July 1998 - XEmacs-Beta - XEmacs Mailing List Archives

21.0-b47 "Uzbek Black": More ielm breakage 26 years, 4 months

Gregory Neil Shapiro

After fixing the startup problem in ielm, I found one other problem. Hitting return on a previously entered command yields an error: Signaling: (wrong-type-argument stringp nil) string-match("\\`[ \n]+\\'" nil) (or (string= string "") (string-match "\\`[ \n]+\\'" string)) ) ielm-is-whitespace(nil) (not (ielm-is-whitespace ielm-string)) ) (if (not (ielm-is-whitespace ielm-string)) (progn (condition-case err ... ...) (if ielm-error-type nil ...) (if ... nil ... ...) (goto-char ielm-pmark) (if ... ...) (if ielm-error-type ... ... ... ...) (setq ielm-output ...))) ) (let (ielm-form ielm-pos ielm-result ielm-error-type (ielm-output "") (ielm-wbuf ielm-working-buffer) (ielm-pmark ...)) (if (not ...) (progn ... ... ... ... ... ... ...)) (setq ielm-output (concat ielm-output ielm-prompt)) (comint-output-filter (ielm-process) ielm-output)) ) ielm-eval-input(nil) (let ((buf ...) ielm-input) (comint-send-input) (ielm-eval-input ielm-input)) ) ielm-send-input() (if (and (< ... 1) (not ...)) (ielm-send-input) (if (and ielm-dynamic-multiline-inputs ...) (save-excursion ... ...)) (newline-and-indent)) ) (let ((state ...)) (if (and ... ...) (ielm-send-input) (if ... ...) (newline-and-indent))) ) (if ielm-dynamic-return (let (...) (if ... ... ... ...)) (newline)) ) (lambda nil "Newline and indent, or evaluate the sexp before the prompt.\nComplete sexps are evaluated; for incomplete sexps inserts a newline\nand indents. If however `ielm-dynamic-return' is nil, this always\nsimply inserts a newline." (interactive) (if ielm-dynamic-return (let ... ...) (newline)))() call-interactively(ielm-return) To reproduce: 1. Start ielm (M-x ielm) 2. At the first "ELISP>" prompt, enter something, for example: (+ 1 1) 3. The result will be printed and you will get another "ELISP>" prompt. Hit C-p twice so you are back on the line that you entered on step 2. 4. Hit return. Here is a patch which fixes the problem: --- ielm.el Sat Jul 11 23:44:59 1998 +++ ielm.el~ Sat Jul 11 23:29:57 1998 ＠＠ -283,8 +283,7 ＠＠ (ielm-output "") ; result to display (ielm-wbuf ielm-working-buffer) ; current buffer after evaluation (ielm-pmark (ielm-pm))) - (if (and (not (null ielm-string)) - (not (ielm-is-whitespace ielm-string))) + (if (not (ielm-is-whitespace ielm-string)) (progn (condition-case err (let (rout) ＠＠ -351,8 +350,7 ＠＠ (setq ielm-:: ielm-:) (setq ielm-: ielm-result)) (setq ielm-output (concat ielm-output "\n")))) - (if (not (null ielm-string)) - (setq ielm-output (concat ielm-output ielm-prompt))) + (setq ielm-output (concat ielm-output ielm-prompt)) (comint-output-filter (ielm-process) ielm-output))) ;;; Process and marker utilities

