Jeremias Maerki [Tue, 5 Aug 2008 06:51:32 +0000 (06:51 +0000)]
Merged revisions 636407-637074,637076-637118,637120-637790,637792-637856,637858-637992,637994-638047,638049-638307,638309-638315,638318-638936,638938-640888,640890-642905,642907-647402,647404-647536,647538-648983,648985-649005,649007-649013,649015-650549,650551-651301,651303-653536,653538-654452,654454-656285,656287-656523,656525-657519,657521-661579,661581-663481,663483-664691,664693-666634,666636-666661,666663-668532,668535-668549,668551-668569,668571-668640,668642-669117,669119-669172,669174-669477,669479-670550,670552-673749,673751-674303,674305-674313,674315-674316,674318-675105,675107-675149,675153-675706,675708-680714,680716-680819,680821-681283,681285,681287-681289,681293-681370,681372-682601 via svnmerge from
https://svn.apache.org/repos/asf/xmlgraphics/fop/branches/fop-0_95
Slight re-phrasing to make it clearer (I believe) that this page only lists changes made since 0.95beta, but that changes from that latter are of course also available in 0.95 final
........
r681330 | vhennebert | 2008-07-31 12:39:50 +0200 (Do, 31 Jul 2008) | 2 lines
Don't be shy Max ;-) Restored alphabetical order in the list of active committers
........
r681350 | jeremias | 2008-07-31 14:09:41 +0200 (Do, 31 Jul 2008) | 1 line
Merged revisions 636407-637074,637076-637118,637120-637790,637792-637856,637858-637992,637994-638047,638049-638307,638309-638315,638318-638936,638938-640888,640890-642905,642907-647402,647404-647536,647538-648983,648985-649005,649007-649013,649015-650549,650551-651301,651303-653536,653538-654452,654454-656285,656287-656523,656525-657519,657521-661579,661581-663481,663483-664691,664693-681285,681287-681290,681292-681298 via svnmerge from
https://svn.apache.org/repos/asf/xmlgraphics/fop/branches/fop-0_95
........
r666635 | jeremias | 2008-06-11 14:18:53 +0200 (Mi, 11 Jun 2008) | 1 line
Some notes on RTF and print output.
........
r668533 | jeremias | 2008-06-17 09:57:55 +0200 (Di, 17 Jun 2008) | 2 lines
Bugzilla #45151:
Note on the "compiling" page that Ant 1.7 is required. The README file already points to this page.
........
r668534 | jeremias | 2008-06-17 09:58:40 +0200 (Di, 17 Jun 2008) | 2 lines
Bugzilla #45146:
Removed duplicate LICENSE.txt and Notice.txt from the JAR.
........
r668570 | jeremias | 2008-06-17 10:54:07 +0200 (Di, 17 Jun 2008) | 1 line
Wrong line number reported in the case of a line overflow.
........
r668641 | jeremias | 2008-06-17 14:59:25 +0200 (Di, 17 Jun 2008) | 1 line
Bugzilla #44412:
Regression fix for empty pages caused by multiple collapsible breaks.
No more empty block areas if a break-before occurs on the first child of an FO to match the behaviour of tables and other FO implementations (clarification by XSL WG pending).
Added an accessor interface for break-before/-after to avoid long if..else lists in BlockStackingLayoutManager.
........
r669173 | jeremias | 2008-06-18 16:07:27 +0200 (Mi, 18 Jun 2008) | 7 lines
Bugzilla #44887:
Fixed potential multi-threading problem concerning the use of DecimalFormat.
Results from performance measurements in a separate test (operation repeated 100'000 times, exemplary):
shared static variable: ~220ms (old choice, problematic!)
always create new instance: ~480ms
ThreadLocal: ~220ms (new choice)
........
r669478 | jeremias | 2008-06-19 15:03:57 +0200 (Do, 19 Jun 2008) | 4 lines
Bugzilla #44412:
Improvements after Vincent's feedback:
Removal of the somewhat ugly Reference.
getBreakBefore() reduced to private visibility.
