November 2010 - Wine-Devel - list.winehq.org

Re: include: Add new Button control styles
by Nikolay Sivov Nov. 1, 2010

Nov. 1, 2010

On 11/2/2010 01:45, André Hentschel wrote: > BS_TYPEMASK is already defined some lines above that > --- > include/winuser.h | 6 +++++- > 1 files changed, 5 insertions(+), 1 deletions(-) > > diff --git a/include/winuser.h b/include/winuser.h > index 4114756..e04e8ad 100644 > --- a/include/winuser.h > +++ b/include/winuser.h > @@ -712,8 +712,12 @@ typedef struct tagWINDOWPLACEMENT > #define BS_GROUPBOX 0x00000007L > #define BS_USERBUTTON 0x00000008L > #define BS_AUTORADIOBUTTON 0x00000009L > +#define BS_PUSHBOX 0x0000000AL > #define BS_OWNERDRAW 0x0000000BL > -#define BS_TYPEMASK 0x0000000FL > +#define BS_SPLITBUTTON 0x0000000CL > +#define BS_DEFSPLITBUTTON 0x0000000DL > +#define BS_COMMANDLINK 0x0000000EL > +#define BS_DEFCOMMANDLINK 0x0000000FL > #define BS_LEFTTEXT 0x00000020L > #define BS_RIGHTBUTTON BS_LEFTTEXT > New styles should go to commctrl.h, and you can't remove BS_TYPEMASK.

1 0

Re: [PATCH 3/3] winmm: MCI_INFO doesn't change the output buffer in case of error.
by Marvin Nov. 1, 2010

Nov. 1, 2010

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=6668 Your paranoid android. === WINEBUILD (build) === Patch failed to apply === W98SE (32 bit mci) === mci.c:835: Test failed: Expect message 0001 from play to 250 wait notify mci.c:842: Test failed: got 0004 instead of MCI_NOTIFY_xyz 0000 from command after close mci.c:903: Test failed: not enough time elapsed 58ms

1 0

Re: [PATCH 2/3] winmm: MCI_SYSINFO doesn't change the output buffer in case of error.
by Marvin Nov. 1, 2010

Nov. 1, 2010

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=6667 Your paranoid android. === WINEBUILD (build) === Patch failed to apply === W98SE (32 bit mci) === mci.c:959: Test failed: mci status position: 70 mci.c:1006: Test failed: mci pause wait returned MMSYSERR 5 mci.c:1020: Test failed: mci pause (space) wait returned MMSYSERR 5 mci.c:1023: Test failed: mci pause wait returned MMSYSERR 5 mci.c:1037: Test failed: Expect message 0004 from play (aborted by close)

1 0

Re: [PATCH 1/3] winmm: mciSendString always returns a response string (albeit empty).
by Marvin Nov. 1, 2010

Nov. 1, 2010

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=6666 Your paranoid android. === WINEBUILD (build) === Patch failed to apply === W98SE (32 bit mci) === mci.c:903: Test failed: not enough time elapsed 58ms mci.c:959: Test failed: mci status position: 58 mci.c:1006: Test failed: mci pause wait returned MMSYSERR 5 mci.c:1020: Test failed: mci pause (space) wait returned MMSYSERR 5 mci.c:1023: Test failed: mci pause wait returned MMSYSERR 5 mci.c:1037: Test failed: Expect message 0004 from play (aborted by close)

1 0

Re: Problems with Test
by James Mckenzie Nov. 1, 2010

Nov. 1, 2010

Charles Davis <cdavis(a)mymail.mines.edu> wrote: > >On 10/31/10 9:23 PM, James McKenzie wrote: >> All: >> >> I'm trying to convert one integer number using a float number factor. >> Here is the code: >> >> int ry = 125; >> double ryf; >> >> ryf = ry/1440; >> >> The result is 0 and should be 0.06666666667 (or even more sixes.) >And why should it be? None of the values in the expression 'ry/1440' are >of type 'double'. They're both 'int's, so the compiler sees it as >integer division instead of FP division. The C language spec doesn't >require the compiler to infer that you wanted FP division when you >assign to a variable of FP type. Ah, the wonders of 'c'. Thank you for the reminder that this is what happens during this division process. Been a long time since I've tried this. > >> What am I doing 'incorrectly'? >Now, if your assignment looked like this: > >ryf = ry/1440.0; > >then it would work correctly. Now the 1440 is of type 'double', so the >compiler emits a floating-point division. > Got it. I tried to cast ry to double, but that did not work. Guess my version of gcc has a problem or I did it wrong. Will try changing the division this evening and see what happens. James McKenzie

2 1

Re: [try 2] crypt32/tests: Add tests for enveloped messages.
by Marvin Nov. 1, 2010

Nov. 1, 2010

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=6658 Your paranoid android. === WNT4WSSP6 (32 bit msg) === msg: unhandled exception c0000005 at 5CF308F4

