December 2001 - XEmacs-Beta - XEmacs Mailing List Archives

Yoshiki Hayashi

Hi, all. Sorry for being absent for quite a while. I recently started reading XEmacs lists again. There are huge pile of unread mail I should read but I hope I'll be able to restart reviewing patches within this year. I still can't spend many hours fixing XEmacs bugs but if you have somethong that falls into the area of my expertise, please resend them and I'll try to fix them. -- Yoshiki Hayashi

7 participants
9 comments

Re: Syntax table issues 22 years, 11 months

Matt Tucker

-- Vin Shelton <acs(a)xemacs.org> spake thusly: > 1. Start up xemacs -vanilla > > 2. Add the following text (comment included) into the *scratch* > buffer: > > ;; This buffer is for notes you don't want to save, and for Lisp > evaluation. ;; If you want to create a file, first visit that file > with C-x C-f, ;; then enter the text in that file's own buffer. (C-x > is the standard ;; XEmacs abbreviation for `Control+x', i.e. hold > down the Control key ;; while hitting the x key.) > ;; > ;; For Lisp evaluation, type an expression, move to the end and hit > C-j. > > (when (or (featurep 'xemacs) window-system) > (require 'mic-paren) > (setq paren-sexp-mode t) > (set (if (boundp 'paren-face) 'paren-face 'paren-match-face) 'bold) > (and (fboundp 'paren-activate) (paren-activate))) > > 3. Evaluate the buffer. > > 4. Put the cursor after the closing paren in 'the x key.)' You will > notice that, even though the parens are balanced and in a comment, > the closing paren turns yellow, indicating a matching open paren > can not be found. A simpler test is: 1. xemacs -vanilla 2. Insert into *scratch*: ;; ( ;; ) 3. Move point to just after ")". 4. Do: "M-:(scan-sexps (point) -1)" --> Error "Unbalanced parentheses" I just tested this with emacs 20.7 and xemacs 21.1 and got the same behavior. The problem is that scan_lists (which scan-sexps uses) won't find a matching parenthesis inside a preceding comment. You might notice that in your test if you insert an open paren at the start of the buffer and then put the cursor after "x key.)", that the open paren _outside_ the comment is matched. The bug is either in mic-paren or paren or, more likely, in syntax.c. Presumably placing the cursor after a parenthesis inside a comment should either match another parenthesis in the same comment or return 'nil, just as it would if placed after a non-parenthesis. Either way, the bug exists in prior versions of Emacs and XEmacs, so shouldn't affect any decision as to whether to roll back the syntax-table changes.

1 participants
0 comments

Does VM support PGP? 22 years, 11 months

Vaclav Barta

Hi, I'm thinking about XEmacs MUAs again, with the view of using ecrypted e-mail. The early results are not encouraging - I can't find any mention of encryption (or PGP) in XEmacs documentation, and VM doesn't seem to do anything at all with my encrypted test messages (but then, my whole setup of VM is suspect)... Do I have a chance? Bye Vasek

2 participants
1 comments

TAGS resolving to .c files in Windows 22 years, 11 months

vinnie_shelton＠notes.teradyne.com

Has anyone else noticed that, in a native Win2K build ("XEmacs 21.4 (patch 6) \"Common Lisp (Windows)\" [Lucid] (i586-pc-win32) of Tue Dec 04 2001 on SHELTCPX"), attempting to find a tag which resolves to a .c file gets an error? Here's what I did: xemacs -vanilla M-. -> e:\acs\build\xemacs-windows-21.4-2001-12-04\TAGS -> gethash --> N ==> Wrong type argument: number-char-or-marker-p, 995\,30731 Here's the backtrace: Signaling: (wrong-type-argument number-char-or-marker-p 995\,30731) find-tag-internal("gethash") #<compiled-function (tagname &optional other-window) "...(80)" [result next tagname local-find-tag-hook find-tag-hook tag-buf find-tag-internal (find-tag find-tag-other-window tags-loop-continue) push-tag-mark pop-to-buffer t switch-to-buffer push-mark run-hooks find-tag nil tag-point last-command other-window tags-loop-scan tags-loop-operate] 4 ("e:\\XEmacs-21.4\\lisp\\etags.elc" . -16542) (byte-code "«'Á?ÂÃ!ÄD?" [current-prefix-arg ... find-tag-tag "Find tag: " nil] 2)>("gethash" nil) call-interactively(find-tag) Here's the relevant entry in the TAGS file: DEFUN ("gethash", Fgethash,gethash,995,30731 Which seems OK. Now it turns out that elhash.c is loaded successfully, but it never gets made into the current buffer, nor is the jump to the proper place successful. Strangely enough, finding tags in .el files works. I'll try to investigate if I can find the time. - vin

