March 2000 - XEmacs-Beta - XEmacs Mailing List Archives

Re: Make gutter redisplay less aggressive 24 years, 8 months

Hrvoje Niksic

Andy Piper <andy(a)xemacs.org> writes: > There were several reasons why redisplay optimisations weren't > kicking in for gutters. This should fix the worst two offenders - > extent changes in the buffer and having more than one gutter on a > frame. This should cause the general flickering people have been > seeing. I would be interested to know if it doesn't The window-change-related flicker is still there. The worst offenders are still: * Frame change, using the WM. * Window change, using C-TAB. * Gnus. Many things in Gnus. For instance, displaying an article, scrolling down by page (SPC in the summary buffer), scrolling down by line (RET in the summary buffer), etc.

2 participants
1 comments

Re: Can we null (or nuke) `format-alist'? 24 years, 8 months

Hrvoje Niksic

"Stephen J. Turnbull" <turnbull(a)sk.tsukuba.ac.jp> writes: > While we're on the subject of `find-file' opacities, is there a good > reason not to make `format-alist' nil by default? The only things > left there by default are the "image/*" formats and "text/enriched". image/* is nuked in 21.2. > "text/enriched" is nearly redundant if you're using a modern MUA > like VM or Gnus, or use tm or SEMI. Probably so. Having examined the current contents of format-alist, I wouldn't mind changing the default to nil.

1 participants
0 comments

Re: Intent to implement Emacs 20 find-file-literally 24 years, 8 months

Jan Vroonhof

Martin Buchholz <martin(a)xemacs.org> writes: > Is there some reason we did not implement this? Is this function > perhaps ideologically blasphemous? I am fairly sure we had it some point. It might also be a good idea to apply the core part of http://www.xemacs.org/list-archives/xemacs-patches/199906/msg00050.html It will work without the EFS part. It would be OK to sync things with a new EFS release, but 7 months is a bit too much of delay. Jan

1 participants
0 comments

Re: A real database for Emacs 24 years, 8 months

Jan Vroonhof

SL Baur <steve(a)beopen.com> writes: > > 1. Can we have this as a module pleeeze? > > Is the module code stable now? Yes. > And, are modules built by default? No. But that is a problem that needs to fixed anyway. > Personally, I'd prefer it not being a module by default, because libpq > is lightweight and linked shared by default and the Lisp binding is > lightweight (~13k with full debugging enabled). Yes, but still libpq installed (there are already people on various lists and groups complaining why the debian XEmacs packages requires the ldap libraries). Of course I have nothing against having it default to non-mule on a plain configure. > asynchronous queries[1] working last night. Cursors should work for > free since they're implemented by the Postgres backend in SQL. I'll Cool... jan

1 participants
0 comments

Re: A real database for Emacs 24 years, 8 months

Jan Vroonhof

Martin Buchholz <martin(a)xemacs.org> writes: > I suggest taking a look at the DBI interface in Perl and Python. DBI > has been quite successful - there's even been a recent book published > on it. Our DB interface should be agnostic about the precise database > being connected to. So people should be able to easily add msql, > Oracle, Informix, DB2, etc... support. More importantly. If you have a standard API that does PostgreSQL there are no licence problems linking to the others. Jan

1 participants
0 comments

Re: Can we null (or nuke) `format-alist'? 24 years, 8 months

Didier Verna

Stephen J. Turnbull wrote: > Since `format-alist' is not documented anywhere that I could cross- > reference from `find-file' (I only found it because I remembered that > insert-file-contents-literally nulls a variable which seemed totally > unconnected to `find-file', so I looked there), we should either nuke > this entirely (yeah, yeah, I know, but dreams are still legal, aren't > they?) or document it. I share your dream ... -- / / _ _ Didier Verna http://www.inf.enst.fr/~verna/ - / / - / / /_/ / EPITA / LRDE mailto:didier＠epita.fr /_/ / /_/ / /__ / 14-16 rue Voltaire Tel. +33 (1) 44 08 01 77 94276 Kremlin-Bicêtre cedex Fax. +33 (1) 44 08 01 99

