November 2002 - Wine-Devel - list.winehq.org

test - please ignore
by Hetz Ben Hamo Nov. 10, 2002

Nov. 10, 2002

1 0

Re: ddraw fix
by Lionel Ulmer Nov. 10, 2002

Nov. 10, 2002

> Probably application bug: it called DIB_DirectDrawSurface_Blt > with .bottom > 36000, seems like DirectDraw sanitized the values. Well, I would prefer the fix to be different... For example, it seems to be valid to have negative top values as long as 'bottom - top' is valid (as it seems that DDraw will clip the blit for you). Problem is that if negative top is valid, then bottom being bigger than dwHeight should be valid too. And your patch will thus prevent this case to work. Did you try to simply return an error message (instead of doing the assert) to see if you see any graphical glitches ? Anyway, I am installing Mingw + the DirectX SDKs to do some test apps on a Win2K laptop. I will try this case to see if I manage to see what real Windows does. Lionel -- Lionel Ulmer - http://www.bbrox.org/

1 0

Re: Regedit from ReactOS
by Dimitrie O. Paun Nov. 10, 2002

Nov. 10, 2002

On November 9, 2002 09:20 pm, Steven Edwards wrote: > Thanks to Robert Dickenson for his work on this. If someone wants to > adopt then here it is. Good stuff. This needs a bit of massaging, but looks good. -- Dimi.

1 0

Re: Fun projects
by Sylvain Petreolle Nov. 10, 2002

Nov. 10, 2002

This is what I did. the gcc included in dev-c++ is a statically linked that doesn't need the cygwin1.dll to run. But that I looked particularily at compiling cygwin under wine. Using the cygwin under is difficult for the moment, as bash and sh both refuse to run today. > I presume this is possible, although the other interesting approach > would be > to run the gcc binary from mingw/cygin under wine itself. I saw that > somebody > got it running recently on the newsgroup; see > http://groups.google.com/groups?q=ming32w+gcc+wine > > Note: I haven't actually done this. These are just ideas :-) Thanks for taking time to answer. ___________________________________________________________ Do You Yahoo!? -- Une adresse @yahoo.fr gratuite et en français ! Yahoo! Mail : http://fr.mail.yahoo.com

