Archive-Date: Wed, 04 Sep 1996 20:42:54 +0200 Sender: owner-emacs-testers@lp.se Date: Wed, 4 Sep 96 11:40:33 PDT From: pf%pinegv@Riem.Com Reply-To: emacs-testers@LISTS.VMS.GNU.AI.MIT.EDU Message-ID: <960904114033.8b7d@Riem.Com> Subject: http, mail list, etc To: emacs-testers@lp.se Wanted to see what the latest was so tried http://vms.gnu.ai.mit.edu. Found a page containing nothing but the string 'Files'. Then tried sending mail to emacs-testers@lists.vms.gnu.ai.mit.edu. Bounced. And have now of course dug up again this address ... which will hopefully work. What's going on? pat ================================================================================ Archive-Date: Thu, 05 Sep 1996 00:02:11 +0200 Sender: owner-emacs-testers@lp.se Date: Thu, 05 Sep 1996 00:02:08 +0200 From: Richard Levitte Reply-To: emacs-testers@LISTS.VMS.GNU.AI.MIT.EDU To: emacs-testers@LISTS.VMS.GNU.AI.MIT.EDU Message-ID: <009A7E45.F18E0260.23@lp.se> Subject: Re: http, mail list, etc > Wanted to see what the latest was so tried >http://vms.gnu.ai.mit.edu. Found a page containing nothing but Drat! I reconfigured my GNU file tree, and apparently made a slight error. No problem, it's fixed. Thank for the notification. >the string 'Files'. Then tried sending mail to >emacs-testers@lists.vms.gnu.ai.mit.edu. Bounced. And have now That's weird. Exactly what was the message you got back? -- R Levitte, Levitte Programming; Spannvägen 38, I; S-161 43 Bromma; SWEDEN Tel: +46-8-26 52 47, (via nation.se) +46-8-728 20 33; No fax right now PGP key fingerprint = A6 96 C0 34 3A 96 AA 6C B0 D5 9A DF D2 E9 9C 65 Check http://www.lp.se/~levitte for my public key. bastard@bofh.se ================================================================================ Archive-Date: Wed, 11 Sep 1996 18:03:07 +0200 Sender: owner-emacs-testers@lp.se Date: Wed, 11 Sep 96 09:01:22 PDT From: pf%pinegv@Riem.Com Reply-To: emacs-testers@LISTS.VMS.GNU.AI.MIT.EDU Message-ID: <960911090122.9a6b@Riem.Com> Subject: changed file characteristics after editing the file with emacs -batch To: emacs-testers@lp.se Hello, Have the following procedure: > ;;(switch-to-buffer (find-file "pf_p_lisp:ha164159.hrd")) > ;;(print command-line-args) > (switch-to-buffer (find-file (nth 3 command-line-args))) > (beginning-of-buffer) > (while (re-search-forward "h...c...p...o...s...t...e...p...s" nil t) > (replace-match "G\0\0\0T\0\0\0I\0\0\0F\0\0\0L\0\0\08\0\0\0H\0\0\0 \0\0\0 " nil t)) > (save-buffer) Run this in batch mode as follows: $ emacs -batch -l procedure_name FILENAME The procedure opens FILENAME; changes the first string to the second and then writes out the file. Lisp appears to do this in way less code than a similar C program. The problem is file characteristics. The file being read in is a binary file (with some ascii in it obviously). When the file is read in it has records of variable length with longest at 530 bytes. When the file is written back out it is of variable length but the longest record being 5460 bytes. So I believe Emacs has removed some of the record boundaries. That is, collapsed what were previously multiple records into a single record. Had never used hexl-find-file and tried using this also, but this is not what I want. What might I do about this? Mapping down onto RMS file characteristics seems to be a pain-in-the-ass whether you do it in C or Emacs Lisp ... pat ================================================================================ Archive-Date: Thu, 12 Sep 1996 18:05:16 +0200 Sender: owner-emacs-testers@lp.se Date: Thu, 12 Sep 1996 18:05:06 +0200 From: Richard Levitte Reply-To: emacs-testers@LISTS.VMS.GNU.AI.MIT.EDU To: emacs-testers@LISTS.VMS.GNU.AI.MIT.EDU Message-ID: <009A845D.6475AEA0.3@lp.se> Subject: Re: changed file characteristics after editing the file with emacs -batch >From: pf%pinegv@Riem.Com >> ;;(switch-to-buffer (find-file "pf_p_lisp:ha164159.hrd")) >> ;;(print command-line-args) >> (switch-to-buffer (find-file (nth 3 command-line-args))) >> (beginning-of-buffer) >> (while (re-search-forward "h...c...p...o...s...t...e...p...s" nil t) >> (replace-match "G\0\0\0T\0\0\0I\0\0\0F\0\0\0L\0\0\08\0\0\0H\0\0\0 \0\0\0 " nil t)) >> (save-buffer) > > Run this in batch mode as follows: > > $ emacs -batch -l procedure_name FILENAME > >The procedure opens FILENAME; changes the first string to the >second and then writes out the file. Lisp appears to do this in >way less code than a similar C program. > > The problem is file characteristics. The file being read in is >a binary file (with some ascii in it obviously). When the file >is read in it has records of variable length with longest at 530 >bytes. When the file is written back out it is of variable >length but the longest record being 5460 bytes. So I believe Is it possible that your replacement loop replaced a few linefeeds with something else? That would explain the different longest record. How about you test by just loading FILENAME in a live Emacs, and do this: (while (re-search-forward "h...c...p...o...s...t...e...p...s" nil t) (replace-match "G\0\0\0T\0\0\0I\0\0\0F\0\0\0L\0\0\08\0\0\0H\0\0\0 \0\0\0 " nil t)) And then look at the result? (Hint: wrap the above in an interactive function, like this: (defun hack () (interactive) (while (re-search-forward "h...c...p...o...s...t...e...p...s" nil t) (replace-match "G\0\0\0T\0\0\0I\0\0\0F\0\0\0L\0\0\08\0\0\0H\0\0\0 \0\0\0 " nil t))) then go to the beginning of the buffer, and do 'M-x hack RET' ) -- R Levitte, Levitte Programming; Spannvägen 38, I; S-161 43 Bromma; SWEDEN Tel: +46-8-26 52 47, (via nation.se) +46-8-728 20 33; No fax right now PGP key fingerprint = A6 96 C0 34 3A 96 AA 6C B0 D5 9A DF D2 E9 9C 65 Check http://www.lp.se/~levitte for my public key. bastard@bofh.se ================================================================================ Archive-Date: Sun, 15 Sep 1996 09:47:41 +0200 Sender: owner-emacs-testers@lp.se Date: Sun, 15 Sep 1996 09:47:37 +0200 From: levitte@lp.se Reply-To: emacs-testers@LISTS.VMS.GNU.AI.MIT.EDU To: emacs-testers@lists.vms.gnu.ai.mit.edu Message-ID: <009A8673.64043780.8@lp.se> Subject: test 2 Relay-Version: ANU News - V6.1B10 04/18/95 OpenVMS VAX; site news.bofh.se Path: levitte Newsgroups: local.gnu.emacs-testers Subject: test 2 Message-ID: From: levitte@lp.se (Richard Levitte - VMS Whacker) Date: 15 Sep 1996 09:47:35 GMT Organization: Levitte Programming Nntp-Posting-Host: www.bofh.se Lines: 7 Sorry... -- R Levitte, Levitte Programming; Spannvägen 38, I; S-161 43 Bromma; SWEDEN Tel: +46-8-26 52 47, (via nation.se) +46-8-728 20 33; No fax right now PGP key fingerprint = A6 96 C0 34 3A 96 AA 6C B0 D5 9A DF D2 E9 9C 65 Check http://www.lp.se/~levitte for my public key. bastard@bofh.se ================================================================================ Archive-Date: Thu, 19 Sep 1996 09:11:53 +0200 Sender: owner-emacs-testers@lp.se Date: Thu, 19 Sep 1996 09:11:50 +0200 From: Richard Levitte Reply-To: emacs-testers@LISTS.VMS.GNU.AI.MIT.EDU To: SEIKENBE@us.oracle.com CC: emacs-testers@lists.vms.gnu.ai.mit.edu Message-ID: <009A8993.0DFCA520.6@lp.se> Subject: Re: Trouble building emacs on Alpha/VMS >I'm having trouble building and running emacs on Alpha/VMS 7.0 with DEC C >V5.3. Are you by chance already aware of the issues? I'm aware of them, but not having access to a 7.0 system hinders me for the moment being. I'm going to do a few tests on a couple of 7.1 system during the weekend. Meanwhile, could you send me your changes? It'd greatly help. >executable image, but it doesn't run properly. It appears to be having >difficulty mapping the .dump file into memory with sys$crmpsc. That's interesting... BTW, Sharon, I'm CC:-ing this message and forwarding yours to