Re: Wine FAQ edits
On Sat, Feb 05, 2011 at 09:49:30PM -0500, Albert Lee wrote: Anyone using the original Bourne sh *interactively* is in for a world of pain anyway (speaking as someone who often had to before the ksh93 replacement). I think relying on tilde expansion can can be considered good practice for the interactive use case. I wouldn't rely on ~ expansion - not unless you detect that it has returned empty string. Although anyone using a system that doesn't support it (most likely Solaris's /bin/sh) will probably know how to avoid the problem, not knowing what is wrong is more problematical. Adding a few 'sh' conditional substitutions to a script might show some people just how powerful the language is - especially if they have onely ever seen windows cmd.com .bat scripts. I've seen some recent systems fail to expand ~ properly (for some definition of the word 'properly') most noticably cygwin on windoes 7. David -- David Laight: da...@l8s.co.uk
Re: Pulling Patch
On Sat, Feb 5, 2011 at 5:48 PM, Shachar Shemesh shac...@shemesh.biz wrote: On 05/02/11 00:24, James McKenzie wrote: Actually, the latest patch is what I don't want reused. And no, you don't put it in the LGPL until it is committed, which I don't expect AJ to do anyway. However, I'm moving in a different direction since my Mac needs more repairs than I'm willing to spend money on. Besides, I've been a big enough pain that my existence here is unwarranted and unneeded. As anyone who attended the last WineConf probably already knows, you have my complete sympathies in that regard. I also doubt very much anyone would use your uncommitted patches against your will, so in that respect, you probably have nothing to worry about. That said, I believe your claim to the right to demand no use is wrong. It is my understanding that by submitting your patches to wine-patches, you have placed them under the LGPL, which is a non-revocable license. Again, in all likely hood, this is a purely hypothetical question. If the LGPL is non-revocable, is code you've placed under it still re-licensable, by you, under another license, as long as you don't revoke the LGPL in the process? ie. could I submit a piece of code to Wine and to another project? Shachar -- Shachar Shemesh Lingnu Open Source Consulting Ltd. http://www.lingnu.com Damjan Jovanovic
Re: Pulling Patch
On 06/02/11 11:13, Damjan Jovanovic wrote: On Sat, Feb 5, 2011 at 5:48 PM, Shachar Shemesh shac...@shemesh.biz mailto:shac...@shemesh.biz wrote: On 05/02/11 00:24, James McKenzie wrote: Actually, the latest patch is what I don't want reused. And no, you don't put it in the LGPL until it is committed, which I don't expect AJ to do anyway. However, I'm moving in a different direction since my Mac needs more repairs than I'm willing to spend money on. Besides, I've been a big enough pain that my existence here is unwarranted and unneeded. As anyone who attended the last WineConf probably already knows, you have my complete sympathies in that regard. I also doubt very much anyone would use your uncommitted patches against your will, so in that respect, you probably have nothing to worry about. That said, I believe your claim to the right to demand no use is wrong. It is my understanding that by submitting your patches to wine-patches, you have placed them under the LGPL, which is a non-revocable license. Again, in all likely hood, this is a purely hypothetical question. If the LGPL is non-revocable, is code you've placed under it still re-licensable, by you, under another license, as long as you don't revoke the LGPL in the process? ie. could I submit a piece of code to Wine and to another project? First, IANAL. You do not give up your copyright when you license code under the LGPL (or any other open source license). You merely provide a license (which is irrevocable). As such, the answer is yes. You can license code for which you own the complete copyrights under as many licenses of any type you wish to as many recipients as you wish, even if the licenses conflict. That said, if the copyright is only for derivative work, then you also need a license for the original work. The only license you have for the original work in the case of Wine is the LGPL, and THAT LICENSE is conditioned upon the fact that you license your own code under the LGPL only. As such, you cannot license changes to wine under another license, despite the fact you have the copyright for it, as that would leave you without the license to create your derivative work in the first place. So the real question is how independent your code is that you wish to submit. As long as you do not copy code from wine, you can submit the same change to as many open source projects as you like (even if their licenses are conflicting), and even use it for a proprietary project. If, however, the code requires Wine code in order to make sense, then you are bound by the LGPL and need to only use the code in a compliant way. Shachar -- Shachar Shemesh Lingnu Open Source Consulting Ltd. http://www.lingnu.com
Re: Wine FAQ edits
On Sat, 5 Feb 2011 21:49:30 -0500 Albert Lee tr...@forkgnu.org wrote: I think relying on tilde expansion can can be considered good practice for the interactive use case. Tilde expansion doesn't work if there are quotes around the path, and if you search the forum, you will find instances of users who did not know that and didn't understand why they were getting No such file or directory messages. $HOME is better because it can be used inside quotes, but what I have found to be most understandable to newbies is simply /home/user/ with a note telling them to substitute their username for user. (Yes, there are people who need to be told that. This is the level of technical knowledge you are writing for in the FAQ.) -- Rosanne DiMesio dime...@earthlink.net
Re: user32: Don't try to move mouse to (-1,0).
On 2/6/11 5:33 AM, Vitaliy Margolen wrote: On 02/05/2011 07:10 AM, Jacek Caban wrote: This trigers screen edge action on KDE, which is 'Present Windows' in default configuration. mouse_event() generates relative mouse movement events unless MOUSEEVENTF_ABSOLUTE is specified. If you don't want a pointer to get near the edge then you should use SetCursorPos() instead. It's enough that we don't try to move the cursor up nor left when it's in the edge. Jacek
Re: d3dx9: Implement D3DXFloat16to32Array and D3DXFloat32to16Array.
Sorry guys. Looks like there are differences between Windows version. Sigh. Will fix. Henri, will you get a chance to look at the patch before I re-send so that I can fix any other defects you might see? I don't want to send too many resends to the list. Thank you so much Misha On Sun, Feb 6, 2011 at 9:32 PM, Marvin test...@testbot.winehq.org wrote: Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at http://testbot.winehq.org/JobDetails.pl?Key=8981 Your paranoid android. === WXPPROSP3 (32 bit math) === math.c:2238: Test failed: Got 7fff, expected 7c00 for index 7. === WVISTAADM (32 bit math) === math.c:2238: Test failed: Got 7fff, expected 7c00 for index 7. === W7PRO (32 bit math) === math.c:2238: Test failed: Got 7fff, expected 7c00 for index 7.
Re: d3dx9: Implement D3DXFloat16to32Array and D3DXFloat32to16Array.
Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at http://testbot.winehq.org/JobDetails.pl?Key=8981 Your paranoid android. === WXPPROSP3 (32 bit math) === math.c:2238: Test failed: Got 7fff, expected 7c00 for index 7. === WVISTAADM (32 bit math) === math.c:2238: Test failed: Got 7fff, expected 7c00 for index 7. === W7PRO (32 bit math) === math.c:2238: Test failed: Got 7fff, expected 7c00 for index 7.