........
r670551 | jeremias | 2008-06-23 14:54:27 +0200 (Mo, 23 Jun 2008) | 3 lines
Use configured source and target resolution for foreign objects in RTF output.
Fixed intrinsic size calculation for SVG images when source resolution is not 72 dpi.
Fixed calculation of px2mm values in the SVG to G2D converter (used by RTF output).
........
r673750 | adelmelle | 2008-07-03 20:25:01 +0200 (Do, 03 Jul 2008) | 1 line
Fix for Bugzilla 45295 (already committed to the trunk with r672010, r672495 and r672496)
........
r674304 | jeremias | 2008-07-06 17:09:10 +0200 (So, 06 Jul 2008) | 18 lines
Bugzilla #43606:
column-width accepts percent values (proportional-column-width and % working)
Table width accepts percent values (100% of the page width as standard)
Nested tables are working
It is also possible to make nested tables with a higher depth than 3 without causing word to crash
Submitted by: Maximilan Aster <maximilian.aster.at.boc-eu.com>
Changes to the patch by Jeremias:
- Patch simplified (reuse of ColumnSetup from the layoutmgr package)
- Percentages and proportional-column-width didn't work properly in my tests -> fixed
- Adjustments for FOP code conventions
Bugzilla #43824:
page-number-citation working but has to be refreshed inside word to show the correct values
Submitted by: Maximilan Aster <maximilian.aster.at.boc-eu.com>
Changes to the patch by Jeremias:
- Adjustments for FOP code conventions
........
r674314 | jeremias | 2008-07-06 18:40:08 +0200 (So, 06 Jul 2008) | 7 lines
Bugzilla #43825:
leader supports fixed percent values for leader-length, most other properties
use-content, leader-pattern-width not implemented
Submitted by: Maximilan Aster <maximilian.aster.at.boc-eu.com>
Changes to the patch by Jeremias:
- Adjustments for FOP code conventions
........
r674317 | jeremias | 2008-07-06 18:44:43 +0200 (So, 06 Jul 2008) | 1 line
Bugfix: -imagein didn't work because the stylesheet didn't get copied into fop.jar.
........
r680715 | jeremias | 2008-07-29 16:37:10 +0200 (Di, 29 Jul 2008) | 1 line
Found a surplus "pop state" which somehow causes a regression (fill color doesn't get properly reset) between 0.94 and 0.95. Shows in Barcode4J's fop-extension-demo.fo. The line has already been there in 0.94 but only shows in 0.95 probably due to an additional save/restore state pair added in PDFRenderer for reference areas after the block-container area refactoring.
........
r680820 | jeremias | 2008-07-29 22:02:14 +0200 (Di, 29 Jul 2008) | 2 lines
Bugzilla #41306:
The AWT Renderer is really not out-of-order capable.
........
r681284 | jeremias | 2008-07-31 08:47:49 +0200 (Do, 31 Jul 2008) | 1 line
Some updates to status.xml (I was not the only one working on the final release).
........
r681290 | jeremias | 2008-07-31 09:44:51 +0200 (Do, 31 Jul 2008) | 2 lines
Website updates for the 0.95 release.
The projectInfo plug-in of Forrest is not ideal for doing a bugfix release. It's not possible to combine the changes for two releases.
........
r681292 | jeremias | 2008-07-31 09:47:13 +0200 (Do, 31 Jul 2008) | 1 line
Disabling the ApacheCon ad for the release build.
........
Reverted revision 670341 and replaced it with a different solution as discussed:
addCharacters() is now deprecated and called by the new characters() method which uses start/length instead of start/end.
Fixed bugs in the AFP and PS extensions (ArrayIndexOutOfBoundsExceptions). They were not properly changed when the interface changed.
