I just wanted to get rid of the duplicated xmlns:r attributes, but seems
model.xml had some inconsistency as well, that's how this refactoring
fixes a bug as a side effect.
Change-Id: I95d753253f03a001c68dbbac7c86740040a180fa
This makes model.xml about 20kb smaller. Most inconsistencies were
introduced when unused tags were removed, like <optional>.
Change-Id: I5e77860cf09bad4cad3e9dd7383371a10386c260
Commit af7d4a5ebf3e6a09cd2079f241dee16aa22e0276 (n#820503: initial MCE
support in writerfilter ooxml tokenizer, 2013-07-03) added these, but in
fact only the namespace-alias part was necessary. Now that oox provides
the list of namespaces to be registered, it can be removed completely.
Change-Id: If8f1b0c07602a0170c63b53d806a5b466d485747
The motivation is that <namespace-alias> elements in model.xml are
redundant, as the same info is available from oox as well. But without
sorting, it's impossible to generate the same output, as the (not
interesting) order isn't the same there.
Change-Id: I634c62e43d1b54100bfa623c6f43dddd34279fb1
See 656ab07b2924dd2703f31f460c8586c375c7878a (Sync oox -> writerfilter
alias for wordprocessingml, 2014-08-10).
Change-Id: I5179961b683473a502693923422171bb274fc845
See 656ab07b2924dd2703f31f460c8586c375c7878a (Sync oox -> writerfilter
alias for wordprocessingml, 2014-08-10).
Change-Id: Iaa28855fbef836768a21d5b3153a1d5c2cd0d685
The reason for adding 'DomainMapper' is to be able to check
during import in these classes whether or not the import is
in a 'styles.xml' file or not.
Also added blank handling for new XML nodes.
Reviewed on:
https://gerrit.libreoffice.org/10870
Change-Id: Ib2acde736d96f30eda40ee327991dc6ae0ed7f25
The motivation is that once all the namespace aliases in
writerfilter/source/ooxml/model.xml match the ones in
oox/source/token/namespaces.txt, then the writerfilter copies could be
dropped.
Change-Id: I1f9abb8bb457189997f28c99b0f6b00660252c14
Change them to choice, what was probably really meant here: allow one of
the values, not a whitespace-separated list of one or more values.
Also, clean up nested choices.
Change-Id: I87bc30b710569e317da05933502b36d8d97e4155
For <value> elements inside <resource> ones, the name attribute and the
text of the element was the same, but our scripts only parsed the text.
So remove the redundant code that's ignored anyway.
Change-Id: I0d96b485c9e095be51571f2d68eb395920c97bc8
Stuff like
NS_ooxml::LN_Value_wordprocessingml_ST_TargetScreenSz_1920x1200 is no
doubt not mapped to anything or quite some VML bits, which are handled
in oox.
Change-Id: Id50af0b61c25794b5d2c34f3deb3f16f3a8ab838
