Turns out MathML is already here, for the most part, so this will
not be too bad at all.
Anyone else think that is very cool?!
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@31979 a592a061-630c-0410-9148-cb99ea01b6c8
performance problems when loading documents with lots of macros.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@31907 a592a061-630c-0410-9148-cb99ea01b6c8
Math manual loads and save correctly it seems but expect some instability period.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@31898 a592a061-630c-0410-9148-cb99ea01b6c8
It would be interesting to know how many headers do we actually use in our code.
Nice task for some script.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@31893 a592a061-630c-0410-9148-cb99ea01b6c8
This patch initializes the buffer_ member of a MathHull inset in most
(but not all) cases. The problems with buffer_ should be greatly
allievated now, but there still can be some corner case.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@31873 a592a061-630c-0410-9148-cb99ea01b6c8
This is another ad-hoc solution for the problem of a zero buffer pointer while reading a document or when using cut&paste.
This simple solution would be suitable for branch.
see also e.g. r29085, r28319, bug #5688.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@31861 a592a061-630c-0410-9148-cb99ea01b6c8
* MathStream{.cpp, h}:
- replace bool dryrun() by enum output that also knows whether the stream is for instant preview
* InsetMathHull.cpp:
- tell the stream whether we use it for instant preview.
* MathString.cpp (write):
- gracefully catch encoding exception for instant preview.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@30895 a592a061-630c-0410-9148-cb99ea01b6c8
This reverts part of r10553. There, an extra '//' was added when the last line was empty. However, this really makes an extra line and that is visible when the lines have a line number. The bug #2067 that would have been fixed by this does not seem to 'exist' anymore.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@30795 a592a061-630c-0410-9148-cb99ea01b6c8
Presumably it fixes
"r30414 unselects insets after a fraction of a second. 1.5.7 doesn't exhibit this bug"
http://www.mail-archive.com/lyx-devel@lists.lyx.org/msg152960.html
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@30536 a592a061-630c-0410-9148-cb99ea01b6c8
guaranteed that LyX is always able to replace $...$ constructs
with \ensuremath{...}.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@30145 a592a061-630c-0410-9148-cb99ea01b6c8
The problem is that, for an unknown reason, an InsetMathHull is used *inside* an InsetMathHull. This is the reason why we have two levels of instant preview. I don't know if that was intended.
When moving the cursor, only the doDispatch() of the enclosing InsetMathHull is
called, not the inner one; so we don't have a chance to detect that the inner InsetMathHull should leave its preview mode.
The new leavePreview() method ensures that the screen will be redrawn if needed with cursor movement.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@27186 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
commands when using math in text mode (such as in $\text{$\phi$}$, where
\phi would be enclosed in \ensuremath{} as it is inside \text{}).
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@26288 a592a061-630c-0410-9148-cb99ea01b6c8
- update labels cache after a label has been erased (bug 4929).
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@25757 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
please check your favourite insets, I am sure I broke _something_...
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@24006 a592a061-630c-0410-9148-cb99ea01b6c8
* src/Buffer.cpp:
- file format change to 324
* development/FORMAT:
- document file format change
* src/insets/InsetNewline.{cpp,h}:
- the former two separate derivates are now merged into one inset
with proper Params and Mailer. This simplifies the structure and allows for
switching between the two kinds.
* src/Text3.cpp:
- The newline insets are now being generated with
newline-insert [newline|linebreak]
* src/factory.cpp:
* src/Text.cpp:
- Newline inset is now a proper inset with opening and closing tag.
* src/FuncCode.h:
* src/LyXAction.cpp:
* src/Text3.cpp:
* src/insets/InsetCollapsable.cpp:
* src/insets/InsetTabular.cpp:
* src/mathed/InsetMathGrid.cpp:
* src/mathed/InsetMathHull.cpp:
- remove LFUN_NEW_LINE, LFUN_LINE_BREAK,
add LFUN_NEWLINE_INSERT
* lib/lyx2lyx/LyX.py:
* lib/lyx2lyx/lyx_1_6.py:
- conversion and reversion routines for newline insets.
* lib/ui/stdmenus.inc:
* lib/ui/stdcontext.inc
- adapt menu and add context menu to switch between newpage variants.
