Age | Commit message (Collapse) | Author |
|
Change-Id: Ia37d19722fc7a4ef46cf6f17fa0e45a12db0e7b3
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/126111
Tested-by: Jenkins
Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>
|
|
for the filter by color options, which have pushed its height up taller
than it used to be. I think it would be better to make more submenus
here to shrink its height down further, but that's not attempted here.
Change-Id: Idb484bd77d2cca6e175d997955b0967720d94ec6
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/126121
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
so its all visible as much as possible under wayland
Change-Id: Iad82f9ab63d57a93361e773a159c2716e13d59f1
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/126120
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
The setting from /org.openoffice.Office.Calc/Input is already cached
in ScModule::m_pAppCfg; synchronizing the two mentioned places was
done in commit 8810812e58f1b014d3156235c1ab5a2d6c73c828.
Another existing cache in ScInputHandler only complicates things,
making the behavior inconsistent when configuring through registry,
com.sun.star.sheet.GlobalSheetSettings, and menu.
So just read the setting when needed from ScModule in methods of
ScInputHandler.
For lok, use ScModelObj::initializeForTiledRendering to set it up
(as was done in 65990058f041c3f1d280a69d411eb4ceacf5a721).
Change-Id: I102d3360aeb45b301f04f4c8bbee1191c23fe0e7
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/126002
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
|
|
Change-Id: I9c708811d9eea04a2421c97a97d7dce14fdc8e63
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/126058
Tested-by: Caolán McNamara <caolanm@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I97ae6145d0139bd33f33d298803c7e787fd717a7
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/126061
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I01af38dd57a645ea0afeaff033ce6d07dfe09535
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/126026
Tested-by: Caolán McNamara <caolanm@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
There's not much point in complicating this by allocating them
dynamically if they're always treated by value anyway.
Change-Id: I8325829201c0ad6c95858916a94c5b4d1d208b1c
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/126024
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
The calculations now follow the documentation for WEEKS in
https://wiki.documentfoundation.org/Documentation/Calc_Functions/WEEKS.
Change-Id: Ifede6e2ec2d9cfb7301fd5eb53bf7b6f187053bc
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125319
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
|
|
Change-Id: Iadf5ff86d043f63355fd30816a9e2011479465c0
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125990
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
|
|
Make ScInputOptions a private ancestor. This allows to drop
OptionsChanged, and avoid potential way to change settings
without committing to configuration.
Drop ScInputOptions::SetDefaults, and use ctor for that.
Change-Id: I8abb7309a72e84ac7c83592f3758f0264cd1cc4e
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125905
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
|
|
Change-Id: I903e8277fea5c223081244ae30dff31bbda7d554
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125904
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
|
|
since the gap in the vcl treeview separator problem is fixed
Change-Id: I105167127588c11895d81dd2aea3c9e119170069
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125952
Tested-by: Caolán McNamara <caolanm@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I5feae2a17eb1ac647f2be4c6936e3a9b0706a333
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125953
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I79e4f404f85846ef85886f8bdbe3d4ba3114a091
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125950
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
do it as an easily backportable change
Change-Id: Ib92e8f1e35e6465b52a69f3a8513068668dc6953
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125949
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I18862328a3239424de80b478a09f568b5dcef1ac
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125948
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I27df936f95ac7727b58ad196201df6d56bceeca9
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125943
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: I45cc9b6c8cbd9205e185bebff97644f53f5cf0af
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125896
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
|
|
The INDEX() offset was calculated with a fix row number derived
from the initial output position, like 35-ROW(). That fails if the
range later is moved or rows inserted/deleted.
Instead, use a cell reference of the cell's current column and
end row, like ROW($A$34)+1-ROW(). 34+1 to not have things shifted
if a row is inserted before 35.
Change-Id: I11276c4acb35505576daa1a20ad49a4fcdbaba49
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125914
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Jenkins
|
|
turned off after drawing the buttons."
This partially reverts LO 5.0
commit 342d84218c17da40d0a4f96bc3b07b536d64517a which comments:
Also - we are actually never call with something else than MAP_PIXEL,
so let's kill the resetting of the mapmode;
let's see if some obscure corner case bites us here.
