August 1998 - XEmacs-Beta - XEmacs Mailing List Archives

Re: X-error in get_mouse_position 27 years

Gunnar Evermann

Didier Verna <verna(a)inf.enst.fr> writes: > It's generally a bad idea to use XSync for something else than > debugging, because it synchronizes the application with the server > itself. XSync will return only when the X server has finished processing all > its requests. The consequence is that it can considerably slow down the > application, because you'll be blocked waiting for all events to be processed, > even those completely unrelated to your problem. OK, so XSync is utterly *evil* :-) > When a synchronization problem occurs, it's almost always the > application's fault (a lack of consistency checking) and always a bad idea to > synchronize with the server. The better solution is to check internally for > consistency. Here it means checking that this window still belongs to a frame > before enquiring the server about an XPointer information. but, how??? x_any_window_to_frame doesn't cut it. In a vanilla XEmacs, if the pointer is in the *scratch* buffer I need to traverse some steps down the Window hierarchy before I get the frame from x_any_window_to_frame. For the child_window returned by XQueryPointer I only get NIL :-( Is this a bug in x_any_window_to_frame or is that only supposed to work for the leaf windows and scrollbars and stuff??? Gunnar

1 participants
0 comments

Re: Bug: Customize open-up.png glyph is getting confused with W3 27 years

SL Baur

SL Baur <steve(a)xemacs.org> writes in xemacs-beta(a)xemacs.org: > I don't see offhand where Custom is doing its glyph searching, but it > should prefer glyphs in etc/custom above all else. Somehow it is > picking up open-up.xpm in etc/w3. O.K. I figured this out. Something needs to change. The function `widget-glyph-find' in wid-edit.el uses a suboptimal algorithm to find glyphs. It puts `widget-glyph-directory' or (locate-data-directory "custom") at the front of a search path, then adds `data-directory-list'. This would be fine except that it then proceeds to search for a list of formats across the whole search path with priority going first to .xpm, then to .gif, .png, .jpg, and .xbm. NOTE 1: I would argue that .png should go to the head, followed by .xpm, .gif, .jpg and .xbm. The end result being that "open-up.xpm" in etc/w3 is found ahead of "open-up.png" in etc/custom when `widget-glyph-find' is called: (widget-glyph-find "open-up" "blahblah") even though etc/custom is supposedly first in the search path. Noting also that the `widget-glyph-cache' only caches the basename, not the path, it seems like all glyph names across XEmacs and Lisp extensions must be unique. I don't believe this is a restriction we can or should enforce. William, Hrvoje, thoughts?

1 participants
0 comments

Re: Macros: Scheme vs Common Lisp 27 years

Clemens Heitzinger

sperber(a)informatik.uni-tuebingen.de (Michael Sperber [Mr. Preprocessor]) writes: > >>>>> "Craig" == Craig Lanning <CraigL(a)internetx.net> writes: > > Craig> At 10:01 AM 8/17/98 +0200, Michael Sperber [Mr. Preprocessor] wrote: > > >> This indeed shows the CONS example to be poor. The point remains the > >> same, however: What if we bind a variable bound in a macro. Say, > >> *features*? What if a temporary identifier in a macro definition > >> shadows a binding in a use? > > Craig> I think I see what you're driving at, but I'm not sure that it's as > Craig> important as it's being made out to be. It's easy enough to use > Craig> MAKE-SYMBOL to create safe variable names for the macro and still > Craig> maintain readability. > > Sure, but there's no guarantee that a given macro has hygienic > behavior. Generally, maintaining hygiene this way is pretty tedious, > which is why programmers usually don't go to the necessary lengths. > > Also, a macro might refer to bindings whose names are free in the > macro definition. There's no guarantee that you'll get the same > bindings for a use that you referred to at the point of definition. Maintaining hygiene is *not* pretty tedious. If you want your CL macro to be hygienic, use (defmacro with-hygienic-variables (var &body body) `(let ,(loop for var in vars collect `(,var (make-symbol ,(symbol-name var)))) ,＠body)) This is due to a posting from Erik Naggum to comp.lang.lisp, and it implements what Craig is talking about some paragraphes above. Therefore, in situations where non-hygienic macros have advantages, use a non-hygienic macro. When you want hygienic variables, use the macro above (or write your own define-hygienic-macro, which uses the above). It is as simple as that. Yours, -- Clemens Heitzinger Student der Techn. Mathematik, Technische Universität Wien, Österreich Vienna University of Technology, Austria

1 participants
0 comments

Bug: Customize open-up.png glyph is getting confused with W3 27 years

SL Baur

I don't see offhand where Custom is doing its glyph searching, but it should prefer glyphs in etc/custom above all else. Somehow it is picking up open-up.xpm in etc/w3. Attached is a *Customize Group ...* buffer showing the wrong glyph in use.

