1 2000-03-23 Tor Lillqvist <tml@iki.fi>
3 * gmodule-win32.c (_g_module_symbol): When looking for symbols in
4 the "main" module we must search both the main program and all
5 currently loaded DLLs. Not only the main program, or even just the
6 DLLs loaded as gmodules. Libglade requires this.
8 Thus we need to get a list of all modules in the current
9 process. There are two alternative APIs to do this: PSAPI and
10 Toolhelp. The former is only available on NT (including Win2k),
11 the latter on Win9x and Win2k. Check which one works, and use
14 First check for the symbol in the main program, and if not found
15 there, in all the modules.
17 Code for using PSAPI and Toolhelp was borrowed from the Dr. Mingw
18 tool written by José Fonseca <em96115@fe.up.pt>. Thanks.
20 2000-03-04 Tor Lillqvist <tml@iki.fi>
22 * gmodule-win32.c: Call g_win32_error_message() to get the error
27 * gmodule.c: Remove LibMain functions that were needed by LCC,
28 which is not a supported compiler.
30 * testgmodule.c (main): Test for G_MODULE_IMPL ==
31 G_MODULE_IMPL_WIN32, not G_OS_WIN32.
33 * gmoduleconf.h.win32: Remove LCC stuff from here, too.
35 Wed Mar 1 05:34:47 2000 Tim Janik <timj@gtk.org>
37 * gmodule-beos.c (_g_module_symbol): do not return NULL symbols.
39 * gmodule-os2.c: removed NetBSD specific defines.
40 (_g_module_self): set an error message for unsupported behaviour.
42 * gmodule-beos.c: many coding style fixups.
46 (_g_module_symbol): bunch of memory leaks plugged.
48 * gmodule-dl.c: make sure the error message returned from dlerror()
49 is always != NULL, by using a wrapper function fetch_dlerror(). based
50 on a patch to fix _g_module_symbol() for NetBSD from Scott Presnell
53 * gmodule-dld.c: minor indentation.
55 * gmodule-win32.c: minor cleanups.
57 * merges from glib-1-2.
59 Sat Feb 19 19:43:29 2000 Tim Janik <timj@gtk.org>
61 * testgmodule.c (main): added test to check that not yet bound symbols
62 in shared libraries of the main module are retrievable, from David Gero.
64 2000-01-13 Martin Baulig <martin@home-of-linux.org>
66 * gmodule.c (g_module_open): Check whether `check_init' is not NULL
67 before we attempt to call it.
69 Sun Oct 3 19:30:52 PDT 1999 Manish Singh <yosh@gimp.org>
72 * testgmodule.c: use G_OS stuff
74 Sat Jul 24 20:47:18 1999 Tim Janik <timj@gtk.org>
76 * merged changes from GLib-1-2 branch.
78 1999-05-06 Tor Lillqvist <tml@iki.fi>
80 * gmodule.c: Remove LibMain.
82 Sat May 1 10:58:57 PDT 1999 Manish Singh <yosh@gimp.org>
84 * Makefile.am: use -avoid-version and -module for test plugins
86 Mon Apr 19 08:43:59 1999 ape@lrdpf.spacetec.no (Asbjorn Pettersen)
88 * Makefile.am (EXTRA_DIST): Add gmodule-os2.c
90 Wed Apr 7 20:12:58 1999 ape@lrdpf.spacetec.no (Asbjorn Pettersen)
92 * gmodule-os2.c: OS/2 specific file for gmodule.
93 * gmodule.c (_g_module_open): Add gmodule-os2.c if G_MODULE_IMPL_OS2.
94 * gmoduleconf.h.in: Add G_MODULE_IMPL_OS2.
96 Fri Apr 23 09:03:48 1999 Tim Janik <timj@gtk.org>
98 * gmodule.c (g_module_symbol): removed inline variable assignment.
99 s/CHECK_ERROR/SUPPORT_OR_RETURN/ to make the code more self descriptive.
101 Wed Feb 10 12:06:30 1999 Tim Janik <timj@gtk.org>
103 * gmodule.c (CHECK_ERROR): be more descriptive on unsupported systems.
105 Wed Feb 10 07:56:33 1999 Tim Janik <timj@gtk.org>
107 * gmodule.c (g_module_error): fixed errernerous code wrt to thread
108 specific error string allocation handling.
110 Thu Jan 21 12:40:11 EST 1999 Jeff Garzik <jgarzik@pobox.com>
112 * gmodule-dl.c (_g_module_build_path):
113 Add braces to eliminate an ambiguous else warning.
115 1999-01-16 Tor Lillqvist <tml@iki.fi>
117 * gmodule-dl.c gmodule-dld.c: In
118 _g_module_build_path, don't add the "lib" prefix and
119 ".so" or ".sl" suffix if already there.
121 * gmodule-win32.c: Likewise for the ".dll" suffix.
123 1998-12-10 Sebastian Wilhelmi <wilhelmi@ira.uka.de>
125 * gmodule.c: Made it MT safe, the g_module_error() is now thread
128 Fri Nov 20 14:43:44 1998 Tim Janik <timj@gtk.org>
130 * gmodule.c (_g_module_build_path): added empty default imlementation
131 so gmodule.so compiles on systems that are not yet supported, fix from
132 Erik Bagfors <bagfors@hpc2n.umu.se>.
134 1998-11-13 Nuno Ferreira <nmrf@rnl.ist.utl.pt>
136 * Makefile.am (INCLUDES): The previous commit message is wrong
137 about the -I$(top_srcdir) being redundant. I put it back.