3 participants
2 comments

[Bug XEmacs-21.4.5] regexp matching won't always work 22 years, 11 months

Christoph Wedler

Hi, I got the following strange behaviour in the regexp matching on both XEmacs/Mule on HP-UX and XEmacs on Windows-NT: the only difference is \$ab\\|dg\$\\| vs ab\\|dg\\| at the beginning... (let ((re "\$\\(\\.\\(\\(ab\\|dg\$\\|[ch]\$\\+\\+\\|pp\\|c\\|h\\|xx\$?\\|tex\$i\\(nfo\$?\\)?\\|el\\|emacs\\|l\\|lsp\\|p\\|java\\|f\$90\\|77\\|or\$?\\|ada\\|p[lm]\\|tcl\\|m\\|scm\\|pm\\|py\\|g\\|s?html\\|ma?k\\)\\)\\|\$[Mm]akefile\\(\\.in\$?\\)\\)$")) (cons (string-match re "sb-image.el") (string-match re "sb-ant.el"))) => (nil . 6) (let ((re "\$\\(\\.\\(ab\\|dg\\|[ch]\\(\\+\\+\\|pp\\|c\\|h\\|xx\$?\\|tex\$i\\(nfo\$?\\)?\\|el\\|emacs\\|l\\|lsp\\|p\\|java\\|f\$90\\|77\\|or\$?\\|ada\\|p[lm]\\|tcl\\|m\\|scm\\|pm\\|py\\|g\\|s?html\\|ma?k\\)\\)\\|\$[Mm]akefile\\(\\.in\$?\\)\\)$")) (cons (string-match re "sb-image.el") (string-match re "sb-ant.el"))) => (8 . 6) BTW, I'll submit a patch to Eric Ludlam for speedbar-0.14beta2... (basically to get my patch to XEmacs from 1999-08-26 back -- my own fault not to realize the "probably"...). - Christoph

2 participants
1 comments

[Bug? XEmacs-21.4.5] default-gutter-visible-p 22 years, 11 months

Christoph Wedler

Hi, I really like the new gutter stuff. Apart from some minor redisplay problem, I have the following problem: I don't want the gutter in the speedbar frame, so I thought I just have to put default-gutter-visible-p nil into `speedbar-frame-plist' as there is default-toolbar-visible-p nil to disable the toolbar there. This indeed makes the gutter invisible, but not for a long time. After changing the window or the buffer in the main frame, the buffer tabs will be displayed in a gutter of the speedbar frame... BTW, I'll submit a patch to Eric Ludlam for speedbar-0.14beta2... which include the above non-gutter-entry. - Christoph

4 participants
3 comments

[Proposed Patch] button2 on ISO8:T in modeline toggles eol type 22 years, 11 months

Christoph Wedler

The following message is a courtesy copy of an article that has been posted to comp.emacs.xemacs as well. Hi, for some time now, I have some coding in my ~/.emacs (appended), which is IMHO quite useful if you work under both Unix and Windows: button2 on the multibyte extension status in the modeline (ISO8, ISO8:T etc) toggles the end-of-line type of the current buffer, i.e., toggles between Unix- (LF) and Dos- (CRLF) line endings. If this is considered generally useful, I would send corresponding patches for coding.el and code-files.el. I need some feedback whether the patches should also include the following patches for command `toggle-buffer-file-coding-system' which does (or should do) the same: 1. Also toggles eol type for new files (technical: switches from eol type nil to crlf, of course not for coding systems like binary). 2. Sets buffer as modified (if not read-only). Why do you want to toggle the eol type if you don't want to save the buffer? BTW, any possibility to get ":dos" instead of ":T" in the modeline? - Christoph (defvar modeline-coding-system-extent (make-extent nil nil) "Extent covering the %C specification of the string in the modeline.") (defvar modeline-coding-system-map (let ((map (make-sparse-keymap))) (define-key map 'button2 'mouse-toggle-buffer-file-coding-system) map)) (set-extent-face modeline-coding-system-extent 'modeline-mousable-minor-mode) (set-extent-property modeline-coding-system-extent 'help-echo "button2 toggles the buffer's end-of-line type") (set-extent-keymap modeline-coding-system-extent modeline-coding-system-map) ;; From mule/mule-init.el, now coding.el: (if (or (featurep 'mule) (featurep 'file-coding)) (let ((modeline (cdr (default-value 'modeline-format)))) ; w/o initial "" (setq-default modeline-format (cons "" (cons (cons modeline-coding-system-extent 'modeline-multibyte-status) (if (eq (car-safe modeline) ; coding.el 'modeline-multibyte-status) (cdr modeline) modeline)))))) (defun mouse-toggle-buffer-file-coding-system (event) (interactive "e") (save-excursion (set-buffer (event-buffer event)) ;; Next 9 lines: like `toggle-buffer-file-coding-system', but also toggles ;; if none of the end-of-line variants is the current one (assuming -unix), ;; should be probably part of `toggle-buffer-file-coding-system' (let ((eol-type (coding-system-eol-type buffer-file-coding-system))) (setq buffer-file-coding-system (subsidiary-coding-system (coding-system-base buffer-file-coding-system) (cond ((eq eol-type 'lf) 'crlf) ((eq eol-type 'crlf) 'lf) ((eq eol-type 'cr) 'lf) (t 'crlf))))) ;; Next line should be part of `toggle-buffer-file-coding-system' (or buffer-read-only (set-buffer-modified-p t)) (redraw-modeline)))