1 participants
0 comments

startup failure using minimal build 24 years, 8 months

Jeff Miller

cvs'd a few minutes ago.... With the following build: ./configure '--with-dialogs=no' '--without-gif' '--with-jpeg=no' '--with-ldap=no' '--with-sound=no' '--without-tty' '- -with-gpm=no' '--with-modules=no' '--with-database=no' '--with-menubars=no' '--package-path=/usr/local/lib/xemacs/packa ges-21.0' '--with-png=no' '--with-tiff=no' '--with-scrollbars=no' '--without-toolbars' '--with-widgets=no' '--with-xface=no' '--with-xpm=no' running src/xemacs fails complaining about init-gutter. I'm guessing it's just a missing ifdef type problem..... Here is the backtrace: [jmiller＠24-216-39-11 xemacs]$ src/xemacs Loading cl-macs... Loading cl-macs...done Initialization error: Symbol's function definition is void: init-gutter This probably means that XEmacs is picking up an old version of the lisp library, or that some .elc files are not up-to-date. exec-directory is "/cvs/xemacs/xemacs/lib-src/" data-directory is "/cvs/xemacs/xemacs/etc/" data-directory-list is ("/usr/local/lib/xemacs/packages-21.0/etc/" "/usr/local/lib/xemacs/packages-21.0/etc/app-defaults/" "/usr/local/lib/xemacs/packages-21.0/etc/auctex/" "/usr/local/lib/xemacs/packages-21.0/etc/bbdb/" "/usr/local/lib/xemacs/packages-21.0/etc/ediff/" "/usr/local/lib/xemacs/packages-21.0/etc/frame-icon/" "/usr/local/lib/xemacs/packages-21.0/etc/gnus/" "/usr/local/lib/xemacs/packages-21.0/etc/gnusrefcard/" "/usr/local/lib/xemacs/packages-21.0/etc/hm--html-menus/" "/usr/local/lib/xemacs/packages-21.0/etc/jde/" "/usr/local/lib/xemacs/packages-21.0/etc/message/" "/usr/local/lib/xemacs/packages-21.0/etc/mule/" "/usr/local/lib/xemacs/packages-21.0/etc/mule-doc/" "/usr/local/lib/xemacs/packages-21.0/etc/pgnus/" "/usr/local/lib/xemacs/packages-21.0/etc/psgml/" "/usr/local/lib/xemacs/packages-21.0/etc/rmail/" "/usr/local/lib/xemacs/packages-21.0/etc/smilies/" "/usr/local/lib/xemacs/packages-21.0/etc/sounds/" "/usr/local/lib/xemacs/packages-21.0/etc/start-files/" "/usr/local/lib/xemacs/packages-21.0/etc/time/" "/usr/local/lib/xemacs/packages-21.0/etc/toolbar-fancy/" "/usr/local/lib/xemacs/packages-21.0/etc/vm/" "/usr/local/lib/xemacs/packages-21.0/etc/w3/" "/usr/local/lib/xemacs/packages-21.0/etc/xemacs-base/" "/cvs/xemacs/xemacs/etc/") doc-directory is "/cvs/xemacs/xemacs/lib-src/" load-path is ("/usr/local/lib/xemacs/packages-21.0/lisp/" "/usr/local/lib/xemacs/packages-21.0/lisp/apel/" "/usr/local/lib/xemacs/packages-21.0/lisp/auctex/" "/usr/local/lib/xemacs/packages-21.0/lisp/bbdb/" "/usr/local/lib/xemacs/packages-21.0/lisp/c-support/" "/usr/local/lib/xemacs/packages-21.0/lisp/calendar/" "/usr/local/lib/xemacs/packages-21.0/lisp/cc-mode/" "/usr/local/lib/xemacs/packages-21.0/lisp/cookie/" "/usr/local/lib/xemacs/packages-21.0/lisp/debug/" "/usr/local/lib/xemacs/packages-21.0/lisp/dired/" "/usr/local/lib