Perhaps my tool is the problem. I'm using TortoiseSVN. And ... that line I suspected is not the problem. The problem is one of my calls to logger(). Which makes no sense to me ... but that appears to be the case.
I'll have a correction committed in a few minutes. Daniel 2009/4/30 Andreas Lehmkühler <andr...@lehmi.de> > > It appears to be the change in PDFStreamEngine. > > > > Interestingly ... PDFStreamEngine does not have a logged changed 761546 > and > > 768122. > > However ... by the time we get to 768122 an additional parameter is in > the > > constructor of TextPosition in the call to processTextPosition. In the > > current revision, it is line 457: > > (int)(fontSizeText * textMatrix.getXScale()), > > > > Now, either I am not using SVN properly, or this was somehow introduced > > without a log of the change to the file being created. > > > > you see the line if you compare PDFStreamEngine.java versions 761546 and > > 768122. But if you just ask 768122 for its changes, that line is not > > highlighted. 768121 (which did not change this file) has the line in > > question. > > > > Ideas anybody? > I've commited that change with version 761547. It is part of a fix for > PDFBOX-450 for details see [1] I'm using eclipse in combination with > subclipse and my diff on the different version works fine. I don't see > neither a version 761546 nor a version 768121. Perhaps your svn tool is > the problem?? > > Andreas > > [1] > > http://mail-archives.apache.org/mod_mbox/incubator-pdfbox-commits/200904.mbox/%3c20090403072831.615592388...@eris.apache.org%3e > > > > > Daniel > > > > On Thu, Apr 30, 2009 at 11:58 AM, Daniel Wilson < > > williamstonconsult...@gmail.com> wrote: > > > >> OK, I've replicated it in my current version -- 770181. Same PDF "for > >> sorted test". > >> > >> 768122 -- failure > >> 767668 -- Pass > >> 768122 again -- failed again. > >> > >> Looks like it's in my commit. > >> > >> Working on resolution ... > >> > >> > >> > >> > >> > >> > >> On Thu, Apr 30, 2009 at 10:37 AM, Brian Carrier < > >> carr...@digital-evidence.org> wrote: > >> > >>> I did a few more random tests to narrow down when the issue was > >>> introduced. rev 767669 works. rev 769594 fails (that build is about > half > >>> way in between 767669 and the current revision). So, the problem lies > >>> somewhere in the 5 commits between those. > >>> > >>> > >>> On Apr 30, 2009, at 10:11 AM, Brian Carrier wrote: > >>> > >>> Hello, > >>>> The "testextract" regression test is failing on my system (w/no local > >>>> modifications) from an out of memory error. It seems to be dying on > the > >>>> Garcia2004_thesis.pdf file. I don't have time this morning to track > down > >>>> which recent checkin started this, but can those who have recently > checked > >>>> stuff in let us know if they ran the "testextract" regression test > before > >>>> they checked in and if it worked for them? > >>>> > >>>> BTW, was the autobuild system changed to run the "testextract" test > after > >>>> each build? > >>>> > >>>> thanks, > >>>> brian > >>>> > >>> > > >
