- Wine-Devel - list.winehq.org

Re: NTFS driver (was: ReactOS GPL vs. proprietary drivers)
by Szakacsits Szabolcs Oct. 22, 2003

Oct. 22, 2003

Hi, Jan Kratochvil <rcpt-ros-kernel.AT.reactos.com_at_jankratochvil.net> wrote: > As GPLed Linux-NTFS still has no NTFS r/w capability I completed the > project for reliable r/w access in Wine way by using MS-Windows > ntfs.sys driver. The goal of the Linux-NTFS project is the provide completely open source code to access, manage NTFS filesystems, not only r/w access whatever way. It's a big difference. There is already a full r/w driver for Linux by Paragon for a while - although it costs money but it doesn't have legal threats. Why the open source write capability isn't achieved? In short, no active development for over a year and before that the code base was completely rewritten, after former developer(s) stopped working on it 3-4 years ago. Otherwise write access is pretty close, I'd say 2-4 months hard work (if one isn't familiar with NTFS that's another 1-3 months). > There are rumours GPL may be problem for linking with proprietary code > (ntfs.sys&ntoskrnl.exe) >From GPL point of view I don't think so. But I could imagine one isn't allowed to use ntfs.sys, ntoskrnl.exe the way he/she wants legally. They need to be checked (plus consider what's legal today can be illegal tomorrow, so keep checking). Or just ask Microsoft. At the same time you could also ask if they have patents on NTFS. I couldn't find any contrary to the rumours. There was a fuss years ago about Microsoft wanted to sue Linux NTFS developers, but the fact seems to be Microsoft threatened only Jeff V. Merkey because he had signed NDA with Microsoft, still he was developing and distributing NTFS tools for Linux. Microsoft denied all of this publicly. You can google up the full story but IMHO it's better just to ask Microsoft. Szaka

3 10

Re: Fix comctl32 conformance test
by Hans Leidekker Oct. 22, 2003

Oct. 22, 2003

On Wednesday 22 October 2003 08:52, Rolf Kalbermatter wrote: > the build process. Seems like MS has to fix a few things in their SDK > to make their import libs match the SDK documentation after those > functions really got available in the first release of W95 and finally > got documented this year. Yes, but what counts for us is what actually exists in their import libs, regardless of documentation. Of course, documentation would probably mean more users of those functions, but (if correct) it would also mean easier implementation for Wine. > Just a thought, can I use Wine import libs in Visual C ;-) ? I haven't tried (I don't have MSVC). But it should be possible. I think Wine is supposed to achieve this type of compatibility as well. You could then - say - use MSVC to write an application that links against a wine-only lib - say - x11drv. -Hans

1 0

wine and Crossover in the press
by Tom Oct. 22, 2003

Oct. 22, 2003

Anyone had a look at this? http://www.theinquirer.net/?article=12207 Tom

2 2

GetProcessWindowStation
by Mathias Berchtold Oct. 22, 2003

Oct. 22, 2003

Hello ... I get the following error when trying to run SmartFTP on Wine. fixme:advapi:RevertToSelf (), stub fixme:advapi:SetThreadToken ((nil), 0xa0): stub (NT impl. only) fixme:win:GetProcessWindowStation (void): stub fixme:win32:GetUserObjectInformationA (0x1 1 0x4078e8e4 12 0x4078e8d8),stub! And afer that I get a Debug ASSERT from the Microsoft Visual C++ Library. File: Allocate.cpp Line: 108 Expression: bSuccess I think the Assert is caused because of the missing function implementations. Do you have any plans to implement them in the future? Wine Information ================ wine --version Wine 20031016 windows="win2k" ("win98" doesn't work because wine "ignores" the unicows.dll) [AppDefaults\\smartftp.exe\\DllOverrides] "ole32" = "native" "oleaut32" = "native" "rpcrt4" = "native" "urlmon" = "native" "comctl32" = "builtin" "wininet" = "builtin" "shdocvw" = "native" "shlwapi" = "native" "crypt32" = "native" "mshtml" = "native" "unicows" = "native" "crypt32" = "native" "cryptui" = "native" "wintrust" = "native" If you need the debug build of SmartFTP to test please contact me. Thanks a lot -Mat SmartFTP.com -----Original Message----- From: wine-devel-admin(a)winehq.org [mailto:wine-devel-admin(a)winehq.org] On Behalf Of Alexandre Julliard Sent: Mittwoch, 22. Oktober 2003 04:20 To: Vincent Béron Cc: Wine devel Subject: Re: Changes to kernel messages lang Vincent Béron <vberon(a)mecano.gme.usherb.ca> writes: > Attached program says so on W2K SP4. There's a message table in > kernel32 for 0x0000, for the others it returns NULL. Is that way to do > it different from yours? Yes, FindResource with language 0 is going to return a resource in all sorts of cases, so it's not a good test. And actually if the resource was really NEUTRAL, then all your FindResource calls should have succeeded. The best way to find out is to dump the resource table of the dll, using something like winedump -x. -- Alexandre Julliard julliard(a)winehq.com

2 1

Re: NTFS driver (was: ReactOS GPL vs. proprietary drivers)
by Fabian Cenedese Oct. 22, 2003

