Re: [Bug 656] New - aol update fails, time conversion possibly broken

[Andreas Mohr]

On Tue, May 07, 2002 at 05:10:18PM -0500, [e-mail] wrote:
> http://bugs.winehq.com/show_bug.cgi?id=656
>
> *** shadow/656	Tue May  7 17:10:18 2002
> --- shadow/656.tmp.9221	Tue May  7 17:10:18 2002
>
> *** 0 ****
> --- 1,43 ----
> - +============================================================================+
> - | aol update fails, time conversion possibly broken                          |
> - +----------------------------------------------------------------------------+
> - |        Bug #: 656                         Product: Wine                    |
> - |       Status: UNCONFIRMED                 Version: unspecified             |
> - |   Resolution:                            Platform:                         |
> - |     Severity: normal                   OS/Version: All                     |
> - |     Priority: P1                        Component: wine-files              |
> - +----------------------------------------------------------------------------+
> - |  Assigned To: [e-mail]                                         |
> - |  Reported By: [e-mail]                                         |
> - |      CC list: Cc:                                                          |
> - +----------------------------------------------------------------------------+
> - |    Milestone: TargetMilestone: ---                                         |
> - |          URL:                                                              |
> - +============================================================================+
> - |                              DESCRIPTION                                   |
> - When you exit aol, it downloads an update, a UTF file, and that update fails to install.
> -
> - I'm almost positive that this is the cause:
> -
> - 080e42d8:Call kernel32.CreateFileA(40566544 "C:\Program Files\America Online
> - 7.0\TOD\ADELETE.AAA",80000000,00000000,00000000,00000003,00000000,00000000) ret=004031a5
> - 080e42d8:Ret  kernel32.CreateFileA() retval=00000040 ret=004031a5
> - 080e42d8:Call kernel32.GetFileTime(00000040,00000000,00000000,40566648) ret=004031b9
> - 080e42d8:Ret  kernel32.GetFileTime() retval=00000001 ret=004031b9
> - 080e42d8:Call kernel32.FileTimeToLocalFileTime(40566648,40566650) ret=004031cb
> - 080e42d8:Ret  kernel32.FileTimeToLocalFileTime() retval=00000001 ret=004031cb
> - 080e42d8:Call kernel32.FileTimeToSystemTime(40566650,40566658) ret=004031d9
> - 080e42d8:Ret  kernel32.FileTimeToSystemTime() retval=00000001 ret=004031d9
> - 080e42d8:Call kernel32.CloseHandle(00000040) ret=00403213
> - 080e42d8:Ret  kernel32.CloseHandle() retval=00000001 ret=00403213
> - 080e42d8:Call kernel32.lstrcmpiA(40566440 "20011008164848",4056633c "20011008214848") ret=0040322c
> - 080e42d8:Ret  kernel32.lstrcmpiA() retval=ffffffff ret=0040322c
> - 080e42d8:Call user32.wvsprintfA(40565e0c,00410708 "   TODPartCheck (%s, %d, %s) failed: %d bytes <%s>\n",4056631c) ret=00402441
> - 080e42d8:Ret  user32.wvsprintfA() retval=0000007a ret=00402441
> - 080e42d8:Call kernel32.lstrlenA(40565e0c "   TODPartCheck (C:\Program Files\America Online 7.0\TOD\ADELETE.AAA, 3, 2001100"...)
> - ret=00402383
> - 080e42d8:Ret  kernel32.lstrlenA() retval=0000007a ret=00402383
> - 080e42d8:Call kernel32.GlobalAlloc(00002042,0000007a) ret=0040239b
> - 080e42d8:Ret  kernel32.GlobalAlloc() retval=4037c4a6 ret=0040239b
> -
> - Which means that one of those *Time*To*Time* functions are broken.

Hmm, yeah, maybe.
I've also suspected these functions lately with some program,
but I quickly went on to check for another cause of the failure.
Seems like it may really have been the file time functions, though...

Well, Oct. 08 2001, 16:48:48 vs. Oct. 08 2001, 21:48:48 sounds like a
5 hour difference to me...
(which almost has to be a bug, since it'd be pretty amazing if such
an accurate difference of exactly 5 hours was by coincidence)
Just assuming that the two dates are supposed to match, of course...

I wish you good luck in analyzing the time conversion functions ! ;-)
(I know that it *will* hurt ;-)