Module: website
Branch: master
Commit: acbf162ec92386c6998cf45a0273d50ea17f797a
URL: http://source.winehq.org/git/website.git/?a=commit;h=acbf162ec92386c6998cf4…
Author: Jeremy Newman <jnewman(a)codeweavers.com>
Date: Wed Apr 30 16:49:01 2008 -0500
fix broken <br /> tags
---
wwn/wn20080428_346.xml | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/wwn/wn20080428_346.xml b/wwn/wn20080428_346.xml
index a6500ed..b18e7bb 100644
--- a/wwn/wn20080428_346.xml
+++ b/wwn/wn20080428_346.xml
@@ -300,7 +300,7 @@ The big ones I can think of are the DIB Engine (an idea which has been
percolating for ages) and .net support. Oh, and the Javascript support
needed to install Photoshop CS3. Longer term, I'd like to see the
user-mode WDM driver framework supported.
-<br><br>
+<br /><br />
Long term, we'll probably find that we have to follow
Windows' example and implement application-specific behavior
to get key apps working perfectly without breaking other apps.
Module: website
Branch: master
Commit: 292655eccdeb713191aebdeb7277fb31d425cb4b
URL: http://source.winehq.org/git/website.git/?a=commit;h=292655eccdeb713191aebd…
Author: Zachary Goldberg <zgs(a)seas.upenn.edu>
Date: Wed Apr 30 11:40:58 2008 -0400
Another typo in Dan Kegel's interview.
---
wwn/wn20080428_346.xml | 12 ++++++------
1 files changed, 6 insertions(+), 6 deletions(-)
diff --git a/wwn/wn20080428_346.xml b/wwn/wn20080428_346.xml
index 6b7a205..a6500ed 100644
--- a/wwn/wn20080428_346.xml
+++ b/wwn/wn20080428_346.xml
@@ -300,6 +300,12 @@ The big ones I can think of are the DIB Engine (an idea which has been
percolating for ages) and .net support. Oh, and the Javascript support
needed to install Photoshop CS3. Longer term, I'd like to see the
user-mode WDM driver framework supported.
+<br><br>
+Long term, we'll probably find that we have to follow
+Windows' example and implement application-specific behavior
+to get key apps working perfectly without breaking other apps.
+Ugly, but true.
+
</p>
</quote>
<p>
@@ -312,12 +318,6 @@ the port to OS X has really opened the door on portability, is there
anything like that on the horizon? Do you ever think someone will have
ntoskrnl loading Windows drivers or hooking into the kernel or something
like that?
-<br />
-<br />
-Long term, we'll probably find that we have to follow
-Windows' example and implement application-specific behavior
-to get key apps working perfectly without breaking other apps.
-Ugly, but true.
</p>
<quote who="DanKegel">
<p>
Module: website
Branch: master
Commit: e3c1271fab4a841295bf8f23b2dfe148c4f6a8d2
URL: http://source.winehq.org/git/website.git/?a=commit;h=e3c1271fab4a841295bf8f…
Author: Zachary Goldberg <zgs(a)seas.upenn.edu>
Date: Tue Apr 29 14:31:55 2008 -0400
Fix Typo in Zumastor.org link
---
wwn/wn20080428_346.xml | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/wwn/wn20080428_346.xml b/wwn/wn20080428_346.xml
index 263c891..6b7a205 100644
--- a/wwn/wn20080428_346.xml
+++ b/wwn/wn20080428_346.xml
@@ -374,7 +374,7 @@ Linux [or OS X], what would it be?
<quote who="DanKegel">
<p>
I'd improve disk storage. For instance, I'd make LVM snapshots more
-space-efficient. (Actually, I *am* doing that; see <a href="http://zumastor.org.)">http://zumastor.org</a>.)
+space-efficient. (Actually, I *am* doing that; see <a href="http://zumastor.org">http://zumastor.org</a>.)
Also, right now, if you put your whole system on an LVM volume,
it's hard to read your disks without actually booting the system
from them; I'd fix that. (Maybe Dan Phillips will fix that, it's certainly
Module: wine
Branch: master
Commit: 3f89e1a0538c9b7dc6c14177c5c08b83aa6887a9
URL: http://source.winehq.org/git/wine.git/?a=commit;h=3f89e1a0538c9b7dc6c14177c…
Author: Roderick Colenbrander <thunderbird2k(a)gmx.net>
Date: Wed Apr 30 10:54:19 2008 +0000
opengl32: Drop the ChoosePixelFormat test as it is very unreliable.
