> | | Hey Joe, is that you? | | Anyway I'm with phantom on braces, I find that program flow is much | easier to follow with the braces 'floating' to the left of their | block, and at the same level as their controling statement. And I | believe that that's much more standard. Also, I've used unindented | braces everywhere, and changing it would be a difficult job. | | You guys might not empathise with me on this, but: if (xxx) dothis; | was very dangerous for the janitor. While I was sorting out some of | the abysmally bad indentation in some modules, the tendancy was to | make this mistake: | | if (xxx) dothis; | ooops_dont_indent_this!!!; | | so, I've used two lines almost everywhere execpt for: | | if (xxx) dox; | if (yyy) doy; | if (zzz) doz; | | for such blocks, I find that the single line form is *much* easier to | read. | | Now for the tough question: | | if (xxx) vs. if (xxx) vs. if (xxx) | dothis; dothis; dothis; | else if (yyy) else else | dothat; if (yyy) if (yyy) | dothat; dothat; | | I like 1) and 2) above. 3) is more logical, technicaly, but less | standard, and leads to too much indentation if there are too many 'else if's | Joe likes 3). What say you Glenn? Š| | ... and I wish Michal would comment, too, but he is very busy right now. | | | ============================================================================ | | Fri, Jan 16, 2004 - 7:57:28 pm | | gregy> | Ray...would you email me a copy of your explain thing that you say | you'll upload to Yahoo....?? | ... | I'm registered...but...I'd rather get a copy from you... | .. | DANG...!!!!!! | I HATE this kinda stuff....now I hafta go d/l again.....mirrors are | supposed to be mirrors....SHEESH...!!!..... | ... | Oh, well...at least glennmcc found the newest stuff..... | ... | | | | ============================================================================ | | Fri, Jan 16, 2004 - 7:59:29 pm | | gregy> | And...I've been offline for about two days...so...I'm just ketchin up | .... | | | ============================================================================ | | Fri, Jan 16, 2004 - 8:01:12 pm | | glennmcc> | Honestly ???? | | 3 looks like S**T | | And with as many 'nested' if,else,if,while,if,elses as there are in this | code..... #3 will very soon have us running off the right side of the | page which will then cause us to have to side-scroll just to see | one......................................................................... | | ============================================================================ | | Fri, Jan 16, 2004 - 9:10:06 pm | | ray> | Gregy, | | I'll send it to you when I'm finished my second pass ... maybe a week? | Š| Glenn, | | I hope I'm not misquoting Joe about 3) above. Me, I think of an | 'if else' stack as rather like a switch -- each test rather like | a 'case' and having the same indentation. | | BTW just today I came across another orphaned file: 'elib.h'. | 'wattcp.h' has: #include , but (now that I have looked at | it more closely) that calls not the 'elib.h' file included with the | package (that would use: #include "elib.h") but some 'standard' | version (Borland or Library) ... so do you know anything about that | file? I've moved it to my file bone yard anyway. | | | | ============================================================================ | | Fri, Jan 16, 2004 - 9:10:31 pm | | gregy> | Hey...!!!....I'm easy..(not cheap..but...easy.....)....you guys | fight it out amongst you.... | .... | Anyway...I'm busy trying to get a handle on that AVA site....and it's | a bit cluttered, and maybe needs new "dir-pointers" in the scripts.. | etc.....he he....happens ever' time ya make a move........but | once this new setup is working...it outa stay stable for, at least, | a week or two..... | | | ============================================================================ | | Fri, Jan 16, 2004 - 9:12:04 pm | | gregy> | Ray........ | | | ============================================================================ | | Sat, Jan 17, 2004 - 12:01:43 am | | gregy> | Ray.....QUESTION....do you have Borland BCC 3.1....??? Is that what | you are compiling your cleaned up code with, to check that it compiles | the same as the legacy code......??? | ... | Someone asked what we were compiling this cleaned up code with, so I | ask... | ... | If you don't have BCC 3.1...I can post it for you to download.... | ... | | | Š| ============================================================================ | | Sat, Jan 17, 2004 - 7:33:47 am | | Joe> | Ray, | | Yep, it's me. | | OK, just in case you missed it, Michal and I both agree that heavily | indented (ie. complex) code such as Arachne should indent by *2* spaces | (at least that's my recollection of the conversation;-). For more | sparse/spacious code *only*, I prefer 3 spaces, Michal prefers 4. OK? | | Also, in case you missed it, not only do I empathise with you about the | 'if (xxx) doyyy' issue, I definitely prefer this as two lines, for the | reasons I explained. The same thought processes you as janitor undergo | also affects coders. Often it takes a second or more with each such | statement to identify the boundary between '(xxx)' and 'doyyy', etc. | | About cascaded 'if ... else' statements, yes I much prefer 3. The only | exception to full indentation is when the cascaded 'if' statements | form a de-facto 'switch/case' statement. BTW, of course you will indent | "off the page" doing this if you don't have a sensible indentation | spacing, that's one of the reasons Arachne should indent by 2 spaces | (it's currently indenting by 1 space, so this will be a very significant | increase in readability). | | | ============================================================================ | | Sat, Jan 17, 2004 - 7:48:52 am | | gregy> | Hi...Joe........The Dude offered to help...even though he claims | only familiarity with C...not any production programming... | ... | I asked him if he would be willing to help me with the "cut & paste" | of Christof's translations into the cleaned code.... | ...I am waiting for his reply.... | ... | I also asked him if he would be interested in the "compiling-checking" | that we will do on the final BASE3 code...before we pronounce it | "cleaned and ready"...... | Will let you all know, here, what he replies.. | | | ============================================================================ | | Sat, Jan 17, 2004 - 9:25:21 am | | Joe> | OK, Gregy. | | BTW, I have just posted an interesting Core beta at ArachneDevelopment Š| ... | | | | ============================================================================ | | Sat, Jan 17, 2004 - 9:39:48 am | | ray> | Gregy, | | Of course I'm using '3.1 ... got it from Glenn ages ago. I wouldn't | trust my own compile checks with anything but identical software to | what Glenn is using. Who ever this Dude is I hope he get's on board! | But there won't be much work to do with compile checks, since everything | I release will be *identical* in compile to the original (feel free | to check up on me, I won't be offended :-) If there is a difference, | it's due to compiler settings (compile for 386 or not, max. speed or | min. size, etc. Oh and the famous "MALOCERR()" thing of course, but | that can be nailed down too. | | ============================================================================ | | Sat, Jan 17, 2004 - 4:49:02 pm | URL | | glennmcc> | Ray, | | | Define *identical* | | Webster says......... | | http://www.webster.com/cgi-bin/dictionary?identical | | | ============================================================================ | | Sat, Jan 17, 2004 - 4:51:30 pm | | glennmcc> | | Are you using definition #2 ??? | | 2: having such close resemblance as to be essentially the same | | | | ============================================================================ | | Sat, Jan 17, 2004 - 7:56:28 pm | | ray> | Nope, close only counts with: //comments that have been shrtn'd to fit. Š| | Given the same compiler settings, If my code differs by one byte, | I'll go back to command.com for a year as penance. With time off for | good behaviour, of course. | | | ============================================================================ | | Sun, Jan 18, 2004 - 3:12:28 pm | | glennmcc> | I'll take that bet. ;-) | | CD to the dir containing your cleaned SRC code. | | Start BC v3.1 | | Choose 'compile' and 'build all' | | Alt+X to exit BC | | After editing nokeyexe.bat so that your current core.exe | will not be over-written.... (rem-out the 2nd line) | | | --- nokeyexe.bat --- | markexe.exe nokey.exe 181000 | xcopy /f nokey.exe c:\arachne\core.exe | rem xcopy core.exe e:\xchaos\arachne\debug\ | ____________________ | | | Run nokeyexe.bat | | Now compare your newly compiled nokey.exe to the core.exe v1.73;GPL | that was in the package you D/Led from my site back in November. | | The files should be exactly the same size and they should differ | from each other in only one respect. | (the 2 bytes containing the date of compile) | | | ============================================================================ | | Sun, Jan 18, 2004 - 3:26:38 pm | URL | | glennmcc> | Front page updated. | (got a couple of surprises) | | | ============================================================================ | | Sun, Jan 18, 2004 - 5:26:22 pm Š| | ray> | Ok, you're on! ... and you loose ;-) | | ... I rebuild and compare .exe's after every pass thru every file, | or after every 10 minutes or so of work (which ever comes first). On | a day like today I will rebuild 20 or 30 times -- I do this so that | I can catch any errors ASAP, and not have to back-track more than | can be helped. I've even assigned the 'fc' batch file to a function | key, to save time. Nope, there won't be one byte different. | | Looking forward to '1.75'! | | | ============================================================================ | | Sun, Jan 18, 2004 - 5:31:28 pm | | gregy> | Joe..Ray...re: using 1.71;ue01 to access/download files from yehawgroup | .... | I ALWAYS use Arachne to access/maintain my YahooGroups....I just d/led | that CORE.UUE....and it come down fine (I haven't unUUEed it, yet)... | ... | I am still regularly using 1.71ue01 (stock, bare, unenhanced....) | and I have NO ...repeat NO....repeat again..NO>>>>> "lockups", EVER.. | ..accessing anything on MyYahoo or MyGroups....only email locks it | up... | ...now....Wizzy is NOT suffering from delusion....he IS actually having | trouble with "lockups"...and maybe some others of you are, also...but.. | ... | I NEVER have these lockups.... | ... | It may be because I ALWAYS run "clean Arachne" from RAMDisk....This | means that my WHOLE Arachne, including CACHE/TEMP/etc is xcopy-ed up | from a fresh install on my HDD, EACH time I fireup my computer...and | ...so...any "smiley" or other bugs that seem to accumulate in Arachnes | run from HDD....do NOT ever affect me.....because my Arachne goes | away each time I shut my computer OFF.... | ...each new fireup is a brandnew, CLEAN, Arachne up-copied from HDD to | RAMDisk.... | ...I noticed Wizzy said he was only running his CACHE and TEMP from his | RAMDisk...well ....that will not prevent accumulating bugs into the | basic Arachne installation.... | | | ============================================================================ | | Sun, Jan 18, 2004 - 5:34:55 pm | | gregy> | Ray........hey...it sounds like glennmcc as ALREADY spotted some | difference between his 173CORE and one compiled from your cleaned | source.... | ...maybe this is just some kind of misunderstanding....but.... Š| ... | So...Glenn....just how much difference do you see....??? | ... | Ray says he's seeing no difference....so...it might be a simple | compiler setting, or something... | ... | ????????? | | | | ============================================================================ | | Sun, Jan 18, 2004 - 5:37:53 pm | | gregy> | I'm asking about this...because I'm about to get my "new" box set up | ...and will start trying to "check-compile" on your files, Ray... | ... | This will give you an additional "cross-checker"....you, yourself asked | for some of us to backstop you, on this, so I thought I'd start doing | it....but...maybe Glenn is already doing this.... | ... | If there's some difference....you can BET I'll run into it...so...he he | he.....I'd like to know what/why, now....before I start driving myself | crazy.... | ... | | | ============================================================================ | | Sun, Jan 18, 2004 - 5:40:09 pm | | gregy> | Also....when I start pasting Christof's translations in (I may be able | to start around 1 Feb, instead of 15 Feb)...I will want to do my own | "check-compile" after I finish each couple of files.....just to be | sure I haven't messed something up.... | .. | So....let's see what the difference is.. | ... | | | ============================================================================ | | Sun, Jan 18, 2004 - 5:43:44 pm | | gregy> | The Dude...Ray...you know him from the old AQC and ArachneList....he | is Stuart Miller...a New Zealander.... | ... | He claims to NOT be a c-coder...but...has had a recent college course | or two in C...so...he should be able to handle the cut & paste with | me... | ...and.... | He is investigating the SSL sources, etc......I told him about your Š| (glennmcc) already getting an ArachneLynx386 fix working for HTTPS | and suggested he discuss this with you.... | | | | ============================================================================ | | Sun, Jan 18, 2004 - 6:44:38 pm | | glennmcc> | ray, | | Do me a favor please. | | Zip-up and Email me your latest compile of nokey.exe to glennmcc@cisnet.com | | We'll just see if it hodls-up to this statement...... | | "Nope, there won't be one byte different." | | | | ============================================================================ | | Sun, Jan 18, 2004 - 8:06:38 pm | | glennmcc> | This is why I say it will not compare as being.... | "Nope, there won't be one byte different." | | | CORE-01 EXE 468088 1-18-04 9:57p | CORE-02 EXE 468088 1-18-04 10:00p | CORE-03 EXE 468088 1-18-04 10:02p | !NOTSAME TXT 646 1-18-04 10:01p | SAME TXT 117 1-18-04 10:03p | | core-01.exe was compiled before making any changes | | core-02.exe was compiled after adding one blank line into html.c | | core-03.exe was compiled after restoring the unaltered html.c | | --- same.txt --- | Comparing files in... | L:\ | with files in... | L:\ | | Comparing CORE-01.EXE with CORE-03.EXE - Files compare ok. | ________________ | | | --- !notsame.txt --- | Comparing files in... Š| L:\ | with files in... | L:\ | | Comparing CORE-01.EXE with CORE-02.EXE - Compare failure. | Offset 46e88h Source = abh Destination = ach | Offset 46f0dh Source = bah Destination = bbh | Offset 47790h Source = 8ch Destination = 8dh | Offset 47a81h Source = d9h Destination = dah | Offset 47db5h Source = 34h Destination = 35h | Offset 47dbch Source = 37h Destination = 38h | Offset 47ee0h Source = 4ah Destination = 4bh | Offset 48560h Source = 1h Destination = 2h | Offset 4870eh Source = 24h Destination = 25h | Offset 48715h Source = 27h Destination = 28h | 10 Mismatches - No more mismatches were present. | ____________________ | | | | ============================================================================ | | Sun, Jan 18, 2004 - 8:11:05 pm | URL | | glennmcc> | As you see... simply inserting one blank line into one file makes the | finished EXE different by 10 bytes. | | You have made many, many changes. | | Unless I miss my guesss..... | Your file and the original will have several thousand differences. | | Something similar in compare results to the corrupted APMs on Bear's site. | | http://www.cisnet.com/glennmcc/download/trashed-apm.txt | | | ============================================================================ | | Sun, Jan 18, 2004 - 8:25:12 pm | | gregy> | ..so..what can we do....??? | ...to make sure the cleaned files are functionally equivalent.... | ... | I mean....I can think of some looooonnngg drawnout procedures...but.. | ... | | | ============================================================================ | | Sun, Jan 18, 2004 - 9:08:08 pm | Š| glennmcc> | That is exactly why I said..... | | Are you using definition #2 ??? | 2: having such close resemblance as to be essentially the same | | I wasn't just joking around. | | I already knew that making even such a (seemingly) insignificant change | as inserting a blank line into a file makes a very significant change in | the end result. | | ============================================================================ | | Sun, Jan 18, 2004 - 10:49:44 pm | | ray> | | gregy> | | Joe..Ray...re: using 1.71;ue01 to access/download files from | | yehawgroup | > snip | | RAMDisk...well ....that will not prevent accumulating bugs into the | | basic Arachne installation.... | | Cool advice, Gregy. I've never tried the ramdisk thing, IIRC, there's | something like a howto somewhere in the docs -- gota try it. Funny thing | is, I can download everything from Hehaw execpt that one file 'JDS173A.ZIP' | It's hard to imagine some local bug on my system that would be that | picky. | | | gregy> | | Ray........hey...it sounds like glennmcc as ALREADY spotted some | | difference between his 173CORE and one compiled from your cleaned | | source.... | | ...maybe this is just some kind of misunderstanding....but.... | | ... | | So...Glenn....just how much difference do you see....??? | | ... | | Ray says he's seeing no difference....so...it might be a simple | | compiler setting, or something... | | There are all sorts of reasons why there might be differences: | -IDE settings | -Command line vs. IDE compile | -Date | - ... and of course the MALLOCERR thing :-( ... that quarantees that | there will be differences, but it is easy to prove that they are of | no consequence. To make a legitimate comparison, MALLOCERR must be | 'turned off'. (I did explain this, didn't I ? ... Or did I just | promise to explain it ? ...) | | | Also....when I start pasting Christof's translations in (I may be able | | to start around 1 Feb, instead of 15 Feb)...I will want to do my own | | "check-compile" after I finish each couple of files.....just to be | | sure I haven't messed something up.... Š| | Good plan; you can't be too careful, trust me on that! | | | | | glennmcc> | | ray, | | | | Do me a favor please. | | | | Zip-up and Email me your latest compile of nokey.exe to | | glennmcc@cisnet.com | | | | We'll just see if it hodls-up to this statement...... | | | | "Nope, there won't be one byte different." | | | Glenn, wait a few more days; I'll be finished then so there'll be no | problems with partial versions floating around. I understand your | ... nuts, its obvious ... I *didn't* explain the MALLOCERR thing did I? :-( | | | This is why I say it will not compare as being.... | | "Nope, there won't be one byte different." | | | | core-01.exe was compiled before making any changes | | | | core-02.exe was compiled after adding one blank line into html.c | | | | core-03.exe was compiled after restoring the unaltered html.c | | | | | | Comparing CORE-01.EXE with CORE-02.EXE - Compare failure. | | Offset 46e88h Source = abh Destination = ach | | Offset 46f0dh Source = bah Destination = bbh | | Offset 47790h Source = 8ch Destination = 8dh | | Offset 47a81h Source = d9h Destination = dah | | Offset 47db5h Source = 34h Destination = 35h | | Offset 47dbch Source = 37h Destination = 38h | | Offset 47ee0h Source = 4ah Destination = 4bh | | Offset 48560h Source = 1h Destination = 2h | | Offset 4870eh Source = 24h Destination = 25h | | Offset 48715h Source = 27h Destination = 28h | | | As you see... simply inserting one blank line into one file makes the | | finished EXE different by 10 bytes. | | | Gentlemen, | | Heh, I start writing before I've read the whole days postings, bad | plan. Now that I'm at the bottom, I see for sure that Glenn is | bumping hard in to the MALLOCERR thing. So that proves that I haven't | discussed it, my apologies, I've wasted your time. Anyway: | | Maddening, isn't it! When I started to clean 'html.c' it drove me | nuts, until Michal (God bless him) figured out what the problem was | the *hard* way, using a disassembled core.exe. So: Š| | On line 435 of arachne.h (original sources), we have the macro | MALLOCERR, one of who's arguments is 'int line'. This calls | mallocerr() on line 64 of 'errors.c'. 