* lib/bind/*.bind:
- adapt to new lfuns.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@23966 a592a061-630c-0410-9148-cb99ea01b6c8
** We show (Eq:bla,#) for labeled formulas
** The first forward delete removes the label
** The second forward delete removes the number
Fixes http://bugzilla.lyx.org/show_bug.cgi?id=2556
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@23762 a592a061-630c-0410-9148-cb99ea01b6c8
"not possible to delete a label of a formula":
Press forward delete at the end of a row and the label is deleted.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@23610 a592a061-630c-0410-9148-cb99ea01b6c8
* Inset and derived class: get rid of getLabelList()
* BufferView::gotoLabel(): Speed up and simplify by using the tocBackend.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@23541 a592a061-630c-0410-9148-cb99ea01b6c8
* InsetMathHull:
- label_: now is a vector of InsetLabel instead of docstring
- addToToc() and updateLabels() update the existing labels.
- nonum_ : use bool instead of int
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@23435 a592a061-630c-0410-9148-cb99ea01b6c8
places, i.e.
- BufferView::setMouseCursor,
- BufferView::mouseEventDispatch
- LyXFunc::dispatch
Before it was called from several dispatch handlers in the middle of the
dispatch process. Because the cursor might get invalidated, this is
not a good idea. We had plenty of crashes and odd behaviours because
of this as nobody really understood those cases.
This fixes (at least) #2069 and #4163.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@23140 a592a061-630c-0410-9148-cb99ea01b6c8
There could be some more clever code to keep the selection. But IMO it's ok like this.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@22864 a592a061-630c-0410-9148-cb99ea01b6c8
* InsetMathHull.cpp (getStatus): disable LFUN_MATH_NUMBER and LFUN_MATH_NONUMBER
when the equation is not in display mode; add a checkbox as needed.
(doDispatch): no need to test for dispatch() anymore.
* InsetMathNest.cpp (getStatus): LFUN_MATH_NUMBER and LFUN_MATH_NONUMBER are not
handled by this class.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@22350 a592a061-630c-0410-9148-cb99ea01b6c8
Now support/* should have no dependencies on src/* anymore.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@21851 a592a061-630c-0410-9148-cb99ea01b6c8
- fileformat change
- tex2lyx support
- rename LFUN_BREAK_LINE to LFUN_NEW_LINE because this is what the lfun does and to avoid confusions
- new menu entry
- new shortcut C-S-Return
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@21778 a592a061-630c-0410-9148-cb99ea01b6c8
* Font::FontBits -> FontInfo
* Font::FONT_XXX -> all enums transfered to FontEnums.h and renamed to FontXxx
I've replaced Font uses with FontInfo were the language() member was not needed, basically all draw() and metrics methods. There's one problematic cases with InsetQuotes which I solved by taking the Buffer main language.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@21240 a592a061-630c-0410-9148-cb99ea01b6c8
This is the first patch in a series meant to remove LTR bias from the code.
Currently, there are many places in the code where "right" and "left" are used,
but where we actually mean "forward" and "backwards" --- which is the same
thing in LTR text, but not in RTL/Bidi text.
See full details in http://permalink.gmane.org/gmane.editors.lyx.devel/97550.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@21128 a592a061-630c-0410-9148-cb99ea01b6c8
In the future, when Cursor has been fixed to work correctly with multiple views of the same Buffer, we could transfer the Undo owner from Buffer to Cursor.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@21031 a592a061-630c-0410-9148-cb99ea01b6c8
- get rid of the cached Dimension. Text inset dimensions are saved in ParagraphMetrics and mathed maintain their own dimension where needed.
- width(), ascent(), descent(): deleted.
- dimension(): now needs a valid BufferView.
- metrics(): now void.
* BufferView::getCoveringInset(): simplify.
* ParagraphMetrics(): now cache inset dimensions.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@20420 a592a061-630c-0410-9148-cb99ea01b6c8
dispatchers in mathed and text only do local work, i.e. only try to
move the cursor with their local knowledge, not globally as before in
the Cursor::goUpDown function before. The old approach used one of
brutefind functions as a last way out to really move the cursor up or
down. The result was some strange behavior that the nearest inset was
selected when jumping from math to text, not one in the next line.
2. removed LFUN_FINISHED_UP/DOWN, as there seem to be no effect and it
seems to be obsolete.
3. target_x is working again, mainly by adding a textTargetOffset_
member to the Cursor which keeps the difference between the target_x
and the actual position of the cursor on screen. As long as
pos+offset=target is fulfilled the target_x is not touched and the
cursor keep the column while going up/down.
4. moved the text cursor up/down code to the Cursor class, in analogy
to the mathed code.
5. For 1 it is necessary to know the cursor position before the
dispatch. So this data is added to the Cursor.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@18569 a592a061-630c-0410-9148-cb99ea01b6c8