5 participants
8 comments

Syntax table issues 22 years, 11 months

Matt Tucker

I'm working this week and next on improving the recent (or not-so-recent) syntax table changes in an effort to get them ready for a 21.4 release. I've heard for awhile that the syntax table performance in XEmacs has been lacking since the patch was applied, but I haven't been able to get enough details to actually track down the problem. If anyone can give me any information on a _reproducible_ case (preferably with -vanilla enabled) that illustrates bugs or poor performance with the syntax table code, I'd appreciate it tremendously. Thanks very much for your time and patience.

2 participants
1 comments

Re: interface improvement (?) 22 years, 11 months

Doug Alcorn

Mats Lidell <matsl(a)contactor.se> writes: > I have instead used shift-clicking with mouse-3 on the mode-line to > kill the buffer. (Implemented with the help of hyperbole.) This > gives IMHO greater control and less risk of making a mistake. Can you post details? I think making a mousable way to kill the current buffer is a Good Thing; however, you're right that putting it on the toolbar is a little risky. It's a good idea to put it on the modeline. Perhaps you're solution should become part of the standard xemacs distribution (with an option of course). -- (__) Doug Alcorn (mailto:doug＠lathi.net http://www.lathi.net) oo / PGP 02B3 1E26 BCF2 9AAF 93F1 61D7 450C B264 3E63 D543 |_/ If you're a capitalist and you have the best goods and they're free, you don't have to proselytize, you just have to wait.

6 participants
8 comments

RE: [Bug? XEmacs-21.4.5] default-gutter-visible-p 22 years, 11 months

Wedler, Christoph

> At 11:09 AM 12/6/01 -0500, Paul Krause wrote: >> This is the best way to get rid of the gutter in tools like speedbar and >> ediff [...] Thanks Paul and Andy for your hints. But anyway, I think the current XEmacs-21.4.5 behaviour is buggy. Since I want to send Eric a patch for speedbar (and this shouldn't include temporarily necessary bug workarounds), I want to know how to make a gutter-less speedbar frame assuming a bug-free gutter display, i.e., I want to make sure that the buggy behaviour is a bug in the implementation and not the specification ;-) I think the visibility of the gutter should be controlled the same way as the visibility of the toolbar, and indeed, there are the 5 corresponding specifiers for this {default,left,right,top,bottom}-XXX-visible-p. But unfortunately, I cannot produce a gutter-free speedbar with speedbar-frame-plist containing `default-gutter-visible-p nil' (as I can produce a toolbar-less speedbar containing `default-toolbar-visible-p nil'). Adding this produces a behaviour which is even stranger than I wrote in my first mail: 1. First, this will result in a speedbar not containing the buffer tabs but reserving space for the gutter. 2. If I also add `top-gutter-visible-p nil' to speedbar-frame-plist, I get the behaviour described in my first mail: no gutter (no space) but only if the buffer where speedbar is invoked from is active; in others, I see the buffer tabs. 3. If I close the speedbar, delete `top-gutter-visible-p nil' from speedbar-frame-plist, I still have the behaviour described in 2. - Christoph

1 participants
0 comments

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

XEmacs-Beta December 2001