5 8

Wine 0.9 TODO v0.4
by Dimitrie O. Paun Nov. 10, 2002

Nov. 10, 2002

*Lots* of updates on this one: http://www.dssd.ca/Wine-0.9-TODO-0.4.html Please note that the above URL is for the _released_ TODO (note that the version appears in the URL). If you want to follow the working version, you can do so at: http://www.dssd.ca/Wine-0.9-TODO.html However, this version is constantly changing; it is relative to the last _released_ version. This is important to keep in mind when looking at the bold items. For example, if you look at the working version right after a new version is released, nothing is going to be bold, as nothing has changed from the released version. To better understand the changes from version to version, you should consult the released versions. The TODO is included below as text, for your commenting convenience. A. WineHQ work Beta available here 1. Website redesign [IN PROGRESS] * worker: Jeremy Newman * status: backend is getting finalized * updated: Nov 7, 2002 2. Move the doc scripts out of documentation into the tools CVS module [TODO] * worker: Jeremy Newman * status: should remind Jeremy to do it :) * updated: Nov 7, 2002 3. Reorganize the navigational menu [DONE] * worker: Francois Gouget & al. * status: merged into new site * updated: Nov 5, 2002 4. Create some really sexy screenshots [IN PROGRESS] * worker: Brian Vincent * status: please send screenshots to the above mentioned maintainer * updated: Nov 9, 2002 * criteria: one app per screenshot, avoid cluttered desktop 5. Rework the FAQ interface [IN PROGRESS] * workers: Dimitrie O. Paun, Keith Matthews, Thomas Wickline * status: released for public review here * updated: Nov 9, 2002 6. Enlist some 'official' distribution maintainers [DONE] * RedHat: Vincent Béron * SUSE: Marcus Meissner * Mandrake: Kye Lewis * Debian: Ove Kaaven * FreeBSD: Gerald Pfeifer * status: completed * updated: Nov 7, 2002 7. Create nice page with apps that run virtually flawless [IN PROGRESS] * workers: Carlos Lozano, Dustin Navea * status: 20% * updated: Nov 7, 2002 * criteria: not MS dlls required, no tricks, etc. B. Documentation work 1. We need to first figure what is out of date. [IN PROGRESS] * worker: Dustin Navea * status: just started, help needed * updated: Nov 9, 2002 2. Fix documentation build process [IN PROGRESS] * worker: Tony Lambregts * status: just started * updated: Nov 9, 2002 3. Move CVS docu from User Guide to Devel Guide [IN PROGRESS] * worker: Bill Medland * status: just started * updated: Nov 7, 2002 4. Merge documentation/wine.texinfo into the Development Guide [TODO] 5. Merge documentation/HOWTO-winelib into the Development Guide [TODO] 6. Do something about documentation/installation-und-konfiguration.german [TODO] 7. Do something about documentation/shell32 [TODO] 8. Do something about documentation/status [TODO] C. Wine configuration 1. Merge configuration into the registry [TODO] 2. Write control panel applets for editing it [TODO] 3. Have decent defaults so we can start control panel without prior configuration [TODO] 4. Have wizard like app to autoconfigure wine [TODO] * proposal: WineSetupTk by Jeremy White 5. Make Wine's DLLs register themselves to avoid manual merging of the winedefault.reg [TODO] 6. Write wine.inf script to setup a new Wine installation [TODO] 7. Have a wineboot script for RunOnce stuff [IN PROGRESS] * workers: Andreas Mohr, Shachar Shemesh * status: 50% * updated: Nov 7, 2002 D. Stabilize utilities 1. Get rid of the init directive from .spec files [DONE] * worker: Alexandre Julliard * status: implemented by this * updated: Nov 7, 2002 2. Make sure the .spec file format is fairly stable [IN PROGRESS] * worker: Alexandre Julliard * status: Yes, I'll do that. * updated: Nov 7, 2002 * what: can we get rid of the rsrc directive? 3. Ensure the various utilities' options are stable [IN PROGRESS] * status: Alexandre fixed winebuild options * updated: Nov 7, 2002 * what: audit the other utilites 4. Get rid of the Perl testing framework [IN PROGRESS] * worker: Dimitrie O. Paun * status: submitted patch for review * updated: Nov 7, 2002 E. Important fixes 1. Window management rewrite, so we can install apps in managed mode. [TODO] * worker: Alexandre Julliard * status: not started 2. Finish DLL separation Phase 1 [IN PROGRESS] * worker: ? * status: ntdll, and kernel not done * updated: Nov 7, 2002 * criteria: DLLs should use only functions exported through .spec file. 3. Detect printing system at runtime [TODO] * worker: Huw Davies * status: Yup, I'm going to do this in the near future. * updated: Nov 7, 2002 4. Detect OpenGL system at runtime [IN PROGRESS] * worker: Lionel Ulmer * status: Patch submitted for review. * updated: Nov 9, 2002 5. Start builtin WinHelp when invoked from apps [IN PROGRESS] * worker: Eric Pouech * status: I have a working version of this * updated: Nov 9, 2002 6. Add macro support to the builin WinHelp [TODO] * worker: Eric Pouech * status: I have this on my todo list * updated: Nov 8, 2002 7. Add CHM support to builtin WinHelp [IN PROGRESS] * worker: ? * status: Cabextract donated to Wine * updated: Nov 9, 2002 8. Good support for VisualBasic apps [TODO] * worker: ? * status: ? * updated: Nov 9, 2002 * tests: submitted by Roderick Colenbrander -- Dimi.

