Age | Commit message (Collapse) | Author |
|
In a build using the system Python during build (i.e., not using
--enable-python=fully-internal) on Fedora 29 (where /usr/bin/python3 is 3.7.1),
UITest_librelogo failed with
> ======================================================================
> FAIL: test_compile_librelogo (compile.LibreLogoCompileTest)
> ----------------------------------------------------------------------
> Traceback (most recent call last):
> File "/data/sbergman/lo-system/core/sw/qa/uitest/librelogo/compile.py", line 128, in test_compile_librelogo
> self.assertEqual(test[1], re.sub(r'(\n| +\n)+', '\n', re.sub(r'\( ', '(', compiled)).strip())
> AssertionError: 'glob[52 chars]_#\n label(_y + _z)\n #_@L_i_N_e@_#\n#_@L_i_N_e@_#\nx(25, 26)' != 'glob[52 chars]_#\n label(_y + _z)\n #_@L_i_N_e@_#\n#_@L_i_N_e@_#\nx(25, ,26)'
> global x
> def x(_y, _z):
> __checkhalt__()
> #_@L_i_N_e@_#
> label(_y + _z)
> #_@L_i_N_e@_#
> #_@L_i_N_e@_#
> - x(25, 26)+ x(25, ,26)? +
>
>
> ----------------------------------------------------------------------
due to an upstream Python change discussed at
<https://bugs.python.org/issue34982#msg329418> "re.sub() different behavior in
3.7".
I am not sure that upstream change really makes sense, despite that being
explicitly confirmed in <https://bugs.python.org/issue34982#msg329420>. But
lets tweak our code to adapt to that anyway. (There may be further places in
LibreLogo.py that would need similar changes; I just fixed enough to make
UITest_librelogo succeed for me.)
Change-Id: I6c8f4b78f63953d582b88037fa56388b50af2b54
Reviewed-on: https://gerrit.libreoffice.org/63038
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
Not sure about this, but just to catch the eye of the pro.
Git blame says 2012, so maybe it's really no more used
Change-Id: I737f3a1b2803f6ce98cb461350db18d06e0bf981
Reviewed-on: https://gerrit.libreoffice.org/62910
Reviewed-by: László Németh <nemeth@numbertext.org>
Tested-by: László Németh <nemeth@numbertext.org>
|
|
with Logo syntax, too.
Mutual recursion, for example drawing dragon curve (see in the
unit test of the commit) doesn't need Python syntax any more
to call the function before its definition.
Change-Id: I93426a8c5be394fb4f1203e0490f7fa8d8491597
Reviewed-on: https://gerrit.libreoffice.org/62926
Tested-by: Jenkins
Reviewed-by: László Németh <nemeth@numbertext.org>
|
|
Change-Id: I9508839d67e5723db5f9155560fdaef333a689f8
Reviewed-on: https://gerrit.libreoffice.org/62911
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
Tested-by: Jenkins
Reviewed-by: Bartosz Kosiorek <gang65@poczta.onet.pl>
|
|
Instead of the incomplete heuristic parenthesis expansion,
now expressions with Logo functions and with own
procedures are parsed completely, solving several issues
with complex Logo expressions. For example, now functions with
more than 1 argument don't need explicit parenthesization.
NOTE: to handle both Logo and Python syntaxes of function calls,
we differentiate the forms "f(x)" and "f (x)". The second form
is handled as Logo syntax, not the Python one:
f x*2 y z -> f(x*2, y, z)
f(x*2, x, z) -> f(x*2, y, z)
f (x*2) y z -> f((x*2), y, z)
so if you want to avoid of the following expansion:
sin 45 + cos 45 -> sin(45 + cos(45))
it's possible to use the following parenthesizations:
sin(45) + cos 45 -> sin(45) + cos(45)
(sin 45) + cos 45 -> (sin(45)) + cos(45)
but not
sin (45) + cos 45 -> sin((45) + cos(45))
Change-Id: Ib0602b47b8b678a352313f471599d44d6904ce17
Reviewed-on: https://gerrit.libreoffice.org/62901
Tested-by: Jenkins
Reviewed-by: László Németh <nemeth@numbertext.org>
|
|
button of the inputbox or messagebox dialog window, instead
of waiting for the next __checkhalt__() in a loop.
Change-Id: I1366ad06152e70321a21e78a626f7a89eb5a7ea0
Reviewed-on: https://gerrit.libreoffice.org/62900
Tested-by: Jenkins
Reviewed-by: László Németh <nemeth@numbertext.org>
|
|
(ASCII and typographical) and for enclosing quotes of string literals.