---
dlls/opengl32/tests/opengl.c | 75 ------------------------------------------
1 files changed, 0 insertions(+), 75 deletions(-)
diff --git a/dlls/opengl32/tests/opengl.c b/dlls/opengl32/tests/opengl.c
index 1a3b770..db9504f 100644
--- a/dlls/opengl32/tests/opengl.c
+++ b/dlls/opengl32/tests/opengl.c
@@ -176,80 +176,6 @@ static void test_pbuffers(HDC hdc)
else skip("Pbuffer test for offscreen pixelformat skipped as no offscreen-only format with pbuffer capabilities has been found\n");
}
-static void test_choosepixelformat(HDC hdc)
-{
- int iPixelFormat;
- int nFormats;
- BOOL found=FALSE;
- PIXELFORMATDESCRIPTOR pfd = {
- sizeof(PIXELFORMATDESCRIPTOR),
- 1, /* version */
- PFD_DRAW_TO_WINDOW |
- PFD_SUPPORT_OPENGL |
- PFD_DOUBLEBUFFER,
- PFD_TYPE_RGBA,
- 32, /* 32-bit color depth */
- 0, 0, 0, 0, 0, 0, /* color bits */
- 0, /* alpha buffer */
- 0, /* shift bit */
- 0, /* accumulation buffer */
- 0, 0, 0, 0, /* accum bits */
- 0, /* z-buffer */
- 0, /* stencil buffer */
- 0, /* auxiliary buffer */
- PFD_MAIN_PLANE, /* main layer */
- 0, /* reserved */
- 0, 0, 0 /* layer masks */
- };
-
- /* Below we test the behavior of ChoosePixelFormat. As documented on MSDN this
- * function doesn't give any guarantees about its outcome. Programs should not
- * rely on weird behavior of the function but unfortunately a few programs like
- * e.g. Serious Sam TSE rely on it.
- *
- * MSDN documents of a few flags like double buffering / stereo that they can be set to DONTCARE.
- * It appears that a 0 value on other options like alpha, red, .. means DONTCARE. The hypothesis
- * is that ChoosePixelFormat returns the first available format which matches the criteria.
- *
- * This test tries to proof the DONTCARE behavior by passing an almost 'empty' pfd to
- * ChoosePixelFormat. The pfd only has some really needed flags (RGBA, window, double buffer) set.
- * Further a 32 bit color buffer has been requested. The idea is that when a format with e.g. depth or stencil bits
- * is returned, while there are also 'better' candidates in the list without them (but located AFTER the returned one)
- * that an option set to zero means DONTCARE. We try to proof this by checking the aux/depth/stencil bits.
- * Proofing this behavior for the color bits isn't possible as all formats have red/green/blue/(alpha), so we assume
- * that if it holds for aux/depth/stencil it also holds for the others.
- *
- * The test below passes at least on various ATI cards (rv250, r300) and Nvidia cards.
- */
-
- iPixelFormat = ChoosePixelFormat(hdc, &pfd);
- if(iPixelFormat) {
- PIXELFORMATDESCRIPTOR pfd_tmp;
- BOOL res;
- int i;
-
- memset(&pfd, 0, sizeof(PIXELFORMATDESCRIPTOR));
- res = DescribePixelFormat(hdc, iPixelFormat, sizeof(PIXELFORMATDESCRIPTOR), &pfd);
-
- nFormats = DescribePixelFormat(hdc, 0, 0, NULL);
- /* Start testing from iPixelFormat, second formats start counting from index=1, so use '<=' */
- for(i=iPixelFormat; i<=nFormats; i++) {
- memset(&pfd_tmp, 0, sizeof(PIXELFORMATDESCRIPTOR));
- res = DescribePixelFormat(hdc, i, sizeof(PIXELFORMATDESCRIPTOR), &pfd_tmp);
- if(!res)
- continue;
-
- /* Check if there is a format which better matches the requirements */
- if((pfd_tmp.cAuxBuffers < pfd.cAuxBuffers) || (pfd_tmp.cDepthBits < pfd.cDepthBits) || (pfd_tmp.cStencilBits < pfd.cStencilBits))
- found = TRUE;
- }
-
- /* When found=TRUE we were able to confirm our hypothesis */
- ok(found == TRUE, "Unable to confirm DONTCARE behavior of unset pixelformatdescriptor flags\n");
- }
-
-}
-
static void test_setpixelformat(HDC winhdc)
{
int res = 0;
@@ -472,7 +398,6 @@ START_TEST(opengl)
ok(res, "wglMakeCurrent failed!\n");
init_functions();
- test_choosepixelformat(hdc);
test_setpixelformat(hdc);
test_colorbits(hdc);
test_gdi_dbuf(hdc);