This fixes bug #8554 and some recently introduced busg:
- Encodings::fromLaTeXCommand() can now handle all combining characters,
not only the one letter ones
- The remainder returned from Encodings::fromLaTeXCommand() must never be
thrown away in tex2lyx, but output as ERT
- No special case for combining diacritical marks needed anymore in parse_text()
- No special cases for accents and IPA combining diacritical marks needed
anymore in parse_text()
- special tipa short cuts may only be recognized if the tipa package is loaded
- Use requirements returned by Encodings::fromLaTeXCommand() instead of
hardcoded registering of tipa and tipax
- Get rid of the name2 variable in parse_text(): We must use name, otherwise
the extra stuff that might have been put into name vanishes
Provide functions for translating to the LyX name
of an encoding from either a LaTeX name or an Iconv
name, with the possibility to specify the package.
This is in anticipation of changing to use the LyX
name of the encoding in the .lyx file format and
allowing multiple lib/encodings entries to have
the same LaTeX name (but different packages!).
The tex2lyx parser needs to worry about the iconv
name of the input encoding, so store that instead
of the latex name.
These encodings were not defined, since they must not be used as document
encodings (the characters {, } and \ may appear in high bytes, and latex
would be confused). However, they are supported by CJK.sty (which uses a
preprocessor to circumvent the limitations of the latex executable). These
encodings are now defined, but used for import in tex2lyx only.
The test case CJK.tex contained fake tests for shift-jis and big5 (the
japanese and chinese characters were entered using the utf8 encoding), and
therefore the wrong interpretation of these encoding looked as if it worked.
The comments about missing iconv support of shift-jis and big5 were wrong as
well (otherwise shift-jis-plain would not work either).
With non-TeX fonts, you can select a 'Non-TeX Font Default' math font, which simply loads unicode-math without actually selecting a math font, this then uses the default math otf font, currently Latin Modern. Other fonts still need to be set manually in the preamble, via \setmathfont.
The implementation suppresses unneeded package requests from unicodesymbols, but the output still uses macros instead of full unicode (both is possible with unicode-math).
The whole thing is a proof of concept, and it needs to be tested. I have tested it with the math manual, which compiles and seems to display correctly if I remove some hardcoded package loadings. OTOH I have not much experience with math.
This addresses #7449 partly.
Math commands need it as well as text commands. At the same time, this
further unifies the checking for termination and fixes cases of wrong
output (e.g. for 0x2005).
* do not terminate macros by {} if not necessary (prevent kerning violation)
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@40891 a592a061-630c-0410-9148-cb99ea01b6c8
now) and move more packages to the new exclude mechanism.
The remaining ones are not so easy.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@40442 a592a061-630c-0410-9148-cb99ea01b6c8
Now you can also require a|b|c, and if any of the features is already used,
no other one will be loaded. The first feature wins if none is already used.
This is required for a part of bug #7811.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@39884 a592a061-630c-0410-9148-cb99ea01b6c8
dialog by only converting text mode latex macros to unicode points.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@30070 a592a061-630c-0410-9148-cb99ea01b6c8
Cyrillic formatted text in mathmode
If some other letters from some script need to be typeset in italic,
it will suffice adding the flag "mathalpha" to the unicodesymbols file.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@28731 a592a061-630c-0410-9148-cb99ea01b6c8
that entries in the unicodesymbols file may not start with '\'.
Maybe the method should be renamed as fromLaTeXCode?
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@27957 a592a061-630c-0410-9148-cb99ea01b6c8
to the method does not match any of the known unicode symbols.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@27953 a592a061-630c-0410-9148-cb99ea01b6c8
What works:
- parsing of inputenc should work
- \inputencoding is acted on in the preamble
What does not work:
- \inputencoding in the text
- all the corner cases I have not considered, and all buggy stuff in the
'what works' paragraph
- InsetLatexAccent are still created, but I do not know when they got added
to the code.
The only notable trick in the code is that I had to disable buffering. Otherwise
the whole text was read before I had a chance to change the encoding...
Finally I remove the artificial limitation that forbid
\usepackage[opt1,opt2]{package1,package2}
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@27592 a592a061-630c-0410-9148-cb99ea01b6c8
file is used, don't add the \lyxmathsym definition to the preamble if we
already are in textmode. This hopefully smashes all weird cases.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@26604 a592a061-630c-0410-9148-cb99ea01b6c8
for entering math accents, such that an umlauted A is translated as \ddot{A},
for example.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@26577 a592a061-630c-0410-9148-cb99ea01b6c8
unicodesymbols file. Also take into account that a unicode symbol
may appear in both mathmode and textmode and thus both the mathmode
and textmode features may be required. As a side effect, the problem
that the \lyxmathsym definition is added to the preamble even when
a unicode symbol appears only in textmode is solved.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@26548 a592a061-630c-0410-9148-cb99ea01b6c8
can be encoded in the current latex encoding before resorting to the
unicodesymbols file.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@25725 a592a061-630c-0410-9148-cb99ea01b6c8
- build an exception list of characters that must be output as LaTeX command in any case
(except in utf8-plain). I.e., the utf8 encodings now also honour the "force" flag.
* lib/unicodesymbols:
- add "force" flag to combine characters. Fixes rest of bug 4946.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@25361 a592a061-630c-0410-9148-cb99ea01b6c8
http://bugzilla.lyx.org/show_bug.cgi?id=1527
* src/mathed/*.cpp:
- Track mode (math or text) when outputting latex code and
properly wrap commands that are in the wrong mode with
either \ensuremath or \lyxmathsym.
* src/mathed/MathParser.cpp:
- Parse \ensuremath and \lyxmathsym, such that exported latex
produces the same result when imported back.
- Replace latex commands with corresponding unicode symbols.
* src/Encoding.cpp:
- Implement Encodings::fromLaTeXCommand() for converting either
a single or a sequence of latex commands to unicode symbols.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@25270 a592a061-630c-0410-9148-cb99ea01b6c8
unicodesymbols file. When the command for a mode is not specified, the
other one is used (taking care that it also works in the "wrong" mode).
This allows to avoid the \lyxmathsym macro when a native math command
is available. Patch from Georg, modifications to the unicodesymbols
file by me.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@25061 a592a061-630c-0410-9148-cb99ea01b6c8
