Age | Commit message (Collapse) | Author |
|
Change-Id: I18b1b0b92b106c5fd96c80472e279a7cfd380235
|
|
|
|
Change-Id: I6d7f279732d9992d584aab96c3a747d6e6130147
|
|
The fix for Bug 76611 caused ~20% performance loss in XLS import.
With optional cloning of ScTokenArray of the shared XLS formulas
depending on the need of address wrapping, it is possible to gain
back near the original performance.
Note: The original patch for Bug 76611 has already got an unit test,
too (see wrapped-refs.xls), and that unit test works with this
improved patch, too.
Change-Id: Ibfb59d1543ef9c4b8a075d5c4e37f77ab451aaa0
|
|
Change-Id: Ia96b7816708e566d59628ae185bdc08023cdbe00
|
|
Change-Id: I2c1dd24fa2215300255ae4e7d50b6cf990b6e843
|
|
in favour of ReadXXX methods
Change-Id: I50a97c1855366dc14c6864da2fb91d1759d4d4db
|
|
Change-Id: I663f70d6324c6dd42208aa3804edfe3680881ea1
|
|
Change-Id: I1ab4e23b0539f8d39974787f226e57a21f96e959
Reviewed-on: https://gerrit.libreoffice.org/12164
Reviewed-by: Noel Grandin <noelgrandin@gmail.com>
Tested-by: Noel Grandin <noelgrandin@gmail.com>
|
|
Change-Id: Ie01e9488c6b8e36c40e6176bc1cac38a93a8087f
Reviewed-on: https://gerrit.libreoffice.org/12194
Reviewed-by: Samuel Mehrbrodt <s.mehrbrodt@gmail.com>
Tested-by: Samuel Mehrbrodt <s.mehrbrodt@gmail.com>
|
|
Change-Id: Ib15413e73409cc33de01fa92a47b9d1237cfc4b2
|
|
also
coverity#736888
coverity#736889
coverity#736890
Change-Id: I3e0d11d12c7da17297ba827f8125b9d99680121b
|
|
also
coverity#706042
coverity#706045
coverity#706046
Change-Id: I173603a638ef3e2d9ca358a4d4ba111910fac05f
|
|
Change-Id: I363c90a71225744bacd0bedccf6938da63e0428f
|
|
Change-Id: I1680d21b8f123ec1d937d699edee62b8ff0d6058
|
|
Change-Id: Ie9ecc13553e089197de24c8a151eb811e40bd56e
|
|
Change-Id: Iade3fedac5d2f8e978b7dd9c30f001d7d1564946
|
|
When importing shared formula tokens.
Change-Id: I7e1a05a78c3a93330476516e0459cffb668e3f66
|
|
Change-Id: I70aad0b38979f45a313b8ac36890fb6c64d11bb0
|
|
Change-Id: I5b00cdf68220282e5d0ff99352e04552cdecce09
|
|
Change-Id: Ib3a442cbb68c23294762561f2911101a087a795e
|
|
Change-Id: Id1dcadcac179c52977e48a6912ce4d5fd542f60c
|
|
Change-Id: Id93fd35fe2482c4b639c25c9dd0d1ce4b4c8aa75
|
|
Because sometimes this may be correct.
Change-Id: Id2c47bb4ad3f91b366a25306169de58bb38c1e81
|
|
Change-Id: I740c0e18eefc64cb5f1da5b88ceb77289790ec37
|
|
Change-Id: I9869d9709f28b68ef7b518527175589d80644668
|
|
Because they are wrong more than half the time.
Change-Id: I24e49c53520442be698e43976d28d0f8477648e6
|
|
This can happen when Excel's original shared formula range is across
multiple columns, in which case we split that into multiple groups (one
per column), and initialize top cell only for the left most column.
This fixes the crasher on the doc from fdo#31296.
Change-Id: I9e4def9836947fc67523f0d99ca981465709b934
|
|
It will make implementing the background compilation of OpenCL kernels
easier, I hope. That needs data accessible through the ScFormulaCell.
Change-Id: I77ad5a94eb13551e6238d5f5285f1c916d7cb992
|
|
It has the potential to slow down the loading as we wait for the
compilation thread to finish when loading has otherwise
finished. Besides, the static kernels are not even used any more.
Keep the static kernel source code and the code that handles it for
now, for reference. Also keep the compilation thread source for future
modifications and re-use.
Change-Id: I7ae6ad08b0ac00a15cc4ce5bc9906972e7e23850
|
|
Instead of a pointer to ScTokenArray, and then clone its instance. If
the token array gets cloned in ctor, take a const reference.
Change-Id: I280fd7eb9eaea9905dbf954a1ace904ab0814dfe
|
|
Change-Id: Ieb8befa4f19ebc31f4afa1370c924b469aa77382
|
|
With this, both ScColumn and ScMatrix store svl::SharedString as their
string values, instead of OUString.
Change-Id: I3faece94d98f774881fd72b3ed5f6143504cd350
|
|
Change-Id: Icbfb7a601a01bf8fb2a30cb49a1522a25e6022f1
|
|
Change-Id: I7b70d5c699819103260c2b248203c41f1d543a79
|
|
Change-Id: I08d7224df62723b70b46981f1b6d61379ab62747
|
|
Change-Id: Ib23a94ee45539c01c3e4c95b3d5983c470950a8b
|
|
Change-Id: I263ef2594080ff7d47d5499c2b62e60e1689d2d6
|
|
Change-Id: I1c9c2a62b2a1ede08fb2f27f541b501dc3ce7b2f
|
|
Take the start row position from the preceding formula record, rather
than believing what's in the shared formula record. The latter can be
wrong, and can be wrong often.
Change-Id: I3a4da110727a7719e5f8cb3e6250c0e1bef04c64
|
|
If we still have 3 MB worth of stream to parse after the first formula
record, we pre-compile OpenCL kernels in a separate thread.
Change-Id: I129d95c26f26e4b1eb989bc2a1764a3f87f71fcb
|
|
It's still a no-op & we need to have a smart mechanism to conditionally
trigger it rather than running it in all documents.
Change-Id: Ia875ebb9405b5de5c5d31418de84c5ca7a62f302
|
|
Change-Id: I205c5b7c1202862ea7129fbd61d54e48cd37b06e
|
|
Many xls documents contain shared formula entries that don't follow
the spec. We still need to be able to handle them.
Change-Id: Ia7d01101a6759967d9ebb27b6540dcb67607d13f
|
|
And do it once when importing xls, xlsx, and ods documents.
Although xls(x) file formats support shared formula natively, it's
still beneficial to re-group it as some old xls documents limit the
length of shared formula span to only 64, or some don't use shared
formulas at all even though they could. And re-grouping just once
is not terribly expensive.
Change-Id: Iff9c605d19baa187553ddab6af8b9fbd4c366d7d
|
|
Change-Id: I70694ee8834b1b2f2ebdfaa90582ccfb19db0210
|
|
No more mapping to range names.
Change-Id: Ic43b6ef35a91fe4d6fff748ebc22969ba4e036db
|
|
Still the first step.
Change-Id: I1897c9c2cd3a5b5245febbfba76e1b088054f578
|
|
Change-Id: If4499dbee52532cc8a95c380bbe840719fe9279b
|
|
Change-Id: I36bc90e8477cba306e107fb6d6e47f5732161df8
|