no reason not to permit this, and using it with PassThru 1 allows for a
very simple Preamble layout-type that acts as ERT for the premable. Of
course, we do have the preamble in Document>Settings, but that allows
the embedding of preamble code in branches.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@36562 a592a061-630c-0410-9148-cb99ea01b6c8
This introduces an isFullUnicode() function to OutputParams that currently only contains XETEX, but later can also contain other fully unicode-aware backends (LUATEX for instance).
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@36445 a592a061-630c-0410-9148-cb99ea01b6c8
The goal is first to remove explicit testing against ERT_CODE,
and second to allow to define layouts with the same properties as ERT insets.
This commit does not change behaviour
* rename OutputParams::verbatim to pass_thru
* disable encoding switch inside pass_thru layouts
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@35091 a592a061-630c-0410-9148-cb99ea01b6c8
- new function hasDigit
* src/Paragraph.cpp (spellcheck):
- ignore words with digits, as in 1.6 (bug 6493).
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@33395 a592a061-630c-0410-9148-cb99ea01b6c8
- comment out code that makes aspell abort
if a word in an unknown language is being checked
(instead of just skipping this word)
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@33387 a592a061-630c-0410-9148-cb99ea01b6c8
weren't taken into account. So the paragraph id management is now going
back to Paragraph.cpp.
Sorry for the trouble.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@32788 a592a061-630c-0410-9148-cb99ea01b6c8
#6415
Enrico please verify that reverse dvi is fixed.
* Text and InsetText: create two private constructors and transfer some
initialisation code from InsetText.
* Paragraph: id generation is transfered to Text. May be transfered to
Buffer in the future, we'll see.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@32766 a592a061-630c-0410-9148-cb99ea01b6c8
Much of the point of this is to allow us properly to handle what LyX does as:
<em>This is <strong>bold and italic</em> and now just bold.</strong>
We output:
<em>This is <strong>bold and italic</strong></em><strong> and now just bold.</strong>
which is valid.
Note how much easier this would have been if emphasis and boldness were insets
rather than ranges. ;-)
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@32086 a592a061-630c-0410-9148-cb99ea01b6c8
can't just output Standard as <p>, or anything else, because we
have structures like:
this is text <branch>and more text</branch> and more
which would then come out as:
<p>this is text <p>and more text</p> and more</p>
So we use the OutputParam html_in_par to try to signal when we
are already in a paragraph. It is expected that we will need to
do some bug fixing here.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@31753 a592a061-630c-0410-9148-cb99ea01b6c8
THis leads to a big speedup of the buffer-write function (>60%). The goal is to make autosave less noticeable.
This patch should be safe as long as transforming a docstring to utf8 is equivalent to transforming the characters one by one.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@31695 a592a061-630c-0410-9148-cb99ea01b6c8
Now inline SpellChecker is faster than ever. Now, I honestly think that we blow out all the competitors (thunderbird, OO, MSWord, etc) :-)
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@30970 a592a061-630c-0410-9148-cb99ea01b6c8
* no one has to know that the current author has by default id 0,
* avoid unnecessary use of the Type enum.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@30862 a592a061-630c-0410-9148-cb99ea01b6c8
Fixes bug #5690: Hebrew file cannot be exported to dvi or pdf.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@30848 a592a061-630c-0410-9148-cb99ea01b6c8
* Move suggestion searching out of check() and onto suggest().
* Cleanup a bit AspellChecker
* Begin Hunspell support (not tested and does not link yet)
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@30824 a592a061-630c-0410-9148-cb99ea01b6c8
This is a hack necessary because \\textcolor is a bit buggy.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@30773 a592a061-630c-0410-9148-cb99ea01b6c8
This patch makes sure that there are minimal changes when loading and saving a file with change tracking.
- the authors are assigned a buffer_id, such that when the file is saved, they get the same id,
- the authorlist is sorted according to the buffer_id,
- the buffer_id is written to the file in the author list (file format change)
- the ids start with 1, because 0 is internally reserved for the current Author,
- when writing the file, the current author is assigned an id if he didn't already have it.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@30756 a592a061-630c-0410-9148-cb99ea01b6c8
accepting changes in a clipboard copy (CutAndPaste.cpp), we set the buffer
of insets (and we reset them later).