5 5

Resolving winelib dll dependencies
by Nix N. Nix Nov. 10, 2002

Nov. 10, 2002

I have an application that consists of a single DLL and the EXE. I have the sources required for the DLL in one directory and I have used winemaker --dll to create the infrastructure for building the winelib dll. It worked. So, now, I have "nix_ctls.dll.so". My problem is that I cannot, for some reason use nix_ctls.dll.so as I would, say comctl32.dll.so . What I mean is this: The sources for the EXE are in another directory. Both are subdirectories of the "project" directory. The name of the EXE is supposed to be "iconmake.exe", if it were compiled under Windoze. So, naturally, my Makefile contains a setting called "iconmake_DLLS". To this setting I have appended "nix_ctls.dll" and to the "iconmake_DLL_PATH" I have appended "-L../nix_ctls". This makes everything work, except the spec thing. This is what that one says: gcc -c -g -O2 -fPIC -D_REENTRANT -DWINELIB -I../common -I. -I/usr/wine/include/wine -o iconmake.spec.o iconmake.spec.c gcc -shared -Wl,-Bsymbolic -o iconmake.so dlg_box.o filedlgs.o mdi_icon.o ../common/helper.o ../common/debugmsg.o ../common/instance.o ../common/dynarray.o mdichild.o toolbars.o iconmake.spec.o -L/usr/wine/lib -lwine -lwine_unicode -lwine_uuid -lm dlg_box.o: In function `PickIconDlg': /home/nix/source/mdi_icon/iconmake/dlg_box.c:396: undefined reference to `InitPicPick' mdichild.o: In function `ChildWndProc': /home/nix/source/mdi_icon/iconmake/mdichild.c:86: undefined reference to `InitEditorWindow' toolbars.o: In function `CreateTheRebar': /home/nix/source/mdi_icon/iconmake/toolbars.c:23: undefined reference to `InitColourControl' Where do I specify "nix_ctls.dll" and the corresponding path "../nix_ctls" in the Makefile so that it will be added to this command as it is added to the command before it, which is: LD_LIBRARY_PATH="/usr/wine/lib:/usr/wine/lib:$LD_LIBRARY_PATH:../nix_ctls" /usr/wine/bin/winebuild -fPIC -o iconmake.spec.c -res mdi_icon.res -L/usr/wine/lib -L../nix_ctls -L/usr/wine/lib/wine -ladvapi32.dll -lcomdlg32.dll -lgdi32.dll -lkernel32.dll -lodbc32.dll -lole32.dll -loleaut32.dll -lshell32.dll -luser32.dll -lcomctl32.dll -lnix_ctls.dll -lwinspool.drv -m gui -exe iconmake dlg_box.o filedlgs.o mdi_icon.o ../common/helper.o ../common/debugmsg.o ../common/instance.o ../common/dynarray.o mdichild.o toolbars.o This command works fine, because the library paths and -l settings are all there. Thanks for your help.