Oct. 22, 2003

Hi I don't know much about NTFS (only using it) and even less about Linux-NTFS. But it was funny following this conversation. >> Filesystem must be the rock solid data storage structure. You must >> know the meaning of each byte (*) for such reliable and interoperable >> filesystem. > >Exactly. Every needed byte is known. Wow! Really? >> (*) You do not need to know the journalling metadata as long as you do not >> support journalling and/or its recovery. > >One of the unknown issues is journaling :) AHA! And not only "The unknown issue" but "One of the unknown issues". Reminds me of Asterix: Whole Europe is part of the Roman Empire. Whole...? Sorry, I couldn't resist, no offense :) bye Fabi

2 1

Xlib and wine and XGrabButton
by Gerhard W. Gruber Oct. 22, 2003

Oct. 22, 2003

I currently write a small desktop app (in case you are curious http://sourceforge.net/projects/launchmenu) and for this I have to research on how to grab mouse events and eventually keyboard as well. Now I thought that this might be usefull for wine when playing games under the DirectX emulation, so I checked the sources for XGrabButton()/XGrabKey() and I couldn't find any references to it. When I tried to play Thief and Thief 2 I always had the problem that the keys are going to the shell where I started it from, instead of going into the game. I wrote a small test app for my project in order to get this feature straight and there I could capture all mouse events (haven't tried keyboard, but it should work equally well) without any interference with the remaining X system. Is this intenttionally that these functions are not used and if so what is the reason for it? Or is there a layer in between wine and X so that wine is not using Xlib API directly? I think that using this should work at least in programs where the entire screen is taken over, like in games, probably other applications well. -- Gerhard Gruber Für jedes menschliche Problem gibt es immer eine einfache Lösung: Klar, einleuchtend und falsch. (Henry Louis Mencken)

3 4

Re: Changes to kernel messages lang
by Alexandre Julliard Oct. 22, 2003

Oct. 22, 2003

Vincent Béron <vberon(a)mecano.gme.usherb.ca> writes: > W2k kernel messagetable is LANG_NEUTRAL, SUBLANG_NEUTRAL only. > > Changelog: > Make the kernel messagetable LANG_NEUTRAL, as native. Are you sure about that one? All the dlls I have checked (Win95, NT4, XP) have the language set to 0x0409. In what case is that causing trouble? -- Alexandre Julliard julliard(a)winehq.com

2 3

Re: More dutch translations
by Alexandre Julliard Oct. 21, 2003

Oct. 21, 2003

Hans Leidekker <hans(a)it.vu.nl> writes: > Changelog: > - More resources translated to dutch. > - Make an example consistent with it's description (wcmd). The wcmd change is wrong, the default prompt really is $p$g. Also your patch got wrapped, you'll have to resubmit. -- Alexandre Julliard julliard(a)winehq.com

1 0

S3 texture compression patent
by Jonathan Wilson Oct. 21, 2003

Oct. 21, 2003

What specificly does this patent cover and what parts of WINE would it affect? (i.e. which bits of windows, if implemented in WINE, would fall under the patent?)

6 7

Re: Summary of translations
by Vincent Béron Oct. 21, 2003

Oct. 21, 2003

Le mar 21/10/2003 à 14:45, Hans Leidekker a écrit : > This is certainly useful for me, as I was just about to start translating > some dlls to dutch. I checked the page you mention above and it showed > there are some parts missing from dlls that already haven been translated > to dutch. I don't think I would have found those otherwise (I wouldn't have > gone looking for them). So I think this tool can certainly prove valuable in > keeping resources up to date. That's why I did it (after I saw Dimi's output). Although now, all French resources seem to be translated... :) Glad to know it helped you too. > > And now for some criticisms. I can understand that you only check languages > for which at least one resource exists because otherwise the table would grow > large and show lots of empty cells for years to come, but the first thing I need > to set out translating is a list of dlls that have translatable resources, which I > now obtain by doing: > > [wine-src] $ find dlls -name "*En.rc" > > Which is crude and depends on a file naming convention. Maybe your tool > could help here too. If there's an html file, the dll contains some resources. Now, not all resources are translatable (icons, cursors, bitmaps we probably don't care, not sure about versioninfo). That can be integrated in the tool too. For now, if there's only a LANG_NEUTRAL version of a resource, it doesn't complain, else it creates the file. > > Second, those filenames for the html output really aren't fit to be served > by apache. They're truncated in a directory listing page so I have to hover > the links to be able to find a particular file. I think you need to either split > them into directories or provide an index page. As Dimi pointed out, all on a single page would be better. Or a couple of pages, with AAFRIKANS-GERMAN on one, HUNGARIAN-SLOVAK, etc. to prevent the table from being too wide, and marking only the relevant languages (ie, those for which at least one translated resource exists). Another thing is languages for which more than one sublanguage is used, such as Chinese. > > Third, I think you should sort the table columns alphabetically by language > name. But they're sorted! Ok, by language ID :) Good idea. > > Lastly, Konqueror doesn't render those popup/layer thingies well, and in > Mozilla I can't can rid of a popup once I clicked a link (well, I can press > the Back button, but that's non-intuitive to say the least). To be honest I > think they're overkill. Make them a regular link. Remember, you said all > comments were welcome ;-) No problem with that. I tried to get the same feeling as Mozilla's Tinderbox, but I can make them regular links instead. Or "tooltips". Vincent

1 0