Discussion: http://markmail.org/message/pdgzf6odxrnhflby
Hacked CommandLineOptions so that it accepts '-' as a specification of stdin/stdout.
Made it work also when infile is specified without any option ('fop - -pdf res.pdf')
TODO Investigate the adoption of Apache Commons CLI
Adrian Cumiskey [Tue, 22 Jul 2008 09:30:22 +0000 (09:30 +0000)]
Desynchronized createFontKey() and removed single use of static TRIPLETS_TYPE in fontLookup() following a suggestion by Jeremias Ingo Maas' threading issue (http://mail-archives.apache.org/mod_mbox/xmlgraphics-fop-users/200807.mbox/%3C000a01c8eb4b$42166e60$a701010a@ebp01422%3E).
Adrian Cumiskey [Mon, 21 Jul 2008 16:48:14 +0000 (16:48 +0000)]
Renamed fname to more correct name fontKey.
Made createFontKey() in FontInfo synchronized which should hopefully fix Ingo Maas's threading problem (http://mail-archives.apache.org/mod_mbox/xmlgraphics-fop-users/200807.mbox/%3C000a01c8eb4b$42166e60$a701010a@ebp01422%3E).
Adrian Cumiskey [Fri, 18 Jul 2008 09:11:10 +0000 (09:11 +0000)]
Ok, I am much happier with this fix now, it doesn't feel like a temporary hack anymore :).
I have revised the fontLookup() algorithm in FontInfo so it now first tries to find matching font triplets by iterating over all the font family names *without substitutions*, if this fails it then iterates over all the font family names looking for matching font triplets *with substitutions* rather then just take the last one in the font family name list with substitutions.
Adrian Cumiskey [Thu, 17 Jul 2008 18:47:58 +0000 (18:47 +0000)]
I'm not really happy with this workaround, it is a bit of a hack (as Symbol and ZapfDingbats are not available) but it does prevent the IllegalArgumentException from being thrown.
Worked around a bug in Xalan 2.4.0 (bundled with Sun Java 1.4.2_xx) where file URLs like file:/C:/.... got converted to file:\C:\.... on Windows causing a FileNotFoundException.
null is not correct here, an empty String needs to be used for the default namespace. Saxon 9 fails with a NullPointerException otherwise. Xalan is more tolerant.
Extracted conversion methods for String to int[] or double[] to a utility class.
Made AreaTreeParser.getAttributeAsXXX() methods static to stress their utility character, and removed the private parseRect() in favor of getAttributeAsRectangle2D().
Added basic-checks.xml: file currently only contains one default check.
Modified testcase2checks.xsl: the added basic-checks.xml will be inserted before the checks particular to the testcase.
Adrian Cumiskey [Thu, 26 Jun 2008 20:49:11 +0000 (20:49 +0000)]
Some basic cleanup.
* Duplicated/redundant PSDictionary, PSPageDeviceDictionary and PSDictionaryFormatException removed and now referenced from xmlgraphics commons.
* Updated xmlgraphics commons jar containing migrated PSPageDeviceDictionary class.
Luca Furini [Mon, 23 Jun 2008 09:24:14 +0000 (09:24 +0000)]
Added support for non-zero borders and padding on page regions.
Testcase included, now disabled because it is supposed to run with relaxed validation switched on.
Switch FOText to use a java.nio.CharBuffer, and implement the CharSequence interface.
TextLayoutManager no longer duplicates the char array, operates on the FOText (charAt(i))
Additionally: endOfNode() for FOText and Character deferred until after white-space handling.
Partial fix of FOEventHandler call sequence (see also Bugzilla #45237):
call startOfNode() after addChildNode()
moved initialization for tableFOs to processNode()
moved finishing code to a finalizeNode() method (which is now also used by AbstractRetrieveMarker)
restored protected status of startOfNode()/endOfNode() in fo.flow.table package
projects / xmlgraphics-fop.git / log