1 0

Re: resend update to sgml documentation
by Dustin Navea Nov. 9, 2002

Nov. 9, 2002

Hmm, wonder where the "generic steps to get output from sgml" came from??? ;) *hint hint* *wink wink* ;P -Dustin --- Tony Lambregts <tony_lambregts(a)telusplanet.net> wrote: > Since the last patch looked really screwed up I am resending with .txt > extention > > Hopfully this makes it easier for others to set up their docbook > environment and encourage more people to write documentation it is not > complete but at least it is a start. > > change log: make the instructions of how to install sgml easier to > understand . > > files changed: > documentation/documentation.sgml > documentation/authors.ent > > -- > > Tony Lambregts > > > Index: documentation.sgml > =================================================================== > RCS file: /home/wine/wine/documentation/documentation.sgml,v > retrieving revision 1.4 > diff -u -r1.4 documentation.sgml > --- documentation.sgml 2 Oct 2001 17:48:16 -0000 1.4 > +++ documentation.sgml 9 Nov 2002 04:00:59 -0000 > @@ -86,6 +86,9 @@ > <para> > Written by &name-john-sheets; <email>&email-john-sheets;</email> > </para> > + <para> > + Modified by &name-tony-lambregts; > <email>&email-tony-lambregts;</email> Nov. 2002 > + </para> > > <sect2 id="writing-docbook"> > <title>Writing Documentation with DocBook</title> > @@ -98,7 +101,123 @@ > DocBook markup looks very similar to HTML markup, although > the names of the markup tags differ. > </para> > + <sect3> > + <title>Getting Started</title> > + <note> > + <title>Why SGML?</title> > + <para> > + The simple answer to that is that SGML allows you > + to create multiple formats of a given document from a > single > + source. Currently sgml is used to create html, pdf and PS > + formats of the Users Guide, Developers Guide, Winelib Users > + Guide Packagers Guide and FAQ. > + </para> > + </note> > + > + <note> > + <title>What do I need?</title> > + <para> > + You need the sgml tools. There are various places where you > + can get them. The most generic way of geting them form > source. > + </para> > + </note> > + > + <note> > + <title>Quick instructions</title> > + <para> > + These are the generic steps to get output from sgml. > + </para> > + </note> > + > + <orderedlist> > + > + <listItem><para> > + Go to <ulink > url="http://www.sgmltools.org">http://www.sgmltools.org</ulink> > + </para></listitem> > + > + <listItem><para> > + Download all of the sgmltools packages > + </para></listitem> > + > + <listItem><para> > + Install them all and build them (./configure; make; make > install) > + </para></listitem> > + > + <listItem><para> > + Switch to your toplevel wine directory > + </para></listitem> > + > + <listItem><para> > + Run ./configure (or make distclean && ./configure) > + </para></listitem> > + > + <listItem><para> > + Switch to the documentation directory > + </para></listitem> > + > + <listItem><para> > + run make_winehq > + </para></listitem> > + > + <listItem><para> > + View wine-doc/index.html in your favorite browser > + </para></listitem> > + > + </orderedlist> > + > + </sect3> > + > + <sect3> > + <title>Getting SGML for various distributions</title> > + <para> > + Most Linux distributions have everything you need already > + bundled up in package form. Unfortunately, each > + distribution seems to handle its SGML environment > + differently, installing it into different paths, and > + naming its packages according to its own whims. > + </para> > + > + <sect4> > + <title>SGML on Redhat</title> > + <para> > + The following packages seems to be sufficient for RedHat 7.1. > You > + will want to be careful about the order in which you install > the > + rpms. > + <itemizedlist> > + <listitem><para>sgml-common-*.rpm</para></listitem> > + <listitem><para>openjade-*.rpm</para></listitem> > + <listitem><para>perl-SGMLSpm-*.rpm</para></listitem> > + <listitem><para>docbook-dtd*.rpm</para></listitem> > + <listitem><para>docbook-style-dsssl-*.rpm</para></listitem> > + <listitem><para>tetex-*.rpm</para></listitem> > + <listitem><para>jadetex-*.rpm</para></listitem> > + <listitem><para>docbook-utils-*.rpm</para></listitem> > + </itemizedlist> > + You can also use ghostscript to view the ps format output and > + Adobe Acrobat 4 to view the pdf file. > + </para> > + </sect4> > + > + <sect4> > + <title>SGML on Debian</title> > + <note> > + <title>Fix me</title> > + <para> > + List package names and install locations... > + </para> > + </note> > + </sect4> > > + <sect4> > + <title>SGML on Other Distributions</title> > + <note> > + <title>Fix me</title> > + <para> > + List package names and install locations... > + </para> > + </note> > + </sect4> > + </sect3> > <sect3> > <title>Terminology</title> > > @@ -1006,47 +1125,6 @@ > </para> > </sect3> > > - <sect3> > - <title>SGML on Redhat</title> > - <para> > - Most Linux distributions have everything you need already > - bundled up in package form. Unfortunately, each > - distribution seems to handle its SGML environment > - differently, installing it into different paths, and > - naming its packages according to its own whims. > - </para> > - <para> > - The following packages seems to be sufficient for RedHat 7.1. > You > - will want to be careful about the order in which you install > the > - rpms. > - <itemizedlist> > - <listitem><para>sgml-common-*.rpm</para></listitem> > - <listitem><para>openjade-*.rpm</para></listitem> > - <listitem><para>perl-SGMLSpm-*.rpm</para></listitem> > - <listitem><para>docbook-dtd*.rpm</para></listitem> > - <listitem><para>docbook-style-dsssl-*.rpm</para></listitem> > - <listitem><para>tetex-*.rpm</para></listitem> > - <listitem><para>jadetex-*.rpm</para></listitem> > - <listitem><para>docbook-utils-*.rpm</para></listitem> > - </itemizedlist> > - You can also use ghostscript to view the ps format output and > - Adobe Acrobat 4 to view the pdf file. > - </para> > - </sect3> > - > - <sect3> > - <title>SGML on Debian</title> > - <para> > - List package names and install locations... > - </para> > - </sect3> > - > - <sect3> > - <title>SGML on Other Distributions</title> > - <para> > - List package names and install locations... > - </para> > - </sect3> > </sect2> > > <sect2 id="emacs-psgml"> > Index: authors.ent > =================================================================== > RCS file: /home/wine/wine/documentation/authors.ent,v > retrieving revision 1.2 > diff -u -r1.2 authors.ent > --- authors.ent 24 Jul 2002 03:00:02 -0000 1.2 > +++ authors.ent 9 Nov 2002 04:01:56 -0000 > @@ -43,6 +43,9 @@ > <!entity name-alex-korobka "Alex Korobka"> > <!entity email-alex-korobka "alex(a)aikea.ams.sunysb.edu"> > > +<!entity name-tony-lambregts "Tony Lambregts"> > +<!entity email-tony-lambregts "tony_lambregts(a)telusplanet.net"> > + > <!entity name-marcus-meissner "Marcus Meissner"> >  > <!entity email-marcus-meissner "Marcus.Meissner(a)caldera.de"> > __________________________________________________ Do you Yahoo!? U2 on LAUNCH - Exclusive greatest hits videos http://launch.yahoo.com/u2