Doing this makes sense because we know this is the only operation on these
out-of-document paragraphs that will require access to a buffer.
Also, this commit gets rid of one explicit test against ERT_CODE and
LISTING_CODE.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@30623 a592a061-630c-0410-9148-cb99ea01b6c8
Basically, insets in cut stack do not have a buffer, and therefore cannot
acess to buffer parameters. What is annoying here is that acceptChanges
requires this buffer params only to be able to read a font in moveItem,
in order to read the buffer language, and I doubt this is really needed...
Another change in this patch is that Inset::getLayout now returns a
plainLayout when the inset does not have a buffer_. This fixes a remaining
crash where dEPM reads isFreeSpacing() for an inset in the clipboard, but
this looks like a fragile situation. And it will not do the right thing when
doing depm in a freespacing inset.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@30605 a592a061-630c-0410-9148-cb99ea01b6c8
Fix the rest of the code, whih implies fixing a few potential bugs where the appendix status was not taken in account.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@30524 a592a061-630c-0410-9148-cb99ea01b6c8
* Counters.cpp (flatLabelString): return a cache of the flattened strings for each used language
* Counters.cpp (theCounter, counterLabel, flattenLabelString): add a lang parameter; in theCounter, populate the cache as needed.
* insets/InsetCaption.cpp:
* insets/InsetFoot.cpp:
* insets/InsetBibitem.cpp:
* insets/InsetCollapsable.cpp:
* Paragraph.cpp:
* Buffer.cpp: pass a language argument to counter methods.
* Paragraph.cpp (translateIfPossible): use the function with same name in gettext.cpp.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@30520 a592a061-630c-0410-9148-cb99ea01b6c8
Paragraph::collectWords() now also collects spellchecker suggested words for inline word completion.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@30240 a592a061-630c-0410-9148-cb99ea01b6c8
NB: I am not sure this escape character preference is something useful to keep. If some language has some special escape characters beside the single quote one we should support that out of the box and not asking the user to define it or them in the preference dialog.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@30210 a592a061-630c-0410-9148-cb99ea01b6c8
- we do not output alignment begin for the default alignment,
so we must not output the end tag (bug 5995)
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@30151 a592a061-630c-0410-9148-cb99ea01b6c8
- do not embrace list items. According to the comment,
this was only done to escape brackets, but brackets
are embraced anyway.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@30112 a592a061-630c-0410-9148-cb99ea01b6c8
routines, one that allows us to "defer" certain material, e.g., a float
that ought to appear in a div but can't simply be put into the middle
of a paragraph.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@30079 a592a061-630c-0410-9148-cb99ea01b6c8
Do not accept the changes in the selection, if it is completely deleted.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@29672 a592a061-630c-0410-9148-cb99ea01b6c8
You'll need to check the continuous spellcheck option in order to test this.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@29465 a592a061-630c-0410-9148-cb99ea01b6c8
are computed recursively to get rid of any \thexxx string by the method
flattenLabelString. The values are cached in Counter objects.
On my mac, the total time (updatelabels+redraw) goes down by 30%. YMMV.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@29342 a592a061-630c-0410-9148-cb99ea01b6c8
- new member internalFontEncoding() that indicates if a language
switches the font encoding internally.
* src/Paragraph.cpp (latexSpecialChar):
- don't call latexSpecialT1 if the internal font encoding isn't T1.
This fixes the output of straight quotation marks in Hebrew and Greek.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@27660 a592a061-630c-0410-9148-cb99ea01b6c8
properly fix bugs 5216 and 5280. The best thing to do would be recognizing
at configure time a buggy iconv and #defining WORKAROUND_ICONV_BUG in
config.h, but I don't know how that could be done.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@27618 a592a061-630c-0410-9148-cb99ea01b6c8
- constify some parameters
- getFontSettings():
- getFirstFontSettings(): return a const ref instead of a copy. This brings a consistent 4% improvement in LateX generation. Might speedup rtl handling too.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@26957 a592a061-630c-0410-9148-cb99ea01b6c8
* Paragraph::insertInset(): check if inset insertion is allowed before insertion.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@26810 a592a061-630c-0410-9148-cb99ea01b6c8