/xemacs/packages-21.0/lisp/edebug/" "/usr/local/lib/xemacs/packages-21.0/lisp/ediff/" "/usr/local/lib/xemacs/packages-21.0/lisp/edit-utils/" "/usr/local/lib/xemacs/packages-21.0/lisp/efs/" "/usr/local/lib/xemacs/packages-21.0/lisp/elib/" "/usr/local/lib/xemacs/packages-21.0/lisp/eudc/" "/usr/local/lib/xemacs/packages-21.0/lisp/footnote/" "/usr/local/lib/xemacs/packages-21.0/lisp/forms/" "/usr/local/lib/xemacs/packages-21.0/lisp/frame-icon/" "/usr/local/lib/xemacs/packages-21.0/lisp/fsf-compat/" "/usr/local/lib/xemacs/packages-21.0/lisp/games/" "/usr/local/lib/xemacs/packages-21.0/lisp/hm--html-menus/" "/usr/local/lib/xemacs/packages-21.0/lisp/igrep/" "/usr/local/lib/xemacs/packages-21.0/lisp/ilisp/" "/usr/local/lib/xemacs/packages-21.0/lisp/ispell/" "/usr/local/lib/xemacs/packages-21.0/lisp/jde/" "/usr/local/lib/xemacs/packages-21.0/lisp/locale/" "/usr/local/lib/xemacs/packages-21.0/lisp/mail-lib/" "/usr/local/lib/xemacs/packages-21.0/lisp/mailcrypt/" "/usr/local/lib/xemacs/packages-21.0/lisp/mh-e/" "/usr/local/lib/xemacs/packages-21.0/lisp/misc-games/" "/usr/local/lib/xemacs/packages-21.0/lisp/mule-base/" "/usr/local/lib/xemacs/packages-21.0/lisp/net-utils/" "/usr/local/lib/xemacs/packages-21.0/lisp/os-utils/" "/usr/local/lib/xemacs/packages-21.0/lisp/pcl-cvs/" "/usr/local/lib/xemacs/packages-21.0/lisp/pgnus/" "/usr/local/lib/xemacs/packages-21.0/lisp/prog-modes/" "/usr/local/lib/xemacs/packages-21.0/lisp/psgml/" "/usr/local/lib/xemacs/packages-21.0/lisp/rmail/" "/usr/local/lib/xemacs/packa ges-21.0/lisp/sh-script/" "/usr/local/lib/xemacs/packages-21.0/lisp/speedbar/" "/usr/local/lib/xemacs/packages-21.0/lisp/supercite/" "/usr/local/lib/xemacs/packages-21.0/lisp/texinfo/" "/usr/local/lib/xemacs/packages-21.0/lisp/text-modes/" "/usr/local/lib/xemacs/packages-21.0/lisp/textools/" "/usr/local/lib/xemacs/packages-21.0/lisp/time/" "/usr/local/lib/xemacs/packages-21.0/lisp/tm/" "/usr/local/lib/xemacs/packages-21.0/lisp/toolbar-fancy/" "/usr/local/lib/xemacs/packages-21.0/lisp/vc/" "/usr/local/lib/xemacs/packages-21.0/lisp/vm/" "/usr/local/lib/xemacs/packages-21.0/lisp/w3/" "/usr/local/lib/xemacs/packages-21.0/lisp/xemacs-base/" "/usr/local/lib/xemacs/packages-21.0/lisp/xemacs-devel/" "/cvs/xemacs/xemacs/lisp/" "/cvs/xemacs/xemacs/lisp/mule/" "/cvs/xemacs/xemacs/lisp/term/") backtrace(external-debugging-output t) # bind (string error debug-on-error stream debugger-args) early-error-handler(error (void-function init-gutter)) # (catch debugger ...) # (unwind-protect ...) # bind (stack-trace-on-signal debug-on-signal stack-trace-on-error debug-on-error) init-gutter() init-post-x-win() # (unwind-protect ...) make-device(x nil) # bind (display) make-x-device(nil) init-x-win() # bind (debugger debug-on-error command-line-args-left) command-line() # (unwind-protect ...) normal-top-level() # (condition-case ... . error) # (catch top-level ...)