'line' is passed the value of | the literal line on which the MALLOCERR macro is found, so even adding | a blank line to a file in which the macro is found will change the | .exe (as Glenn has found out). | | It's really a very cool debugging thing, since if the macro is called, | it points to the very line in the sources on which it is called. But | that is the only change -- the value of a line number argument, which | is of zero consequence. | | Now, when cleaning, I needed to find a way to make sure that any 'fc' | changes were bugs, *not* just MALLOCERR arguments so: I 'turned the | macro off' ... I've redefined it as "puts(howdy!)" for the duration of | the clean. Obviously, I had to do exactly the same thing in the | compile with the original sources! Thus I am comparing 'apples and | apples'. Once I'm finished, and there are no 'fc' differences, then I | will 'repair' MALLOCERR, and all will be well -- the .exe will be very | different, but we will have the certainty that the only differences | are line number arguments. | | Thus, before we can compare .exe's we must not only synchronize | compiler settings, we must both disable MALLOCERR, and both do it in | exactly the same way! | | Sorry for not talking about this sooner, I had no idea you were | bumping into this for the first time, I assumed that you would have | run into the problem long ago and figured out what it was. | | | | | | ============================================================================ | | Sun, Jan 18, 2004 - 11:47:06 pm | | Joe> | Gregy, | | I have e-mailed the other Glenn, sending him a copy of the "cookie-safe" | core, I hope he tries it - if it fixes his lock-ups then I'll know it's | 100% effective. I also suggested his problem may be due to him having | a GeoCities web site, this may give him different treatment by Yahoo. | | ============================================================================ | | Sun, Jan 18, 2004 - 11:53:51 pm | | Joe> | Ray, | | I have a theory. Perhaps the 'JDS173A.ZIP' file is a problem because it Š| was the first file created in the 'Joe' directory? Can you download | 'html.c' from the 'Ray' directory? IIRC, that was the first file created | in that directory. Anyway, per your suggestion, I have renamed the file | in question to lower case, so you can try your theory too. | | ============================================================================ | | Mon, Jan 19, 2004 - 12:03:24 am | | Joe> | Glenn, Ray, Gregy, | | FWIW, I remember when first compiling with Ray's "html.c", I ended up | with (if I'm not mistaken) 11 bytes different in the executable. | | And Ray, you did briefly mention the "mallocerr" thing earlier : | http://groups.yahoo.com/group/ArachneDevelopment/message/73 | | | ============================================================================ | | Mon, Jan 19, 2004 - 12:53:19 am | | gregy> | Well....hey....some of us have MALLOC-leakage....... | ... | I now dimly remember it..... | ... | he he...prolly forget it by tomorrow...... | ... | ..so...expect some squawks from me when I start compile-checking my | pastes.... | ... | (just for the first few days, I predict....sommat happens after I've | "run the same trail" a couple times)..... | | | ============================================================================ | | Mon, Jan 19, 2004 - 2:42:44 pm | | ray> | Joe, Gregy, | | Thanks, boys, I feel a wee bit better knowing that I did at least | mention it in passing. The bad news of course is that the M.. thing | means that we've got to nail down the vertical spacing of the | base1 code, since (as we've seen), something as simple as adding a | line will result in different .exe's (very bad for comparisons :(). | Ironicaly, vertical spacing was the thing I spent the least time | worrying about, since it seemed to be the most arbitrary. I'm thinking | that once I'm done, I'll post the lot, then you guys should | either proof read and make changes and/or get *really* picky and | suggest changes that can be GPEP'ed, which I'll make and then *that* | will be the base1 code ... sound like a plan? Š| | BTW, also, some thought should be given to the lead lines of each file. | At pressent about 2/3 of the files have what you might call | 'Michael's standard leader" at the top, but some have nothing. | Should Michael's leaders be removed? I'd say yes, tentatively. I'd | like a lead line something like: | | CORE.1; see CORE1.TXT. Proofread by Joe, 19-01-04. CZECH not translated. | | 'CORE1.TXT' Would contain all relevant information about the | project as a whole. What say ye? | | | ============================================================================ | | Mon, Jan 19, 2004 - 4:57:54 pm | | ray> | Joe, | | It just gets more curriouser: | | html.c OK | INETSRT.C OK | JDS173C.ZIP OK | jds173a.zip NO! | | (Using '1.73) | | | | ============================================================================ | | Mon, Jan 19, 2004 - 5:29:37 pm | | glennmcc> | ray, | | So........... | | | Are you using definition #2 ??? | 2: having such close resemblance as to be essentially the same | | | | As to copiler settings. | | They are 'locked-in' by nokey.prj | | If you are not compiling from inside of the IDE using nokey.prj | and choosing compile/build all................ | | You are doing it wrong. | Š| You can not claim "identical" and you will not get "identical" | if you do not follow these EXACT steps to-the-letter. | | Absolutely, positively NO variance from these outlined steps. | | Sun, Jan 18, 2004 - 3:12:28 pm | glennmcc> | I'll take that bet. ;-) | | CD to the dir containing your cleaned SRC code. | | Start BC v3.1 | | Choose 'compile' and 'build all' | | Alt+X to exit BC | | After editing nokeyexe.bat so that your current core.exe | will not be over-written.... (rem-out the 2nd line) | | --- nokeyexe.bat --- | markexe.exe nokey.exe 181000 | xcopy /f nokey.exe c:\arachne\core.exe | rem xcopy core.exe e:\xchaos\arachne\debug\ | ____________________ | | Run nokeyexe.bat | | Now compare your newly compiled nokey.exe to the core.exe v1.73;GPL | that was in the package you D/Led from my site back in November. | | The files should be exactly the same size and they should differ | from each other in only one respect. | (the 2 bytes containing the date of compile) | | | | | ============================================================================ | | Mon, Jan 19, 2004 - 5:30:39 pm | | glennmcc> | Time for dinner...... | | BBL to see how much ruckus I've caused. | | ============================================================================ | | Mon, Jan 19, 2004 - 5:31:16 pm | | gregy> | I'd say...keep Michael's old headers, until we get through BASE1... | ..then we can make more radical changes to the code for BASE2...and | then maybe finalize removals/additions for BASE3... Š| ... | | | | ============================================================================ | | Mon, Jan 19, 2004 - 5:39:48 pm | | gregy> | Ray....I think you're thinking of TOO much of a jump from BASE0 to | BASE1....I think that when you are done....THAT is BASE1....and.. | when I and TheDude get done inserting Christof's translations...THAT | is BASE1T.....and Christof's translated-code will be put back in as | BASE0T (legacy with translations)... | ... | Then...we can proof read, and do further cleanup, and maybe some | removal/addition....and TNAT will be BASE2... | ... | Then REAL picky and more cleanup and removal/addition....and THAT | will be BASE3.... | | During this time....our being "in the cleaning process" does NOT | preclude you guys (especially glennmcc) from issuing a new "version" | based on the legacy BASE0 or BASE0T....there are enhancements that | glennmcc wants to get "out there"....and Joe, also....and Michal has | done some things with WWWMAN or INSIGHT (or both)....so...there's no | reason to keep these things back while we labor to get a good cleaning | on the legacy code... | | | ============================================================================ | | Mon, Jan 19, 2004 - 5:42:36 pm | | gregy> | Ray...I hafta agree with glennmcc, about keeping the code so it | compiles to the same as the legacy....at least for BASE1...and...any | future changes in CORE.EXE that appear as differences from BASE0 | compiled CORE...must be fully documented and accessed for their impact | and effect on the operation of CORE... | | | ============================================================================ | | Mon, Jan 19, 2004 - 10:25:29 pm | | ray> | Glenn, Gregy. | | | As to copiler settings. | | They are 'locked-in' by nokey.prj | | If you are not compiling from inside of the IDE using nokey.prj | | and choosing compile/build all................ | | You are doing it wrong. | | You can not claim "identical" and you will not get "identical" Š| | if you do not follow these EXACT steps to-the-letter. | | Absolutely, positively NO variance from these outlined steps. | | We're not connecting somehow on this; I'm not quite sure I know what | you are getting at. Compiler settings are anything but 'locked in', | you can fool with them to your hearts delight. What *is* true | (obviously) is that the settings must be identical to produce a | meaningful .exe comparison. But with *any* settings, you can still | verify the accuracy (compile wise that is) of the cleaned code by | simply making the needed change to MALLOCERR in the original code, | making the exact same change to the cleaned code, compiling them both | with the same settings, and comparing the results ... which *will* be | identical (not close ... identical). (Then reversing the M... change | in both as the last step, of course, to get the 'working' .exe wich | *won't* be identical). Then, if you like your .exe's at tight as | possible you do a final, command line compile to get the smallest .exe | possible. Try the above with the few cleaned modules you have and | you'll see what I mean :-) I repeat: I will not release one line | of cleaned code unless the whole thing is identical in compile to the | originals (except for date, MALLOCERR, blah, blah) | | BTW, Glenn, why not just rename 'nokey.prj' to 'core.prj' so that you | get 'core.exe' directly without having to rename it? ... that's what | I've done, it just saves one step. | | | I'd say...keep Michael's old headers, until we get through BASE1... | | ..then we can make more radical changes to the code for BASE2...and | | then maybe finalize removals/additions for BASE3... | | Sure, just wanted you guys to think about it. | | | Ray....I think you're thinking of TOO much of a jump from BASE0 to | | BASE1....I think that when you are done....THAT is BASE1....and.. | | when I and TheDude get done inserting Christof's translations...THAT | | is BASE1T.....and Christof's translated-code will be put back in as | | BASE0T (legacy with translations)... | | Then...we can proof read, and do further cleanup, and maybe some | | removal/addition....and TNAT will be BASE2... | | Then REAL picky and more cleanup and removal/addition....and THAT | | will be BASE3.... | | No disagreement! ... when I'm done, that's 'base1' ... ETA about 1 week -- b | before I send it down the pike, there's no reason why I can't do a few | more 'project wide' changes, if you guys tell me in time! | | | | ============================================================================ | | Tue, Jan 20, 2004 - 12:43:38 am | | Joe> | Ray, Gregy, | | Well, in the next day or two, I'm just going to delete 'jds173a.zip' Š| and upload it again. I'll let you know when to try again. Gregy, have | you tried downloading this file with Arachne (I know I have, without | a problem, so Ray's situation seems bizarre). | | Also, I have been having discussions with Yahoo about the need for | Javascript to acquire a Yahoo ID. They say they'll review it, so | there's a *small* chance they may fix this. Ray, did you need JS | to get your Yahoo ID or did the special "profile request" stuff allow | you to create one without needing JS? (Incidentally, Glenn doesn't need | JS to get a Yahoo ID, he's had one since 1997 .... he he ;-) | | ============================================================================ | | Tue, Jan 20, 2004 - 12:56:40 am | | Joe> | All, | | I did a test yesterday ... I replaced "__LINE__" with "0" in the | MALLOCERR() definition (which seemed a sensible substitute for the | purpose of EXE comparison) and compiled (a) the original sources and | (b) those in which Christof had made translations. Apart from needing | "//" in one place in one file (compilation error), both resulted in | byte-identical EXE files. :-) | | Also, I'd be very wary of releasing code compiled from the command | line (unless it used the makefile I posted a while ago, which was | derived from 'nokey.prj'), as there can be subtle problems in data | structures, depending on compiler settings and optimizations, as | compared to the original IDE-based compilation. In other words, it | is safer to "do it Glenn's way" ... | | ============================================================================ | | Tue, Jan 20, 2004 - 1:12:28 am | | Joe> | Gregy, | | Thanks for your assistance at a4dos. I only used the term "maintainer" | to put things in perspective for Florian. Although you aren't really | the "maintainer", you certainly are the "coordinator"! | | BTW, if you have ever read the disclaimers I use with my software, | you will know that I am very cautious about describing the "status" | (for lack of a better word) of said software. So I will be the first | to describe any software I release as "very alpha", if that is the | case. Since I haven't described the Cores I've released that way, | it means I have every confidence they are on par with 1.73;GPL and | you should have no reluctance to try them. The only caution I have | issued is for the first core I released, on the basis that this was | compiled with BC 3.0 (including updates), rather than the more | familiar BC 3.1. That said, I have seen no evidence of any concerns | about using BC 3.0 in my own testing. You may be aware that Kevin | had a crash with the first Core, but then his Config.Sys, Autoexec.Bat, Š| Arachne.Bat and Arachne.Cfg were all a mess, so I wouldn't read much | into that. | | | ============================================================================ | | Tue, Jan 20, 2004 - 10:08:29 am | | glennmcc> | ray, | | As I said, The compiler settings are locked-in to nokey.prj | | Have a look at it. | | CD to the dir containing the SRC code. | | Start BC and open the project 'window'. | | Take a looke at the settings for the project itse;f and for each | individual file. | | They are 'locked-in' to the needed settings for memory model, CPU family, | static/dynamic linking, etc,etc,etc.... | | When you compile using nokey.prj.... the compiler settings are locked-in. | | | Now as to why we can't simply remname nokey.prj as core.prj | | Try it... see what happens. | | ============================================================================ | | Tue, Jan 20, 2004 - 10:25:39 am | | glennmcc> | All of that about compiler settings is of-coarse assuming that you ARE | using _my_ Borland configuration files that were included in in the ZIP | that I sent you. | | If you have changed the settings..... your bad. | | ============================================================================ | | Tue, Jan 20, 2004 - 3:44:18 pm | | ray> | | Joe, | | Also, I have been having discussions with Yahoo about the need for | | Javascript to acquire a Yahoo ID. They say they'll review it, so | | there's a *small* chance they may fix this. Ray, did you need JS | | to get your Yahoo ID or did the special "profile request" stuff allow | | you to create one without needing JS? (Incidentally, Glenn doesn't Š| | need | | ... no problem getting in, or downloading any file except jds173a.zip ! | | | I did a test yesterday ... I replaced "__LINE__" with "0" in the | | MALLOCERR() definition (which seemed a sensible substitute for the | | purpose of EXE comparison) and compiled (a) the original sources and | | (b) those in which Christof had made translations. Apart from needing | | "//" in one place in one file (compilation error), both resulted in | | byte-identical EXE files. :-) | | Yeah, that's the ticket, Joe, all you have to do to get a usable | .exe comparision is disable MALLOCERR() one way or another. (I found | it easiest to just redefine the macro itself in 'arachne.h'). | | | Also, I'd be very wary of releasing code compiled from the command | | line (unless it used the makefile I posted a while ago, which was | | derived from 'nokey.prj'), as there can be subtle problems in data | | structures, depending on compiler settings and optimizations, as | | compared to the original IDE-based compilation. In other words, it | | is safer to "do it Glenn's way" ... | | Hmmmm. This talk of 'subtle problems' is interesting ... good to | know exactly what these might be! Right now I'm using a core.exe | compiled from the command line (I made my own makefile using 'prj2mak'. | Anyway, there's no doubt that we | all have to sing the same tune when it comes to comparisons. | | Glenn, | | | As I said, The compiler settings are locked-in to nokey.prj | > ... | | When you compile using nokey.prj.... the compiler settings are | | locked-in. | | I'll take another look, but I had no trouble changing things at all. I | changed to 386 optimization and changed the output directory to suit | myself. (Of cource I've still got the original, too). | | | Now as to why we can't simply remname nokey.prj as core.prj | | Try it... see what happens. | | ... Did it a long time ago ... can't see any problem at all, just get | 'core.exe' directly. Am I missing something? | | | | | | | | ============================================================================ | | Tue, Jan 20, 2004 - 3:45:38 pm | Š| ray> | ... but caution must rule the day ... I'll go back to the original | downloaded 'core.exe' and see if that fixes things vis. downloading. | | | ============================================================================ | | Tue, Jan 20, 2004 - 5:07:39 pm | | glennmcc> | What about all of those #IFDEF NOKEY statements. | | If you do not compile using nokey.prj none of those will be compiled. | | | | ============================================================================ | | Tue, Jan 20, 2004 - 5:45:10 pm | | ray> | AHHHH, but they will! :-))) | | | | | ============================================================================ | | Tue, Jan 20, 2004 - 5:49:44 pm | | ray> | | ... Which reminds me to ask: There is alot of functionality | in Arachne that is conditional on the OVRL define. Does anyone know | what's so special about this? It obviously reffers to the 'DOS overlay' | compiler setting (as opposed to the 'DOS standard' setting), but | one might think that the 'standard' would be ... well ... standard. | | | ============================================================================ | | Tue, Jan 20, 2004 - 6:42:05 pm | | Joe> | Ray, | | You missed my question ... did you need JS to acquire your Yahoo ID | or did the special "profile request" stuff allow you to create one | without needing JS? | | ============================================================================ | | Tue, Jan 20, 2004 - 6:44:28 pm | | Joe> Š| Ray, | | The OVRL stuff is for using overlays. If you can (eg. with BC 3.X), | this lets you have more memory available. If you can't, the code | compiles with smaller buffers and other stuff to conserve memory. | Something like that, I've not tried compiling non-overlay mode ... | | | ============================================================================ | | Tue, Jan 20, 2004 - 6:51:04 pm | | glennmcc> | | Explain please, just HOW the lines contained within the #ifdef nokey | conditionals will get compiled when nokey.prj is not being used. | | re:"in Arachne that is conditional on the OVRL define." | | As I suggested with renaming nokey.prj as core.prj................ | | Rename nokey.prj as ovrl.prj and see what happens. ;-) | | ============================================================================ | | Tue, Jan 20, 2004 - 9:11:49 pm | | ray> | Joe, | | Sorry, ya, I missed your point. And the answer is ... I don't know. | I set up my Yahoo ID a long time ago and did it under Winder$ because | at the time Arachne wouldn't handle it. | | Glenn, | | All the 'gobal defines' (if that's what they should be called) are | defined within the project file itself; the name of the .prj has | no effect if I'm not mistaken. Well, I've already said it ... so | now I'll go and check and see if I'm wrong ... | | | | | ============================================================================ | | Tue, Jan 20, 2004 - 9:33:52 pm | | ray> | ... and ... I'm half wrong. Yeah, the .exe's are different becuz they | contain name info. after all (I thought they didn't, and I'll bet | that a command line compile ommits that info) ... but the main point | is that the name of the project file has no bearing on which | conditionals are active: I tried core.pjr noname.prj ray.prj and | posix.prj and they were identical appart from the name bytes. Š| | But, gotta admit, that that's another source of comparison difference | that I wasn't expecting. So now we've got to look out for: | | -date | -name of project file | -MALLOCERR | -compiler settings. | | But that's go to be the full list, no? | | | | | | ============================================================================ | | Tue, Jan 20, 2004 - 9:49:55 pm | | gregy> | Joe, Ray....jds173a.zip...???....that's that little 6Kbyte file, | right....??? | ... | It downloaded fine, first time.... | ... | | | ============================================================================ | | Tue, Jan 20, 2004 - 10:41:53 pm | | glennmcc> | In the voice of Jar Jar Binks...... | | Mygiveup :((( | | ============================================================================ | | Tue, Jan 20, 2004 - 11:28:30 pm | | Joe> | Gregy, | | Yep, thanks for the info. I guess it's just Ray who has some sort | of evil spell over his computer ... | | | ============================================================================ | | Tue, Jan 20, 2004 - 11:32:00 pm | | Joe> | Glenn, | | I've had no response so far from MP about Miniterm sources, I guess Š| he has lost all interest in his baby. :-( | | Would you mind asking him? Perhaps you'll have better luck ... | | ============================================================================ | | Wed, Jan 21, 2004 - 3:35:49 pm | | ray> | Joe, Gregy, | | Yup, must be an evil spell ... like 4dos or FAT32 or something. I'll | try again from a 'Glenn-aproved' type install (FAT16, DRDOS, COMMAND.COM | etc.) if that don't bag it, then I'm stumped. | | Re Miniterm: That's just the dialer, no? Didn't I read somewhere in | passing that LSPPP can be used as a dialer too? | | | Glenn, | | Glad to see you are surendering . | | | | ============================================================================ | | Wed, Jan 21, 2004 - 3:37:55 pm | | ray> | Oh, | | Joe, you don't suppose that the jds... thing could be some sort of | ... na ... some sort of glitch on the part of my ISP? ... they have | the most bomb-proof anti spam filters. | | | | ============================================================================ | | Wed, Jan 21, 2004 - 4:33:18 pm | | Joe> | Ray, | | Great minds think alike! ;-) | | Just last night it occurred to me that your ISP was operating as a | proxy and caching stuff. Something must have gone wrong with your | first download attempt and now every time you try, the ISP is giving | you back the same crap. I sometimes get that here with our company's | proxy server. | | In case that's not it, I have now deleted the file and uploaded again, | so you might try one last time when convenient. Š| | ============================================================================ | | Wed, Jan 21, 2004 - 4:37:55 pm | | Joe> | Ray, | | Well, DOSPPPD (EPPPD) also has its own dialler. Yet Arachne uses | _its_ own dialler instead, because it is better integrated and | provides the terminal window capability ... | | ============================================================================ | | Wed, Jan 21, 2004 - 4:42:35 pm | | glennmcc> | Joe, | | Me have better luck getting a reply from Michael ??? | | Hahahahah .... you're funny. | | I've tried..... I can't get a reply from him either. | | IMO, | Michael seems perfectly content to just let Arachne die. :(( | | | ============================================================================ | | Wed, Jan 21, 2004 - 4:50:53 pm | | glennmcc> | re: lsppp as dialer. | | | No problem. :)) | | | --- in my current arachne.cfg --- | (dialer line split for this posting) | (my PCI modem does not have a comport # assigned to it so I need to give | lsppp the IRQ # and base port address instead of the com port #) | | [profile] | Profile 4234057L.ACF | | Dialer @lh lsppp.exe /x /i:4 /b:bx800 /d:330-423-4057 | /B:115200 /V:60 /U:glennmcc@cisnet.com /P:******>PPP.LOG | | Connection @lsppp.exe /?>NUL | Hangup @lsppp.exe /u>NUL | _________________________________ | Š| | | ============================================================================ | | Wed, Jan 21, 2004 - 11:11:19 pm | | Joe> | Glenn, | | Hmmm ... maybe MP himself has died? :-/ | | Anyway, about LSPPP dialling : | a) Yes, it's easy enough to configure it for dialling but it's not | going to provide the terminal-window capability, so it's not a drop-in | replacement for Miniterm. | b) Although it dials fine for me, there are some people whose modems | it won't dial on, although they have no trouble with Miniterm. You | may recall someone complaining at arachne-development (IIRC) about | that when I was encouraging the use of LSPPP. My response was along | the lines of "Sure, in an ideal world, it would be good if the LSPPP | dialler worked for every situation, however LSPPP's main function is | as a PPP packet driver, and personally I wouldn't care if the dialler | stuff was removed altogether". So if we were to delete the terminal | window option and use LSPPP for dialling too, we could be facing a | new problem, which is not what we want right now. | | ============================================================================ | | Thu, Jan 22, 2004 - 4:41:08 pm | | ray> | Joe, | | Is lsppp open source? | | | ============================================================================ | | Thu, Jan 22, 2004 - 4:50:05 pm | | ray> | Joe, | | Tried jds173a.zip again with no luck :( But ... inspite of your | reload of that file it still shows the "f4.grp.yahoofs.com" message | at the bottom of the screen when Arachne attempts to download it -- | that's the same message as before. All the other files have f2, f3 | etc. etc. Maybe it's something with 'f4' God know what, though. | I'l like you to try this: Upload exactly the same file, but to | a different name, and leave the existing file in place, that should | nail down wether it's the 'f4' or the contents or name of the file | itself. This is soooo strange. | | | ============================================================================ Š| | Thu, Jan 22, 2004 - 6:11:29 pm | | Joe> | OK Ray, | | Will do that shortly ... | | About LSPPP, remember you "complaining" to me about my instructions | in LSPPPCFG, to "unzip LSPPP into the Arachne directory"? Something | along the lines of "Hey, there's no need to unzip the source files | into the Arachne directory". I think that answers your question. | | ============================================================================ | | Thu, Jan 22, 2004 - 7:25:36 pm | | ray> | Joe, | | Yep, that it does. Anyway, isn't that an opportunity for us to | bypass the Miniterm problem and integrate lsppp? Seems to be the | best choice for packet driver and if it dials too, then all's that's | needed is some sort of terminal function, no? | | | | ============================================================================ | | Thu, Jan 22, 2004 - 8:04:09 pm | | glennmcc> | Why do we need a "terminal function" ? | | ============================================================================ | | Thu, Jan 22, 2004 - 9:22:52 pm | | ray> | Glenn, | | ... you don't *ask* questions ... you *answer* them . | | ... but seriously, isn't 'terminal mode' when we see the connection | strings? ... and isn't that important sometimes? | | Joe, | | The test file ... no problemo! ... I dunno what to make of it; why | the trouble with just that one other file? We now know it isn't the | content of the file. The only other options seem to be the name is | a problem, or the "f4" is a problem. Either way it's very strange ... | probaly not worth spending any more time on, but I'll keep | wondering ... | Š| | ============================================================================ | | Thu, Jan 22, 2004 - 9:27:55 pm | | ray> | Joe, | | Just tried to download the 'bad' file again: I'm thinking that this | is what it's got to be -- something in the blasted cookie! Is there | some way to test this? | | Yeah, try that new core .... | | | | | | ============================================================================ | | Thu, Jan 22, 2004 - 9:36:04 pm | | ray> | Joe, | | Results with new core ... same problem as before :-( | But, the new core reduces my memory to "143+" from "160+" ! | | | ============================================================================ | | Fri, Jan 23, 2004 - 12:18:37 am | | Joe> | Ray, | | Err, if you suspect a cookie problem, sign-out, delete or hide your | cookie jar, clear your cache, then try again. But my bet is your ISP | is simply caching this URL with some crap that went haywire first time. | | As for the 1.73J3 core, that will only prevent lock-ups with excessively | long cookies, which doesn't sound like your particular problem anyway, | it won't make disfunctional cookies work any better. You didn't have a | lock-up problem, so you won't see any difference there. As for the | memory, that's because 1.73J3 uses the standard build, not 386+387 | optimized as I believe your previous core was. | | ============================================================================ | | Fri, Jan 23, 2004 - 3:13:38 pm | | Joe> | Ray, | | When I said "1.73J3 uses the standard build", I meant the existing Š| Core of 1.73J3 uses the "standard build". Of course, you can build | it yourself using 386/387 optimization, to get back to "160+". | | | | ============================================================================ | | Fri, Jan 23, 2004 - 6:28:19 pm | | ray> | Joe, | | Tried deleting cookies, no diff ... to hell with it! I'll keep thinking | about the problem, but I'm not going to waste any more of your time | talking about it. If you get any ideas, though, let me know. | | | ============================================================================ | | Sat, Jan 24, 2004 - 1:19:07 pm | | ray> | Joe, Glenn, | | School time: | | In 'int09.c' we see this: void STvect(in No, void interrupt(*fn)()) | | ... why would a function take a *void* function as | an argument? ... since a void fuction returns nothing, what does | "void interrupt(*fn)" return as a value? | | | | ============================================================================ | | Sat, Jan 24, 2004 - 3:54:16 pm | | gregy> | Ahh...you guys have had a busy couple of days, while I was offline.. | ... | ... | BTW...has anyone seen Michal since he last posted here and on the | yehaw group....??? | ... | | | | ============================================================================ | | Sat, Jan 24, 2004 - 3:57:14 pm | URL | | glennmcc> | Ray, Š| | Do you mean to tell me that you do not understand something as simple as tha | | | Well, so that you do not feel totally alone. | | | Untill today.... I have never looked inside of int09.c | | Just now looked at it. | | From what I can tell by just a quick glance. | | The functions do not return anything. | (as you most likely already suspected) | | | int09.c is basically keyboard scanning code. | | Thanks to Christof..... we can now figure it out better. | | I just got his latest translations and int09.c is in there :))) | | ---- int09.ctr --- | //Testing ALT + TAB | #include | #include | | #define ALTSCAN 0x38 | #define TABSCAN 0x0F | #define EXTSCAN 0xE0 | | //GLOBALS | unsigned char g_Scan=0,g_PrevScan=0; | char g_Alt =0; | | int g_AltTab=0; | int g_PrtScr=0; | | void interrupt (*OldKeyboard)(void); | void interrupt (*OldPrtScr)(void); | | void InstalAltTab(void); | void ReleaseAltTab(void); | void InstalPrtScr(void); | void ReleasePrtScr(void); | | //Internal ppg. | void STsetvect(int No, void interrupt (*fn) () ) | { | disable(); | poke(0, No*4+2, FP_SEG(fn)); | poke(0, No*4, FP_OFF(fn)); | enable(); | } Š| | void interrupt (*STgetvect(int no))() | { | static void far *Res; | disable(); | Res = MK_FP(peek(0, (no<<2)+2), peek(0,no<<2)); | enable(); | return( Res ); | } | | void AcceptKey(void) | { static char kbval; | kbval = inportb(0x61); | outportb(0x61,kbval|0x80); | outportb(0x61,kbval); | outportb(0x20,0x20); | return; | } | | void interrupt NewPrtScr() | { | g_PrtScr=1; | } | | //propousti vsechny stisky klavesnice, pouze testuje + | //Pozn: odkomentovanim AcceptKey(), lze nepropustit TAB po ALT | //tr.: releases/lets pass all keys, only tests + | // note: by uncommenting AcceptKey(), you can hook TAB after ALT | void interrupt NewKeyboard() | { | g_PrevScan = g_Scan; | g_Scan = inportb(0x60); | switch( g_Scan ) | { case ALTSCAN : g_Alt = g_PrevScan==EXTSCAN ? 2 : 1; //press ALT le | break; | case ALTSCAN|0x80 : g_Alt = 0; //release ALT | break; | case TABSCAN : if(g_Alt) //press TAB | { g_AltTab = 1; //combination + | //AcceptKey(); //polknu TAB ?? | //tr.: do I swallow TAB ?? | //return; //nevolam puvodni ?? | //tr.: don't I call the original one ? | ReleaseAltTab(); | } | break; | default:; | } | //printf(">>> %02x\n",g_Scan); | OldKeyboard(); | } | | //EXTERNAL PPG: | //Instalation : | void InstalAltTab() Š| { | OldKeyboard = STgetvect(0x09); | STsetvect(0x09, NewKeyboard); | } | | //Deinstallation: | void ReleaseAltTab() | { | STsetvect(0x09, OldKeyboard); | } | | void InstalPrtScr() | { | OldPrtScr = STgetvect(0x09); | STsetvect(0x05, NewPrtScr); | } | | //Deinstallation: | void ReleasePrtScr() | { | STsetvect(0x05, OldPrtScr); | } | ___________________ | | ============================================================================ | | Sat, Jan 24, 2004 - 3:59:40 pm | URL | | glennmcc> | Michal is still alive-and-well and posting at Freelists.org | | ============================================================================ | | Sat, Jan 24, 2004 - 7:14:14 pm | | gregy> | yeah.....I just read his post about the MIME variables to coollist... | ...and... | I posted his reply on the AVA FAQ-DISC Board....... | ... | I just accessed that URL you posted....and I see that the same kind of | crap is going on on that freelists list....that was going on on the | coollist list..... | ... | ..just a WHOLE bunch of yak-yak-yak, without anyone of the whiners | ever DOing anything....except contend some more, and yak some more... | ...that's why I never did/will subscribe to it.. | ... | heck...I couldn't even get anyone to even POST to the FAQ-etc boards.. | ...but...they could all COMPLAIN to the coollist list.... | ... | ...so...it's just "mush on, regardless".....in the near future, we | will have all the collateral ArachneCommunity sites' URLs collected | at ArachneVirtualAve....and most of the functions of a working Š| community site.. | ... | HOWEVER....so MANY of them NEVER even read the posts....even Wiz is | still taking about setting up a "free" upload/download site....even | though I already announced the availibility of the AVA site.... | ...now... | IF someone wants to set up something alternate, then fine...but...when | it sounds like they haven't even, ever, heard of the AVA site...well | ...hmmmmmmm (things that make me go hmmmmmmm..........) | | | | ============================================================================ | | Sat, Jan 24, 2004 - 9:00:49 pm | URL | | glennmcc> | AVA site ???? | | What's that ???? | | ============================================================================ | | Sun, Jan 25, 2004 - 9:40:49 am | | ray> | Gregy, | | Yeah, did I miss something vis. "AVA" ? | | Glenn, | | Yes, a hearty thanks to Christof. Even what comments we have will | make a huge difference to figuring things out. | | As far as my question goes: Way back in my C book-learnin' days I | probably knew what a void argument was good for -- but that war | 20 years agee when I war yung. | | | | | | ============================================================================ | | Sun, Jan 25, 2004 - 6:28:08 pm | | ray> | Blast! | | Wrapping things up in the cleaning department, and almost the | last thing to do was to convert all the tabs to spaces, as agreed | | ... SPROINGGGG ... Š| | any tabs found on any line position that is not modulo 4, expand to | the wrong number of spaces! Nuts, I thought that the 'optimal fill' | setting on the compiler would prevent that -- always use spaces | unless a 'full' tab was being inserted. Now I've gotta find a way | to replace those errant tabs 'inteligently'. | | And don't you guys gloat! | | | | | ============================================================================ | | Tue, Jan 27, 2004 - 12:16:02 am | | Joe> | Ray, | | AFAIK (remember, I am NOT a C coder;-), that void parameter effectively | means "for this parameter, pass me an untyped pointer to a routine". (In | C everything's a pointer, unless it's an integer, and even then, the two | are often interchangable, depending on the implementation!) | | Back to spaces, check with Michal, but I'm sure he preferred 2 space | indentation for complex code such as Arachne, like I do. This was we | can indent properly _and_ still fit lots of code on a line ... | | ============================================================================ | | Tue, Jan 27, 2004 - 4:13:48 pm | | ray> | Joe, | | Ya, that rings a bell! | | About spaces: I just wrote a filter to replace tabs with however | many spaces are needed ... seems to work OK ... but I used 4 spaces :-( | and it's all tested and zipped and ready to send to Gregy. I'm | thinking to upload as is, let Gregy chew on it for a bit, what with | moving the translations and all, then, if he or anyone else has | suggestions for further 'global' changes, I can do it then. But we're | thinking much alike -- I don't know if I could ever get used to | 2 space indentation, but OTOH, durring my last pass through the code, | I was really getting to doubt my commitment to 4 ... I'm thinking | 3 is what we need. Anyway, you guys look it over and let me know. | | | ============================================================================ | | Tue, Jan 27, 2004 - 4:15:27 pm | | ray> | Oh, Joe, not that I'm much into that sort of thing, but what's happening Š| on the old coollist that made you decide to quit? Haven't been on it | for months myself since I couldn't afford the distraction. | | | | ============================================================================ | | Wed, Jan 28, 2004 - 8:19:26 pm | | ray> | Gregy, | | It's commin' down the wire ... | | | ============================================================================ | | Thu, Jan 29, 2004 - 10:00:32 pm | | ray> | ... a strange silence decended ... | | | ============================================================================ | | Thu, Jan 29, 2004 - 11:07:53 pm | | Joe> | Boo!!! | | | ============================================================================ | | Fri, Jan 30, 2004 - 1:49:49 pm | | ray> | AHHH! | | | Were's Gregy? I'm sweating bullets as to wether or not he can | comple my code with no problems ... got a bet with Glenn riding on it | ... | | | | ============================================================================ | | Fri, Jan 30, 2004 - 4:27:49 pm | | glennmcc> | You will loose the bet because of the stipulations. | | My bet was that it had to compare as identical to _mine_. | | Not as being identical to some other of _yours_. Š| | | | ============================================================================ | | Fri, Jan 30, 2004 - 4:34:29 pm | | glennmcc> | Sun, Jan 18, 2004 - 3:12:28 pm | glennmcc> | I'll take that bet. ;-) | | CD to the dir containing your cleaned SRC code. | | Start BC v3.1 | | Choose 'compile' and 'build all' | | Alt+X to exit BC | | After editing nokeyexe.bat so that your current core.exe | will not be over-written.... (rem-out the 2nd line) | | --- nokeyexe.bat --- | markexe.exe nokey.exe 181000 | xcopy /f nokey.exe c:\arachne\core.exe | rem xcopy core.exe e:\xchaos\arachne\debug\ | ____________________ | | Run nokeyexe.bat | | Now compare your newly compiled nokey.exe to the core.exe v1.73;GPL | that was in the package you D/Led from my site back in November. | | The files should be exactly the same size and they should differ | from each other in only one respect. | (the 2 bytes containing the date of compile) | | | ============================================================================ | | Fri, Jan 30, 2004 - 4:36:22 pm | | glennmcc> | Now........ will you win or lose the bet that I made ? | (comparing your new compile to my original) | | ============================================================================ | | Sat, Jan 31, 2004 - 7:20:57 pm | | ray> | Glenn, | | I 'win' over here (exept for mallocerr aguments, of course), but all Š| that matters is that I win over at your place too! And we won't know | that until you try a compile with the cleaned code, and that won't | happen until Gregy uploads it. | | Gregy, where are you? | | | ============================================================================ | | Sat, Jan 31, 2004 - 7:36:35 pm | | glennmcc> | You don't get it.................... | | | Identical == 3 bytes different between mine and yours. | (the 3 bytes which show the date it was comopiled) | | No....... "except for mallocerr" | | Your stipulation was......... And I quote. | | Sat, Jan 17, 2004 - 7:56:28 pm | ray> | Nope, close only counts with: //comments that have been shrtn'd to fit. | | Given the same compiler settings, If my code differs by one byte, | I'll go back to command.com for a year as penance. With time off for | good behaviour, of course. | | ____________ | | I won't hold you to a 'not one byte' stpulation. | | I WILL hold you to those 3 bytes which are the date of compile. | | If yours varies from my original by more than those 3 bytes...... you lose t | | | ============================================================================ | | Sat, Jan 31, 2004 - 10:25:18 pm | | ray> | Glenn, | | Looks like we have a misunderstanding. I thought I made it clear, | of and on, that the mallocerr thing was the one exeption, but that | with mallocerr 'turned off' then never mind the 3 bytes -- compile | on the same day, and the .exe's will indeed be identical (and are over | here). With mallocerr 'turned on' an identical .exe is not possible | for resons discussed, but since the differences are just line number | arguments, there are no *errors* ... and that's the best that | can be done by me or by anyone. Why would I try to achieve that | which is known to be impossible and which dosen't matter anyway? Š| Anyway, if that wasn't your understanding, sorry, we had different | ideas on what the point at issue was. | | | | | | | ============================================================================ | | Sat, Jan 31, 2004 - 11:24:32 pm | | glennmcc> | No problem Ray, | | Now.... we are on the same page. | | As is now shown by your statement above. | | "Why would I try to achieve that which is known to be impossible | and which dosen't matter anyway?" | | You ARE using definition #2 | | __________________ | | Sat, Jan 17, 2004 - 4:51:30 pm | glennmcc> | | Are you using definition #2 ??? | 2: having such close resemblance as to be essentially the same | | | ___________________ | | | The files do not, and will not compare as "identical". | | But rather they will be "essentially the same". | | Therefore, | . | . | . | . | . | . | . | . | . | . | . | . | . | . Š| . | . | . | . | . | . | . | . | . | . | . | . | . | . | . | . | . | . | . | . | | You lose the bet. | | You must now dump 4Dos and start using command.com | | You may go back to 4Dos on Feb 1, 2005 | | | | ============================================================================ | | Sun, Feb 01, 2004 - 12:11:45 pm | | ray> | AHHH! | | Clemency! | | | | ============================================================================ | | Sun, Feb 01, 2004 - 2:13:51 pm | URL | | glennmcc> | Nope........ no mercy. | | http://www.webster.com/cgi-bin/dictionary?Clemency | | It's 1 year with out 4Dos. | And since this is a leap year.... that's 366 days of using only command.com | | ============================================================================ | Š| Mon, Feb 02, 2004 - 9:07:02 pm | | ray> | My lawyer says we're taking it to court. | | | | ============================================================================ | | Tue, Feb 03, 2004 - 4:51:27 pm | | glennmcc> | One of my good friends is a judge. | | We'll put it in his venue and you'll lose again. | | | ============================================================================ | | Tue, Feb 03, 2004 - 8:16:27 pm | | ray> | Ha! | | More Yankee imperialism. | The case will be tried up here, so you're going to have to learn | to speak either French or Punjabi. | | | | ============================================================================ | | Sat, Feb 07, 2004 - 8:39:24 am | | ray> | Glenn, | | Who is this Howard Wulf that Christof mentioned? Sounds like | a powerful coder. | | | ============================================================================ | | Sat, Feb 07, 2004 - 9:06:34 am | | Joe> | Good, we need more coders. The number of feature requests and bug | fix requests is growing faster than can presently be accomodated. | | We will also need to consolidate the various efforts, including the | translations and reformatting, some time soon. Some sort of round- | -robbin approach may be workable, where we each add our changes to | a common code base when we are ready, to prevent the different changes | accumulating to a point where integration becomes difficult. I have | been doing a significant number of changes since 1.73J3, in stages, Š| to eventually support full-size cookies. Nothing beneficial presently, | so I've just been testing this stuff myself, rather than inflicting | it on unsuspecting users. | | One thing about the code formatting troubles me however. In reading the | TXT files, am I right to believe "out-dentation" will be a permanent | feature of the new code base? I had assumed it was an interim measure | while sorting things out. Otherwise, I must say I don't favour this as | a permanent feature of the code (why else do you think I have argued | at every opportunity that two-space indentation was the most appropriate | for this code?). Anyway, give this some thought and ask Michal for his | oppinion if you get a chance before I do... | | Also, I've had no feedback on the BC 3.0 Core I originally uploaded at | "hypermart.net", does anyone have any good/bad experience with it? BTW, | I consider the compiler issue as rather important, since a GPL project | without a viable compiler is ultimately doomed. | | | | ============================================================================ | | Sat, Feb 07, 2004 - 10:58:29 am | | Joe> | Gregy, | | I've just tried Arachne to FTP-upload a file to your 'joe' directory. | No go. No error, but the file just doesn't appear. I know you discussed | the details with Glenn on how to FTP-upload with Arachne, and although | I couldn't find these just now, I tried all the different permutations | in the FTP-upload dialogue box anyway. Well, I find HTTP uploading | much easier and more successful anyway, even with DOS. FTP uploading | is not 'my cup of tea' (being a coffee drinker;-). | | Anyway, it's way past my bedtime. The temperature's dropped to 28.5C, | so perhaps I'll manage to get some sleep now (too hot earlier). Ah, | the joys of Summer! | | | | ============================================================================ | | Sat, Feb 07, 2004 - 11:57:11 am | | ray> | Hi Joe, | | ... 28.5C ... Summer ... Joe, where do you live? | | | | | We will also need to consolidate the various efforts, including the | | translations and reformatting, some time soon. Some sort of round- | | -robbin approach may be workable, where we each add our changes to | | a common code base when we are ready, to prevent the different changes Š| | accumulating to a point where integration becomes difficult. I have | | been doing a significant number of changes since 1.73J3, in stages, | | to eventually support full-size cookies. Nothing beneficial presently, | | so I've just been testing this stuff myself, rather than inflicting | | it on unsuspecting users. | | Nuts, Gregy would have been just the man to handle the sort of thing | you're talking about here; there's no question that coordination will | become vital, as well as some sort of protocol for documenting changes | to the code. Speaking of which: I finally got around to looking at your | posted changes vis. time logging. It looks like good stuff, but I sure | would prefer the use of 'Glenn style' documentation. With Glenn's | changes, you can see what he's done line by line, so there's no doubt | what's new and what isn't. Also, he always marks the beginning and the | end of any area that's been changed, so you know where you have to look. | IMHO this sort of approach could be vital to keeping things organized. | | | One thing about the code formatting troubles me however. In reading | | the | | TXT files, am I right to believe "out-dentation" will be a permanent | | feature of the new code base? I had assumed it was an interim measure | | while sorting things out. Otherwise, I must say I don't favour this as | | a permanent feature of the code (why else do you think I have argued | | at every opportunity that two-space indentation was the most | | appropriate | | for this code?). Anyway, give this some thought and ask Michal for his | | opinion if you get a chance before I do... | | I hear you. It was one of those things that I used as an ad hoc | solution to outrageously long lines. It's not till one gets to the | end of a project like the clean that one can look back and see that | solutions that looked good at the time haven't panned out as expected | -- you know what I mean. | | There are 38 'outdents' of more than a line or two in the whole | 108,000 lines of the core sources, and only 16 files have them. One | could argue that this isn't too high a price to pay for keeping the 4 | space indentation consistently throughout the whole thing, but it's a | judgement call. (And only diamonds are forever ;-) Myself, once I | got used to the idea, I didn't find the outdents to be obnoxious at | all, since many of them involve blocks that are so large you can't see | from one end to the other anyway. Also, some of the outdents are due | more to ultra-long logical lines than they are due to indentation, so | changing the indentation spacing would only solve the problem some of | the time. Take a look at some of the real world examples in the code | and let me know what you think. Of course MT and Glenn should be | asked their opinion and as always consensus will rule the day. I'll | keep a list of whatever changes are desired for the next edition (if | one looks like it might be useful). | | ... And there's the translations to move over, I was sure hoping that | Gregy would handle that and give me a break. | | | Also, I've had no feedback on the BC 3.0 Core I originally uploaded at | | "hypermart.net", does anyone have any good/bad experience with it? Š| | BTW, | | I consider the compiler issue as rather important, since a GPL project | | without a viable compiler is ultimately doomed. | | It worked for me, but as far as I know all the rest of us are using | '3.1. I'm a long way from being an expert on this sort of thing but I do | know that a company can only claim damages from copyright infringement | if it can show that it has actively attempted to prevent such | infringement, and I rather doubt that Borland would have any intention | of enforcing it's copyright on us, even if it knew what we were doing. | | Also, a claim for damages must prove that the intellectual property in | question is of current commercial value to the claimant, and since | '3.1 is no longer marketed by Borland I'd say we're safe on that score | too. ... just my 2 cents worth. Finaly, since Arachne is itself freeware, | Borland would have no one to make a claim against even if it wanted to | do so -- no one has made money off of its product. | | OH, almost forgot: Did you do a build using the cleaned sources and verify | that the .exe is identical? (mallocerr "off", naturaly) ... gota be | sure that what we uploaded isn't corrupt! | | | | ============================================================================ | | Sat, Feb 07, 2004 - 1:29:52 pm | | gregy> | Nope...haven't died...yet........JUST all three of my modems | went belly up......so cannot get online from home, right now... | ... | hope to have replacement modems by next Thursday....will check in here | probably next Tuesday, and Thursday....\ | ... | I'm on the University computers, right now.... | ... | JOE... | ... | Yes...!!!....we need to coordinate the code-deltas....I have a little | preliminary proposal, which I couldn't upload/post because of the blitzed | modems...so....as soon as I get back online at home...I will post/upload | the preliminary proposal......prolly next weekend....until then...please | be patient...\\ | ... | JOE....yes....Glennmcc clued me in to all the Arachne FTP-upload techniques. | can't get it to work, either....like you...no | errors...just that the file refuses to appear on the server......NOTE..\ | ...it LOOKS like it uploads, statusbar indicates it uploads...but...no | file is up it the directory...so....until I get it figured out...again, | please be patient.... | | | ============================================================================ | Š| Sat, Feb 07, 2004 - 1:35:03 pm | | gregy> | Ray....yeah....there was a strange silence........and...in the | process of being silenced....I lost a WHOLE load of mail....and your | 400Kbyte+ email was one of them.....:((( | ... | HOWEVER....please do NOT resend, until I tell you to...no sense in having | my mailbox stuffed full.... | | ============================================================================ | | Sat, Feb 07, 2004 - 4:25:37 pm | | glennmcc> | OK, | Since the status indicator swoed thayt the file was being uploaded..... | It _was_ in-fact uploaded. | | But.... to the wrong directory. | | Have a look one directory up from where you intended it to be. | If it's not there........ go up one more.... not there.... up one more. | etc, etc, etc.... | | It _is_ there somewhere. :) | | ============================================================================ | | Sat, Feb 07, 2004 - 5:26:41 pm | | ray> | Welcome back Sarge! ... I was worried about you. | | I'll resend when you want, or you can just grab it from Yehaw. | | | ============================================================================ | | Sat, Feb 07, 2004 - 6:32:06 pm | | glennmcc> | Speaking of that...... | | I can't grab anything from Yehaw becuase I can't | sign-up for a user name without a JS browser. :((( | | ============================================================================ | | Sat, Feb 07, 2004 - 7:23:47 pm | | ray> | Glenn, | | If you want I'll send it to you direct. Š| | | ============================================================================ | | Sat, Feb 07, 2004 - 7:49:18 pm | | glennmcc> | Would you ? Please ? | | | ============================================================================ | | Sat, Feb 07, 2004 - 8:06:45 pm | | ray> | 'commin down the wire. | | ============================================================================ | | Sat, Feb 07, 2004 - 9:28:46 pm | | glennmcc> | Got it...... | | Thank you very much. :) | | ============================================================================ | | Sun, Feb 08, 2004 - 7:21:59 am | | phantom> | Ray, | | I live in what we here call "God's own country"! :-) | Try to peek in the 'pix' directory of my website for a big clue. | | Roger on the comments. You'll note that jds173a has a "begin" and | "end" of changes indicated. I think the others involve individual | line changes, but I'll look into it ... | | | | ============================================================================ | | Sun, Feb 08, 2004 - 7:30:15 am | | phantom> | Glenn, | | If you need any assistance with Yahoo, I'm happy to help. Your | Yahoo ID from 1997 is still valid, even though mail will have | been disabled (and probably deleted) through non-use. You don't | need to "reactivate" your Yahoo ID to use it at Yahoo Groups, | you only need to do that for Yahoo Mail and perhaps some of the | other Yahoo services (probably Yahoo Briefcase, Geocities). Š| | I can also add you manually to the ArachneDevelopment mailing | list, so it will accept postings from you. If you would like | that, just tell me if you prefer to view messages via the web | interface or whether you would like them to go directly to your | e-mail inbox. | | | ============================================================================ | | Sun, Feb 08, 2004 - 7:39:42 am | | phantom> | PS. | | It goes without saying that Yahoo Mail is not accessible with | Arachne (not sure about Briefcase and Geocities). The point I | was trying to make is that you won't need JS to sign-up your | existing Yahoo ID at Yahoo Groups (or access the Group resources). | | As for Yahoo Mail and HTTP-uploading, Lynx 2.8.5dev.16 works | just fine (for DOS), if you need these capabilities. Sometimes | when I try to login using Lynx, I get a screen saying I haven't | been logged in yet. I don't know why I sometimes get this, but | it's incorrect (probaly a cache thing), if I "goto" a Yahoo | URL where I need to be logged in, it recognizes that I am in fact | logged in and gives me the access I need ... | | In summary, DOS isn't as useless as some would like us to think. :-) | | | ============================================================================ | | Sun, Feb 08, 2004 - 7:47:36 am | | phantom> | Ray, | | Getting back to your posting ... | | We all know how we'd like copyright law to work, unfortunately | (as law often does), it doesn't operate that way. | | As for recompilation, I mentined some time ago that I substituted | "0" for "__LINE__" in that macro and got identical EXE's ... so | I'm confident that your checks are adequate and haven't repeated | this test in recent times. | | | | ============================================================================ | | Sun, Feb 08, 2004 - 9:32:38 am | | phantom> Š| Oops! | | Due to one of many changes by hypermart, the 'pix' directory no | longer shows a default page. Try the 'mirror' directory instead ... | | | ============================================================================ | | Sun, Feb 08, 2004 - 10:13:18 am | | ray> | Joe, | | No need for any hints mate -- you don't have to be a Kiwi to know | where "God's own" is :-) | | North Island, I take it. | | We're on the same wave lenght vis. __LINE__ but I just wanted you | to check using the whole cleanned package, since IIRC you didn't have | everything last time -- but thanks for the vote of confidence! | | | | ============================================================================ | | Sun, Feb 08, 2004 - 5:34:22 pm | | glennmcc> | phantom, | | There's one problem. | | I do not remember what my Yahoo ID was way back in 1997 :( | | What was that username ? and how did you find it ? | | | ============================================================================ | | Sun, Feb 08, 2004 - 7:44:09 pm | | Joe> | Ray, | | North What??? I hope you're kidding. Anyway, my web site won't | disappear for a couple more weeks, so you can take a peek whenever | your curiosity gets the better of you. | | ============================================================================ | | Sun, Feb 08, 2004 - 7:45:26 pm | | Joe> | Glenn, Š| | You have a Yahoo ID of "glennmcc", of course. How could you | possibly not remember that? | | ============================================================================ | | Sun, Feb 08, 2004 - 8:01:35 pm | | ray> | Joe, | | Hmmm, that one picture isn't much help ... is that one of the | abutments of the Sydney harbour bridge? If so then you're from Oz. | ... and I thought only Kiwi's had the right to call their country | 'God's own' 'would have bet money on it. | | | | | ============================================================================ | | Sun, Feb 08, 2004 - 8:03:50 pm | | Joe> | Hi Ray, | | Did you check the name of the JPEG? | | ============================================================================ | | Sun, Feb 08, 2004 - 8:05:27 pm | | Joe> | PS. Glenn, | | BTW, if you should want/need a new Yahoo ID, I can create one | for you with a temporary password. You can then change the | password without needing JS. | | ============================================================================ | | Sun, Feb 08, 2004 - 9:14:40 pm | | ray> | Joe, | | Yeah! ... I still have it. After almost 30 years as a postie, I | pride myself on having seen so many postcards that I can identify | just about anything that's ever been on one, even if the only real | clue is the abutment of a bridge. | | Sorry about thinking you were a mere New Zelander | | | Š| | | ============================================================================ | | Mon, Feb 09, 2004 - 5:56:34 pm | | ray> | Did a wee bit of