July 2009 - Wine-devel - list.winehq.org

Re: Suggestions about the website of www.winehq.org
by Dan Kegel 15 Jul '09

15 Jul '09

On Mon, Jul 13, 2009 at 9:47 PM, <axel_xia(a)sohu.com> wrote: > I am a Chinese user for Fedora and have be trying to use wine for very long > time. Based on my experience in wine evaluation, may I provide my > suggestions on wine development & prospect? Yes, please! But please don't cross-post. Your question probably belongs on wine-devel or wine-users. cc'ing five different mailing lists is not a good idea. > 1) There are many many discussions in the internet about the Chinese font > sets issue in wine (wine very very slow, or can not display Chinese > correctly, & etc). However, most popular Linux distributions almost have no > issues for Chinese character display nowadays, the Chinese version of $MS > windows are perfectly compatible with Chinese character and even all the > softwares for $MS windows use default $MS Chinese fonts without any issues. > So, is that possible for wine/crossover to directly use/share Chinese fonts > (even all fonts of different languages) in linux distributions? As far as I know, the MS chinese fonts are copyrighted, and their license terms do not permit redistribution. (The "microsoft core fonts" were distributed under a very permissive license, and are more or less available on all linux systems, but they don't have good chinese fonts.) If you know of any legally downloadable chinese fonts, please let me know and I'll at least add them to winetricks. If you're saying that Linux already comes with great Chinese fonts, and Wine isn't using them properly, then that sounds like a bug we need to fix. There are two bugs open right now for CJK fonts: http://bugs.winehq.org/show_bug.cgi?id=13829 (fonts aren't found properly) http://bugs.winehq.org/show_bug.cgi?id=16325 (fonts don't look good) Perhaps you should file one for the slow fonts issue. > 2) Is that possible to have someone from winehq.org to verify $MS software > applicable in wine or not(at lease which .dll should be copy into > ~./wine/.../system32)? I found thousands of $MS softwares are posted on > AppDB in www.winehq.org but only checked by evaluators. So I wonder if > possible for winehq.org people to validate these programs and even better to > tell visitors step by step to run each software in wine. I think it is > better to post the correct steps of $MS software emulation in AppDB because > we need to rely on the offical website. It is harmful that users don't think > AppDB is believable then less and less people will come to www.winehq.org > for help day by day. Sadly, we don't have enough manpower to keep appdb up to date ourselves; we rely on volunteers. Our goal is for there to be no steps needed beyond running setup and starting the app, so that's what Wine developers tend to spend their time on, and we leave appdb updating to volunteers. That said, it would be helpful for somebody to make a pass over the popular apps in the appdb and update the howtos, including converting ad hoc installation steps into winetricks steps as appropriate. Any volunteers? > 3) Is that possible to open accounts in Forums of winehq website for common > users? And also is that possible to have different forums in Chinese or in > other languages established worldwide? That's a good idea. We can also link to any other Wine forums that already exist in other languages; do you know of any in Chinese? Thanks again for writing. We've known for a long time that we have some issues with Asian languages, but since most of our developers are not fluent in Asian languages, we haven't made much progress on it. We rely on feedback like the messages you've been sending lately. - Dan

2 1

Wine Appdb is down
by nn 15 Jul '09

15 Jul '09

Hi, Wine Appdb is down ---------------------------------- Warning: mysql_connect() [function.mysql-connect]: Can't connect to local MySQL server through socket '/var/run/mysqld/mysqld.sock' (2) in /home/winehq/opt/appdb/include/query.php on line 82 Database error, please try again soon: Can't connect to local MySQL server through socket '/var/run/mysqld/mysqld.sock' (2) ---------------------------------- ____________________________________________________________________________________ Access Yahoo!7 Mail on your mobile. Anytime. Anywhere. Show me how: http://au.mobile.yahoo.com/mail

4 3

Re: wpp: add I/O hooks [try 2]
by Juan Lang 15 Jul '09

15 Jul '09

Hi Matteo, it'd be easier to review your patch if you sent it along with a subsequent patch that made use of the I/O hooks. As it is, it's hard to know why this is necessary. Cheers, --Juan

2 1

Re: [4/4] WineD3D: Don't call glGetError if nobody is going to see our fixme
by Chris Robinson 14 Jul '09

14 Jul '09

It may still be important to do this, if glGetError is ever used directly to check for errors. Errors that are normally caught by checkGLError would then be left for a later call to glGetError to pick up.

2 1

Re: winhttp(1/12): Implement WinHttpGetDefaultProxyConfiguration
by Alexandre Julliard 14 Jul '09

14 Jul '09

Juan Lang <juan.lang(a)gmail.com> writes: > +static inline void copy_char_to_wchar_sz(const BYTE *src, DWORD len, WCHAR *dst) > +{ > + const BYTE *begin; > + > + for (begin = src; src - begin < len; src++, dst++) > + *dst = *src; > + *dst = 0; > +} That looks very wrong. Is there really evidence that Unicode chars are truncated? > + l = RegQueryValueExW( key, WinHttpSettings, NULL, &type, NULL, &size ); > + if (!l && type == REG_BINARY && size >= 5 * sizeof(DWORD)) > + { > + BYTE *buf = HeapAlloc( GetProcessHeap(), 0, size ); > + > + if (buf) > + { > + DWORD *magic = (DWORD *)buf; > + DWORD *unknown = magic + 1; > + DWORD *flags = unknown + 1; > + DWORD *len = flags + 1; Please define an appropriate structure. -- Alexandre Julliard julliard(a)winehq.org