Several bug reports have indicated artifacts on screen,
all coming from the use of autofilter,
and in these cases the map mode is 0 (Map100thMM),
so I restored the whimsically removed logic.
Change-Id: I9bb5d7418fc4083ed71139a64329b11fa1e4cc13
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125879
Tested-by: Jenkins
Reviewed-by: Justin Luth <jluth@mail.com>
|
|
inline some hot methods, and use OUStringLiteral to avoid repeated
construction of an OUString, saves about 3%
Change-Id: I6d09ade4e08a94ed5dd49a8882e26bb5786409d5
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125873
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: I5ba98b3651470f7c86b9aca9359d563e64058552
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125874
Tested-by: Jenkins
Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>
|
|
Reverts patch I29d90b413e6f6f6e8b73d30ac47d644a69d15046,
ad8edac43e73555bc2055514300c5b81a1bb04ea and partially
Ibdee348d84a8b96d8f698e7fc6db89eb24422f5d
Change-Id: Iba5fd2d2a65d813c2d5ba85e66a7c3e07b726ca5
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125871
Tested-by: Jenkins
Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>
Reviewed-by: Heiko Tietze <heiko.tietze@documentfoundation.org>
|
|
Change-Id: I7c6c60489a58378238ff63ff9c6b45f3a7aa4bf7
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125866
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Setting "Same content on first page" in the Page Style
dialog on either the Header or Footer tab resulted
disabled header/footer after reload. This was due to
flawed logic when assigning SC_UNO_PAGE_HDRON/FTRON
value: "leftness" was checked for but not "first pageness".
Regression from commit b9993369b4011cd8f367e439f5248096827f1efc
"tdf#121715 XLSX: support custom first page header/footer".
Co-authored-by: Daniel Arato (NISZ)
Change-Id: I8ee4c9f8145913d9dc5f4b66cdbc2352f2e405f0
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125367
Tested-by: László Németh <nemeth@numbertext.org>
Reviewed-by: László Németh <nemeth@numbertext.org>
|
|
I assume the threading was originally limited only to real cores
because that's how many FPUs there are in the CPU. But Calc
dynamically allocates many data structures (ScFormulaCell most
notably), which means they end up scattered all over the memory,
and processing them often stalls on memory reads, leaving time
for other HT core to run. For data-heavy documents (such as using
VLOOKUP) using HT cores saves 30%, but I can see 10+% gains
even for compute-heavy documents.
Change-Id: I302b17e55f27541606d922a1f0b27005ce0f0ba8
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125801
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
the timing has to be just right to get this to happen
Change-Id: I1ae81bf5c79b7f8b3fdf82b3fab94d5a3b6076cc
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125816
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
This makes autofilter even with tdf#136838 almost instanteous.
Change-Id: I94b4b6d6ab6f8e73312d88c8b88c0f393707f117
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125795
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
The tdf#136838 comment #2 document has MaterialNumber column
that is actually strings (even though they look like numbers),
but autofilter creates ByValue query items (for whatever reason).
So extend the conditions for fast searching to include this case,
otherwise it'd be handled by the generic slower code.
Change-Id: I0fe192b99cd2999282db53ba98587b712c42c762
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125794
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
Change the compareByString() function to a template with a bool
parameter that will disable all the checks, allowing the compiler
to effectively generate just a small inline function. And then
compute the bool value once and call the fast version if true.
With comment #2 in tdf#136838 this actually saves 40% of time.
Change-Id: I5a5190f19a1df163b28e527090ec880e6de5bbda
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125768
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
Autofilter with large documents can create queries that have thousands
of items. Searching all of those for every cell using the generic
algorithm can be quite slow. First try an optimized search for this
case that skips most of the complications and just tries to find
in the query items an exact match for the cell. This significantly
speeds up tdf#133867 or attachment from comment #2 in tdf#136838.
Change-Id: I2bba18da6a101c76398d8c42c4306c53682899c1
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125746
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
When using autofilter, the query entry can have thousands of items,
and fetching of the cell string value is not trivial in some cases.
I'm not doing this for numeric values, as those depend on query
items and also seem fairly fast.