optimization today by replacing the 'Piip()' function | in 'misc.c' with direct calls to 'printf("%c", 7)' ... saves 880 | bytes and .exe (command line MAKE) is down to 434,304; and | my 'dos free' is up to 161+ ... gotta love it. | | | | ============================================================================ | | Thu, Feb 12, 2004 - 6:20:04 pm | | ray> | Gregy still having problems? | | | ============================================================================ | | Fri, Feb 13, 2004 - 12:13:45 am | | Joe> | Hmmm ... how does one destroy three (IIRC) modems at once??? | Maybe the PC is the problem, the serial port or something is dodgy? | | Speaking of memory issues, does anyone else get XSWAP crashes on | occasion? I had one and, a bit latter, full lock-up when I was looking | around at Simtel.Net the other day, I wonder if others get that. | Cookies don't seem to be the problem, as these seem to be all less | than 300 bytes ... | | Glenn, have you had a chance to try registering at ArachneDevelopment | since our last discussion here? As I explained earlier, there are | _some_ things I can do to assist you, if you have difficulties. | | | ============================================================================ | | Fri, Feb 13, 2004 - 1:17:16 pm | | ray> | Joe, | | I was thinking the same thing vis. Gregy: hard to believe it's | modem trouble. | | hL7! | | ============================================================================ | Š| Fri, Feb 13, 2004 - 4:39:35 pm | | glennmcc> | Joe, | Xswap crashes at simtel.net are most likely due to... | 1) multiple nested tables (many deep) | 2) java script that is not correctly commented-out | (some sections commented-puit while others are not) | (a page with some out and others in will not even be 'fixed' by using | the ignorejs feature) :(( | | As for yehaw.... nope ... no luck. | I can't evemn get it to send me a new password via email because I | nolonger have the email account that I used to signup for glennmcc as a | user name. (I used my old netzero email account which no longer exists) | | So... screw it.... I'll just have to fire-up Linux Netscape and get a | new username. | | BBL | | ============================================================================ | | Fri, Feb 13, 2004 - 5:55:44 pm | | ray> | All, | | Here is a strange little mystery: Since the boss is still away, I | thought I'd spend a couple of days converting some of the | while() loops in the code into for() loops (where the logic is | identical of course). Anyway, each time I convert a loop, and re-build | inside the IDE, the .exe increases by 16 bytes, which is strange if | the logic is the same, but what is stranger is that if I do a | command line MAKE, this dosen't happen -- the .exe's stay the same. | This is the first time I've ever seen the IDE behave in a different | way that the command line compiler. Any thoughts? | | Glenn, | | So what did you think of the code I sent you? Good? Bad? Ugly? | | | Hey Joe, did you ever crack Christof's lsppp problem? | | | | | ============================================================================ | | Fri, Feb 13, 2004 - 7:02:36 pm | | glennmcc> | Inside of the IDE is the one and ONLY 'right way' to compile. | Š| As to the code you sent........... I'll leave my judgement till I've | looked at it a little more. | | ============================================================================ | | Fri, Feb 13, 2004 - 7:35:35 pm | | ray> | ... I NEEEEEeeeeew you'd say that. | | | ============================================================================ | | Sat, Feb 14, 2004 - 9:57:30 pm | | glennmcc> | Well, since you saw that one coming... I guess we're good to go then, huh? | | You're gonna drop that command line compile crap and start doing it the righ | right ? | | | | | ============================================================================ | | Sun, Feb 15, 2004 - 12:43:35 am | | mht> | I really like mysteries like this while/for issue ;-) | | Ray, do you get different .OBJ when you convert a single loop, or | different .EXE only? Can I have such two .OBJs (if different) for | comparison (ones coming from a possibly smallest C source preferred, | of course). | | | ============================================================================ | | Sun, Feb 15, 2004 - 11:11:30 am | | ray> | Glenn, | | My crime-spree continues! You'll never take me alive. | ... Nothin' to loose since I broke bail over the 4DOS thing. | | | | | | | ============================================================================ | | Sun, Feb 15, 2004 - 1:23:30 pm Š| | ray> | Michal, | | The while/for mystery has gotten even more mysterious because it | has dissapeared! When I started converting I was just comparing | .exe's, not .obj's in the IDE, and each loop changed increased the | length of the .exe by 16 bytes. Then, once I figured out that the | command line 'make' was producing identical .exe's, I stopped using | the IDE altogether ... but when I was finished, I did a final build | using the IDE ... and the difference had vanished! I must have | done something wrong, but I don't know what. Anyway, I've just | sent you the code with for() loops, so you can play with it | yourself if you want to. | | | | ============================================================================ | | Sun, Feb 15, 2004 - 4:11:24 pm | | Joe> | Glenn, | | Yes, Simtel seems worse than any other site for crashing Arachne, it's | the first one to cause me complete lock-up for quite a while. However, | I do get the odd, unpredictable XSWAP crash "once in a blue moon" | elsewhere and was wondering if others get that sometimes. I've decided | to increase the "minimum free memory" of Arachne from 75000 to 80000 | bytes - I hope that helps (Arachne changes the way it allocates XSWAP | stuff depending on how much free memory it has). | | As for your "glennmcc" ID, it's a shame you may need a new one, I'm | sure you'd prefer to keep this. If your ID had your birth date and/or | a "hint question & answer" associated with it, you can e-mail Yahoo | for assistance. Just explain that you no longer remember the password | and that the old e-mail address is no longer valid. With luck, they | will verify that the old e-mail address is no longer valid and, if you | can convince them you really are "glennmcc", they can give you a new | password. Well, at least there's no harm in trying ... | | ============================================================================ | | Sun, Feb 15, 2004 - 4:14:32 pm | | Joe> | Ray, | | That "for/while" thing is interesting. All I can comment is that | apparently it is easier for compilers to optimize "while" loops | than "for" loops ... | | ============================================================================ | | Sun, Feb 15, 2004 - 7:02:07 pm Š| | glennmcc> | No problem Joe. | | Since I was born in 1959......... I am now glennmcc1959@yahoo.com | | | | ============================================================================ | | Sun, Feb 15, 2004 - 7:12:57 pm | | ray> | Well I'll be ... Glenn is just a kid ;-))) | | | | ============================================================================ | | Sun, Feb 15, 2004 - 7:23:50 pm | | glennmcc> | Watch it .... you old fart. | | BTW, | I just created my own directory in the files area. | | And I used Arachne v1.75 to do it. :))) | | ============================================================================ | | Sun, Feb 15, 2004 - 7:26:46 pm | | glennmcc> | But... after 4 crashes in a row.... | | I am never trying to visit my.yahoo.com ever again. | | All of those nested tables within tables within tables within tables | and all of that very badly written and incorrectly commented java script | is more than Arachne can handle. :((( | | | ============================================================================ | | Sun, Feb 15, 2004 - 7:47:04 pm | | ray> | Speaking of old farts, I wish Gregy would check in more often, I | feel isolated from my chain of command when he's away. | | And: | | I've been keeping a log of all the times I get the famous stray | bytes problem in the editor. I'm thinking we should all do the Š| same thing and then compare notes. I've already noticed something | interesting: the number of stray bytes is always 4, and they are | always either all ASCII lower ( < 127 ) or all ASCII upper. If others | have noticed the same thing, it's gotta be a clue to what's going | on, no? | | | | | ============================================================================ | | Tue, Feb 17, 2004 - 9:10:24 pm | | Joe> | Glenn, | | FYI, now that you have a Yahoo ID, you can get the ArachneDevelopment | list messages sent directly to "glennmcc[at]cisnet.com", rather than to | your default Yahoo e-mail address. I know you have currently selected | to view messages via the Yahoo Groups web interface only, but don't | think that's your only alternative option. If you want to do this, you | will have to fire-up Netscape one last time, log-in to Yahoo, click the | "My Account" link, click the "edit" button a little bit down the page, | add your 'cisnet' e-mail address as one of your "alternative" e-mail | addresses and click on the "primary" bullet next to it. It's simpler to | do than to explain, but I hope I've explained it OK. | | BTW, the U.S. "My Yahoo" stuff doesn't seem much different to the Oz | "My Yahoo" page, except we get a "Cathy" cartoon as well. I don't get | any crashes with Arachne acccessing "My Yahoo", FWIW. | | ============================================================================ | | Tue, Feb 17, 2004 - 9:18:16 pm | | glennmcc> | Joe, | | Are you CRAZY ??????? | | I would NEVER do anything so foolish as to give Yahoo access to my main | email address. | | I gave them glennmcc@humanoddity.com | | | ============================================================================ | | Tue, Feb 17, 2004 - 9:38:56 pm | | geezer> | Well so much for that ... I had thought it would be a good idea | to rationalize the header structure of Arachne. arachne.h is a monster | that calls headers that call headers that call headers. It include | many headers that it doesn't 'need', just to pass them in bulk to Š| various .C files. Anyway, I thought I'd remove these included includes | from arachne.h and then include them as needed in whatever .C files | wanted them -- sound like a good idea? Anyway I was having some luck | doing this but the results were the oposite of what I expected: instead | of compile time going down, it went through the roof, and the lines | compiled counter went from 108,000 to 230,000! It's counter | to common sense but it seems to be that the 'pre-compiled header' | setting (which is on) is happier when the number of included files | is smaller, even if the content of those files is larger. | | Live and Learn. | | | ============================================================================ | | Wed, Feb 18, 2004 - 4:24:51 pm | | Joe> | Glenn, | | OK ... if you're paranoid ... | | I don't see how giving Yahoo your e-mail address so you can receive | and *send* list messages via SMTP, is any different to sending any | other mailing list your e-mail address to do so. For me, I really like | the flexibility Yahoo gives me. I can use either their e-mail (web-mail) | system or my own for list messages and/or use the Yahoo Groups web | interface. I can even get them to redirect any mail sent to my Yahoo | web-mail address to my main e-mail address. (To get this particular | feature as a free service however, you have to agree to receive ad-mail | from at least one chosen category. Yet, even though I've agreed to this, | they have never sent me a single ad-mail:-). They don't send me junk, | sell my e-mail address to others, or any of the other nonsense some | people like to tell you about Yahoo. They even answer e-mails I send | them, which is a courtesy most other companies don't bother with. | | Anyway, I just thought I'd mention that mailing list option, in case | you weren't aware of it, it's a great convenience without any "strings | attached". | | ============================================================================ | | Wed, Feb 18, 2004 - 4:30:38 pm | | Joe> | Geezer, | | Interesting. Modularity isn't one of C's strong points (although C | coders seem blissfully unaware of this). I guess this is just another | manifestation of this fact. | | | ============================================================================ | | Wed, Feb 18, 2004 - 11:05:39 pm Š| | ray> | Joe, | | I can vouch for what you say about Yahoo; I don't really like the | outfit, and there are those sign in irritations, but I belong to | several of their groups and they have never sent me any garbage | either. | | And I've started to figure out this header thing -- you are right: | C doesn't seem to behave sensibly on this point, but I've made | quite a bit of progress: got my 'build all' down to 84,000 lines | in 26.4 seconds, from 108,000 lines in 33 seconds with the original | code, so that's progress. It's as I figured -- Borland wants as few | headers as possible in each file, but it doesn't care how big those | headers are (long live 'arachne.h' ;-) and it does much better if | the headers are always in the same order, if possible. | | | | ============================================================================ | | Wed, Feb 18, 2004 - 11:10:35 pm | | ray> | Almost forgot: | | Joe, | | I've got this irritating little problem with Borland: When I'm in | the IDE, I set the output directory to my ramdisk, and all's well, | but even after using 'prj2mak', on the command line, 'make' | can't find the arachne libraries unless I copy them to the output | directory on my ramdisk ... not a big deal, but it would be nice | to fix. I've tried tinkering with my .mak file, but with no luck. | Any ideas? | | | ============================================================================ | | Thu, Feb 19, 2004 - 4:16:52 pm | | mht> | Joe, | what is placed after the -L option in the copy of .cfg file placed | at the end of your .mak file? Have you tried to change this? | | | | ============================================================================ | | Thu, Feb 19, 2004 - 6:47:22 pm | | glennmcc> | I got problems. :( Š| | Some how I screwed-up my /etc/XF86Config file | and now I can't startx :(( | | Anybody got a working copy of XF86Config for a | Sis315 AGP vid card and Slackware v7 | | I start Linux in 1024x768 framebuffer consoll mode | and none of the configs I have tried will work. | | Keeps screwing-up my monitor and all that happens | is the little box bouncing around with the message | that the horiz sync was out of range. :((( | | No option at that point but Cntrl+Alt+backspace | ant then reboot the system. | | ============================================================================ | | Thu, Feb 19, 2004 - 9:34:28 pm | | Joe> | Ray, Michal, | | I'll check it out this weekend and report back ... | | Glenn, | | Sorry, my video card's a Cirrus Logic PCI card and my Linux is | Mandrake and I'm using an old version of X, so my XF86Config | file won't be much use to you. Try reducing your vertical refresh | rate, this will correspondingly reduce your horizontal sync rate. | | ============================================================================ | | Thu, Feb 19, 2004 - 9:56:41 pm | | glennmcc> | Thanks Joe, I'll give that a try. | | | | ============================================================================ | | Thu, Feb 19, 2004 - 9:56:51 pm | | ray> | Glenn, | | Looks like a fresh install. | | ... down to 67,000 lines of code for a 'build all'. | | | ============================================================================ Š| | Thu, Feb 19, 2004 - 10:47:00 pm | | glennmcc> | Joe, | | I got her back up and running. | | I'm here with X and Linux NS right now. | | Tried reduceing vert refresh but that did not do the trick. | | All I had to do was delete XF86Config and since I run the consoll in framebu | mode....... | the framebuffer Xserver is now being used automatically. :))) | | ============================================================================ | | Thu, Feb 19, 2004 - 10:48:36 pm | | glennmcc> | Ray, | | Is that from the command line ? | | Or from inside the IDE ? | | | ============================================================================ | | Fri, Feb 20, 2004 - 2:47:04 pm | | ray> | Glenn, | | ... IDE. As far as I know the command line compiler dosen't | give out that information. OTOH, to do accurate build-time measurements | I use a batch file on the command line that does this: | | timer on | make -fcore.mak -B | timer off | | | | ============================================================================ | | Fri, Feb 20, 2004 - 9:19:33 pm | | ray> | EEEEE-HAAAA ... 65,884 lines in 25 seconds. | | But that's probably as good as she'll get. | | Š| ============================================================================ | | Fri, Feb 20, 2004 - 9:26:31 pm | | gregy> | I'm back....!!!....I think (Lord Willing) permanently........ | I might mention a few (of the multitude) of problems I ran into, but | the primary one was my stumble-bummish incompetence........it's | really humbling to see yourself doing things like I've done in the past | three weeks... | ... | ... | NOTE - I will start on the CutPof Christof's translations this | weekend...has Christof finished....??? | ... | RAY....I will email you for another set of your cleaned up BASE1 | source files....since I am sure I lost those about 3 weeks ago, when | this whole mess started....BUT...don't send them until I email you for | them.....thanks... | ... | I'll post more tomorrow....right now...I just wanna get this whole mess | back together and go to bed..... | ... | | | | ============================================================================ | | Fri, Feb 20, 2004 - 9:27:52 pm | | gregy> | hmmmmm....can't type, either........"...CutPof Christof's | translations...."..... | ... | | | ============================================================================ | | Fri, Feb 20, 2004 - 9:28:47 pm | | gregy> | hmmmm....that's the PHP doing that...."Cut and Paste".... | | | ============================================================================ | | Sat, Feb 21, 2004 - 11:56:02 am | | ray> | 53,972 lines in 24.8 seconds ... | | Let's see now ... at about 10,000 lines per day, I sould have Arachne | down to 0 lines in 0 seconds in about 6 days -- that's what I call | tight, fast code. | Š| | | | ============================================================================ | | Sat, Feb 21, 2004 - 2:25:12 pm | | mht> | Almost any code can be optimized and made one byte shorter. | (I noticed this several times when I had to stuff new features into | a very limited space, like a boot sector.) | Applying this 'infinite code shrinkability theorem' sufficiently many | times, you get zero. Obviously false, but funny... | | | ============================================================================ | | Sat, Feb 21, 2004 - 4:22:56 pm | | ray> | Michal, | | Thanks for liking my joke . | è¹Úƒ! | But seriously, I think it is good to optimize the header structure | as much as possible. It looks like in the originals, 1/2 of the | lines compiled are due to the compiler having to re-do the headers | so often. The project I'm working on now has reduced compile time | by 25% due to better use of 'pre-compiled headers'. | | | ============================================================================ | | Sun, Feb 22, 2004 - 3:54:21 pm | | Joe> | Ray, | | I checked that "output directory" issue and it's like this : | The output directory specifies where the *.OBJ files are written (also | *.EXE ...), so this is where the linker expects to find them after the | compiler pass is finished. However, *.LIB files are a collection of | *.OBJ files, so the linker also expects to find the *.LIB files in the | same "output directory", even though these are pre-compiled object | files, not the result of the compilation pass. That's what seems to | be happening when you use 'make', just a little bug I suppose. | | | ============================================================================ | | Sun, Feb 22, 2004 - 3:59:16 pm | | Joe> | Michal, | Š| The "-L" option in 'nokey.cfg' depends on which version of the compiler | I am using at the time, and corresponds to the "LIBPATH = ..." line | near the top of the makefile. Since the 'nokey.cfg' file is created | each time you run 'make', there is no need to manually place anything | in it, like the "-L" option or whatever. | | ============================================================================ | | Sun, Feb 22, 2004 - 5:16:02 pm | | ray> | Joe, | | Thanks, that makes some sense; strange that the IDE doesn't have the | same problem. Like you said, a wee bug. | | | ============================================================================ | | Sun, Feb 22, 2004 - 7:12:05 pm | | ray> | ... 