To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=56715
------- Additional comments from [EMAIL PROTECTED] Mon Nov 7 07:08:54 -0800 2005 ------- Looking at http://www.go-oo.org/patches/64bit/trivial-*.diff: [There are some places where now implicit conversion from sal_Int32 to long is used; I think that is ok, assuming that sal_Int32 is a (not necessarily proper) subset of long for all platforms we are interested in.] [There are some places where numeric operations are now done with a different, potentially "smaller" promoted type than before (after changing operands from long to sal_Int32); I think that is ok, as those places are all related to computations with screen/pixel sizes, which should probably not overflow on neither sal_Int32 nor long.] canvas/source/vcl/textlayout.cxx 1.6 l. 320: return type of OffsetTransformer::operator() must be sal_Int32, already fixed in 1.6 dbaccess/source/ui/tabledesign/TEditControl.hxx 1.14 l. 109: why change declaration of ClipboardInvalidator from ULONG param to sal_uInt32 param when (a) the corresponding definition in .cxx is not changed correspondingly, (b) the use of the param in the .cxx is only to pass it to another function as ULONG (AutoTimer.setTimeout), and (c) the only call to ClipboardInvalidator ctor uses an argument (500) that can be either ULONG or sal_uInt32? dbaccess/source/ui/tabledesign/TEditControl.hxx 1.14 l. 137: why not change 2nd param of GetTotalCellWidth from USHORT to sal_uInt16 to match overriden EditBrowseBox::GetTotalCellWidth; why not change return value of GetTotalCellWidth definition in .cxx from ULONG to sal_uInt32; in definition of GetTotalCellWidth in .cxx, return value of GetTextWidth (long) is implicitly converted to sal_uInt32, which potentially is a problematic narrowing conversion extensions/source/bibliography/datman.hxx 1.13 l. 209: moveRelative dead code? filter/source/pdf/pdfexport.cxx 1.44 l. 499: assigning GetActionCount() (ULONG) to sal_uInt32 nCount is potentially a problematic narrowing conversion sch/source/ui/unoidl/ChXChartDocument.cxx 1.53: why change one instance of "long nTranslation" to "sal_Int32 nTranslation" sfx2/inc/docinf.hxx 1.7: why change nReloadSecs and Get/SetReloadDelay from ULONG to sal_uInt32? Get/SetReloadDelay are actually used in combination with sal_Int32, long, ... in various places sfx2/inc/docinf.hxx 1.7: what is the reason for changing lTime and GetTime from long to sal_Int32 (and for *not* changing SetTime) so3: what is the reason for changing "office format" from long to sal_Int32 in some places (but *not* in the declaration of SvFactory::IsIntern) svtools/source/config/cacheoptions.hxx, svtools/source/config/cacheoptions.cxx: changing sizes from sal_Int32 to sal_Size does not work that easily; for one, mnGrfMgrTotalSize for example cannot be sal_Size (>>=, <<=), for another, goodies/source/graphic/grfmgr.cxx for example uses the result of GetGraphicManagerTotalCacheSize() where a ULONG is expected -- are those places checked for bad conversions? svtools/source/edit/svmedit.cxx 1.30 l. 1233--1234: changing nHeight/nLineHeight from long to sal_Int32 leads to potentially problematic narrowing conversions svtools/source/edit/svmedit.cxx 1.30 l. 1373: chaning nOnePixel from long to sal_Int32 leads to a potentially problematic narrowing conversion I did not (yet) look at trivial-svx.diff, trivial-vcl.diff, and trivial-xmloff.diff. We should clarify first whether the patches are already in a good enough quality (which I am not sure about from my above findings). --------------------------------------------------------------------- Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
