Age | Commit message (Collapse) | Author |
|
Change-Id: I12ef5e6257db337c4432b251bc92107a2c2ea88b
Reviewed-on: https://gerrit.libreoffice.org/64557
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: I9fb8366634b31230b732dd38a98f800075529714
Reviewed-on: https://gerrit.libreoffice.org/64510
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
... othervise the next drag event will insert the new range immediately
after the just inserted, resulting in invalid reference.
Regression after commit 42aece949ec96c775b31216bddd723aa5321e966
don't replace content of formula field when selecting range, tdf#58635
Change-Id: I7dfdc1ce041ebadf4406dbb605676d87cd489efa
Reviewed-on: https://gerrit.libreoffice.org/64574
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
Tested-by: Mike Kaganski <mike.kaganski@collabora.com>
|
|
Move XRefreshable Java tests to C++ for ScDDELinkObj.
Change-Id: I121d9fe6a66c549e402b0cf6fd54fea1201e443f
Reviewed-on: https://gerrit.libreoffice.org/64511
Tested-by: Jenkins
Reviewed-by: Jens Carl <j.carl43@gmx.de>
|
|
Change-Id: Ie171eea334de65b993554146c11adb95e0dbb85e
Reviewed-on: https://gerrit.libreoffice.org/64245
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
It's silly to just return NaN just because the function can't handle
something. If nothing else, at least a proper error should be reported
(or in this case, the fallback to the core function will take care of it).
Change-Id: I9c971082f4c5c9836318cf63d15fa7c278274273
Reviewed-on: https://gerrit.libreoffice.org/64244
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
Change-Id: I885a1b70556311f5a0862fec24a3596dbeada5d6
Reviewed-on: https://gerrit.libreoffice.org/64243
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
sc_subsequent_filters_test in ScFiltersTest::testFunctionsODS() also
checks "SERIESSUM(2;3;2;3)", which the opencl code does not handle.
At least bail out gracefully instead of returning 0.
Change-Id: I154dca8cc437a6225b4eb98012232d80683f0114
Reviewed-on: https://gerrit.libreoffice.org/64242
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
Otherwise the yielddisc unittest fails.
Change-Id: I2ebb0fbf9d4aea6678c91d0289418991aa92dbb5
Reviewed-on: https://gerrit.libreoffice.org/64241
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
Now 20 MRU plus...
Change-Id: I069021bf1f42da793f7ec79a7e0eacbea519bc14
Reviewed-on: https://gerrit.libreoffice.org/64515
Tested-by: Jenkins
Reviewed-by: Heiko Tietze <tietze.heiko@gmail.com>
|
|
Move XNamed Java tests to C++ for ScDDELinkObj and refactor class
XNamed. ScDDELinkObj doesn't allow to set the name, so test that a
exception of type RuntimeException is thrown.
Change-Id: Ie110c3c0ea22d205d4f2c4575f3db3676dda9ce3
Reviewed-on: https://gerrit.libreoffice.org/64497
Tested-by: Jenkins
Reviewed-by: Jens Carl <j.carl43@gmx.de>
|
|
Change-Id: I803e82841898ee9507a1874d0436de69ce28f55b
Reviewed-on: https://gerrit.libreoffice.org/64387
Tested-by: Jenkins
Reviewed-by: andreas_kainz <kainz.a@gmail.com>
|
|
Change-Id: I594e7beb3fb320b70ffe4dd98fee46c833f443dc
Reviewed-on: https://gerrit.libreoffice.org/64453
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: Ie2b921612352dbb83477a4d9f6d70795a4703664
Reviewed-on: https://gerrit.libreoffice.org/64381
Tested-by: Jenkins
Reviewed-by: andreas_kainz <kainz.a@gmail.com>
|
|
Otherwise the mround unittest fails in opencl mode.
Change-Id: I6c8866495bbb44ff7dfc075ba4aac58d80dac7c7
Reviewed-on: https://gerrit.libreoffice.org/64240
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
Change-Id: Id8253537025cc373c1ff183c0059158489e11750
Reviewed-on: https://gerrit.libreoffice.org/64239
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
149a4d7566 removed NeedParallelReduction() even though the code should
be there. I'm not sure about GenSlidingWindowFunction(), since the base
class is a template, but better revert that one too.
Also revert that part of 73e6a7975b that removed the subsequently
no longer used functions.
Change-Id: I932ffd58f9528ec840c4575078f8356640eb5420
Reviewed-on: https://gerrit.libreoffice.org/64249
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
Both of these are trivial and quite common mathematical operations.
Unary minus is a bit complicated in that unary '-' is the same symbol
as binary '-' and so there needs to be a hack to expand the '-'
from the list to both of the opcodes ocSub and ocNegSub.