Change-Id: I3caf3b707afa1fb41ba3afe9ff12ebce7ce63847
Reviewed-on: https://gerrit.libreoffice.org/62384
Tested-by: Jenkins
Reviewed-by: László Németh <nemeth@numbertext.org>
|
|
from LibreLogo to Python
Change-Id: I39df100a13efe3573b33cb856701cbeb0d95954d
Reviewed-on: https://gerrit.libreoffice.org/62364
Tested-by: Jenkins
Reviewed-by: László Németh <nemeth@numbertext.org>
|
|
for program running and command name expansion, and for following fixes:
tdf#106792: regression in line length and continuous line drawing
tdf#100941: line breaking by "magic wand"
tdf#120422: program lines are different paragraphs by "magic wand"
Also add function __is_alive__() to LibreLogo.py to check
LibreLogo program termination via XScript API.
Change-Id: If884b3fd608a6e8077be853eb2dd17fbdfff2011
Reviewed-on: https://gerrit.libreoffice.org/62263
Tested-by: Jenkins
Reviewed-by: László Németh <nemeth@numbertext.org>
|
|
Format program lines as paragraphs, instead of a single paragraph
with line breaks:
- basic debug feature "jump to the bad line" works after formatting
- fix 2-page editing area: page break before the LibreLogo program
Now formatting of program lines doesn't depend on the actual regular
expression setting of Search & Replace functionality of Writer,
so this is the intended fix for tdf#100941 "LibreLogo: 'magic wand'
icon inserts incorrect '\n' characters instead of paragraph breaks".
NOTE: setting also AlgorithmType, not only AlgorithmType2 prevents
crashing of LibreOffice at opening Search & Replace dialog after
usage of the "magic wand" icon.
partial revert of the commit b1a6d157683b8182089ed5854179c8da8c416304
Resolves: tdf#100941 LibreLogo: replace literal '\n' with newline
Change-Id: I34f581278fdae8d41967800d05662e37b731b59d
Reviewed-on: https://gerrit.libreoffice.org/61610
Reviewed-by: László Németh <nemeth@numbertext.org>
Tested-by: László Németh <nemeth@numbertext.org>
|
|
related to measurement changes, for example we got only
5.7 cm long line for "FORWARD 10cm", because the processed
value measured in 0.0100 mm instead of the correct twips
(pt/20 ~ 0.0176mm).
Note: only the line drawing was shorter, the turtle path
didn't change, so the turle could draw only dashed line
for multiple FORWARD or BACK, instead of a continuous one.
regression from
commit 36bade04d3780bc54c51b46bb0b63e69789658a5
tdf106792 Get rid of SvxShapePolyPolygonBezier
Change-Id: I16d75dbdadef5af9c545abc86575490559b3d54c
Reviewed-on: https://gerrit.libreoffice.org/61145
Tested-by: Jenkins
Reviewed-by: László Németh <nemeth@numbertext.org>
|
|
Update:
python3 LibreLogo.py LibreLogo*.properties
Change-Id: Ib631f53b47f1f00b14338534cc82d94f33c48233
Reviewed-on: https://gerrit.libreoffice.org/44203
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Andras Timar <andras.timar@collabora.com>
|
|
git grep -l "[ _\.]THROUGHT" | xargs sed -i 's/THROUGHT/THROUGH/g'
git grep -l -i "[ _\.]THROUGHT" | xargs sed -i 's/throught/through/g'
In ENUMs: THROUGHT = THROUGH (preserved as valid alternate spelling)
In ooxmlexport8 - unit test confirms THROUGH = THROUGHT
Change-Id: Iae0fef9a8adcb96761989f38903a24ffb1b91e77
Reviewed-on: https://gerrit.libreoffice.org/35998
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
Isn’t it better to just use the desired characters directly,
given that “\uHEX”-style references are so error-prone?
Change-Id: I307b1fad7a9e0a23a38925e3c84829f823decf6e
|
|
Change-Id: Ic6c41fbcc36c11a7528cde0986593a39c2d6738b
Reviewed-on: https://gerrit.libreoffice.org/34803
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: I280b6cd02a98037a71701a6a7a540c87de22d07f
Reviewed-on: https://gerrit.libreoffice.org/31268
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
|
|
Bunch of these were setting C++ or Make modes and icky tabs...
Also, reportedly Emacs can figure out to enable python-mode
automatically.
Change-Id: I50072488fb92cb4d27aa3f74f717a28ae3967543
|
|
by simple SearchAlgorithms2's ABSOLUTE(=1), rather than using
SearchAlgorithms' REGEXP(=1).
BTW avoid RowDirection because it is for Calc only.