1 participants
0 comments

Re: [bug] Editting gzipped files broken. 27 years

SL Baur

Karl M Hegbloom <karlheg(a)inetarena.com> writes in xemacs-beta(a)xemacs.org: > I get an error about a bad CRC when I try to view a manual page that > I just editted. It's a compressed file, and I've got XEmacs set up > to do the auto compression thing. It's compiled with file > codeing... could it be an interaction with that? Yes. All decompression modes are broken wrt to coding-systems. crypt.el is probably the least broken (it will work if the decompressed file is compatible with the default coding system and doesn't require any conversion). jka-compr is the most broken.

1 participants
0 comments

Re: 28 bit ought to be enough for anyone. 27 years

SL Baur

Kenichi Okuyama <okuyama(a)trl.ibm.co.jp> writes in xemacs-beta(a)xemacs.org: >>>>>> "sb" == SL Baur <steve(a)xemacs.org> writes: sb> With the old GNU malloc in XEmacs, the mmapping sb> version of the relocating allocator can be used. > So... world is moving faster then I thought :p ;; The mmapping version of the relocating allocator has been default ;; with the old GNU malloc since XEmacs 19.14. :-) sb> What is PHKmalloc? If it's portable and faster, I have no problem sb> replacing the old GNU malloc with it, or adding it as an additional sb> option. > It's default malloc routine used on FreeBSD since 2.2. (current > stable version is 2.2.7) Authored by Poul-Henning Kemp. I don't know > whether it's faster nor portable than Doug Lea malloc. As long as > what I read what's written above, I beleave their basic idea are > same. > # That means, on both FreeBSD and Linux, we simply can use default > # malloc. That's quite good news. So long as we dump, you will need to do some work to get an mmapping system malloc to work. mmap is poison prior to dumping so it must be disabled in temacs. There are well-defined places to do this. Grep for DOUG_LEA_MALLOC in the sources. The calls to mallopt(M_MMAP_MAX, %d) turn mmapping off if the second param is 0, and turn it on if the second param is non-zero.

1 participants
0 comments

Re: gnus mail borken under cygwin 27 years

Adrian Aichner

>>>>> "Andy" == Andy Piper <andyp(a)parallax.co.uk> writes: Andy> I use smtpmail for sending mail and gnus nnfloder for Andy> reading mlist stuff. If I do M-x mail and send some mail, Andy> all works fine. If I do this from gnus it says it is sending Andy> it , but I don't get the usual smtp exchange in the echo Andy> area and the mail is not actually sent (my apologies to Andy> those who might have thought I was being more antisocial Andy> than usual). Andy, the crucial bit must be somewhere in there since I was able to send mail from gnus (see Message-ID: <ur9y8692v.fsf(a)ecf.teradyne.com> sent to you only). Good Luck, Adrian 35 lines matching mail\|message\|smtp\|gnus in buffer .emacs. 4: '(gnus-play-startup-jingle nil) 9: '(nnfolder-directory "~/Mail" t) 12: '(nnmail-treat-duplicates (quote warn)) 17: '(display-time-mail-balloon-gnus-split-width 25) 28: '(gnus-server-alist (quote (("" nnfolder "" (nnfolder-directory "~/Mail/") (nnfolder-active-file "~/Mail/active") (nnfolder-get-new-mail nil) (nnfolder-inhibit-expiry t)))) t) 29: '(gnus-async-prefetch-article-p (quote gnus-async-unread-p)) 31: '(display-time-mail-balloon-show-gnus-group t) 32: '(gnus-read-active-file (quote some)) 41: '(gnus-activate-foreign-newsgroups t) 46: '(mail-host-address "ecf.teradyne.com" t) 47: '(gnus-suppress-duplicates t) 65: '(send-mail-function (quote smtpmail-send-it) t) 70: '(display-time-form-list (quote (load mail))) 71: '(gnus-activate-level 1) 75: '(display-time-ignore-read-mail nil) 84: '(cvs-shell "/Cygnus/B19/H-i386-cygwin32/bin/sh.exe" t) 86: '(gnus-use-long-file-name nil) 90: '(toolbar-news-reader (quote gnus)) 91: '(gnus-asynchronous t) 99: '(mime-edit-split-message nil t) 100: '(toolbar-mail-reader (quote gnus)) 105: '(message-send-mail-function (quote smtpmail-send-it)) 110: '(mime-editor/split-message nil t) 117: '(gnus-summary-ignore-duplicates t) 118: '(message-signature-file "~/.signature") 119: '(nnfolder-active-file "~\\Mail\\active" t) 122: '(display-time-mail-balloon-subject-width 30) 130: '(user-mail-address "aichner(a)ecf.teradyne.com" t) 131: '(query-user-mail-address nil) 132: '(gnus-check-new-newsgroups (quote ask-server)) 136: '(message-default-organization "Teradyne GmbH" t) 137: '(exec-path (quote ("c:/CYGNUS/B19/H-i386-cygwin32/bin/" "c:/usr/local/bin/" "c:/PERL/bin/" "c:/WINNT/system32/" "c:/WINNT/" "c:/cvs/" "c:/PROGRA~1/Teradyne/IG-XL/bin/" "c:/program files/devstudio/sharedide/bin/ide/" "c:/program files/devstudio/sharedide/bin/" "c:/program files/devstudio/vc/bin/" "\"C:/PROGRAM FILES/XNFS\"/" "c:/export/home/tmp/CVSroot/xemacs-20/lib-src/"))) 138: '(gnus-message-archive-group nil) 141: '(smtpmail-default-smtp-server "engine.ecf.teradyne.com") 143: '(message-signature t) Andy> I'm guessing others aren't seeing this - so maybe its some Andy> interaction with smtpmail? Andy> andy Andy> --------------------------------------------------------------------------- Andy> " .sigs are like your face - rarely seen by you and uglier Andy> than you think" Dr Andy Piper, Technical Architect, Parallax Andy> Solutions Ltd mail: andyp(a)parallax.co.uk web: Andy> www.parallax.co.uk/~andyp -- Adrian Aichner Applications Engineer Teradyne GmbH Semiconductor Test Group Telephone +49/89/41861(0)-208 Dingolfinger Strasse 2 Fax +49/89/41861-217 D-81673 MUENCHEN E-mail aichner(a)ecf.teradyne.com