2 2

transl question
by Aric Stewart 14 Jul '09

14 Jul '09

I am looking at the Japanese report from transl and i see something odd. looking at DIALOG MSGBOX for user32... http://source.winehq.org/transl/resource.php?lang=011%3A00&resfile=dlls%2Fu… the STYLE is reported as different, but when I look at the source they appear identical. Is it because the Japanese resource has FONT 9, "MS Shell Dlg" specified? Should that show up as an Error to be corrected? -aric

3 2

Re: 2/2 winex11: add XRender based GetSrcAreaStretch
by Aurimas Fišeras 14 Jul '09

14 Jul '09

On 07/11/2009 10:08 PM, Roderick Colenbrander wrote: > Hi, > > This patch adds a XRender based GetSrcAreaStretch. A lot of work is > offloaded to X (less round trips are needed) and it can take advantage > of hardware acceleration. On Nvidia and AMD drivers blt performance > has improved by 20% in CrystalMark2004 while more optimizations can be > added later on. I get a similar performance increase on intel driver too.

2 1

Transl: Getting rid of the conf directory (Part 1)
by Paul Vriens 14 Jul '09

14 Jul '09

Hi, In an effort to make conf more dynamic (and less depending on hardcoded configuration) I'd like to get rid of the conf directory. Let's make a first distinction here. We currently have a conf directory used by the scripts/* files and one by the php/* files. I'd like to discuss the scripts/conf directory first. Once that is done we can start discussing the php/conf directory The scripts conf directory ========================== The conf directory shows all the languages currently known by transl to be in use for translations. (We get a warning if a language is found in the source and we don't have a conf file for it.) The most prominent use of the conf directory is to create a list of languages that will show up on the index page. The conf file could hold a parameter "[ignore-sublang]". This parameter is used to change a LANG_XXX, SUBLANG_* into LANG_XXX, SUBLANG_NEUTRAL. So, if we want to get rid of the hardcoded conf directory we need some logic that replaces the above. The logic should hold the following rules: 1. LANG_XXX, SUBLANG_XXX_YYY is the highest level possible so we don't have to do anything here. 2. LANG_XXX, SUBLANG_DEFAULT (only 1 sublanguage in our code) should show LANG_XXX, SUBLANG LANG_XXX_YYY. 3. LANG_XXX, SUBLANG_DEFAULT (multiple sublanguages in our code) should show all possible LANG_XXX, SUBLANG_XXX_YYY. 4. LANG_XXX, SUBLANG_NEUTRAL (only 1 sublangugage in our code) should show LANG_XXX, SUBLANG_XXX_YYY. 5. LANG_XXX, SUBLANG_NEUTRAL (multiple sublanguages in our code) should mean show all possible LANG_XXX, SUBLANG_XXX_YYY. (This logic is already somehow in place when the Portuguese issue was solved.) I propose the following changes: 1. The first thing that should happen before any parsing is done is to get a list of all the different languages Wine is translated in. 2. Automatically create a scripts/conf directory that is based on the list found in step 1. This is temporary until the php part is converted as well. These conf files don't need the [hide] parameter or the [ignore-sublang] parameter anymore. 3. Change some of the php scripts to get rid of the two parameters mentioned in step 2. This basically means that we are not going to show pages anymore for the neutral languages (English is the exception). Note for step 2. The php scripts rely on the php/conf (copy of scripts/conf) directory for retrieving the name of a language. The kernel32 resources also have these names albeit sometimes a little different. I've noted the following differences: conf-dir kernel32 ======== ======== Chinese (Simplified) Chinese (PRC) Chinese (Traditional) Chinese (Taiwan) English (US) English (United States) English (Neutral) English Portugese (Brazilian) Portuguese (Brazil) Portugese (Portugal) Portuguese (Portugal) Do we want to go with the kernel32 names or should we have some kind of name override? To prove my point I just looked at the list of translations we have, created the necessary files in scripts/conf, got rid of the 2 parameters and ran the Master.sh script to regenerate everything. The results were as expected and looked basically the same as on our transl page. After changing the index page to not use the [hide] parameter anymore but some other logic it looked exactly the same. Thoughts, ideas, suggestions, questions on all of this? (As said he php/conf directory can/will be discussed in a second email once this is "solved"). -- Cheers, Paul.

2 8

Re: Add stub for IoOpenDeviceRegistryKey
by Paul Chitescu 14 Jul '09

14 Jul '09

+ FIXME( "stub:\n"); + return STATUS_SUCCESS; This is incorrect, since you are adding a stub you must not return success as it would imply that *DevInstRegKey is filled with a valid handle. You must either return an error or provide a full valid implementation (unlikely). Anything else is your private hack, unsuitable for including in the public wine repository since it could crash some drivers.

1 0

Re: [2/8] wbemprox: Add a stub implementation of IWbemLocator.
by Alexandre Julliard 14 Jul '09

14 Jul '09

Hans Leidekker <hans(a)codeweavers.com> writes: > diff --git a/dlls/uuid/uuid.c b/dlls/uuid/uuid.c > index 18b4d55..785d690 100644 > --- a/dlls/uuid/uuid.c > +++ b/dlls/uuid/uuid.c > @@ -76,6 +76,7 @@ DEFINE_GUID(GUID_NULL,0,0,0,0,0,0,0,0,0,0,0); > #include "isguids.h" > #include "objsafe.h" > #include "perhist.h" > +#include "wbemcli.h" There doesn't seem to be any Wbem clsid in the Windows libuuid. -- Alexandre Julliard julliard(a)winehq.org

1 0