the Emacs testers list. Perhaps should you subscribe? Just send the word SUBSCRIBE in a message to emacs-testers-request@lists.vms.gnu.ai.mit.edu. -- R Levitte, Levitte Programming; Spannvägen 38, I; S-161 43 Bromma; SWEDEN Tel: +46-8-26 52 47, (via nation.se) +46-8-728 20 33; No fax right now PGP key fingerprint = A6 96 C0 34 3A 96 AA 6C B0 D5 9A DF D2 E9 9C 65 Check http://www.lp.se/~levitte for my public key. bastard@bofh.se ================================================================================ Archive-Date: Thu, 19 Sep 1996 09:13:53 +0200 Sender: owner-emacs-testers@lp.se Date: Thu, 19 Sep 1996 09:13:50 +0200 From: Richard Levitte Reply-To: emacs-testers@LISTS.VMS.GNU.AI.MIT.EDU To: emacs-testers@lists.vms.gnu.ai.mit.edu Message-ID: <009A8993.55EC7A40.29@lp.se> Subject: Problems on VMS 7.0 I seem to recall someone else having problems with VMS 7.x. I just got the following message (I've already answered it, and cc:-ed the answer to this list): Message-ID: <199609181851.LAA10301@mailsun2.us.oracle.com> Date: 18 Sep 96 10:52:51 -0700 From: "Sharon Eikenberry" To: admin@vms.gnu.ai.mit.edu Subject: Trouble building emacs on Alpha/VMS MIME-Version: 1.0 X-Mailer: Oracle InterOffice (version 4.0.2.1.40) Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="US-ASCII" I'm having trouble building and running emacs on Alpha/VMS 7.0 with DEC C V5.3. Are you by chance already aware of the issues? I get fatal compilation errors, which I've worked around. I have an executable image, but it doesn't run properly. It appears to be having difficulty mapping the .dump file into memory with sys$crmpsc. Thanks, --Sharon Eikenberry seikenbe@us.oracle.com (617) 684-5556 -- R Levitte, Levitte Programming; Spannvägen 38, I; S-161 43 Bromma; SWEDEN Tel: +46-8-26 52 47, (via nation.se) +46-8-728 20 33; No fax right now PGP key fingerprint = A6 96 C0 34 3A 96 AA 6C B0 D5 9A DF D2 E9 9C 65 Check http://www.lp.se/~levitte for my public key. bastard@bofh.se ================================================================================ Archive-Date: Thu, 19 Sep 1996 15:13:00 +0200 Sender: owner-emacs-testers@lp.se Date: Thu, 19 Sep 1996 08:14:32 -0500 Message-ID: <199609191314.IAA00222@public-9.rose-hulman.edu> From: "Matthew X. Economou" To: emacs-testers@lists.vms.gnu.ai.mit.edu Subject: Problems on VMS 7.0 References: <009A8993.55EC7A40.29@lp.se> Reply-To: emacs-testers@LISTS.VMS.GNU.AI.MIT.EDU -----BEGIN PGP SIGNED MESSAGE----- >>>>> "RL" == Richard Levitte writes: RL> I seem to recall someone else having problems with VMS 7.x. I RL> just got the following message (I've already answered it, and RL> cc:-ed the answer to this list): I'm the other person. We're running OpenVMS 7.0 on our AlphaServer 2100 4/200. When Emacs is started, I immediately get the error message: Warning: past acceptable memory limits Killing some buffers may delay running out of memory. However, certainly by the time you receive the 95% warning, you should clean up, kill this Emacs, and start a new one. Also, there was a slight problem with strdup being in the std. C libraries (and the Emacs sources defining a function called strdup), but that was easy to work around (renamed the function to emacs_strdup, or something like that). I have the original post I mailed you, if you like. -----BEGIN PGP SIGNATURE----- Version: 2.6 Comment: Processed by Mailcrypt 3.4, an Emacs/PGP interface iQCVAwUBMkFHLijh7SVEyfhxAQFf9QQAlb3E0hlwPEXJSsGDDrRnapZdBiCrBJE6 Ivg376Gb2vu9hP4GMuXcjDtZ2QwYJUIGmBwADZN9zgAMR4IlEZIcPkIbsRNKBKh+ Gs4W4EVW2haSy5aZbwjV533uW36saOoOv23SWoGC2JzvZKrvck5+04GaKhkl2Js6 vhDH+QcpekI= =84pO -----END PGP SIGNATURE----- -- Matthew X. Economou - Finger/WWW for public key. - Geek of Computer Science(!) econommx@rose-hulman.edu - http://www.rose-hulman.edu/~econommx - Random quote: "They say absolute power corrupts absolutely, but