1 0

Fix for bug #688 (documentation building)
by Vincent Béron Nov. 9, 2002

Nov. 9, 2002

This one I'm not sure which one is better. So I include both patches, but only one should be applied. Functionnally, both produce the exact same html files, although the fix1 creates (by the system's db2html) a directory with images (which are unused). So I'd tend to go with fix2. Please comment, and I'll post the good one to wine-patches. Vincent

4 6

RFC: rpcss skeleton prototype #2
by Greg Turner Nov. 9, 2002

Nov. 9, 2002

This one implements "pseudo-fork" behavior, a main workhorse loop, pseudo-debug log emitter, and command-line interpretation. In particular I'm looking for comments/criticism/flames/etc for: o my handling of NO_NAMELESS_STRUCT... works for me, but how correct will this be, say, under MSVC with Microsoft headers? o I differentiate between "server mode" and "lazy mode"... In lazy mode I assume that I'll have at least one incoming connection (if not I time out after 20 minutes); after that, once all endpoints are unmapped (all RPC-specific code is unimplmented, but this is what RPCSS_empty() is supposed to detect) I time out after a short delay (20 sec.) In server mode, I just wait around forever until I'm told to terminate (unimplemented, will be a command-line arg in the tradition of "wineserver -k", processed via the pipe)... reasonable? o Pseudo-debug log emitter: should I be using stderr? is this basically an OK approach? What is the hex number that most real wine debug logs begin with? I'd like to look the same. o Make/autoconf BS o the pseudo-fork crud (RPCSS_spork() and the /^ cmd-line arg.) should I be worried about accidentally invoking the real rpcss.exe? is this an unacceptably stupid approach in general? thanks for your help. diff -ur -x CVS -x 'bigdif*' ../wine.test/configure.ac ./configure.ac --- ../wine.test/configure.ac 2002-11-03 11:50:15.000000000 -0600 +++ ./configure.ac 2002-11-01 18:53:32.000000000 -0600 @@ -1501,6 +1501,7 @@ programs/regedit/Makefile programs/regsvr32/Makefile programs/regtest/Makefile +programs/rpcss/Makefile programs/uninstaller/Makefile programs/view/Makefile programs/wcmd/Makefile diff -ur -x CVS -x 'bigdif*' ../wine.test/programs/Makefile.in ./programs/Makefile.in --- ../wine.test/programs/Makefile.in 2002-11-03 11:50:15.000000000 -0600 +++ ./programs/Makefile.in 2002-11-01 18:53:18.000000000 -0600 @@ -17,6 +17,7 @@ regedit \ regsvr32 \ regtest \ + rpcss \ uninstaller \ view \ wcmd \ @@ -39,6 +40,7 @@ progman \ regedit \ regsvr32 \ + rpcss \ uninstaller \ wcmd \ wineconsole \ @@ -56,6 +58,7 @@ progman \ regedit \ regsvr32 \ + rpcss \ uninstaller \ wcmd \ wineconsole \ @@ -67,6 +70,7 @@ # Symlinks to apps that we want to run from inside the source tree SYMLINKS = \ + rpcss.exe \ wineconsole.exe \ winedbg.exe @@ -115,12 +119,16 @@ # Rules for symlinks +rpcss.exe$(DLLEXT): rpcss/rpcss.exe$(DLLEXT) + $(RM) $@ && $(LN_S) rpcss/rpcss.exe$(DLLEXT) $@ + wineconsole.exe$(DLLEXT): wineconsole/wineconsole.exe$(DLLEXT) $(RM) $@ && $(LN_S) wineconsole/wineconsole.exe$(DLLEXT) $@ winedbg.exe$(DLLEXT): winedbg/winedbg.exe$(DLLEXT) $(RM) $@ && $(LN_S) winedbg/winedbg.exe$(DLLEXT) $@ +rpcss/rpcss.exe$(DLLEXT): rpcss wineconsole/wineconsole.exe$(DLLEXT): wineconsole winedbg/winedbg.exe$(DLLEXT): winedbg --- /dev/null 1969-12-31 