Change-Id: I01fa2e0cbce8f770d1ed079c0dba74830402a895
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125735
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
Change-Id: I80d5a9c1ac9b05273f0237b966ad23a11a17a0c5
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125798
Tested-by: Jenkins
Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>
|
|
Change-Id: I59fdcf48fb0da9d944b00a140e043e4ea5fb1ae5
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125789
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Jenkins
|
|
Change-Id: I9b0cef7a8a536a8db7d0b840dfe69714fdbf2cd4
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125779
Tested-by: Jenkins
Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>
|
|
Change-Id: I78d754d36c7ea41f7e3259be331d23b8a25c480f
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125771
Tested-by: Jenkins
Reviewed-by: Samuel Mehrbrodt <samuel.mehrbrodt@allotropia.de>
|
|
Change-Id: Ibc2eb80d7cc0cfc22c7fea0dc7ebe495ae0927c6
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125765
Tested-by: Jenkins
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
|
|
instead of random number
Change-Id: I89f3be2d6bba02f1fbe52065fab81caded2656f3
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125699
Tested-by: Jenkins
Reviewed-by: Samuel Mehrbrodt <samuel.mehrbrodt@allotropia.de>
|
|
cp workdir/LinkTarget/Executable/fftester instdir/program
LD_LIBRARY_PATH=`pwd`/instdir/program valgrind instdir/program/fftester `pwd`/sc/qa/unit/data/slk/pass/ofz41299-1.slk slk
==805505== Conditional jump or move depends on uninitialised value(s)
==805505== at 0x1D66BAA5: ScRawToken::CreateToken(ScSheetLimits&) const (token.cxx:354)
==805505== by 0x1D30D115: ScCompiler::CompileString(rtl::OUString const&) (compiler.cxx:4831)
==805505== by 0x1DC68FE0: ScImportExport::Sylk2Doc(SvStream&) (impex.cxx:2174)
==805505== by 0x1DC654E4: ScImportExport::ImportStream(SvStream&, rtl::OUString const&, SotClipboardFormatId) (impex.cxx:390)
==805505== by 0x1DB84091: TestImportSLK (docsh.cxx:3404)
==805505== by 0x406657: sal_main_with_args(int, char**) (fftester.cxx:485)
==805505== by 0x40389D: main (fftester.cxx:97)
Change-Id: If0cbab2b1b41dacba3b2871b09e31e2d8675ba82
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125677
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
The commits introducing this part of code have tests attached,
so it's fairly easy to check that this part of code is in fact
not necessary, but if in doubt there's the option of simply asking
the author of the code.
Change-Id: I3b754090ac462d3ae8e4676fc1643570ec53aac1
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125695
Tested-by: Jenkins
Reviewed-by: Eike Rathke <erack@redhat.com>
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
Change-Id: I1597a8e77199445bf377dbe54adc3134bb04fd51
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125748
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
|
|
Otherwise for #ref!e:o NextNewToken() bailed out with an
uninitialized maRawToken.
Change-Id: I8b3b399a5a3636ccda3104ba2a5971e5e9a8fc02
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125736
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Jenkins
|
|
Change-Id: Ia1badb975f4effca668a5c2a41cfef3beccf8e2b
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125739
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
This may speed up some lookups.
Change-Id: I13a39abc43c64eb758fa55a6cdbda3e83e179121
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125743
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
This is strictly speaking not necessary if callers would always
initialize to nullptr, but I guess this is better.
Change-Id: Ica88395e0f615afbd58efa71e5d64bfdd2a37319
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125734
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
When not querying for the test equal condition, it's pointless
to keep searching for it to become true only to throw it away.
Change-Id: Ie861bac141f80025e95753fb8b1202498df17383
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125733
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
To be extended in following commits.
Change-Id: I09eaf0f5cfdbbe0aff254231dbb983fd7f72cc5c
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125732
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
Change-Id: I2910231c9cf205ce697616e660e49056c221e89d
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125731
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
so the submenu pops down when the mouse moves back into its parent.
This removes layer of InterimItemWindow and for gtk (under wayland) the
autofilter and its submenu can use native GtkPopovers
Change-Id: I4c8c3d2087566f508d75eda72e272d9f7fba2f00
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125428
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|