hey, it's as fun as hell!" -- allen@raptor.business.swt.edu ================================================================================ Archive-Date: Thu, 19 Sep 1996 15:17:54 +0200 Sender: owner-emacs-testers@lp.se Message-ID: <9609191249.AA26250@us2rmc.zko.dec.com> Date: Thu, 19 Sep 96 09:06:33 EDT From: Private Instigator Reply-To: emacs-testers@LISTS.VMS.GNU.AI.MIT.EDU To: emacs-testers@lists.vms.gnu.ai.mit.edu CC: seikenbe@us.oracle.com, eiche@star.ENET.dec.com Subject: RE: Problems on VMS 7.0 Richard or Sharon, W/r/t build errors... can you characterize the errors? W/r/t create and map section... first cut obviously would be some conflict with 64bitness in 7.0. BTW Richard, the problem I had sending you a log file generated by set host 0/log using VMSmail and SMTP% turned out to be an UCX problem. UCX was looking at a null at the beginning of each VMS command line and decided the line was empty. Being fixed to look at RMS's idea of the record length. dee ================================================================================ Archive-Date: Thu, 19 Sep 1996 16:09:23 +0200 Sender: owner-emacs-testers@lp.se Date: Thu, 19 Sep 1996 16:09:18 +0200 From: Richard Levitte Reply-To: emacs-testers@LISTS.VMS.GNU.AI.MIT.EDU To: emacs-testers@LISTS.VMS.GNU.AI.MIT.EDU Message-ID: <009A89CD.60349200.1@lp.se> Subject: Re: Problems on VMS 7.0 >From: "Matthew X. Economou" >I'm the other person. We're running OpenVMS 7.0 on our AlphaServer Thanks. >I have the original post I mailed you, if you like. No need. There are mail archives, you know. :-) -- R Levitte, Levitte Programming; Spannvägen 38, I; S-161 43 Bromma; SWEDEN Tel: +46-8-26 52 47, (via nation.se) +46-8-728 20 33; No fax right now PGP key fingerprint = A6 96 C0 34 3A 96 AA 6C B0 D5 9A DF D2 E9 9C 65 Check http://www.lp.se/~levitte for my public key. bastard@bofh.se ================================================================================ Archive-Date: Thu, 19 Sep 1996 16:38:59 +0200 Sender: owner-emacs-testers@lp.se Date: Thu, 19 Sep 1996 16:38:54 +0200 From: Richard Levitte Reply-To: emacs-testers@LISTS.VMS.GNU.AI.MIT.EDU To: emacs-testers@LISTS.VMS.GNU.AI.MIT.EDU Message-ID: <009A89D1.82733700.3@lp.se> Subject: RE: Problems on VMS 7.0 >From: Private Instigator >W/r/t build errors... can you characterize the errors? I've given a 7.1 Alpha a try a few days ago. It looked like TPU was imcompatable with itself. I'll come back on this later. >W/r/t create and map section... first cut obviously would >be some conflict with 64bitness in 7.0. Hmm, that's weird... Emacs uses only the P0 space, and that one is exactly the same on 7.0 as on 6.x... The *shouldn't* be a difference. Could there be a difference when it comes to maping a global section? Emacs creates such a beast and puts the read-only part of the .DUMP file in there if the section does not already exist. Thus, several instances of Emacs can share that part as well, decreasing memory usage in the process. >BTW Richard, the problem I had sending you a log file generated >by set host 0/log using VMSmail and SMTP% turned out to be an Thanks. >UCX problem. UCX was looking at a null at the beginning of >each VMS command line and decided the line was empty. Being >fixed to look at RMS's idea of the record length. Weird... -- R Levitte, Levitte Programming; Spannvägen 38, I; S-161 43 Bromma; SWEDEN Tel: +46-8-26 52 47, (via nation.se) +46-8-728 20 33; No fax right now PGP key fingerprint = A6 96 C0 34 3A 96 AA 6C B0 D5 9A DF D2 E9 9C 65 Check http://www.lp.se/~levitte for my public key. bastard@bofh.se ================================================================================ Archive-Date: Tue, 24 Sep 1996 20:01:12 +0200 Sender: owner-emacs-testers@lp.se Message-ID: <9609241607.AA04842@us2rmc.zko.dec.com> Date: Tue, 24 Sep 96 13:51:25 EDT From: Private Instigator Reply-To: emacs-testers@LISTS.VMS.GNU.AI.MIT.EDU To: emacs-testers@lists.vms.gnu.ai.mit.edu CC: eiche@star.ENET.dec.com Subject: RE: Problems on VMS 7.0 >I've given a 7.1 Alpha a try a few days ago. It looked like TPU >was imcompatable with itself. I'll come back on this later. Please do. >Hmm, that's weird... Emacs uses only the P0 space, and that one >is exactly the same on 7.0 as on 6.x... The *shouldn't* be a >difference. That's not what I mean. We made substantial changes to the executive and I/O portions of VMS for 64 bits. The changes cause ripples out into the compilers, RTLs and other system components. Occasionally these changes introduce a bug, hard as that is to believe :-) >Could there be a difference when it comes to maping a global section? >Emacs creates such a beast and puts the read-only part of the .DUMP >file in there if the section does not already exist. Thus, several >instances of Emacs can share that part as well, decreasing memory >usage in the process. There are 64 bit versions of those services, yes. I doubt emacs has been changed to use them. If the service is somehow buried in an RTL routine, I suppose you could be getting to it that way. It'd be nice to have more info on the failure, so I don't have to speculate. >>UCX problem. UCX was looking at a null at the beginning of >>each VMS command line and decided the line was empty. Being >>fixed to look at RMS's idea of the record length. >Weird... Not really too weird. SMTP apparently doesn't deal well with some control characters, hence MIME. We tried Multinet yesterday and it failed in a slightly different way from UCX. Cheers, dee ================================================================================ Archive-Date: Wed, 25 Sep 1996 21:18:18 +0200 Sender: owner-emacs-testers@lp.se Date: Wed, 25 Sep 1996 21:18:11 MET DST Message-ID: <12873.41363.405647v0.1b7.LEVITTE@devil.bofh.se> From: Richard Levitte - VMS Whacker Reply-To: emacs-testers@LISTS.VMS.GNU.AI.MIT.EDU To: emacs-testers@lists.vms.gnu.ai.mit.edu CC: calle@ctrl-c.liu.se Subject: Send mail directly through an SMTP server Once upon a time, Richard Flower made a package to send mail from Emacs on VMS by talking directly to the SMTP server. It has been present in the beta directory (ftp://vms.gnu.ai.mit.edu/gnu-vms/software/beta/smtp.el and corresponding in the mirrors) since, but I haven't heard anything since (I remember there was a discussion on how appropriate this really was, since you could as well have to send mail through DECnet or UUCP). Richard's implemention was designed to be called from the vms-sndmai.el package that comes with Emacs for VMS. A few days ago, Calle Englund started hacking together another, simpler package of the same kind, called sendmail-smtp.el, now also present in the beta directory. It is of slightly different design, as it is meant to replace vms-sndmai.el altogether, and thus, no temporary files are produced, which has the effect of speeding up the process. Packages like these are only usefull to those of you who, like me, send your mail through SMTP exclusivelly. This doesn't mean you can't send mail to everyone, at least if you use MX, which can be used to send mail through all kinds of protocolls. I don't know how other similar packages do... Anyway, If you want to try one of these out, please do so. If you want to help hack, please do so, and contribute patches. If you want your reply to reach Calle as well, please CC: to him. He's not on this list, at least right now. (I'm currently testing sendmail-smtp.el, as well as help hacking it) -- R Levitte, Levitte Programming; Spannvägen 38, I; S-161 43 Bromma; SWEDEN Tel: +46-8-26 52 47, (via nation.se) +46-8-728 20 33; No fax right now PGP key fingerprint = A6 96 C0 34 3A 96 AA 6C B0 D5 9A DF D2 E9 9C 65 Check http://www.lp.se/~levitte for my public key. bastard@bofh.se