1 participants
1 comments

21.0-b47 "Uzbek Black": ielm doesn't work 26 years, 4 months

Gregory Neil Shapiro

Trying to start ielm results in the lisp backtrace below. A possible fix is attached. Signaling: (setting-constant : nil) (setq : nil) ) inferior-emacs-lisp-mode() (save-excursion (set-buffer (get-buffer-create "*ielm*")) (inferior-emacs-lisp-mode)) ) (if (comint-check-proc "*ielm*") nil (save-excursion (set-buffer ...) (inferior-emacs-lisp-mode))) ) (lambda nil "Interactively evaluate Emacs Lisp expressions.\nSwitches to the buffer `*ielm*', or creates it if it does not exist." (interactive) (if (comint-check-proc "*ielm*") nil (save-excursion ... ...)) (pop-to-buffer "*ielm*"))() call-interactively(ielm) command-execute(ielm t) execute-extended-command(nil) call-interactively(execute-extended-command) recursive-edit() byte-code("..." [pop-to-buffer debugger-buffer erase-buffer t 50 print-length print-escape-newlines standard-output backtrace debugger-mode re-search-forward "\n[* ] debug(" 1 debugger-reenable debugger-args (lambda debug) "Entering:\n" debug backtrace-debug 3 delete-char ?\* 0 exit "Return value: " debugger-value prin1 ?\n ?\ error "Signaling: " "Beginning evaluation of function call form:\n" nil message "" buffer-read-only inhibit-trace recursive-edit] 3) debug(error (undefined-keystroke-sequence [#<keypress-event control-X> #<keypress-event control-A>])) Here is the patch: --- xemacs/xemacs-21.0-b47/packages/lisp/xemacs-devel/ielm.el Sun Jan 25 20:24:59 1998 +++ ./ielm.el Sat Jul 11 23:29:57 1998 ＠＠ -301,15 +301,15 ＠＠ (if (ielm-is-whitespace (substring ielm-string ielm-pos)) ;; need this awful let convolution to work around ;; an Emacs bug involving local vbls and let binding - (let ((:save :) - (::save ::) - (:::save :::)) + (let ((ielm-:save ielm-:) + (ielm-::save ielm-::) + (ielm-:::save ielm-:::)) (save-excursion (set-buffer ielm-working-buffer) (condition-case err - (let ((: :save) - (:: ::save) - (::: :::save) + (let ((ielm-: ielm-:save) + (ielm-:: ielm-::save) + (ielm-::: ielm-:::save) (ielm-obuf (current-buffer))) (setq ielm-result (eval ielm-form)) (setq ielm-wbuf (current-buffer)) ＠＠ -346,9 +346,9 ＠＠ (setq ielm-output (concat ielm-output "*** " ielm-error-type " *** ")) (setq ielm-output (concat ielm-output ielm-result))) ;; There was no error, so shift the ::: values - (setq ::: ::) - (setq :: :) - (setq : ielm-result)) + (setq ielm-::: ielm-::) + (setq ielm-:: ielm-:) + (setq ielm-: ielm-result)) (setq ielm-output (concat ielm-output "\n")))) (setq ielm-output (concat ielm-output ielm-prompt)) (comint-output-filter (ielm-process) ielm-output))) ＠＠ -386,9 +386,9 ＠＠ * \\[comint-dynamic-complete] completes Lisp symbols (or filenames, within strings), or indents the line if there is nothing to complete. -During evaluations, the values of the variables `:', `::', and `:::' -are the results of the previous, second previous and third previous -evaluations respectively. +During evaluations, the values of the variables `ielm-:', `ielm-::', and +`ielm-:::' are the results of the previous, second previous and third +previous evaluations respectively. The current working buffer may be changed (with a call to `set-buffer', or with \\[ielm-change-working-buffer]), and its value ＠＠ -436,12 +436,12 ＠＠ (setq fill-paragraph-function 'lisp-fill-paragraph) ;; Value holders - (setq : nil) - (make-local-variable ':) - (setq :: nil) - (make-local-variable '::) - (setq ::: nil) - (make-local-variable ':::) + (setq ielm-: nil) + (make-local-variable 'ielm-:) + (setq ielm-:: nil) + (make-local-variable 'ielm-::) + (setq ielm-::: nil) + (make-local-variable 'ielm-:::) ;; font-lock support (make-local-variable 'font-lock-defaults)