1 0

[do not apply] Re: crypt32/tests: Add tests for enveloped messages.
by Alexander Morozov Nov. 1, 2010

Nov. 1, 2010

Please do not apply this patch. Use [try 2]

1 0

Re: [PATCH 2/2] comdlg32/tests: Add GetSaveFileName extension tests.
by Marvin Nov. 1, 2010

Nov. 1, 2010

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=6657 Your paranoid android. === W98SE (32 bit filedlg) === filedlg.c:1094: Test failed: Filename is beef.txt, expected deadbeef === WNT4WSSP6 (32 bit filedlg) === filedlg.c:1094: Test failed: Filename is beef.txt, expected deadbeef === W2KPROSP4 (32 bit filedlg) === filedlg.c:1094: Test failed: Filename is beef.txt, expected deadbeef === WXPPROSP3 (32 bit filedlg) === filedlg.c:1094: Test failed: Filename is beef.txt, expected deadbeef === W2K3R2SESP2 (32 bit filedlg) === filedlg.c:1094: Test failed: Filename is beef.txt, expected deadbeef === WVISTAADM (32 bit filedlg) === filedlg.c:1094: Test failed: Filename is beef.txt, expected deadbeef === W2K8SE (32 bit filedlg) === filedlg.c:1094: Test failed: Filename is beef.txt, expected deadbeef === W7PRO (32 bit filedlg) === filedlg.c:1094: Test failed: Filename is beef.txt, expected deadbeef === W7PROX64 (32 bit filedlg) === filedlg: unhandled exception c0000005 at 750B4C19 === W7PROX64 (64 bit filedlg) === filedlg.c:1094: Test failed: Filename is beef.txt, expected deadbeef

1 0

Re: [4/4] d3d9: Add a sRGB format test
by Henri Verbeet Nov. 1, 2010

Nov. 1, 2010

On 1 November 2010 11:41, Stefan Dösinger <stefandoesinger(a)gmx.at> wrote: > > Am 01.11.2010 um 10:41 schrieb Henri Verbeet: > >> On 31 October 2010 22:53, Stefan Dösinger <stefan(a)codeweavers.com> wrote: >>> + static const struct >>> + { >>> + float x, y, z; >>> + float u, v; >>> + } >>> + quad[] = >>> + { >>> + {-1.0, -1.0, 0.1, 0.0, 0.0}, >>> + {-1.0, 1.0, 0.1, 1.0, 0.0}, >>> + { 1.0, -1.0, 0.1, 0.0, 1.0}, >>> + { 1.0, 1.0, 0.1, 1.0, 1.0} >>> + }; >> Not that I care all that much, but notice that those literals are >> actually doubles. > Now that I've fixed a number of those I guess we should stick to caring about that. You should probably do the Clear()'s too then.

1 0

Re: [PATCH 2/4] include: Add macros for retrieving control message headers.
by Alexandre Julliard Nov. 1, 2010

Nov. 1, 2010

Erich Hoover <ehoover(a)mines.edu> writes: > @@ -185,6 +185,25 @@ VOID WINAPI GetAcceptExSockaddrs(PVOID, DWORD, DWORD, DWORD, struct WS(sockaddr) > BOOL WINAPI TransmitFile(SOCKET, HANDLE, DWORD, DWORD, LPOVERLAPPED, LPTRANSMIT_FILE_BUFFERS, DWORD); > INT WINAPI WSARecvEx(SOCKET, char *, INT, INT *); > > +/* > + * Macros for retrieving control message data returned by WSARecvMsg() > + */ > +#define WSA_CMSG_DATA(cmsg) ((UCHAR*)((WSACMSGHDR*)(cmsg)+1)) > +#define WSA_CMSG_FIRSTHDR(mhdr) \ > + ((size_t) (mhdr)->Control.len >= sizeof(WSACMSGHDR) \ > + ? (WSACMSGHDR *) (mhdr)->Control.buf : (WSACMSGHDR *) 0) > +#define WSA_CMSG_ALIGN(len) (((len) + sizeof (size_t) - 1) \ > + & (size_t) ~(sizeof (size_t) - 1)) > +static inline WSACMSGHDR *WSA_CMSG_NXTHDR(WSAMSG *msg, WSACMSGHDR *cmsg) > +{ > + if ((size_t)(msg)->Control.len<sizeof(WSACMSGHDR)) return 0; > + cmsg = (WSACMSGHDR*)((unsigned char*)cmsg + WSA_CMSG_ALIGN(cmsg->cmsg_len)); > + if ((unsigned char*)(cmsg+1) > ((unsigned char*)msg->Control.buf + msg->Control.len) > + || ((unsigned char*)cmsg + WSA_CMSG_ALIGN(cmsg->cmsg_len) > ((unsigned char *)msg->Control.buf + msg->Control.len)) ) > + return 0; > + return cmsg; > +} This should go into ws2def.h. Also WSA_CMSG_NXTHDR is supposed to be a macro, not an inline function. -- Alexandre Julliard julliard(a)winehq.org

1 0