Apache OpenOffice (AOO) Bugzilla – Full Text Issue Listing |
Summary: | sw/qa/unoapi sw.SwXTextGraphicObject::com::sun::star::beans::XPropertySet failure | ||
---|---|---|---|
Product: | Writer | Reporter: | Stephan Bergmann <stephan.bergmann.secondary> |
Component: | code | Assignee: | AOO issues mailing list <issues> |
Status: | CONFIRMED --- | QA Contact: | |
Severity: | Trivial | ||
Priority: | P3 | CC: | issues, orw |
Version: | DEV300m84 | ||
Target Milestone: | --- | ||
Hardware: | All | ||
OS: | All | ||
Issue Type: | DEFECT | Latest Confirmation in: | --- |
Developer Difficulty: | --- |
Description
Stephan Bergmann
2010-07-07 15:35:04 UTC
mine it took me a long time to figure out that the reason for the failure is not the big fat (but legitimate) IllegalArgumentException on "WrapInfluenceOnPosition", but the line: 19: LOG> setting property 'VertOrientPosition' failed the setPropertyValue test first gets a value, then sets a different value, and then gets the value again. the VertOrientPosition apparently fails because the values from both getPropertyValue calls are the same. for VertOrientPosition the value ends up in SwFmtVertOrient::SetPos(). the problem seems to be that the layout code also sets this value: #0 SwFmtVertOrient::SetPos (this=0xbf9e2cc4, nNew=4) at ../../../inc/fmtornt.hxx:72 #1 0x063de85b in objectpositioning::SwToCntntAnchoredObjectPosition::CalcPosition (this=0xbf9e2ef4) at /net/x42-so29/export/home/ms216673/hg/swuno/sw/source/core/objectpositioning/tocntntanchoredobjectposition.cxx:417 #2 0x05f72b21 in SwFlyAtCntFrm::MakeObjPos (this=0xb14838ec) at /net/x42-so29/export/home/ms216673/hg/swuno/sw/source/core/layout/flycnt.cxx:1542 #3 0x05f7cbb2 in SwFlyFreeFrm::MakeAll (this=0xb14838ec) at /net/x42-so29/export/home/ms216673/hg/swuno/sw/source/core/layout/flylay.cxx:248 so what i guess happens is that setPropertyValue writes the value, then the timer runs and re-layouts the frame, computing the same position as before, then getPropertyValue returns the same value as the first time. oh, and in my DEV300m83 i've not managed to reproduce the failure, because in the test framework there is a filter that first filters out properties that behave in exactly the way that the setPropertyValue would regard as failure! (the qadevOOo/tests/java/ifc/beans/_XPropertySet.java:602: if ( isWritable && isNotNull ) canChange = isChangeable(name); so if i set a breakpoint then the timer will goes off and this check filters out the property, so the setPropertyValue test will never test it. @sb: does this failure occur "often" for you, or just once? any ideas what should be done to test timing-sensitive properties? @mst: experienced it once (after a few successful runs), then disabled it, so no idea how frequently it occurs now experienced a similar "[sw.SwXTextEmbeddedObject::com::sun::star::beans::XPropertySet::setPropertyValue()] is testcode: [setPropertyValue()] - PASSED.FAILED" due to "setting property 'VertOrientPosition' failed" now experienced a similar "[sw.SwXTextFrame::com::sun::star::beans::XPropertySet::setPropertyValue()] is testcode: [setPropertyValue()] - PASSED.FAILED" due to "setting property 'HoriOrientPosition' failed" worked around for now by disabling affected tests as <http://hg.services.openoffice.org/cws/sb126/rev/7ea26b3e91c1>; please revert when fixing this issue Reset assigne to the default "issues@openoffice.apache.org". |