1 participants
0 comments

21.0-b47 "Uzbek Black": Disappearing cursor 26 years, 4 months

Gregory Neil Shapiro

An xemacs session running under an xterm loses the cursor on reverse isearches. To reproduce: - Start xemacs with: xemacs -q -nw - In a new buffer, type: a: b: c: - With the cursor after the 'c:', type C-r followed by ':' and your cursor should disappear so you have no idea where the isearch is matching. This is happening under Solaris 2.6 running under OpenWindows. uname -a: SunOS monkeyboy.sendmail.org 5.6 Generic_105181-06 sun4m sparc ./configure '--site-prefixes=/usr/local' '--with-database=dbm,berkdb' '--with-pop' '--debug' '--site-includes=/usr/local/bind/include' '--site-libraries=/usr/local/bind/lib' XEmacs 21.0 "Uzbek Black" configured for `sparc-sun-solaris2.6'. Where should the build process find the source code? /usr/local/src/xemacs/xemacs-21.0-b47 What installation prefix should install use? /usr/local What operating system and machine description files should XEmacs use? `s/sol2.h' and `m/sparc.h' What compiler should XEmacs be built with? gcc -g -O3 -Wall -Wno-switch Should XEmacs use the GNU version of malloc? yes Should XEmacs use the relocating allocator for buffers? yes What window system should XEmacs use? x11 Where do we find X Windows header files? /usr/dt/include /usr/openwin/include Where do we find X Windows libraries? /usr/dt/lib /usr/openwin/lib Additional header files: /usr/local/bind/include Additional libraries: /usr/local/bind/lib Additional prefixes: /usr/local Runtime library search path: /usr/local/pilot/lib:/usr/local/bind/lib:/usr/local/samba/lib:/usr/local/lib:/usr/openwin/lib:/usr/dt/lib:/usr/share/lib:/lib:/usr/ccs/lib:/opt/SUNWspro/lib:/usr/lib:/usr/local/ssl/lib Compiling in support for XAUTH. Compiling in support for XPM images. Compiling in support for X-Face message headers. Compiling in support for JPEG image conversion. Compiling in support for PNG image conversion. Compiling in support for TIFF image conversion. Compiling in native sound support. Compiling in support for Berkeley DB. Compiling in support for DBM. Compiling in support for ncurses. Compiling in support for CDE. Compiling in support for ToolTalk. Compiling in EXPERIMENTAL support for Drag'n'Drop ( CDE ). Compiling in support for proper session-management. Using Lucid menubars. Using Lucid scrollbars. Using Motif dialog boxes. Compiling in DLL support. movemail will use "dot-locking" for locking mail spool files. Using POP for mail access Compiling in extra code for debugging. Compiling in code for checking XEmacs memory usage.