18:00:00.000000000 -0600 +++ ./programs/rpcss/Makefile.in 2002-11-03 18:51:44.000000000 -0600 @@ -0,0 +1,15 @@ +TOPSRCDIR = @top_srcdir@ +TOPOBJDIR = ../.. +SRCDIR = @srcdir@ +VPATH = @srcdir@ +MODULE = rpcss.exe +APPMODE = cui +IMPORTS = user32 kernel32 + +C_SRCS = \ + rpcss_main.c + +(a)MAKE_PROG_RULES@ + +### Dependencies: + --- /dev/null 1969-12-31 18:00:00.000000000 -0600 +++ ./programs/rpcss/README 2002-11-03 18:52:12.000000000 -0600 @@ -0,0 +1,23 @@ +rpcss README + +rpcss, copyright 2002, Greg Turner +rpcss is to be distributed under the LGPL/X11-MIT licenses +See the Wine License for further information. + +Wine needs a server whose role is somewhat like that +of rpcss.exe in windows. + +Enjoy. + +KNOWN BUGS / TODO: + + o Service hooks are unimplemented (if you bother to implement + these, also implement net.exe, at least for "net start" and + "net stop" (should be pretty easy I guess, assuming the rest + of the services API infrastructure works (which net.exe should + assume, even if the API is lean or bugged))). + + o Is supposed to use RPC, not random kludges, to map endpoints. + + o Who knows? Whatever (sane) things rpcss does, we ought to at + least think about doing... but what /does/ it do? --- /dev/null 1969-12-31 18:00:00.000000000 -0600 +++ ./programs/rpcss/rpcss.h 2002-11-03 18:50:26.000000000 -0600 @@ -0,0 +1,43 @@ +/* + * Copyright 2002 Greg Turner <gmturner007(a)ameritech.net> + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this library; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + */ + +#ifndef __WINE_RPCSS_H +#define __WINE_RPCSS_H + +#include <stdio.h> + +#include "windows.h" + +/* yet another crude debug logger substitute (just what wine needed!) */ +#define RPCSS_DEBUG(args...) \ + do { \ + if (in_debug_mode) {\ + printf("rpcss.exe.so:%s: ", __FUNCTION__); \ + printf(args); \ + } \ + } while (FALSE) + +extern BOOL in_debug_mode; + +extern BOOL RPCSS_is_server(); +extern void RPCSS_be_server(); +extern void RPCSS_server_terminate(); + +#endif /* __WINE_RPCSS_H */ + +/* end of header */ --- /dev/null 1969-12-31 18:00:00.000000000 -0600 +++ ./programs/rpcss/rpcss_main.c 2002-11-04 01:16:29.000000000 -0600 @@ -0,0 +1,308 @@ +/* + * rpcss (main.c) + * + * Copyright 2002 Greg Turner <gmturner007(a)ameritech.net> + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this library; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + */ + +#include <assert.h> + +#include "rpcss.h" +#include "winnt.h" + +/* in seconds */ +#define RPCSS_LONG_LAZY_TIMEOUT 1200 +#define RPCSS_SHORT_LAZY_TIMEOUT 20 + +/* are we the "fork child"? */ +static BOOL secret_fork_child; + +/* do we dump debug info to the console? */ +static BOOL in_debug_mode; + +/* are we in server mode, with no timeout? */ +static BOOL in_server_mode; + +/* was server mode terminated? */ +static BOOL server_mode_was_terminated; + +/* when do we just give up and bail? (UTC) */ +static SYSTEMTIME lazy_timeout_time; + +#if defined(NONAMELESSSTRUCT) + #define FILETIME_TO_ULARGEINT(filetime, ularge) \ + ( ularge.s.LowPart = filetime.dwLowDateTime, \ + ularge.s.HighPart = filetime.dwHighDateTime, \ + ularge ) /* r-value */ + #define ULARGEINT_TO_FILETIME(ularge, filetime) \ + ( filetime.dwLowDateTime = ularge.s.LowPart, \ + filetime.dwHighDateTime = ularge.s.HighPart, \ + filetime ) /* r-value */ +#else + #define FILETIME_TO_ULARGEINT(filetime, ularge) \ + ( ularge.LowPart = filetime.dwLowDateTime, \ + ularge.HighPart = filetime.dwHighDateTime, \ + ularge ) /* r-value */ + #define ULARGEINT_TO_FILETIME(ularge, filetime) \ + ( filetime.dwLowDateTime = ularge.LowPart, \ + filetime.dwHighDateTime = ularge.HighPart, \ + filetime ) /* r-value */ +#endif /* NONAMELESSSTRUCT */ + +#define TEN_MIL 10000000LL + +/* returns time remaining in seconds */ +long RPCSS_GetLazyTimeRemaining() +{ + SYSTEMTIME st_just_now; + FILETIME ft_jn, ft_ltt; + ULARGE_INTEGER ul_jn, ul_ltt; + + assert(! in_server_mode); /* then why are you calling? */ + + GetSystemTime(&st_just_now); + SystemTimeToFileTime(&st_just_now, &ft_jn); + FILETIME_TO_ULARGEINT(ft_jn, ul_jn); + + SystemTimeToFileTime(&lazy_timeout_time, &ft_ltt); + FILETIME_TO_ULARGEINT(ft_ltt, ul_ltt); + + return (ul_ltt.QuadPart - ul_jn.QuadPart) / TEN_MIL; +} + +void RPCSS_SetLazyTimeRemaining(long seconds) +{ + SYSTEMTIME st_just_now; + FILETIME ft_jn, ft_ltt; + ULARGE_INTEGER ul_jn, ul_ltt; + + RPCSS_DEBUG("(seconds == %ld)\n", seconds); + + assert(! in_server_mode); /* then why are you calling? */ + assert(seconds >= 0); /* negatives are not allowed */ + + GetSystemTime(&st_just_now); + SystemTimeToFileTime(&st_just_now, &ft_jn); + FILETIME_TO_ULARGEINT(ft_jn, ul_jn); + + /* we want to find the time ltt, s.t. ltt = just_now + seconds */ + ul_ltt.QuadPart = ul_jn.QuadPart + seconds * TEN_MIL; + + /* great. just remember it */ + ULARGEINT_TO_FILETIME(ul_ltt, ft_ltt); + if (! FileTimeToSystemTime(&ft_ltt, &lazy_timeout_time)) + assert(FALSE); +} + +#undef FILETIME_TO_ULARGEINT +#undef ULARGEINT_TO_FILETIME +#undef TEN_MIL + +BOOL RPCSS_is_server() +{ + return in_server_mode; +} + +void RPCSS_be_server() +{ + RPCSS_DEBUG("\n"); + in_server_mode = TRUE; +} + +void RPCSS_server_terminate() +{ + assert(in_server_mode); + server_mode_was_terminated = TRUE; +} + +BOOL RPCSS_work() +{ + RPCSS_DEBUG("\n"); + Sleep(1000); + return FALSE; /* we didn't do anything */ +} + +BOOL RPCSS_empty() +{ + BOOL rslt = TRUE; + + /* FIXME */ + + RPCSS_DEBUG("%s\n", rslt ? "TRUE" : "FALSE" ); + return rslt; +} + +BOOL RPCSS_is_ready_to_die() +{ + return + ( + RPCSS_empty() && + ( + (in_server_mode && server_mode_was_terminated) || + ( + (! in_server_mode) && + (RPCSS_GetLazyTimeRemaining() <= 0) + ) + ) + ); +} + +void RPCSS_main_loop() { + BOOL did_something_new; + + RPCSS_DEBUG("\n"); + + for (;;) { + did_something_new = FALSE; + + while ( (! did_something_new) && (! RPCSS_is_ready_to_die()) ) + did_something_new = (RPCSS_work() || did_something_new); + + if ((! did_something_new) && RPCSS_is_ready_to_die()) + break; /* that's it for us */ + + if (did_something_new) { + /* if we were in server mode, but terminated, then, + because we did something new, fall back to lazy mode */ + if (in_server_mode && server_mode_was_terminated) + in_server_mode = server_mode_was_terminated = FALSE; + + /* if we aren't in server mode, set the lazy timeout */ + if (! in_server_mode) + RPCSS_SetLazyTimeRemaining(RPCSS_SHORT_LAZY_TIMEOUT); + } + } +} + +BOOL RPCSS_ProcessArgs( int argc, char **argv ) +{ + int i; + char *c; + + for (i = 1; i < argc; i++) { + c = argv[i]; + while (*c == ' ') c++; + if ((*c != '-') && (*c != '/')) + return FALSE; + c++; + switch (*c++) { + case 's': + case 'S': + in_server_mode = TRUE; + break; + case 'd': + case 'D': + in_debug_mode = TRUE; + break; + case '^': + /* "secret" hidden command-line arg means we are the "fork child" */ + secret_fork_child = TRUE; + break; + default: + return FALSE; + break; + } + while (*c == ' ') c++; + if (*c != '\0') return FALSE; + } + + return TRUE; +} + +void RPCSS_Usage() +{ + printf("\nWine RPCSS\n"); + printf("\nsyntax: rpcss [/d] [/s]\n\n"); + printf(" /d: debug mode (print debug messages to std. output)\n"); + printf(" /s: server mode (don't time out)\n\n"); +} + +void RPCSS_Spork() +{ + /* Because we are a wine app, we can't fork. This is + a trick to invoke ourselves, so we can provide a + similar effect of creating a "background" process */ + + PROCESS_INFORMATION pi; + STARTUPINFO si; + char *oldcmdline, *newcmdline; + + /* create a string to invoke ourselves */ + oldcmdline = GetCommandLineA(); + newcmdline = LocalAlloc(LPTR, strlen(oldcmdline) + 10); /* more than enough */ + assert(newcmdline); /* must succeed */ + + sprintf(newcmdline, "%s /^", oldcmdline); + + ZeroMemory(&pi, sizeof(PROCESS_INFORMATION)); + ZeroMemory(&si, sizeof(STARTUPINFO)); + si.cb = sizeof(STARTUPINFO); + + if (!CreateProcess( + NULL, // executable + newcmdline, // command line + NULL, // process security attributes + NULL, // primary thread security attributes + FALSE, // no inherit handles + 0, // creation flags + NULL, // use parent's environment + NULL, // use parent's current directory + &si, // STARTUPINFO pointer + &pi // PROCESS_INFORMATION + )) + assert(FALSE); + + LocalFree(newcmdline); +} + +int main( int argc, char **argv ) +{ + /* + * We are invoked as a standard executable; typically, we act in a + * "lazy" manner. We open up our pipe, and hang around until we have + * nothing left to do, and then silently terminate. When we're needed + * again, rpcrt4.dll.so will invoke us automatically. + */ + + if (!RPCSS_ProcessArgs(argc, argv)) { + RPCSS_Usage(); + return 1; + } + + if (! in_debug_mode) { + if (secret_fork_child) + FreeConsole(); + else { + /* pseudo-fork */ + RPCSS_Spork(); + return 0; + } + } else if (secret_fork_child) { + /* this should never happen... perhaps they typed /d /^ on cmd line */ + RPCSS_Usage(); + return 2; + } + + if (! in_server_mode) + /* we want to wait for something to happen, and then + timeout when no activity occurs... but we will be + extra patient on the first wait */ + RPCSS_SetLazyTimeRemaining(RPCSS_LONG_LAZY_TIMEOUT); + + RPCSS_main_loop(); + + return 0; +} -- gmt "The purpose of government is to rein in the rights of the people" --President Bill Clinton, MTV interview, 1993

7 15

Re: help support
by Sylvain Petreolle Nov. 9, 2002

Nov. 9, 2002

CHM is a special extension of IE. The patch to make any windows compatible with CHM (except Win 3.1) is downloadable at http://www.microsoft.com/downloads/release.asp?ReleaseID=30328 To handle and generate CHM I found a french page (tools are in english). http://www.jurixt.com/informatique/hlp_compile_aide.htm Utility to decompile CHM and many many more : http://www.keyworks.net/keytools.htm Google power ! ___________________________________________________________ Do You Yahoo!? -- Une adresse @yahoo.fr gratuite et en français ! Yahoo! Mail : http://fr.mail.yahoo.com

1 0