2 participants
1 comments

A real database for Emacs 24 years, 8 months

SL Baur

Attached are patches to XEmacs 21.1.10 to add runtime support for the PostgreSQL relational database. I've tested it against both 6.5.3 and 7.0beta1. Mulish database stuffs are known to cause crashes, however anything US-ASCII based seems to be O.K. Multiple and networked connections work, as do connecting to a version of the postmaster different from the version of the library linked to. I will be adding support for asynchronous queries, non-US ASCII character sets and documentation before merging into the 21.2 source tree. If you installed Postgres from source RPMs, it will be automatically autodetected. If you installed into the default location (/usr/local/pgsql), or somewhere else, you will need to add the base directory to --site-prefixes (eg. --site-prefixes=/usr/local/pgsql). sample session: (setq P (pq-connectdb "host=zaizen.m17n.org port=25432")) #<PGconn zaizen.m17n.org:25432 steve/steve> (setq R (pq-exec P "SELECT * FROM pg_database;")) #<PGresult PGRES_TUPLES_OK - SELECT> (require 'pg-util) pg-util (pg-util-query-results R) (("datname" "datdba" "encoding" "datpath") ("template1" "128" "1" "template1") ("steve" "129" "1" "steve") ("japanese" "129" "1" "japanese") ("plpgsql_test" "128" "1" "plpgsql_test") ("postgres" "128" "1" "postgres") ("pltcl_test" "128" "1" "pltcl_test")) (setq R (pq-exec P "This is a bad query;")) #<PGresult PGRES_FATAL_ERROR - > (pq-pgconn P 'pq::ERROR-MESSAGE) "ERROR: parser: parse error at or near \"this\" " (setq R (pq-exec P "CREATE TABLE test_table (f1 text, f2 int4);")) #<PGresult PGRES_COMMAND_OK - CREATE> (setq R (pq-exec P "INSERT INTO test_table VALUES ('string', 42);")) #<PGresult PGRES_COMMAND_OK - INSERT 38570 1> Questions, comments are welcome. There are three new files: src/postgresql.h, src/postgresql.c, and lisp/pg-util.el. ChangeLog: 2000-03-06 SL Baur <steve(a)musashimaru.m17n.org> * configure.in: add autodetection of PostgreSQL runtime libraries src/ChangeLog: 2000-03-06 SL Baur <steve(a)musashimaru.m17n.org> * config.h.in: Add symbols HAVE_POSTGRESQL and HAVE_POSTGRESQLV7 * inline.c: Include postgresql.h lrecord stuffs to placate buggy GCCs. * emacs.c (main_1): Call postgres initialization code. * postgresql.h: New file. PostgreSQL RDBMS support. * postgresql.c: New file. Index: configure.in =================================================================== RCS file: /usr/CVSroot/XEmacs/xemacs/configure.in,v retrieving revision 1.138 diff -u -r1.138 configure.in --- configure.in 2000/02/09 01:40:53 1.138 +++ configure.in 2000/03/06 03:51:59 ＠＠ -497,6 +497,7 ＠＠ with_sparcworks | \ with_tooltalk | \ with_ldap | \ + with_postgresql | \ with_pop | \ with_kerberos | \ with_hesiod | \ ＠＠ -2698,6 +2699,55 ＠＠ fi fi +dnl Autodetect PostgreSQL +dnl The default installation location (non-Linux) is /usr/local/pgsql; +dnl a different prefix can be selected at build/install time. If PostgreSQL +dnl is installed into a different prefix, that prefix must be specified in +dnl in the --site-prefixes flag. +dnl The default RPM-based Linux installation location is /usr. +AC_CHECKING(for PostgreSQL) +postgres_includes_found=no +save_c_switch_site="$c_switch_site" +dnl First check site prefixes +if test "$with_postgresql" != "no"; then + AC_CHECK_HEADER(libpq-fe.h,postgres_includes_found=yes) +fi +dnl test for Linux-style installation in /usr +if test "$postgres_includes_found" = "no" -a "$with_postgresql" != "no" -a \ + -d "/usr/include/pgsql"; then + c_switch_site="$c_switch_site -I/usr/include/pgsql" + AC_CHECK_HEADER(libpq-fe.h,postgres_includes_found=yes) + if test "$postgres_includes_found" != "yes"; then + c_switch_site="$save_c_switch_site" + fi +fi +if test "$postgres_includes_found" = "no" -a "$with_postgresql" != "no" -a \ + -d "/usr/local/pgsql/include"; then + c_switch_site="$c_switch_site -I/usr/local/pgsql/include" + AC_CHECK_HEADER(libpq-fe.h,postgres_includes_found=yes) + if test "$postgres_includes_found" != "yes"; then + c_switch_site="$save_c_switch_site" + fi +fi + +dnl last check -- can we link against libpq? +if test "$postgres_includes_found" = "yes"; then + AC_CHECK_LIB(pq,PQconnectdb,with_postgresql=yes,with_postgresql=no) +fi +if test "$with_postgresql" = "yes"; then + AC_CHECK_LIB(pq,PQconnectStart,with_postgresqlv7=yes,with_postgresqlv7=no) +fi +if test "$with_postgresql" = "yes"; then + AC_DEFINE(HAVE_POSTGRESQL) + if test "$with_postgresqlv7" = "yes"; then + AC_DEFINE(HAVE_POSTGRESQLV7) + fi + XE_PREPEND(-lpq, LIBS) + XE_ADD_OBJS(postgresql.o) +else + c_switch_site=$save_c_switch_site +fi + dnl ---------------------- dnl Graphics libraries dnl ---------------------- ＠＠ -4060,6 +4110,7 ＠＠ test "$with_ns_ldap" = yes && echo " Compiling in support for LDAP (Netscape SDK)." test "$with_ldap" = yes -a "$with_umich_ldap" = no -a "$with_ns_ldap" = no && echo " Compiling in support for LDAP (Generic)." +test "$with_postgresql" = yes && echo " Compiling in support for PostgreSQL." test "$with_ncurses" = yes && echo " Compiling in support for ncurses." test "$with_gpm" = yes && echo " Compiling in support for GPM (General Purpose Mouse)." Index: src/config.h.in =================================================================== RCS file: /usr/CVSroot/XEmacs/xemacs/src/config.h.in,v retrieving revision 1.53 diff -u -r1.53 config.h.in --- config.h.in 1999/07/07 03:10:32 1.53 +++ config.h.in 2000/03/06 03:52:08 ＠＠ -357,6 +357,11 ＠＠ /* Do we have Netscape LDAP SDK library */ #undef HAVE_NS_LDAP +/* Do we wish to link against the PostgreSQL RDBMS run-time library? + */ +#undef HAVE_POSTGRESQL +#undef HAVE_POSTGRESQLV7 + /* Do you have the Xauth library present? This will add some extra functionality to gnuserv. */ #undef HAVE_XAUTH Index: src/emacs.c =================================================================== RCS file: /usr/CVSroot/XEmacs/xemacs/src/emacs.c,v retrieving revision 1.93 diff -u -r1.93 emacs.c --- emacs.c 1999/12/04 04:13:05 1.93 +++ emacs.c 2000/03/06 03:52:09 ＠＠ -1062,6 +1062,10 ＠＠ syms_of_eldap (); #endif +#ifdef HAVE_POSTGRESQL + syms_of_postgresql (); +#endif + /* Now create the subtypes for the types that have them. We do this before the vars_*() because more symbols may get initialized here. */ ＠＠ -1437,6 +1441,10 ＠＠ #ifdef HAVE_LDAP vars_of_eldap (); +#endif + +#ifdef HAVE_POSTGRESQL + vars_of_postgresql(); #endif /* Now initialize any specifier variables. We do this later Index: src/inline.c =================================================================== RCS file: /usr/CVSroot/XEmacs/xemacs/src/inline.c,v retrieving revision 1.12 diff -u -r1.12 inline.c --- inline.c 1998/05/24 05:40:44 1.12 +++ inline.c 2000/03/06 03:52:09 ＠＠ -67,6 +67,11 ＠＠ #include "eldap.h" #endif +#ifdef HAVE_POSTGRESQL +#include <libpq-fe.h> +#include "postgresql.h" +#endif + #ifdef HAVE_TOOLBARS #include "toolbar.h" #endif