1 participants
0 comments

re: [OK, sort of] Uzbek Black on linuxppc 26 years, 4 months

Nobu Toge

I meant to mean Uzbek Black, not Uzbek White. Sorry. - N.Toge

1 participants
0 comments

[OK, sort of] Uzbeck White on LinuxPPC 26 years, 4 months

Nobu Toge

This is a follow-up to my previous report concerning the build failure of Uzbeck White (b47) on LinuxPPC kernel 2.1.24. I am using gcc-2.7.2-2O, have been seeing gcc -o temacs dies with very useless status message with b47 Uzbeck White. I never saw this error with b46 Toggenburg. After some looking into, I notice that apparently this version (2.7.2-2O) of gcc on linuxppc emits some garbage when a symbol name exceeds 31 (or so) characters. As people moved from b46 to b47, some patches were applied to winslots.h. In one place a symbol that was previously called vertical-divider-draggable-p was replaced by vertical-divider-always-visible-p. Reflecting this, there is a variable called Vvertical-divider-always-visible-p (34 char long) introduced in window.c. Apparently this screws up the gcc behavior and it pukes on me when it tries to link temacs. K&R says that a C-compiler is supposed to differentiate symbol names up to 31 chars, and beyond that is up to each implementation. I don't know if the behavior above of gcc-2.7.2-2O is legal or not strictly speaking, but it sounds to me like a bug rather than an allowed feature. A quick work-around is to shorten the symbol name as I indicated in the attached patch file. But the XEmacs maintainers may as well *not* implement this patch and add an INSTALL/README warning against the use of gcc-2.7.2-2O on linuxppc. Then Uzbek White does build and run fine, including mule/canna support, if I this patch, though. Fukui san reported that egcs 1.0.3a built Uzbek White OK. I haven't tried that out myself. If I find time during this week end, I will give a try. I also recognize that there is gcc 2.8.something RPM floating around for linuxppc/mklinux, but I don't know how that one fares. Best, - Nobu Toge (toge(a)accad1.kek.jp) Index: xemacs-20.0/src/winslots.h diff -u winslots.h winslots.h --- winslots.h Sun Jul 12 12:16:50 1998 +++ winslots.h Sun Jul 12 12:19:38 1998 ＠＠ -60,7 +60,7 ＠＠ /* Spacing between outer egde of divider border and window edge */ WINDOW_SLOT (vertical_divider_spacing, EQ); /* Whether vertical dividers are always displayed */ - WINDOW_SLOT (vertical_divider_always_visible_p, EQ); + WINDOW_SLOT (vertical_divider_always_visi_p, EQ); #ifdef HAVE_SCROLLBARS /* Width of vertical scrollbars. */ Index: xemacs-20.0/src/scrollbar.c diff -u scrollbar.c scrollbar.c --- scrollbar.c Tue Jun 30 15:36:01 1998 +++ scrollbar.c Sun Jul 12 12:23:21 1998 ＠＠ -603,7 +603,7 ＠＠ completely at either window edge, we do this always, as users usually do not reposition scrollbars 200 times a second or so. Do you? */ - if (NILP (w->vertical_divider_always_visible_p)) + if (NILP (w->vertical_divider_always_visi_p)) MARK_FRAME_WINDOWS_STRUCTURE_CHANGED (XFRAME (WINDOW_FRAME (w))); else MARK_WINDOWS_CHANGED (w); Index: xemacs-20.0/src/window.c diff -u window.c window.c --- window.c Sun Jul 12 12:39:46 1998 +++ window.c Sun Jul 12 12:41:22 1998 ＠＠ -69,7 +69,7 ＠＠ Lisp_Object Vmodeline_shadow_thickness; /* Whether vertical dividers are draggable and displayed */ -Lisp_Object Vvertical_divider_always_visible_p; +Lisp_Object Vvertical_divider_always_visi_p; /* Whether a modeline should be displayed. */ Lisp_Object Vhas_modeline_p; ＠＠ -742,7 +742,7 ＠＠ return 0; /* Always if draggable */ - if (!NILP (w->vertical_divider_always_visible_p)) + if (!NILP (w->vertical_divider_always_visi_p)) return 1; #ifdef HAVE_SCROLLBARS ＠＠ -5618,8 +5618,8 ＠＠ some_window_value_changed, 0, 0); - DEFVAR_SPECIFIER ("vertical-divider-always-visible-p", - &Vvertical_divider_always_visible_p /* + DEFVAR_SPECIFIER ("vertical-divider-always-visi-p", + &Vvertical_divider_always_visi_p /* *Should XEmacs always display vertical dividers between windows. When this is non-nil, vertical dividers are always shown, and are ＠＠ -5628,12 +5628,12 ＠＠ This is a specifier; use `set-specifier' to change it. */ ); - Vvertical_divider_always_visible_p = Fmake_specifier (Qboolean); - set_specifier_fallback (Vvertical_divider_always_visible_p, + Vvertical_divider_always_visi_p = Fmake_specifier (Qboolean); + set_specifier_fallback (Vvertical_divider_always_visi_p, list1 (Fcons (Qnil, Qt))); - set_specifier_caching (Vvertical_divider_always_visible_p, + set_specifier_caching (Vvertical_divider_always_visi_p, slot_offset (struct window, - vertical_divider_always_visible_p), + vertical_divider_always_visi_p), vertical_divider_changed_in_window, 0, 0); uname -a: Linux ptaro.kek.jp 2.1.24 #122 Thu Apr 9 10:27:39 EST 1998 ppc unknown ./configure '--cflags=-g -O2 -fsigned-char' '--dynamic=yes' '--error-checking=none' '--debug=no' '--with-dialogs=athena' '--with-site-lisp' '--prefix=/usr' '--with-xfs' '--with-canna' '--with-mule' '--package-path=/usr/lib/xemacspack' '--with-x' XEmacs 21.0 "Uzbek Black" configured for `powerpc-unknown-linux'. Where should the build process find the source code? /home/toge/XEmacs/xemacs-21.0-b47 What installation prefix should install use? /usr What operating system and machine description files should XEmacs use? `s/linux.h' and `m/powerpc.h' What compiler should XEmacs be built with? gcc -g -O2 -fsigned-char Should XEmacs use the GNU version of malloc? yes Should XEmacs use the relocating allocator for buffers? yes What window system should XEmacs use? x11 Where do we find X Windows header files? /usr/X11/include Where do we find X Windows libraries? /usr/X11/lib Compiling in support for XAUTH. Compiling in support for XPM images. Compiling in support for X-Face message headers. Compiling in support for JPEG image conversion. Compiling in support for PNG image conversion. Compiling in support for TIFF image conversion. Compiling in native sound support. Compiling in support for Berkeley DB. Compiling in support for GNU DBM. Compiling in support for ncurses. Compiling in support for GPM (General Purpose Mouse). Compiling in Mule (multi-lingual) support. Compiling in XIM (X11R5+ I18N input method) support. Using raw Xlib to provide XIM support. Using XFontSet to provide bilingual menubar. Compiling in support for Canna on Mule. Compiling in support for OffiX. Compiling in EXPERIMENTAL support for Drag'n'Drop ( OffiX ). Compiling in support for proper session-management. Using Lucid menubars. Using Lucid scrollbars. Using Athena dialog boxes. Compiling in DLL support. movemail will use "dot-locking" for locking mail spool files.