Change-Id: I556e20e161c22dc89919afffbf904c1ef1d552ff
Reviewed-on: https://gerrit.libreoffice.org/64362
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
Some error checking, but also simply use "if(cond) code;" rather than
"if(!cond) try-to-set-benign-values-that-will-break-things-nevertheless".
Change-Id: Ic0c296273ba4c174dba5e96eadeeee4005988142
Reviewed-on: https://gerrit.libreoffice.org/64342
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
The data vector may not have all values (if last cells are empty).
In that case GetRefRowSize() is larger than GetArrayLength().
Other cases in this function had a similar check, this one was
missing it for some reason (with the 'gid0' removed, since it's
a fixed position).
Change-Id: I01692b51d6ed22ee6db419508cacf556fb9e644d
Reviewed-on: https://gerrit.libreoffice.org/64237
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
Test::testFormulaRefUpdateRange could trigger this, leading to recursion
that wasn't handled properly by the code, since it wasn't expected
to happen at late time (ScDependantsCalculator should have already
caught it). This is all caused by the fact that FetchVectorRefArray()
fetches also all rows before the given rows (to make the caching simpler
I suppose). But that fetching could lead to Interpret() calls.
Therefore, make ScDependantsCalculator in OpenCL mode check also all
rows above.
Change-Id: Iaecc105663df21b01443759287cec605470d34a5
Reviewed-on: https://gerrit.libreoffice.org/64236
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
These functions were removed in 149a4d7566, although they quite clearly
do serve a purpose, and then e9586cf0b5 removed the no longer used
variables.
Without this, ParallelReductionVectorRef class simply does not work.
E.g. SUM(B1:B250) would not sum all the cells, but just one (testcase
is Test::testFormulaRefUpdateRange with OpenCL forced).
Change-Id: Iad1aeba2fbae2b4fc1d4236030adbb9ce6d972a1
Reviewed-on: https://gerrit.libreoffice.org/64235
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
This is a debug tool and it's better to get it this way rather than
the code doing something else e.g. because of a typo.
Change-Id: I5ac7e7bd6ecaa15fe935d53df63e3c369e0f8f16
Reviewed-on: https://gerrit.libreoffice.org/64234
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
Change-Id: If965471464b4d3b2ded41951f8078d2e4dcaad10
Reviewed-on: https://gerrit.libreoffice.org/64233
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
PRODUCT() when there are no arguments (or all cells are empty) is 0.
Change-Id: I4bcb9afe84d08833526255da0c61f6847d68ea36
Reviewed-on: https://gerrit.libreoffice.org/64232
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
Table operations do some tricks with reusing a cell with different values,
and this confuses opencl handling, which operates on a group as a whole.
Possibly relevant only with SC_FORCE_CALCULATION=opencl,
sc_subsequent_filters_test fails in testDataTableOneVarXLSX().
Change-Id: I55e0d238061c8a45a2388ccf6c769ff6a1fddbcc
Reviewed-on: https://gerrit.libreoffice.org/64230
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
No idea why somebody thought pow() would be better than a mere division.
Without this the unittest fails because of a slightly different result.
Change-Id: Ibbde8b872049be310ac98389634aab6eab154ca6
Reviewed-on: https://gerrit.libreoffice.org/64229
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
Otherwise the sqrtpi unittest fails in opencl mode. This value is less
precise, but consistency should be more important here, not to mention
that float doesn't have that high precision anyway.
Change-Id: I8e85b7029d33932c81009e5d48e0fe2164c68dcf
Reviewed-on: https://gerrit.libreoffice.org/64228
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
That's how the scaddins implementation works, and the unittests fail
in opencl mode without this.
Change-Id: I08ea7dd0a0222408abf899d2da649be53f314a4c
Reviewed-on: https://gerrit.libreoffice.org/64227
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
It fails sc_financial_functions_test with SC_FORCE_CALCULATION=opencl.
Change-Id: If43928df8b1382eb7518fc99b6eee1759d822058
Reviewed-on: https://gerrit.libreoffice.org/64225
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
The last argument is optional and defaults to 0 according to docs.
Change-Id: I2e3aeb1da13b119e03e5bca18a63447e0fdc77fa
Reviewed-on: https://gerrit.libreoffice.org/64224
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
Found with bin/find-unneeded-includes
Only removal proposals are dealt with here.
Change-Id: I5a4951e983078ab6b6cbf0e96b00a20584b85e20
Reviewed-on: https://gerrit.libreoffice.org/64398
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
|
|
Move XElementAccess Java tests to C++ for ScLabelRangesObj.
Change-Id: Icf49fb7e8c24b169e4fe33ffed8cc1412d21f9e8
Reviewed-on: https://gerrit.libreoffice.org/64432
Tested-by: Jenkins
Reviewed-by: Jens Carl <j.carl43@gmx.de>
|
|
Move XIndexAccess Java tests to C++ for ScLabelRangesObj.