46,034 lines in 23.95 seconds ... that's as good as she gets, but | it's still a 26% improvement in 'buld all' time over the original code. | | | | ============================================================================ | | Sun, Feb 22, 2004 - 9:06:52 pm | | ray> | Hey Gregy, you haven't dissapeared again, have you? | | | | ============================================================================ | | Mon, Feb 23, 2004 - 2:44:22 am | | gregy> | nope...just getting reorganized again.......has Christof finished | his translations....??? | | | ============================================================================ | | Mon, Feb 23, 2004 - 6:52:06 am | | ray> | I dunno, you're the coordinator ;-) | | | | Š| ============================================================================ | | Sat, Feb 28, 2004 - 7:15:03 am | | gregy> | sheesh...talk about the blind leading the blind...I haven't been doing | very good coordination lately...... | ... | I just emailed him, today...I'll know this weekend | ... | | | ============================================================================ | | Sat, Feb 28, 2004 - 7:16:55 am | | gregy> | Ray....okay...send me your finished(?) cleaned-up BASE1 files.... | ... | Please email-attach them in two or three groups, so the emails don't | get too big and take too long to d/l.... | ... | I'll look for them as soon as I read your notice on here... | ... | | | ============================================================================ | | Sat, Feb 28, 2004 - 9:28:14 am | | ray> | You're behind the times! ... base1a is done, and I'm almost finished | base 1b (optimized header structure and combining of some files for | faster compile). | | I'll send you base1a in 3 different zips: english.zip czech.zip and | notes.zip. | | | | | | ============================================================================ | | Sun, Feb 29, 2004 - 7:56:44 pm | | gregy> | he he...well...I kinda "said" that.....I was, at least, 4 weeks "behind | times"...... | ... | I got them...I haven't looked at them...but...I think I understand | that "english" is files that have no Czech comments...??? | that "czech" is files that do have Czech comments....??? | that "notes" is your notes about what you did to clean-up the files...?? | Š| about 308Kbytes, 83Kbytes, and 22Kbytes... | | | ============================================================================ | | Sun, Feb 29, 2004 - 7:59:20 pm | | gregy> | Ray....and....I think that what you are now calling BASE1b is what I | was thinking of as BASE2....since you are making significant changes | to the layouts/composition of the files....??? | .... | We can put the translations into them, also...but...I think the first | translations should go into the BASE1a code, which is the code that | is closest to the legacy BASE0 code we got thru GLP..... | ... | Do you all agree....??? | ... | | ============================================================================ | | Sun, Feb 29, 2004 - 8:00:14 pm | | ray> | Yup, that's it. BTW Gregy, before you start to work on the | translations, let me know -- I've got a few suggestions for you. | | | | ============================================================================ | | Sun, Feb 29, 2004 - 8:03:54 pm | | gregy> | Now....I've just gotten back online, as I said.....and....then my ISP | got hit with a REAL bad DDoS attack, that blocked all their servers, | clogged the email servers completely, and even overloaded the WWW | DNS servers assigned to our domains....so....they've gotten it pretty | much straightened out, now...except one of the DNS servers has been | taken offline permanently....(complete switchout)...and I can not | access the web.....but...my POP3 still responds to requests very very | slowly (takes for ever for it to start the POP3 download)...but...it | downloads fast once it starts.... | .. | However....things are "possible" again...... | ... | | | ============================================================================ | | Sun, Feb 29, 2004 - 8:04:50 pm | | gregy> | hey...Ray........ | ... Š| Yeah..!!!...any suggestions are very welcome... | ... | go ahead... | | | ============================================================================ | | Sun, Feb 29, 2004 - 8:06:37 pm | | ray> | Oooops, you wrote just as I posted. | | I'd like the others to look at 'base1b' (now that you're back, you | can take charge of the naming) ... the changes don't affect the | function of the code at all, just speed up the compile and | move some functions around wich reduces the number of files. I like | it and if the others do to, then I'd say the translations should go | into the most recent files. | | | | ============================================================================ | | Sun, Feb 29, 2004 - 8:07:27 pm | | gregy> | ooops...two posts back...."I can NOW access the web"..... | | | ============================================================================ | | Sun, Feb 29, 2004 - 8:09:12 pm | | ray> | I'll talk to you privately. (no secrets, just to save traffic here). | | | ============================================================================ | | Sun, Feb 29, 2004 - 8:09:21 pm | | gregy> | okay...read my post...we can put the translations into both BASE1a | and BASE1b (which I think we should call BASE2..???).... | ... | After that...the translations will be "in all the code"....so we will | have them whereever we go... | ... | agreed...?? | | | | ============================================================================ | | Sun, Feb 29, 2004 - 8:13:34 pm Š| | gregy> | ah...come over to | http://www.hi-line.net/brd/misc/mindex.php | | | ============================================================================ | | Sun, Feb 29, 2004 - 8:15:47 pm | | gregy> | ooops, again....... | http://www.hi-line.net/~gfeig/brd/misc/mindex.php | | | ============================================================================ | | Sun, Feb 29, 2004 - 8:24:16 pm | | gregy> | as I said...my ISPs email servers are still responding VERRRYYYY slow, | and sometimes timeout before they decide to give me the POP3 d/l... | so..if you send me an email...you'd better post me here, to let me | know it's there...so I can go up on their webemail server, and read | it online... | ...or...go over to my MISC board and we can chat there.. | | | | ============================================================================ | | Mon, Mar 01, 2004 - 12:07:51 am | | Joe> | Yup, I agree with Greggy, the re-arranged stuff (Ray) should be called | Base2 ... | | ============================================================================ | | Mon, Mar 01, 2004 - 12:47:35 am | | gregy> | well...let's discuss this........we wouldn't wanna jump in too | hastily..... | ... | I will try to start pasting Christof's translations into Ray's BASE1a | that he sent me....try to start that by tomorrow... | ... | BTW...once we get things sort of stabilized (i.e. translations into | both BASE1 and BASE2)...I suggest that we just ship around (email | attachments, server uploads, etc) JUST the deltas (I mean only the | files that have changes in them.....).... | .... | For new, numbered releases, we can put all the source in a package, | for download, but for stuff done in the interim....we should just Š| post the deltas...and everyone already has the other, unchanged, files | ... | | | ============================================================================ | | Mon, Mar 01, 2004 - 5:41:48 pm | | gregy> | Christof has finished about 350Kbytes of files...he says he will get | the rest as soon as he can....maybe not for another month.....but.. | I've got plenty to do, with just those...... | ... | I notice Ray's files seem to be all CR/LF line terminations, and | Christof's also, while the originals from Michael are all have just | the unix-stile ^J line termination....so...I guess most compilers | (including BCC3.51) are "line terminator agnostic"....so...I won't | hafta worry about setting up my OS/2 EPM ProgrammingEditor to do | anything special....however....I'm pretty sure I can set it up to | save in "all one style"....and so I can convert from any style to | another. I also have DOS utilities that can do this....but...I | think is makes no difference on most compilers... | | | | ============================================================================ | | Mon, Mar 01, 2004 - 7:51:23 pm | | ray> | All, | | Gregy, you're the coordinator so whatever you name things is fine with | me (Base1b == Base2). | | I still think it would save time to translate directly into BASE2, but | I won't bitch about it. | | Do you have/use any GREP program? If not may I recomend MINITERM! | | Do you have or can you get a 2nd, monochrome monitor and card? If so, | I could set you up with my dual monitor TSR which is very handy for | the translation moving work because you can look at both copies of the | files (Christoff's and Base 1a /2) full screen at the same time. | | | ============================================================================ | | Mon, Mar 01, 2004 - 7:53:39 pm | | ray> | Joe, | | Off topic ... I hear the weather down is OZ is bloody awful. | Š| | ============================================================================ | | Mon, Mar 01, 2004 - 8:45:00 pm | | ray> | Gregy, | | If I didn't mention it already Ron Clarke has volunteered to help you. | And maybe Hans-Georg too, if you twist his arm ;-) | | | | | ============================================================================ | | Mon, Mar 01, 2004 - 10:56:17 pm | | Joe> | Greggy, | | Although some compilers are EOL agnostic, the same can't be said for | all the various tools I use. The first thing I do when I import a new | set of code is convert everything to CR/LF. This nonsense all comes | about due to the FTP text vs. binary mode stuff. | | | ============================================================================ | | Mon, Mar 01, 2004 - 11:06:47 pm | | Joe> | Ray, | | I've heard you mention MINITRUE once or twice, so I expect this is | what you meant, rather than MINITERM. Some day I'll have to see what | this MINITRUE utility is all about ... | | As for the weather, it's dropped maybe ten degrees in Sydney as of | a week or so back, but otherwise is quite OK. There were some severe | storms in Queensland earlier and there's a cyclone in Western Australia | at the moment, so I expect that's what you may be referring to ... | Fortunately, Sydney is far away from these events. | | | ============================================================================ | | Tue, Mar 02, 2004 - 1:55:53 pm | | ray> | Joe, | | I'd heard that there's been a long term drought -- very severe. | | MINITRUE is an open source GREP clone. VERY nice IMHO, but it's the Š| only one I've ever used, so my opinion is not worth too much. Still, | it is amazing what it can do. | | | | ============================================================================ | | Tue, Mar 02, 2004 - 1:57:30 pm | | ray> | Ignorance is bliss (sometimes!) ... I'd never even given the | new line problem a thought. | | | ============================================================================ | | Tue, Mar 02, 2004 - 4:08:20 pm | | Joe> | I also have the original (?) GREP source code. I compiles OK under BC | with minor changes. It has advantages and disadvantages over Borland's | implementation, I mainly use Borland's because it supports filename | wildcards and because of habit. | | Yep, there's been a severe drought for about four years, the last two | years have been particularly bad, although the last few months have | seen some relief. Agricultural output in some sectors has been very | badly affected and consequently the current account balance, although | the rising A$ and the world economic slump has also played its part. | | ============================================================================ | | Wed, Mar 03, 2004 - 3:13:11 pm | | ray> | BASE2 ready. | | | ============================================================================ | | Wed, Mar 03, 2004 - 3:46:39 pm | | glennmcc> | OK, kewl. | | Where can we go to download it ? | | ============================================================================ | | Wed, Mar 03, 2004 - 5:21:44 pm | | ray> | I guess Gregy or Joe will upload it. I'll wait to hear from one or | both of them. | Š| BTW, Joe, don't give up on me vis. Lynx -- I'm gonna start working with | it sortly, but for the next few days I'm going to be too busy to | do anything computer-wise at all. | | | ============================================================================ | | Wed, Mar 03, 2004 - 9:51:00 pm | | gregy> | Ray, et al.....there's no reason why we can't paste into BASE2, after | we paste into BASE1....... | ... | ...and...Christof hasn't finished all the files, yet....I've got about | 350K of them....but....I'm not gonna wait for more.....he'll prolly | have them done by the time I get done with bASE1.... | | | | ============================================================================ | | Wed, Mar 03, 2004 - 10:01:52 pm | | gregy> | Ray....I have other GREP-type utilities...especially in OS/2...but.. | I'll take you up, SOMEDAY, on shipping me that MINITRUE....but..not | now.......I've so many things to think about, that I can't | keep the current ones straight..... | .. | Joe...yes...I always like to run a CR/LF util on all my source files | ....makes them listable in ALL my editors... | ... | Ray.....RE: Dual-monitors....I have that capability in my OS/2 (in | fact, I'm now snooping around for a dual-head VGA card)...but..REALLY | ...when I paste....I like to have two windows side-by-side (even in | DOS, I run two DOS windowed sessions)...with the "pastee" on the left | and the "paster" on the right.....then I keep the files line-aligned | as I paste down thru..... | ..hmmm..hope I haven't confuzzeled you.......anyway...this is | the way I done it for years....and it sorta seems most automagic for | my way of doing things... | .. | | | ============================================================================ | | Wed, Mar 03, 2004 - 10:04:32 pm | | ray> | Gregy, | | ... sure there is! ... your time! ... We've gotta think of a way | to do it the second time automagicaly. Joe, would VC be the right | tool? | Š| | ============================================================================ | | Wed, Mar 03, 2004 - 10:04:43 pm | | gregy> | oh...BTW...I discovered sompin t'other day....when you view a ^J | line-terminated source file in Arachne, then save it to another file.. | ...hey...!!!.....voila...!!!....Arachne automagically puts CR/LFs in | there..... | | | | ============================================================================ | | Wed, Mar 03, 2004 - 10:06:06 pm | | gregy> | hmmmm....Ray..???.........how do you mean...."your time".. | ...?? | | | ============================================================================ | | Wed, Mar 03, 2004 - 10:07:22 pm | | gregy> | ..and...for pasting into BASE2...I was kinda thinking of seeing if | Ron might do some of that.... | ... | | | ============================================================================ | | Wed, Mar 03, 2004 - 10:10:40 pm | | ray> | Gregy, | | OS/2 freak eh? ... Never used it myself, but a professor friend of mine | says that he is still angry to this day over the way IBM let it | die -- says it was/is wonderful. | | Anyway, whatever works for you -- sounds like you've got it | well in hand ( ... Sheesh, maybe he could give *me* some pointers ;-) | | | | ============================================================================ | | Wed, Mar 03, 2004 - 10:12:25 pm | | ray> | By 'your time' I mean *your* time ... don't want to see you have to | do it twice, is all. Š| | | ============================================================================ | | Wed, Mar 03, 2004 - 10:13:57 pm | | gregy> | OH..!!!....and...I was thinking as a mathematician (one of my college | majors)..again....about the "file-consolidation" project you're doing | in BASE2(BASE1b)..... | ..math-ters ALWAYS think in terms of "limits"...these things are the | ridiculous end points of any problem... | ... | With this problem....our limits are: | BOTTOM===one file for each line of code | MAXIMUM===all lines of code in one humongous file | | We now have somewhere in between those.....ahhhhhh....when you go to | consolidate files....keep in minds the "changeablity" factor for | ArachneCORE.....we don't wanna consolidate files so that we can't | easily find broken-apart functions for new changes.... | .... | ...I mean....maybe we should be careful that we don't get ourselves | in such a "consolidated" condition that we have to do "step-wise | decomposition" all over again, just to make some changes to ONE small | function..... | | | ============================================================================ | | Wed, Mar 03, 2004 - 10:22:12 pm | | gregy> | he he...well...as you can see...I was gonna try to sneak in some | "delegation" functions....by seeing if I can take Ron up on his kind | offer.... | ... | Anyway...I'm not gonna panic...I'm just keepin on keepin on.....and | it'll get done as soon as it gets done.... | ... | And....sometimes "doin it twice" can result in some improvements the | second time around.... | ... | RE:...your professor...."dead OS/2"...???....I hope to tell you that | IBM hasn't "died it"...yet....we're still getting updates, new drivers, | etc from them....and they HAVE licensed SerenitySystems to continue | upgrading it, and marketing it as "eCommstation"....(eCs1.1 is the | current version).....however...it's not a "poor man's OS"....all the | new stuff is between 100-200 bucks a version (upgrades are cheaper).. | ....but....IBM did stop "shrinkwrapped" sales (i.e. retail sales)... | ...all they offer now is their SoftwareChoice (which my 2-year sub | runs out this April).... | ... | What has happened, due to this IBM no-shrinkwrap, is that the largest | retailer of IBM OS/2 is now eBAy........ Š| ..he he.. | | | | ============================================================================ | | Wed, Mar 03, 2004 - 10:24:08 pm | | gregy> | ...re-reseller...... | | | ============================================================================ | | Thu, Mar 04, 2004 - 12:30:49 am | | Joe> | Ray, | | Well ... remember our very early discussions about limiting the | scope of changes so that file comparisons would remain feasible? | The consensus was that this wasn't a priority, so in effect, the | ability to use VC to selectively incorporate changes is diminished. | At this stage, only you can judge to what extent this is so, with the | Base2 stuff. Generally, VC is an extremely good tool for selectively | incorporating changes. You can simply Alt-Next or Alt-Previous to any | difference and delete it. When you have the desired mixture of "old" | and "new" stuff (you can't delete common stuff), you can save it to | file by pressing F2. Have a play ... I'm sure you'lll find this a very | powerful capability, although only practical when file differences are | not excessive. | | ============================================================================ | | Thu, Mar 04, 2004 - 5:58:07 pm | | ray> | Gregy, | | Math major eh? ... now I know that you and I have two languages in | common. | | | ..math-ters ALWAYS think in terms of "limits"...these things are the | | ridiculous end points of any problem... | | ... | | With this problem....our limits are: | | BOTTOM===one file for each line of code | | MAXIMUM===all lines of code in one humongous file | | | | We now have somewhere in between those.....ahhhhhh....when you go to | | consolidate files....keep in minds the "changeablity" factor for | | ArachneCORE.....we don't wanna consolidate files so that we can't | | easily find broken-apart functions for new changes.... | | .... | | ...I mean....maybe we should be careful that we don't get ourselves Š| | in such a "consolidated" condition that we have to do "step-wise | | decomposition" all over again, just to make some changes to ONE | small | | function..... | | Yes, sober thoughts. The diminishing returns curve is very steep for | this, | so I've only done the most glaringly obvious ones (12); the rest will | have to wait for a deeper understanding. The optimum result curve | will | have a rather large flat spot near it's apex, but I'm estimating that | eventually it would be ideal to have something in the order of 50 | files | in the project, though anything from say 30 - 70 should be acceptable. | | | | And....sometimes "doin it twice" can result in some improvements the | | second time around.... | | Yeah, but there's the value of your time to consider. With what Joe | says below, I'm thinking that if you translate into base1a, it should | be a piece of cake to import the translations into base2 using VC. | I'd use Ron to help you with base1a -- let's see if I can automate | the step to base2. | | | RE:...your professor...."dead OS/2"...???....I hope to tell you that | | IBM hasn't "died it"...yet....we're still getting updates, new | | drivers, | | etc from them....and they HAVE licensed SerenitySystems to continue | | upgrading it, and marketing it as "eCommstation"....