2 participants
1 comments

Re: Intent to convert all XSET... functions 24 years, 8 months

Jan Vroonhof

Martin Buchholz <martin(a)xemacs.org> writes: > It's true that there will be the urge to just do > > bar (Lisp_FOO_to_Object (alloc_lcrecord_type (Lisp_FOO, &lrecord_FOO))); > > which will lose horribly. But someone will submit a patch out of the > blue to implement conservative gc for the stack, right? It happened > for mule support on Windows (thanks Tomonori!). I am still thing it might be a big win to use a C++ compiler to catch GCPRO mistakes (if we really switched to C++ we could even have objects GCPRO themselves). In fact I think a C++ compiler could catch mistakes like the above at compile time. The idea is to make function arguments and return values to have different classes with no implicit conversion between them. In the same way you could catch most if not all calls to functions that can GC from functions that cannot. [I still haven't tried this so there may be some snag somewhere.] Jan

3 participants
5 comments

Re: C++ vs. GCPRO 24 years, 8 months

Jan Vroonhof

Martin Buchholz <martin(a)xemacs.org> writes: [My first post seems missing from the list. Sorry if you see this twice] > I'd like you to prove me wrong - in this case I would support the > permanent switch to C++. Here is something I prepared earlier (ahem). All these structs would be typedef'ed in a real use of course. I think you can do this checking as long as you allow about 6 or so different types that under C would all just be the same thing. g++ -c gcpro.C gcpro.C: In function `void incorrect()': gcpro.C:74: conversion from `rv_lispobj' to non-scalar type `funarg_lispobj' requested Here is gcpro.C typedef unsigned int lispobj_internal; struct funarg_lispobj; struct stack_lispobj; /* Function returnvalues */ struct rv_lispobj { lispobj_internal i; // Here would come contructurs from most lisp_obj }; struct ungcable_lispobj { lispobj_internal i; ungcable_lispobj(unsigned int in) { i = in; }; private: ungcable_lispobj(); }; /* Function arguments for functions that cannot GC */ struct nogcarg_lispobj { lispobj_internal i; nogcarg_lispobj(const struct ungcable_lispobj in); nogcarg_lispobj(const struct funarg_lispobj in); nogcarg_lispobj(const struct stack_lispobj in); nogcarg_lispobj(const struct rv_lispobj in); }; struct funarg_lispobj { lispobj_internal i; funarg_lispobj(const struct stack_lispobj in); // Would check in.gcprod_flag // Note the missing contructor from rv_lispobj; // Note the missing contructor from nogcarg_lispobj; }; struct stack_lispobj { lispobj_internal i; bool gcpro_flag; stack_lispobj() { gcpro_flag = false; }; stack_lispobj(const struct rv_lispobj in) { i = in.i; }; }; struct ungcable_lispobj zero(0); struct rv_lispobj Fcons(struct nogcarg_lispobj car, struct nogcarg_lispobj cdr); // Can GC struct rv_lispobj bar(struct funarg_lispobj arg); void correct1() { stack_lispobj obj = Fcons(zero,zero); obj.gcpro_flag = true; // Would be part of GCPRO; bar(obj); } void correct2() { Fcons(Fcons(zero,zero),zero); } void incorrect() { bar(Fcons(zero,zero)); }

2 participants
1 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 March 2000