139 Wed Nov 11 23:23:22 EST 1998 Jeff Garzik <jgarzik@pobox.com>
141 * Makefile.am : INCLUDES is the right way to add to CFLAGS, not
142 DEFS. Also there are bugs with '+=' in makefiles.
143 Got rid of DEFS line by moving G_LOG_DOMAIN setting into INCLUDES.
144 Removed redundant -I from INCLUDES.
146 Tue Oct 27 04:00:11 1998 Tim Janik <timj@gtk.org>
148 * testgmodule.c (main): changed the #ifdef WIN32 test to NATIVE_WIN32,
149 this needs to be more constistent throughout the code, do we go for
150 NATIVE_WIN32 or WIN32?
152 * gmodule.c (LibMain): special cased the #ifdef __LCC__ case for
153 NATIVE_WIN32, since lcc maybe used on other platforms as well.
154 * libgplugin_a.c (LibMain):
155 * libgplugin_b.c (LibMain):
156 likewise. not sure i like this special requirement for lcc in here.
158 * gmodule-dl.c (_g_module_build_path):
159 feature empty "" directories and prepend the module name with "lib".
161 * gmodule-dld.c (_g_module_build_path):
162 * gmodule-win32.c (_g_module_build_path):
163 feature empty "" directories.
165 * we need some more magic in the _g_module_build_path variants
166 so we don't append/prepend lib and .so, .sl or .dll for those names
167 that already contain it.
169 * applied patch from Tor Lillqvist for g_module_build_path() and
172 1998-10-20: Tor Lillqvist <tml@iki.fi>
174 * gmodule/gmodule-win32.c:
177 * gmodule/gmodule.c gmodule/gmodule.h:
178 Added the funcion g_module_build_path that builds the path to
179 a module file, decorating the name according to the system's
180 conventions. Added the Windows implementation.
182 * gmodule/libgplugin_a.c gmodule/libgplugin_b.c:
183 Added LibMain for LCC-Win32.
185 * gmodule/testgmodule.c:
186 Handle Windows dll names.
188 1998-10-25 Raja R Harinath <harinath@cs.umn.edu>
190 * gmodule.h: Remove `#pragma }'.
191 * gmoduleconf.h.in: Likewise.
193 Wed Oct 21 19:58:27 1998 Tim Janik <timj@gtk.org>
195 * gmodule.c (g_module_symbol): fixed a string pointer bug that could
196 cause garbage error messages from g_module_symbol() for systems that
199 Mon Sep 21 01:54:48 1998 Tim Janik <timj@gtk.org>
202 * gmodule.c: renamed old _de_init functionality to _unload.
203 modules are now expected to export:
204 G_MODULE_EXPORT const gchar* g_module_check_init (GModule *module);
206 G_MODULE_EXPORT void g_module_unload (GModule *module);
207 returning a string other than NULL from g_module_check_init() will
208 prevent the module from being loaded. a call to g_module_make_resident()
209 from g_module_unload() will prevent the module from being unloaded and
210 still make it resident.
212 Thu Sep 17 06:34:22 1998 Tim Janik <timj@gtk.org>
215 * gmodule.c: implemented g_module_make_resident() which can be
216 used to make modules resident.
217 fixed a buglet about the optional "g_module_de_init" function in
218 modules, which could get invoked twice on very obscure occasions.
220 Tue Sep 15 14:57:30 1998 Owen Taylor <otaylor@redhat.com>
222 * Makefile.am: Update to libtool-1.2b,
223 change library versioning scheme to drop LT_RELEASE
224 from the -l line, while keeping it in the soname.
226 Thu Aug 20 07:08:16 1998 Tim Janik <timj@gtk.org>
228 * gmodule.c: provide no operation implementation for the underlying
229 _g_module_* functions, so we at least compile on systems that have
230 neither of G_MODULE_IMPL_DL or G_MODULE_IMPL_DLD.
232 Mon Aug 17 03:41:52 1998 Tim Janik <timj@gtk.org>
235 * gmodule.c (g_module_open): changed the return type for the
236 GModuleCheckInit function to be a string, describing the error
238 (g_module_symbol): show the failing symbol on error messages.
240 Fri Aug 14 02:24:39 1998 Tim Janik <timj@gtk.org>
242 * Makefile.am: feature the G_LOG_DOMAIN macro to set the log domain
243 to "GModule" upon compilation. we currently have to add this definition
244 to the DEFS variable.
245 * testgmodule.c: we need an ugly #undef G_LOG_DOMAIN at the start
246 of this file currently, since automake doesn't support per target
249 Mon Aug 10 03:35:57 1998 Tim Janik <timj@gtk.org>
251 * gmodule.c: minor changes to internal interface.
253 * gmodule-dld.c: put some comments into the files, and provided
254 better error checking for shl_findsym(). whish i had a system to
257 Mon Aug 10 02:18:31 1998 Tim Janik <timj@gtk.org>
259 * Makefile.am (lib_LTLIBRARIES): for now, skip the dependency on
260 -lglib for libgmodule-1.1.la, libgplugin_a.la and libgplugin_b.la
261 since this clashes with inter-library-dependencies for not installed
262 libraries. glib-config takes care of this for the usuall case, but
263 there needs to be a better way...
265 Sun Aug 9 15:57:38 1998 Tim Janik <timj@gtk.org>
267 * testgmodule.c: test program for GModule.
269 * libgplugin_b.c: test plugins for testgmodule.c.
272 * gmodule.c: GModule library implementation, which is basically
273 a wrapper about system specifc dynamic loading facilities.
276 Sun Aug 9 10:31:05 1998 Tim Janik <timj@gtk.org>
278 * ChangeLog start for gmodule.