(eCs1.1 is the | | current version).....however...it's not a "poor man's OS"....all the | | new stuff is between 100-200 bucks a version (upgrades are | cheaper).. | | ....but....IBM did stop "shrinkwrapped" sales (i.e. retail sales)... | | ...all they offer now is their SoftwareChoice (which my 2-year sub | | runs out this April).... | | That is very interesting! I had no idea that OS/2 was still kicking | -- | gotta tell Boyd. | | | | Joe> | | Ray, | | | | Well ... remember our very early discussions about limiting the | | scope of changes so that file comparisons would remain feasible? | | The consensus was that this wasn't a priority, so in effect, the | | ability to use VC to selectively incorporate changes is diminished. | | At this stage, only you can judge to what extent this is so, with | the | | Base2 stuff. Generally, VC is an extremely good tool for selectively | | incorporating changes. You can simply Alt-Next or Alt-Previous to | any | | difference and delete it. When you have the desired mixture of "old" Š| | and "new" stuff (you can't delete common stuff), you can save it to | | file by pressing F2. Have a play ... I'm sure you'll find this a | very | | powerful capability, although only practical when file differences | are | | not excessive. | | | Yes, with the first clean making sometimes dozens of changes per line, | using VC would be hopeless, but from base1a on, I'm thinking VC will | be | a God-send, since the changes will tend to be localized (with base2 | it's | just different headers and some blocks of code moved into different | files). | I'm thinking that using it to incorporate the translations should be | a snap. Will start fooling with it next week. | | | | ============================================================================ | | Thu, Mar 04, 2004 - 5:59:56 pm | | ray> | Joe, Gregy, | | Does either of you want me to sent you base2? Glenn want's a look. | Or should I send it to him directly? | | ============================================================================ | | Thu, Mar 04, 2004 - 8:38:21 pm | | Joe> | Ray, | | You can try e-mailing them to me and I'll upload them to | ArachneDevelopment. However, I should warn you our mail | server is having trouble ATM, so it's possible e-mail may | not reach me until IT finish sorting things out. | | ============================================================================ | | Thu, Mar 04, 2004 - 9:25:09 pm | | ray> | Joe, | | OK, I'll send them tomorow. | | | | ============================================================================ | | Sat, Mar 06, 2004 - 5:14:29 am Š| | gregy> | Ray....do you have BASE2 (Base1b..??) already done...??? | ... | If so...please email-attach them to me in the same chunks as last | time... | .. | NOTE - This weekend I will set up the directories on the new Community | Homesite...ArachneVirtualAve....and put the legacy code up there, the | legacy translations, and whatever BASE2 (1b..??) that Ray sends me. | .... | QUESTION - I have INSIGHT and WWWMAN sources....I think I also have | the XLOPIF library source (glennmcc, didn't you send me that..??) | ...so...are there translations to be done in those two/three sets of | files.....??? | ... | ALSO - Michal did some tweaking/cleaning in WWWMAN, didn't he...??? | Where is that source....??? | ... | I want to get all these legacy sources together on the AVA site, so | we can just point people to them. Then I want to put the BASE2 up, | then the BASE3 (if required).....then start putting the new distros | in another directory(s)....for people to download. We have 150Mbytes | to use, there, there's no pinch like there is on this site...(my own | personal, with only 10Mbytes..)... | | | ============================================================================ | | Sat, Mar 06, 2004 - 8:30:05 am | | ray> | Gregy, Joe, | | Will send BASE2 later today. | | | ============================================================================ | | Sat, Mar 06, 2004 - 12:12:32 pm | | ray> | Joe, Gregy, | | Just sent it. One big zip for Joe, 4 smaller ones for Gregy. | | Looking forward to feedback, especialy if you guys have the same | improvement in build speed as I did (25%). | | | ============================================================================ | | Sat, Mar 06, 2004 - 2:36:25 pm | | glennmcc> Š| Gregy, | | IIRC, | Yes... I sent the Xlopif SRC code to you. | (if not, let me know and I will) | | As for wwwman and insight. | | Michal's latest changes to wwwman are in the SRC code 'package' on my | site. | | The most recent changes that I made to insight are on my site as-well. | | ============================================================================ | | Sat, Mar 06, 2004 - 3:55:05 pm | | ray> | Shoulda said so before Glenn, I had no idea that either you or Michal | had new stuff posted. | | | ============================================================================ | | Sat, Mar 06, 2004 - 4:47:20 pm | | glennmcc> | It's not 'new' per-say. | | The changes that we made last Nov are the newest in both wwwman and | insight. | | | ============================================================================ | | Sat, Mar 06, 2004 - 7:48:01 pm | | gregy> | thanks...glennmcc...I thought that's what they were....sort of small | "cleanup" of bugs, etc in the then-existing programs... | ... | Oh...BTW.....wasn't it November 2-3, 2003....when Michael turned the | source out to GPL...??? | ... | I'm asking because I'm hacking together a small addendum to "Arachne | History" to post the whole history on AVA... | .. | | | ============================================================================ | | Sat, Mar 06, 2004 - 10:24:11 pm | | ray> | Gregy, Joe, Š| | Burn that garbage I sent you! ... Wrong stuff. Will send the | right stuff tomorrow or Monday. Sorry about that, I need a rest, | I'm getting sloppy. | | | ============================================================================ | | Sun, Mar 07, 2004 - 3:01:21 pm | | glennmcc> | Here you go Gregy, | | It was Nov 8, 2003 | | http://www.cisnet.com/glennmcc/ar173gpl/ar_gpl-1.txt | http://www.cisnet.com/glennmcc/ar173gpl/ar_gpl-2.txt | http://www.cisnet.com/glennmcc/ar173gpl/ar_gpl-3.txt | http://www.cisnet.com/glennmcc/ar173gpl/ar_gpl-4.txt | http://www.cisnet.com/glennmcc/ar173gpl/ar_gpl-5.txt | http://www.cisnet.com/glennmcc/ar173gpl/ar_gpl-6.txt | | | | ============================================================================ | | Sun, Mar 07, 2004 - 4:16:30 pm | | Joe> | Ray, | | No problemo ... burnt! Relax, get some sunshine | | ============================================================================ | | Sun, Mar 07, 2004 - 5:29:08 pm | | ray> | Thanks Joe, | | I need to spend some quality time with my cello. Once I get | base2 uploaded I'm taking a break for sure. | | | ============================================================================ | | Tue, Mar 09, 2004 - 8:08:36 am | | gregy> | Ray....I just got four(4) email-attachments....were those base2....?? | ... | | | | ============================================================================ Š| | Tue, Mar 09, 2004 - 8:19:14 am | | gregy> | glennmcc.....thanks....I d/led them and they will be preserved for | posterity as "deathless prose"...... | ... | I also, just d/led everything off your /ar173gpl/page.....and I also | ripped-off your .HTM page, itself.... | .. | The .HTM page will be slightly modified with the AVA URLs, and a link | at the bottom will reference your original /ar173gpl/page, and also | Baastian's ar173gpl page. | | The files will be filed in /SOURCE/... /VERSIONS/... /ALPHAS/... | /ETC/subdirectories on AVA.... | .... | I am setting that up, to move most of interim stuff off this site | onto the AVA site.... | ... | We can still maintain working directories here, but we have lots more | space (150Mbytes) there. | | GLENNMCC....also...can you give me a link to the LOPIF graphics | library | source...?? And any other sources for things we distribute with | Arachne.....like Miniterm and Bernies APMmanager, modified EPPPD, etc. | ... | I know that some of these have no source....and we can "fake" it by | just disassembling them, and publishing that...(no one said EVERYTHING | GPL must be in C language.........)... | ... | ...and stuff like jpIRC I have Jim Patterson's TurboPascal source, | which he released to PC (it's on his site)....I think we're covered | as long as we reference things in a "Credits Page"....... | | | ============================================================================ | | Tue, Mar 09, 2004 - 5:04:05 pm | | glennmcc> | Gregy, | | re:..........and I also ripped-off your .HTM page, itself.... | | | Grab a fresh copy of the just now updated D/L page. | http://www.cisnet.com/glennmcc/ar173gpl/index.htm | | It now includes the link to 'lopif.zip' | | ============================================================================ | | Tue, Mar 09, 2004 - 5:14:55 pm Š| | glennmcc> | Ok, scratch that. | | If you just went and got a new one..... go get it again. | | It now has a link to apm.zip | | Sorry... that's all. (no gots miniterm SRC) :( | | | ============================================================================ | | Tue, Mar 09, 2004 - 5:19:12 pm | | glennmcc> | Oh........ | | Remember to remove the code at the bottom for my page hit counter. | | ============================================================================ | | Wed, Mar 10, 2004 - 7:48:18 pm | | gregy> | Thanks for the reminder....we wouldn't want hits on that page posted | over at AVA to show up on your counter........actually...what | I want it for is to post it kinda "undercover" unlisted behind the | other files....and only put a link up to it if yoursite, Bear'ssite, | and Baastian'ssite all go down..... | ....I want to keep MAIN DOWNLOAD sites as Bear's and Baastian's.... | your site is backup for theirs...and the ArachneCommunitySite on AVA | will be a "backup backup"..... | | | ============================================================================ | | Wed, Mar 10, 2004 - 7:53:49 pm | | gregy> | Got LOPIF and APM....thanks..... | ... | Now, these are officially (I just said so...!!!.......) the | BASE0 sources for those two sets of files....and I will immediately | upload them into the AVA BASE0 subdirectory... | .. | I will post the dirs of that area over there, perhaps later tonight.. | ... | FWIW...the AVA site will NOT give an authorized listing of any | directory that is accessed simply by /DIRECTORY/(without INDEX.HTML) | ...so...I need to make HTML dirs to put in there as INDEX.HTML.... | ...but....I'm not worrying about that at this moment... | .... | As for MINITERM.....I'll disassembly it, sometime in the future, and | we can upload/post that as its source.. Š| ... | | | | ============================================================================ | | Wed, Mar 10, 2004 - 9:01:43 pm | | glennmcc> | IMO, | | Miniterm should be 'ditched' in favor of Lsppp | | BTW, | I can't even use miniterm on this new machine. | The PCI modem does not have a comport # | | And since miniterm requires a comport # to send the dial string | to...... | | I must use Lsppp so that I can send the info to the | base port address and IRQ which are used by the PCI modem. | | | This is how I dial now...... via IRQ 4 and base port address bx800 | | | [profile] | Profile 4234005L.ACF | Connection @lsppp.exe /?>NUL | Hangup @lsppp.exe /u>NUL | ;Hangup @termin.com 0x60>NUL | | [dialer] | DialPage file:pppglenn.htm | | Dialer @lh lsppp.exe /x /i:4 /b:bx800 /d:330-423-4005 /B:115200 /V:60 | /U:glennmcc@cisnet.com /P:*****>PPP.LOG | TerminalWindow @NUL | UseTerminal No | | | ============================================================================ | | Thu, Mar 11, 2004 - 1:43:29 am | | Joe> | Eh???? | | Miniterm does NOT require a port number! | | In Arachne, click the "PPP setup" button for the internal dialler, | then click the "non-standard port" bullet, then fill-in whatever | address and IRQ are appropriate. That's it. What this does in | Arachne.Cfg is set the Port number to 0, with the Base and IRQ Š| settings as selected. | | As I have indicated before, not everyone's modem works with the | LSPPP dialler, so switching to this for dialling is just going | to cause us headaches. Not to mention the terminal window for | debugging ... | | ============================================================================ | | Fri, Mar 12, 2004 - 5:39:23 pm | | glennmcc> | Sorry... it does not work. | | This the setup I just now tried. | | [profile] | Profile 4234005m.ACF | Connection @epppd.exe>>PPP.LOG | Hangup @termin.com 0x60>NUL | | [dialer] | DialPage file:pppglenn.htm | | Dialer @MINITERM.EXE>PPP.LOG | TerminalWindow @MINITERM.EXE | UseTerminal No | Port 0 | Irq 4 | Base bx800 | Mode 8N1 | Speed 115200 | | | --- this is the resulting ppp.log --- | | MINITERM 1.65 - external dialer for PPP/SLIP drivers. (c)2000 Arachne | Labs. | * Reading configuration file L:\ARACHNE\ARACHNE.CFG | | Error when communicating with port. | MINITERM failed to connect, exiting. | | * Leaving MINITERM and trying SLIP/PPP. Press + to abort. | PPPD status: IP address set to | _____________ | | | As you see.... Miniterm can't seem to handle port address bx800 | | | | ============================================================================ | | Fri, Mar 12, 2004 - 5:49:24 pm Š| | | According to Joe DaSilva, | | Not one person who followed the instructions in his Lspppcfg 'package' | has had it fail to work. | ______________________________________________________________________ | | From: "da Silva, Joe" | To: "'arachne@xxxxxxxxxxxxx'" | Date: Thu, 11 Mar 2004 16:24:23 +1100 | Arachne at FreeLists---The Arachne Fan Club! | | Greg, | | The reason this doesn't work for you is that you decided | to make up your own installation method, rather than | follow the one given in the instructions. The fact that you | found the line "@epppd.exe>>PPP.LOG" in your | Arachne.Cfg file proves that you have not installed this | stuff per instructions (basically, just type "cfg-ppp lsppp"). | | PUHLEASE follow the instructions contained in the | LSPPPCFG package! This package has been downloaded | by *hundreds* of Arachne users over the past three or so | years (no, that is not an exaggeration) and not one of these | (that followed the instructions) has ever reported that they | didn't work. | | I can understand your caution, which resulted in you trying | to "do things your own way", however please note that the | LSPPPCFG package saves back-up copies of all the files | it modifies anyway, and will restore these for you if you | should wish to do so. | | The instructions in the LSPPPCFG package correspond to | "Step 3a" below, for a single configuration. Try this first, to | satisfy yourself it all works, before proceeding with "Step 3b" | through "Step 3d" ... | | Joe. | | ============================================================================ | | Fri, Mar 12, 2004 - 5:54:06 pm | | | Or was this only refering to Lsppp when used for the packet driver | but not as the dialer as-well ? | | ============================================================================ | | Fri, Mar 12, 2004 - 6:01:28 pm | | ray> Š| Glenn, | | No info. on core.exe arguments? :( | | | | | ============================================================================ | | Fri, Mar 12, 2004 - 6:05:41 pm | | ray> | Joe, | | Just tried to download your lattest core again with no luck, but | I noticed that it's the download entry that starts with 'f4.grp ...' | again! That was the same number that didn't work before. What is | it about 'f4' ? | | | ============================================================================ | | Fri, Mar 12, 2004 - 9:15:27 pm | | glennmcc> | Hmmm... I got it just fine. | | This page, | http://www.esperance-linux.co.uk/misc/UserAgent.php | | Shows my current user agent string to be....... | xChaos_Arachne/4.1.73J4;GPL (DOS x86;WATTCP/1.05; 800x600,HiColor; | www.arachne.cz) | | ============================================================================ | | Fri, Mar 12, 2004 - 10:19:49 pm | | phantom> | Glenn, | | Re Miniterm : You stated this wouldn't work because it required a | Com Port number. This didn't sound right, so I quickly did an | experiment, selecting "non standard port" and explicitly specifying | the IRQ and Base address. This worked just fine for me, which proves | Miniterm *can* work this way. Why it doesn't work for you is a | mystery, perhaps Miniterm does a "sanity check" on the base address | and thinks $800 doesn't seem credible? If so, this would be a trvial | thing to fix, if only we can convince MP to find a spare monitor to | resurrect his DOS box and locate the missing source code (MP did | eventually answer and was surprised the source was missing, but his | DOS box was decommissioned). | | | ============================================================================ Š| | Fri, Mar 12, 2004 - 10:31:43 pm | | phantom> | Greggy, | | Correct. Since its first release over three years ago, not one | user has reported that my LSPPPCFG package didn't work for them, | unless they decided not to follow the instruction. | | This refers to the combination of Arachne, LSPPP, LSPPPCFG and | Miniterm, as configured by CFG-PPP (part of the LSPPPCFG package). | | Beware however that not all versions of LSPPP work properly, so in | the LSPPPCFG documentation I recommend particular versions to use. | For example, version 0.75, which was bundled with Arachne 1.73;GPL, | doesn't work with my ISP due to LCP bugs. Similarly, version 0.9 | is virtually unusable due to other problems. | | So, my remark relates specifically to my LSPPPCFG package, and | does not imply anything about LSPPP vs Miniterm, even though | LSPPPCFG specifically uses Miniterm because it provides better | integration with Arachne. | | | ============================================================================ | | Fri, Mar 12, 2004 - 10:37:45 pm | | phantom> | Glenn, | | I just noticed something. In your posting above, you are specifying | a base address of "bx800". You repeat this in closing, so it doesn't | seem to be a typo. Well, "bx800" is not a valid hex. number in 'C' | notation, is it? Please try again using "0x800" (I'm assuming this | is what you should have used). | | | ============================================================================ | | Fri, Mar 12, 2004 - 10:48:35 pm | | phantom> | Ray, | | Hmmmm ... what is it with you and 'f4'??? | | Perhaps it's similar to the problem I have here with the WebBoy | URL Glenn posted the other day? It simply doesn't seem to exist. | This (at this end, anyway) isn't an Arachne issue. I can't access | that URL with IE either. | | The only thing I could suggest, to resolve this mystery, is to try | another browser (eg. lynx386 or one of the "mainstream" browsers). Š| Perhaps another ISP (with a different DNS) might work (if you have | access to an alternative ISP via a friend or something). | | | | ============================================================================ | | Fri, Mar 12, 2004 - 11:03:31 pm | | phantom> | Ray, | | Well, I just tried my own advice ... | I _can_ access that WebBoy URL using a different ISP! | | As you are aware, Yahoo seem to have re-organized the Yahoo Groups | network so, since you say you couldn't even sign-in now, this may | be a different problem to your 'f4' problem. For about two days | earlier this week, I couldn't access the AD group, even with IE. | | In summary, if you can sign-in, this sounds like your traditional | 'f4' problem (see previous posting), if you can't, this is a new | problem and hopefully will resolve itself in a couple of days. | | | ============================================================================ | | Sat, Mar 13, 2004 - 2:47:25 am | | gregy> | re:...MINITERM.EXE source....hey...it's only 51K and we don't need | the original source....we can disassemble and use the assembly | language for the source.....as I said, before....source doesn't need | to be C.... | | | ============================================================================ | | Sat, Mar 13, 2004 - 9:45:50 am | | ray> | I'll get in touch with Michael and send him a bloody money order to | buy a monitor for his DOS box FGS. | | | | | ============================================================================ | | Sat, Mar 13, 2004 - 2:34:49 pm | | glennmcc> | Nope.... the actuall base port address is b800 | | Lsppp.exe works with bx800 Š| | Miniterm does not. | | I will try xb800 | | BRB | | | | | | ============================================================================ | | Sat, Mar 13, 2004 - 2:39:45 pm | | glennmcc> | Nope... xb800 did not work either. | | However..... 