1 participants
0 comments

Re: Macros: Scheme vs Common Lisp 27 years

Michael Sperber

>>>>> "Craig" == Craig Lanning <CraigL(a)internetx.net> writes: Craig> You can build any size macro with relative ease by building it incrementally Craig> and using MACROEXPAND to check it. Sure. I don't want to write macros by trial, however. I want to write them by design. Craig> Macros do not really reference bindings. They alter the lexical environment Craig> by making potentially complex substitutions. >> >> That's what they do in Common Lisp. In Scheme, it's different. Craig> How are they different in Scheme? Macros in Scheme obey the exact same scoping rules as the rest of the language. In: (define-syntax foo (syntax-rules () ((foo) bar))) BAR always refers to the binding of BAR lexically visible in the macro definition. I'm not sure I understood your question right. -- Cheers =8-} Chipsy Friede, Völkerverständigung und überhaupt blabla

1 participants
0 comments

Re: Macros: Scheme vs Common Lisp 27 years

Michael Sperber

Hi Craig, Many thanks for the package system explanations. Indeed, all this going back-and-forth between symbols and names confuses me. >>>>> "Craig" == Craig Lanning <CraigL(a)internetx.net> writes: Craig> It is typically considered bad programming style (in any language) to Craig> make common use of global variables. This is not true. It is bad programming style to use *non-encapsulated* global variables. Of course, a functional programming style is usually preferable. This, however, excludes dynamic binding as well. Craig> One of the main benefits of dynamically binding global Craig> variables is to use them as global parameters which can be Craig> modified within an execution thread and appear as global values Craig> to all the code in the thread, but be completely invisible to Craig> all other threads. I'm not arguing against the benefits of dynamic binding. I am, however, arguing that it is bad design in Common Lisp that a single declaration may inadvertently change the meaning of all bindings to the same naming, even though they may be totally unrelated. >> Of course, this mechanism is necessary to keep compatibility with >> older Lisps. From a fresher perspective, I conjecture it would be >> better to make dynamic scoping explicit everywhere by introducing >> extra binding and reference operators. Craig> I cannot remember ever needing a nondynamic global variable. You're missing the point. The point is that introducing a dynamic global variable changes the meaning of program constructs which don't have anything to do with the global variable at all. -- Cheers =8-} Chipsy Friede, Völkerverständigung und überhaupt blabla

1 participants
0 comments

Re: Macros: Scheme vs Common Lisp 27 years

Michael Sperber

>>>>> "Craig" == Craig Lanning <CraigL(a)internetx.net> writes: Craig> If the macro is expanded at top level then there are no lexical bindings Craig> between it and whatever binding it may reference at top level. Your Craig> revised example had a lexical binding between the macro use and top level. Yes, that's exactly the point. Craig> That still doesn't explain how using macroexpand limits the Craig> complexity of the macros that you can write. Using in itself doesn't. However, if I *need* macroexpand to write the macro and check the quoting/unquoting and the hygiene, there is a limit at least to my perception as to what size macro expansions I can still grok. Maybe it gets better with practice. Craig> Macros do not really reference bindings. They alter the lexical environment Craig> by making potentially complex substitutions. That's what they do in Common Lisp. In Scheme, it's different. -- Cheers =8-} Chipsy Friede, Völkerverständigung und überhaupt blabla

2 participants
1 comments

2025

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 August 1998