Re: [Fwd: Worsening of Win32 layout]

From: Stephen Viles (sviles_abi@iinet.net.au)
Date: Thu Sep 18 2003 - 12:46:50 EDT

  • Next message: Robert Wilhelm: "Re: [Fwd: Worsening of Win32 layout]"

    PS. All of the below tests were done _without_ Tomas' one-line patch,
    ie. with gr_Win32Graphics.cpp, line 484 still reading:

    gcpResult.lpDx = pCharAdvances; // Distances between character cells

    Thanks to Oliver for warning me (via IRC) of the regression.

    Cheers - Stephen.

    19/09/03 2:37:50 AM, Stephen Viles <sviles_abi@iinet.net.au> wrote:

    >gr_Win32Graphics.cpp, lines 496-499:
    >
    >if(XAP_App::getApp()->theOSHasBidiSupport() == XAP_App::BIDI_SUPPORT_GUI)
    > placementResult = GetCharacterPlacementW(m_hdc, (LPCWSTR) currentChars, iLength, 0, &gcpResult, 0);
    >else
    > placementResult = GetCharacterPlacementW(m_hdc, (LPCWSTR) currentChars, iLength, 0, &gcpResult, GCP_REORDER);
    >
    >Observations on my Windows 98SE platform:
    >
    >1. placementResult is always 0 after the call to GetCharacterPlacementW,
    >indicating that GetCharacterPlacementW function has failed:
    >http://msdn.microsoft.com/library/en-us/gdi/fontext_4l84.asp
    >"Return Values
    >If the function succeeds, the return value is the same as the return value
    >from GetTextExtentPoint32, the width and height of the string in logical units.
    >If the function fails, the return value is zero."
    >
    >2. Line 499 is executed. Forcing call of GetCharacterPlacementW with flags == 0 in line 497
    >(by commenting out lines 496,498,499) still results in placementResult == 0.
    >
    >It may not be relevant, but I am a bit confused that the test in line 496 shows
    >XAP_App::getApp()->theOSHasBidiSupport() != XAP_App::BIDI_SUPPORT_GUI
    >and earlier test in line 477 shows XAP_App::getApp()->theOSHasBidiSupport()
    >!= XAP_App::BIDI_SUPPORT_NONE. XAP_Win32App::_setBidiOS appears to set
    >only BIDI_SUPPORT_GUI or BIDI_SUPPORT_NONE. BIDI_SUPPORT_FULL is not set:
    >
    >Searching for: BIDI_SUPPORT_FULL
    >af\xap\xp\xap_App.h(175): typedef enum {BIDI_SUPPORT_NONE, BIDI_SUPPORT_GUI, BIDI_SUPPORT_FULL} BidiSupportType;
    >text\fmt\xp\fp_TextRun.cpp(105): s_bBidiOS = (XAP_App::getApp()->theOSHasBidiSupport() ==
    >XAP_App::BIDI_SUPPORT_FULL);
    >Found 2 occurrence(s) in 2 file(s)
    >
    >I can't find the actual value of XAP_App::getApp()->theOSHasBidiSupport()
    >due to gdb problems ("Internal error: `this' is not an aggregate").
    >
    >Cheers - Stephen.
    >
    >19/09/03 12:11:40 AM, <msevior@physics.unimelb.edu.au> wrote:
    >>
    >>Oh well, maybe that little patch needs to be reverted.
    >>
    >>Anyone got other ideas? Maybe I did it wrong.
    >>
    >>Martin
    >>
    >>-------- Original Message --------
    >>Subject: Worsening of Win32 layout
    >>From: "Dr. Oliver Schurr" <schurro@rz.uni-potsdam.de>
    >>Date: Thu, September 18, 2003 6:29 pm
    >>To: Martin Sevior <msevior@physics.unimelb.edu.au>
    >>
    >>Dear Martin
    >>
    >>that little change to gr_Win32Graphics.cpp worsened the win32 layout,
    >>inthat it returned to the ever-changing linebreaks upon ZOOOMING. Both
    >>higher and lower level ZOOMING changes the linebreaks here.
    >>
    >>Thanks
    >>Oliver
    >
    >
    >
    >
    >
    >

    This archive was generated by hypermail 2.1.4 : Thu Sep 18 2003 - 12:58:24 EDT