Change-Id: I50ab460110ed43befb3e378e94f4fda0f2777f4d
Reviewed-on: https://gerrit.libreoffice.org/27250
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
|
|
(a regression caused by the PyUNO changes of
commit af8143bc40cf2cfbc12e77c9bb7de01b655f7b30)
Change-Id: I62112555a997f8f57059de21e0f0b174f3c673c6
|
|
Change-Id: I8d1767c64fcdf68084590c7faf869700e7b6ddf3
|
|
Change-Id: Ib0492a20a83deba2dc4a59a4277072cf140b12c6
|
|
We don't have a mechanism to use different icon set for extensions; let's do
it at some stage, but until then, please let's not modify the icons to breeze
only.
This reverts commit 48ecae47a5d1911296549fd31378ff07b402d2ab.
|
|
Change-Id: I8e576afc2d7d57b6b3d56788e69fc5e1fa3c39ab
|
|
Change-Id: I8f622680ddd31d0a7048c14c85932ae495ae0f5e
|
|
Example: drawing square within a circle:
PENUP
REPEAT 4 [
FORWARD 100
POINT
BACK 100
RIGHT 360/4
] FILL
CIRCLE 200
Change-Id: Ica3ce44306fc985717ff73e8a3dec5dddf69f19b
|
|
Change-Id: I0781e4d8d9a9c2c9b8ef9968babeca8274d126ec
|
|
by using make's file function via the var2file wrapper instead of
using loooooong echo statements
Change-Id: Ie81007ad7de8c4e9f4d07724fba3cbe0e93f821d
|
|
Feature test macros that govern conditional compilation should be
defined in config_*.h include files, not on the compilation command
line.
Change-Id: I40575a4762fd2564f10927b6f38a112dd9f9a3d7
|
|
Change-Id: Ia6edb02b871a41828758ba5fd5376c811c4084cc
|
|
Change-Id: I7ef59f9ce589ab0d68b02c1b2ba61c061f9c51ae
|
|
Bug report: http://openscope.org/browse/OOO-837
Change-Id: I5c17ed6059107ec8fc12bf9f411e52b22f131065
|
|
Change-Id: I48055d84fd078a131ed8f620575a462fcb101019
|
|
Change-Id: Iffefd979cbe5c3ba8daafa92e33e9a288d2d906c
|
|
Change-Id: I7c25cd94f8a1ca339f7423c26f21f13c7a68906d
|
|
Change-Id: I7fe8c26b7ca93319658c14abd1142f1623141ce8
|
|
Change-Id: I201c498b0ff002ed92c2fcf9847ac1efc50b3fde
|
|
Change-Id: I2bc45e4ba63f5faaee7389bcd9d7b3f563503186
|
|
Change-Id: I2e85a795064ee9e58f3389aec94c942f6ae77612
|
|
Change-Id: I3c51ba693caa80c8b530a9eee932a48a125e2eca
|
|
Change-Id: I2204002533bbb3e7c801b3228b0310a42b19a882
|
|
Change-Id: Icb0d195ba82b023d370847242b4e3b5546fa0320
|
|
Change-Id: Id7dc7a2853a8c56ee56eab55c078650e16c278fd
|
|
Change-Id: I807dda255b741996480116fab22377b39bf963b4
|
|
Change-Id: I3d501900f4a2bc2424f4133e4fb9f3efe8b0c510
|
|
Change-Id: I3c05c5f7e1721a20e6eab12e2aa620aa917b7378
|
|
Change-Id: If0f37d480a745a4d245c4c6cf114374223fda610
|
|
Add more FOO_FOR_BUILD variables and some gb_Foo_for_build functions.
Get rid of gb_INSTROOT and gb_DEVINSTALLROOT, just use INSTROOT.
Change-Id: Iee531b02d14fae41edb68ad589a5dec829a60255
|
|
Introduced gb_INSTROOT, which is the same as $(INSTDIR) except for Mac OS X,
where it is $(INSTDIR)/LibreOffice.app/Contents. Most stuff ends up there (so
most occurrences of $(INSTDIR) have been replaced with $(gb_INSTROOT)), but SDK-
related stuff goes to $(INSTDIR)/$(gb_Package_SDKDIRNAME). (And
GeneratedPackage needed to be made more flexible, to allow for packages that go
into either of those two places.)
For Android and iOS, gb_INSTROOT probably still needs to be set.
The most obvious missing thing yet to make instdir work for Mac OS X is the
instdir/*/LibreOffice.app/Contents/ure/ vs.
instdir/*/LibreOffice.app/Contents/ure-link/ split.
Change-Id: I4478edd27b14c92c96d92d5169bdca3ec50d78f5
|
|
Change-Id: Icebd47d57a830986c371a62ec7dab81ca8fd930a
|
|
Change-Id: Ibb1cb41f526b4117e2d7ef39a2101286511bf48f
|