0xb800 does work :))) | | I am online right now with this setup. | | [profile] | Profile 4234005M.ACF | Connection @epppd.exe>>PPP.LOG | Hangup @termin.com 0x60>NUL | | [dialer] | DialPage file:pppglenn.htm | Dialer @MINITERM.EXE>PPP.LOG | TerminalWindow @MINITERM.EXE | UseTerminal No | Port 0 | Irq 4 | Base 0xb800 | Mode 8N1 | Speed 115200 | InitString ATf50 | DialString ATDT | PhoneNumber 330-423-4005 | Autologin No | WaitFor1 | Response1 | WaitFor2 | Response2 | WaitFor3 | Response3 | | | ============================================================================ | | Sat, Mar 13, 2004 - 2:40:09 pm | | glennmcc> | Now I will try lspppcfg Š| | BBL | | | | ============================================================================ | | Sat, Mar 13, 2004 - 2:55:28 pm | | glennmcc> | There we go. | | You get yet another tesimonial for a job well done. | | [profile] | Profile 4234005j.ACF | Connection @SUB-PPPD | Hangup @termin.com 0x60>NUL | | [dialer] | DialPage file:pppglenn.htm | | Dialer @MINITERM.EXE>PPP.LOG | TerminalWindow @MINITERM.EXE | UseTerminal No | Port 0 | Irq 4 | Base 0xb800 | Mode 8N1 | Speed 115200 | InitString ATf50 | DialString ATDT | PhoneNumber 330-423-4005 | Autologin No | WaitFor1 | Response1 | WaitFor2 | Response2 | WaitFor3 | Response3 | | | I do have 2 suggestions for future versions of Lspppcfg | | | 1) please leave the dialer page as it was before. | (I had to manually edit arachne.cfg to put it back to pppglenn.htm) | | 2) make one more command line option for cfg-ppp | | ex: | cfg-ppp lsppp lh | | This would prevent the need for the 2nd 'manual edit' that I did.... | Š| --- in sub-pppd.bat -- | lh lsppp.exe >> PPP.LOG | ______________________ | | | ============================================================================ | | Sat, Mar 13, 2004 - 4:06:04 pm | | glennmcc> | I just fund ne more 'manual edit' that I needed to do which I'll | sugjest | yet a 3rd command line option to added for. | | cfg-ppp lsppp lh ppp | | So that we don't need to edit arachne.cfg to use | | IP_Address PPP | | | So.... | | cfg-ppp lsppp | [use defaults] | (no lh and use BOOTP) | | cfg-ppp lsppp lh | (use lh and BOOTP) | | cfg-ppp lsppp ppp | (no lh and use PPP) | | cfg-ppp lsppp lh ppp | (use lh and PPP) | | Sound do-able ? | | ============================================================================ | | Sat, Mar 13, 2004 - 6:29:19 pm | URL | | glennmcc> | re"Perhaps it's similar to the problem I have here with the WebBoy | URL Glenn posted the other day? It simply doesn't seem to exist." | | As you already found by using a different ISP.... yes it does exist. | | URL above is the screen-cap I made just a little while ago. | | Now, I wonder if your 'regular' ISP is blocking it..... or | is it simply not being found by your 'regular' DNSs | | Might try using ISP #1 with the DNSs from ISP #2 Š| (and vice-versa) | | | ============================================================================ | | Sat, Mar 13, 2004 - 6:43:15 pm | | glennmcc> | Hey, | | I'm just back from 'the other place'. | | Does no-one here visit there anymore ??? | | ============================================================================ | | Sun, Mar 14, 2004 - 2:14:20 am | | gregy> | re: "Other Place"...see my post on my MISC... | ... | re:...this "fix" you guys have for Arachne not finding that b800 | addy....hey..!!!....this MAY be a hint as to how to get Arachne to | find the OS/2 packet-driver addy....AND....hold on....it may be a | hint for getting Arachne to find the IEShivaDialer packet-driver | addy.... | ...at least it's an "avenue of investigation"..... | | | ============================================================================ | | Sun, Mar 14, 2004 - 1:32:53 pm | | glennmcc> | Yes......... not just 2, but perhaps 3 birds with 1 stone. | | Thank you Joe for telling us about that. | | Had you not told me that Arachne set the com# to 0 and then used just | the IRQ and port address instead..... I never would have tried | miniterm again. | | | ============================================================================ | | Sun, Mar 14, 2004 - 4:07:16 pm | | Joe> | Glenn, | | Glad you now like Miniterm as much as I do. ;-) | | As for your suggestions re. LSPPPCFG, | 1. The dial page change was an oversight that I fixed about two | versions ago, so you might like to download the current version Š| for future use. | 2. The "LH" option has been available for quite a while. The syntax | is "CFG-PPP LH LSPPP". | 3. I'll think about a "PPP" option, although I consider this to be | "crippled", since the whole point of LSPPPCFG is to simplify things | for users, so eliminating the DNS requirements by using "BOOTP" | with | LSPPP was the natural choice. | | ============================================================================ | | Sun, Mar 14, 2004 - 4:13:17 pm | | Joe> | Glenn, | | I forgot to mention, once LSPPPCFG is installed, all configuration | changes for dial-up can be done via the Arachne GUI, no need to | edit 'Arachne.Cfg'. | | The URL access problem ... I think it's a DNS problem. No way to swap | DNS between my two ISP's however, they seem to override my settings | anyway. I can even use "junk" DNS addresses and nothing changes. :-/ | | ============================================================================ | | Mon, Mar 15, 2004 - 6:29:08 am | | gregy> | Joe...that's strange....I can use any of 6 sets (of 2) DNS numbers | which my ISP has for it's various areas around here....and I have | some "backup" DNSes from FreeNIC....which are scattered all over... | ... | As far as I can tell...MOST ISPs do NOT block you browser's requests | to ANY valid DNS number......so....what you are experiencing is... | "strange"..... | | | | ============================================================================ | | Mon, Mar 15, 2004 - 4:48:07 pm | | ray> | Joe, | | Tried to get CORE4.ZIP using Firebird (W98) and got: | | "The connection was refused" msg! | | Anyway, this seems to prove that the problem has nothing to do with | Arachne. Lets repeat the same experiment we did last time: copy | core4.zip to someothername.zip and let me see if I can download that | -- | it worked last time ... Š| | | ============================================================================ | | Mon, Mar 15, 2004 - 5:08:24 pm | | glennmcc> | ray, | | Was that other file named 'core?.zip' ? | | If so...... | Perhaps your ISP is blocking files with the word 'core' in their name. | | As an anti 'hardcore porn' tactic. | | See if you can get this. | http://www.angelfire.com/linux/glennmcc/downloads/core4.zip | | | ============================================================================ | | Mon, Mar 15, 2004 - 6:18:49 pm | | ray> | Glenn, | Glenn, | | Creative thinking, but what you posted downloaded just fine, so that's | not it. I'm fairly sure it has something to do with "f4. ..." since | every time something has refused to download, that has been the | starting number. | | | | | ============================================================================ | | Mon, Mar 15, 2004 - 6:21:20 pm | | ray> | Now that's interesting -- in the post above I most certainly did not | type Glenn twice -- that was added for me by some force unknown. | | Let's see if it happens again: | | Glenn, | | | | ============================================================================ | | Mon, Mar 15, 2004 - 6:37:04 pm | | glennmcc> Š| rayray, | | That's OK... I never tire of seeing my name in print. ;-) | | | ============================================================================ | | Mon, Mar 15, 2004 - 6:39:22 pm | | glennmcc> | But serioously... | | Unless I miss my guess....... | You just saw one of the 'seldom seen' memory leaks in Arachne's | internal editor. | | It not only places srange characters in to the textarea.... | But it sometimes duplicates lines. | | | ============================================================================ | | Mon, Mar 15, 2004 - 6:44:25 pm | | Joe> | Well, I had an idea ... I used a Whois service to convert the WebBoy | UMC URL's domain name to its corresponding IP address. Then I used | that IP address in the URL ... it still didn't work, said "host not | found" or similar! So, the problem isn't DNS look-up after all, it's | the "routing table" (or whatever it's called). One ISP can't reach | this URL, the other can. | | I wonder if Ray's 'f4' problem is something like my 'WebBoy UMC' | problem? At least it's not an Arachne problem, but seems to correlate | with a particular ISP. | | Later today, I'll upload Core4 again to AD, under a different name. | If this ends up on a different server (ie. not 'f4'), I'll bet Ray | will be able to download it ... | | ============================================================================ | | Mon, Mar 15, 2004 - 7:48:36 pm | | ray> | Joe, | | Yup, I bet I will be able to too. Anyway I got the goods from Glenn, | but what are those two extra .exe's for? Isn't it just a question | of dropping in the new core? BTW I like the little batch file! -- | it's a good example of taking a little extra time to make things | idiot proof. I'm going to keep it just as an example in my .bat file | archives. | | Š| ============================================================================ | | Mon, Mar 15, 2004 - 11:15:05 pm | | Joe> | Ray, | | As I stated at AD earlier, the new Core uses a new Cookies.Lst | file format (to achieve full size cookies without blowing the | available memory away). Therefore, I have provided utilities to | convert between the old and new format. You will notice that the | 'cook2new' utility is invoked by that batch file. | | BTW, silly question, you _did_ sign-in to Yahoo before attempting | to download the Core4.Zip file using Firebird, didn't you? | | | ============================================================================ | | Mon, Mar 15, 2004 - 11:26:46 pm | | Joe> | PS. The need to convert the Cookies.Lst file was the main reason | for making a batch file for Core4. | | ============================================================================ | | Mon, Mar 15, 2004 - 11:29:04 pm | | Joe> | Ray, | | OK, I have uploaded Core4.Zip as Test2.Zip, so you might like to | try downloading again, when convenient. This upload ended up on | 'f1', so hopefully it will work for you. | | ============================================================================ | | Tue, Mar 16, 2004 - 4:27:25 pm | | glennmcc> | Joe, | | Could I ask a BIG favor of you ? | | May I have a copy of you SRC code changes need to do those 'full size | cookies' ? | | And BTW, | By 'full size', do you mean the entire 4096 bytes allowed by the RFC ? | | | ============================================================================ | | Tue, Mar 16, 2004 - 5:35:55 pm Š| | Joe> | Glenn, | | Good timing, I intend to upload the source code tonight. | | As for "full size", yes, that does mean 4096 bytes. At least, that's | how much the cookie code can handle. The http buffer is either 4096 | or 2048 bytes (was 4000 or 2000), so this sets the ultimate limit | to cookie size handling. Also, if there are multiple cookies | applicable | to a domain and path, their combined limit is 4096 bytes. In other | words, the old 800 byte limit has effectively become 4096 bytes. | | | ============================================================================ | | Tue, Mar 16, 2004 - 6:36:37 pm | | glennmcc> | FANTASTIC !!!!!! :))) | | ============================================================================ | | Tue, Mar 16, 2004 - 9:04:58 pm | | ray> | Joe, | | Yes, logged in before attempting download (is there any other way?) | Will try downloading test2.zip and let you know. | Thanks for clarification vis. small .exe's. | Looking forward to playing with your new core, and checking out | sources. | | | ============================================================================ | | Tue, Mar 16, 2004 - 9:09:22 pm | | ray> | Joe, | | Yup, test2.zip downloads w.o. any problem -- go figure. | | ============================================================================ | | Wed, Mar 17, 2004 - 6:05:35 pm | | ray> | Gregy, | | Do you have a rough ETA for completion of moving the translations into | BASE1a ? | Š| | ============================================================================ | | Wed, Mar 17, 2004 - 7:32:51 pm | | glennmcc> | Hello everyone, | | _MY_ copy of Insight.exe now says this at the bottom of the | page......... | | Arachne Insight 3.85 GNUpyright (G)1997,2000 Arachne Labs | | What does yours say ? | | | ============================================================================ | | Wed, Mar 17, 2004 - 8:01:27 pm | | ray> | Nothing half so nice. | | | ============================================================================ | | Thu, Mar 18, 2004 - 12:33:01 am | | Joe> | Glenn, | | Dunno ... never used it! | | ============================================================================ | | Thu, Mar 18, 2004 - 10:08:13 pm | | ray> | Joe, | | 'J4 running fine, and Dos memory == 164+ :-) | | | | ============================================================================ | | Fri, Mar 19, 2004 - 1:04:41 am | | Joe> | Ray, | | Excellent! The problem with getting full-size cookie support into | Arachne in the past was doing so without excessive memory wastage, | so I put a lot of effort into achieving this in such a way that I | could actually reduce Arachne's memory requirements (which were Š| bordering on unacceptable, IMHO). | | ============================================================================ | | Fri, Mar 19, 2004 - 12:03:57 pm | | ray> | Joe, | | Yeah, on first glance one might have thought that all that was | needed to increase cookie size was to change the value of a #define | or two -- it's apparent that you've taken a much higher road -- will | look at the sources this weekend and I'm expecting to learn alot :-) | | | ============================================================================ | | Sun, Mar 21, 2004 - 10:03:29 pm | | ray> | Gregy, where are you? | | | | ============================================================================ | | Sun, Mar 21, 2004 - 11:49:37 pm | | gregy> | I are back again.......but I come withOUT yet any ETA...sorry... | ......just getting start has been something of a struggle... | ...he he...not because it's such a difficult job....just that it's | had to be "squoooooooshed" thru difficult circumstances... | | | ============================================================================ | | Mon, Mar 22, 2004 - 12:56:46 pm | | ray> | Gregy, | | OK, but if you do get some gradual idea of an ETA, let me know, so | that I can coordinate the end of work on '2a' with the end of the | translations into '1a' -- that way I'll be free to help with the | translations into 2a. | | | ============================================================================ | | Tue, Mar 23, 2004 - 5:42:01 am | | gregy> | wilco...(willcomply).... | Š| | | ============================================================================ | | Tue, Mar 23, 2004 - 2:30:46 pm | | ray> | roger that. | | | ============================================================================ | | Wed, Mar 24, 2004 - 10:44:30 pm | | ray> | Gle | | Found the blasted thing. Strange that it only happened after using | ^R, though. Life gets a little harsher once one leaves the security | of identical compiles behind. I'm glad I'm neurotic about commenting | and dating everything -- find the last 'good' .exe, find the first | 'bad' .exe bracket the date range, seach for changes dated | between those, and it's got to be there -- and was :) | | | ============================================================================ | | Wed, Mar 24, 2004 - 10:45:59 pm | | ray> | ... Damned editor bugs ... | | that was: Glenn | | ... | | come to think of it, what's wrong with Gle anyway ?? | | | | ============================================================================ | | Thu, Mar 25, 2004 - 2:04:53 pm | | glennmcc> | The same thing that's wrong with 'ra' | | ============================================================================ | | Thu, Mar 25, 2004 - 6:22:46 pm | | ray> | Na, you missed the gag: Gle pronounced GLEE. | | Come to think of it, my inittials are R.A. and at work, when I sign Š| off on my registered letters with them, if the clerk thinks I've been | a bit too big for my britches, she does call me "RA the sun god". | | | ============================================================================ | | Thu, Mar 25, 2004 - 8:30:28 pm | | glennmcc> | Nope... did not "miss the gag" | | We both got it. | | I _was_ refering to "RA the sun god". | | You see ??? | | Great minds DO think alike. | | | | ============================================================================ | | Thu, Mar 25, 2004 - 9:56:22 pm | | ray> | Well I'll be. | | ... but hopefuly not *too* alike -- the tension between oposing | viewpoints can be very fruitful as long as folks don't get | defensive. | | ============================================================================ | | Sat, Mar 27, 2004 - 8:43:20 am | | ray> | Sgt. Feig! | | Report for duty! | | | | ============================================================================ | | Sat, Mar 27, 2004 - 11:42:46 pm | | gregy> | I ARE already "on duty"...... | | | | ============================================================================ | | Sun, Mar 28, 2004 - 11:37:07 pm Š| | ray> | G, | è¹Úƒ! | Well then report, soldier! | | | ============================================================================ | | Mon, Mar 29, 2004 - 4:54:50 am | | gregy> | well...do you want the progress report....??? | ... | ..in inches....or in centimeters......??? | ... | ? | | | ============================================================================ | | Mon, Mar 29, 2004 - 1:27:23 pm | | ray> | Yup, | | ... in FP (files processed). | | | ============================================================================ | | Mon, Mar 29, 2004 - 9:29:48 pm | | ray> | Just in case anyone's interested, there are 407 functions | defined in Arachne (for those of you who might like statistics -- | baseball fans and that sort of person ...) | | | ============================================================================ | | Mon, Mar 29, 2004 - 11:44:21 pm | | gregy> | Ray....I like that kinda stuff...as I said....I was trained as a | mathematician...... | ...but... | ....also....those kindsa things may turn out to be "useful info" for | future developers... | | | ============================================================================ | | Tue, Mar 30, 2004 - 3:01:37 pm | Š| ray> | Yup, | | My snotty comments about baseball were roughly aimed in your direction | ;-) | | ... So how's it going with the translations? | | | ============================================================================ | | Thu, Apr 01, 2004 - 5:35:58 pm | | ray> | Interesting how we go from doing all our buisiness here, to doing | it all on Hehaw, then back again. | | So no comment on my function renaming scheme. The question now is: | does silence confer consent? | | ============================================================================ | | Fri, Apr 02, 2004 - 12:54:38 am | | Joe> | OK, Ray, | | You have spurred a response about the renaming ... although in my | haste :-(in a rush, ATM, busy) I initially responded to the wrong | thread. Arghh! | | ============================================================================