1 participants
0 comments

Warning about Gnus-5.6.24, tm and XEmacs/Mule 26 years, 4 months

SL Baur

Gnus-5.6.24 breaks `binary' .elc compatibility and tm in some fashion I haven't figured out yet. The workaround is to bytecompile tm against the new Gnus sources.

1 participants
0 comments

Re: No PSGML menus on WinNT? (Re: psgml-xemacs.el needs a load hook) 26 years, 4 months

SL Baur

Adrian Aichner <aichner(a)ecf.teradyne.com> writes: >>>>>> "APA" == Adrian Aichner <aichner(a)ecf.teradyne.com> writes: > Steven, > the follwing still holds true with the latest > emacs-version "21.0 \"Uzbek Black\" XEmacs Lucid" and > psgml-version "1.0.1" > on Windows NT (native build). > This may be a race-condition. > When I > edebug-defun sgml-xemacs-get-popup-value (in psgml-xemacs.el) > then the menus DO insert their text in the HTML buffer! What does `sgml-xemacs-get-popup-value' have to do with it? I see the right mouse button bound to `sgml-tags-menu' and it appears to be functioning normally.

1 participants
0 comments

Re: No PSGML menus on WinNT? (Re: psgml-xemacs.el needs a load hook) 26 years, 4 months

Adrian Aichner

>>>>> "APA" == Adrian Aichner <aichner(a)ecf.teradyne.com> writes: Steven, the follwing still holds true with the latest emacs-version "21.0 \"Uzbek Black\" XEmacs Lucid" and psgml-version "1.0.1" on Windows NT (native build). This may be a race-condition. When I edebug-defun sgml-xemacs-get-popup-value (in psgml-xemacs.el) then the menus DO insert their text in the HTML buffer! Any hints? Adrian >>>>> "sb" == SL Baur <steve(a)xemacs.org> writes: APA> Hello Steven, hello All, APA> is anybody successfully using menus in PSGML under XEmacs 21.0 on APA> Windows NT? APA> I can only edit HTML documents using PSGML (and TEMPO) keybindings on APA> WIndows NT. The context-sensitive HTML memus do pop up, but no of the APA> m inserts any text in the buffer; even though I have a validating APA> (Next Trouble Spot returns OK) HTML 3.2 document (with DTD parsed). APA> All this under APA> XEmacs21.0 pre2, PSGML 1.0.1 APA> Any hints? I'm willing to invest-igate. -- 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: Beta names 26 years, 4 months

SL Baur

Remek Trzaska <remek(a)npac.syr.edu> writes: > Is this question still open? Yes. I'm personally not very enthusiastic about it, but the People Have Spoken and the 22 series will be named after Jazz musicians. We still need names for 21.1 (~12) and 21.2 (~20). > How about Roman emperors (50-something entries): > http://www.clark.net/pub/jltoth/roman_emperors.html > I tried the Greek gods as well, but the longest list I was able to find > contained only *32* of them, including Sirens, some Heroes, etc. ;-( Hmm, Greek gods would kind of cool.

2 participants
1 comments

Re: Default OffiX support to off 26 years, 4 months

Olivier Galibert

[This may be of interest to people on xemacs-beta too] Hrvoje Niksic <hniksic(a)srce.hr> writes: > Offhand, this patch looks wrong. Hrvoje, there is a part of configure.in you seem to have missed: -*) dnl If FOO is a boolean argument, --FOO is equivalent to dnl --FOO=yes. Otherwise, the value comes from the next dnl argument - see below. opt=`echo '' $arg | sed -e 's:^ ::' -e 's:^-*$.*$$:\1:'` val="yes" valomitted=yes Boolean variables in configure.in are actually tristate, "yes", "no" or empty. Empty means that the user did not specify anything in the configure command line. Specifying "--with-xxx" without a parameter will give a "yes" in $with_xxx. Specifying "--without-xxx" gives "no" in $with_xxx. -test -z "$with_offix" && with_offix=yes +test -z "$with_offix" && with_offix=no This has to be interpreted as changing "if the user didn't specify anything, default to yes" to "if the user didn't specify anything, default to no". Exactly the effect intended. OG.

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 July 1998