Change-Id: I448561be2395beb9448ffcfc29434a94468166cb
Reviewed-on: https://gerrit.libreoffice.org/64431
Tested-by: Jenkins
Reviewed-by: Jens Carl <j.carl43@gmx.de>
|
|
Change-Id: I5629c4e0bbe29390db5350b05523409c48bbb84d
Reviewed-on: https://gerrit.libreoffice.org/64434
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: Ie69be492f41e42714edea252af9054414b0b841a
Reviewed-on: https://gerrit.libreoffice.org/64134
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Jenkins
|
|
Change-Id: Ifdf1a152f6bc2e2f6edae97a5191120f630f7e49
Reviewed-on: https://gerrit.libreoffice.org/64374
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: I790f957c9ebc6d87a94a355c30215245aaabeb30
Reviewed-on: https://gerrit.libreoffice.org/64345
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
|
|
Move XElementAccess Java tests to C++ for ScAnnotationsObj.
Change-Id: Ife8c2d51a3aae409fa008c5ad2515a8976481771
Reviewed-on: https://gerrit.libreoffice.org/64407
Tested-by: Jenkins
Reviewed-by: Jens Carl <j.carl43@gmx.de>
|
|
Move XIndexAccess Java tests to C++ for ScAnnotationsObj.
Change-Id: Icbd34ffe4b4911754045e4a5dc511b3fb25d4dd0
Reviewed-on: https://gerrit.libreoffice.org/64404
Tested-by: Jenkins
Reviewed-by: Jens Carl <j.carl43@gmx.de>
|
|
Rename testGetByIndex to testIndex of sheet::XSheetAnnotations test
cases so it won't collide with testGetByIndex of
container::XIndexAccess.
Change-Id: Ic48ffe954c9f4f00b68066ad5d0d2df89cc4dfd7
Reviewed-on: https://gerrit.libreoffice.org/64402
Tested-by: Jenkins
Reviewed-by: Jens Carl <j.carl43@gmx.de>
|
|
Change-Id: I27e5e4604cd999d778eb84976b3bea0ef35122ee
Reviewed-on: https://gerrit.libreoffice.org/64353
Tested-by: Jenkins
Reviewed-by: Eike Rathke <erack@redhat.com>
|
|
Move XIndexAccess Java test to C++ for ScAreaLinksObj.
Change-Id: I374b7301caf67c900b7a64920d2ac5b642a8c6fd
Reviewed-on: https://gerrit.libreoffice.org/64376
Tested-by: Jenkins
Reviewed-by: Jens Carl <j.carl43@gmx.de>
|
|
... ever since commit 9ae5a91f7955e44d3b24a3f7741f9bca02ac7f24
V560 A part of conditional expression is always false: nRangeNo < nRCount.
The value range of unsigned short type: [0, 65535].
Actually, simple if (nRangeNo < nRCount) would be enough, given that if
sal_uInt16 is less than another sal_uInt16, it cannot be equal to greatest
sal_uInt16. But let's keep this redundant comparison, in case types or
define changes later.
Change-Id: I886a0bc847a8c30966dea3ea0670bcee9b43c7f1
Reviewed-on: https://gerrit.libreoffice.org/64021
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
|
|
Change-Id: I735ca7d70045b278d6b885cc2de35bf76885b487
Reviewed-on: https://gerrit.libreoffice.org/64258
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
|
|
Move XElementAccess Java tests to C++ for ScAreaLinksObj.
Change-Id: I89beb6a1c05d853e94ebbedb031d87484cb84b38
Reviewed-on: https://gerrit.libreoffice.org/64373
Tested-by: Jenkins
Reviewed-by: Jens Carl <j.carl43@gmx.de>
|
|
Move XElementAccess Java tests to C++
for ScCellsObj.
Change-Id: If721582a36ef5a82fba20d4ec7e0806d1d0524d9
Reviewed-on: https://gerrit.libreoffice.org/64365
Tested-by: Jenkins
Reviewed-by: Jens Carl <j.carl43@gmx.de>
|
|
Change-Id: I76b440e72a80f20d05af477a38e9695cb17cb979
Reviewed-on: https://gerrit.libreoffice.org/64354
Tested-by: Jenkins
Reviewed-by: Jens Carl <j.carl43@gmx.de>
|
|
This is a followup of e22ab5e6f6b0ea49231ca454a567133996306116.
Change-Id: I7052e113173204f7cfcd40622f52c5d4e14b3aa9
Reviewed-on: https://gerrit.libreoffice.org/64139
Tested-by: Jenkins
Reviewed-by: Eike Rathke <erack@redhat.com>
|
|
Move XEnumerationAccess Java tests to C++ for ScCellsObj.
Change-Id: I3ec78142db7cfa924bf5663ca44cca80b6a0e556
Reviewed-on: https://gerrit.libreoffice.org/64315
Tested-by: Jenkins
Reviewed-by: Jens Carl <j.carl43@gmx.de>
|