[Libreoffice-commits] core.git: Changes to 'feature/perfwork3'

[Kohei Yoshida]

New branch 'feature/perfwork3' available with the following commits:
commit 6cfa01aaf226130c655eef38ba9c42c7715475b5
Author: Kohei Yoshida <kohei.yoshida at collabora.com>
Date:   Tue Oct 14 17:40:40 2014 -0400

    AdoptTextMinSize() calls NbcAdjustTextFrameWidthAndHeight() at the end.
    
    No need to call this again in the caller.
    
    Change-Id: Ib36853cf5a0720082275e0490b4443cd9e0bd007

commit 09d2b89c5552204aff0001f039fb9d76afb4792d
Author: Kohei Yoshida <kohei.yoshida at collabora.com>
Date:   Tue Oct 14 17:19:56 2014 -0400

    Reduce scope level & Annotate the code a bit.
    
    Change-Id: I6ebf849243000920b1c1f7e67d2dc81fcee9e16b

commit 79b5057d62004ed5ab53c3c30f074e8dc021d0a0
Author: Kohei Yoshida <kohei.yoshida at collabora.com>
Date:   Tue Oct 14 16:29:04 2014 -0400

    Remove local variables that are no longer used.
    
    Armin removed their uses in 120e469d176026ceb59abbf74d2ad255323cbc9a.
    
    Change-Id: Id620466929ca3a66dd1c261e81aacd533a44cd76

commit 0e3afe9bd08c6c857703019c630559b902821db9
Author: Kohei Yoshida <kohei.yoshida at collabora.com>
Date:   Tue Oct 14 16:14:41 2014 -0400

    Method descriptions.
    
    Change-Id: Iece61424e09fc36768889fde2c848ed2b0722701

commit 866769b83a3c4028fd86a7c33749645220aa33ed
Author: Kohei Yoshida <kohei.yoshida at collabora.com>
Date:   Tue Oct 14 15:37:14 2014 -0400

    Explain what these are.
    
    Change-Id: Ibb7b830c4216967a5877224c702ad4b3e1fe0b34

commit e9b1d1214404ed86adcbdb7451fc0618dca96731
Author: Kohei Yoshida <kohei.yoshida at collabora.com>
Date:   Tue Oct 14 11:35:29 2014 -0400

    Take the current axis alignment state.
    
    Change-Id: I02c28f44af0a1510bdb780d1e6c70561b7ba24e5

commit b26bf647427784906d542574d7abebf6ad547878
Author: Kohei Yoshida <kohei.yoshida at collabora.com>
Date:   Mon Oct 13 19:32:15 2014 -0400

    More on documenting the methods.
    
    Change-Id: Ibdbc1b26521e56de5dcba1b9572bb0221fc0fff5

commit e266abe7c3a0e7055281b974fe3df03a23e3fa6b
Author: Kohei Yoshida <kohei.yoshida at collabora.com>
Date:   Mon Oct 13 15:24:33 2014 -0400

    Scope level reduction & explain what the max axis index really is.
    
    The concept of axis index and dimension index is quite confusing in the
    chart2 code....
    
    Change-Id: I035af0a225396bd5394089d6f584f78dcaf0ad31

commit 99769e581a60423623547747f806dc13e8e92d99
Author: Kohei Yoshida <kohei.yoshida at collabora.com>
Date:   Mon Oct 13 15:11:51 2014 -0400

    Explain what doAutoScaling() does.
    
    Change-Id: I15ca9ca0c3a22094f0a969adf2b3f1145d116052

commit 7af340d2bb196a0b4ceb64fca0daaa323be4690d
Author: Kohei Yoshida <kohei.yoshida at collabora.com>
Date:   Mon Oct 13 21:57:39 2014 -0400

    Make VCartesianAxis::get2DAxisMainLine() truly const method.
    
    No internal state gets modified by this method now.
    
    Change-Id: I10475ad3483cb3a92e41208e1a55a6b925c10223

commit 3894bec2a9b2077366ff90be8969c97312177cfc
Author: Matúš Kukan <matus.kukan at collabora.com>
Date:   Thu Oct 9 15:22:54 2014 +0200

    package: Add ZipOutputEntry to isolate deflating of streams.
    
    Preparation commit for deflating streams in parallel.
    We still use the same single XOutputStream (ByteChucker :-) for
    sequential writing but this can now be changed more easily.
    
    Change-Id: Idf26cc2187461660e31ac2e12c4708e761596fb2

commit 885ebddd12a954abc79d91e010941e876ecc516d
Author: Kohei Yoshida <kohei.yoshida at collabora.com>
Date:   Fri Oct 3 16:43:25 2014 -0400

    Inlining make no sense for virtual functions.
    
    Change-Id: I6392eaceb0544b7faa9a0c726acf6619d971dbb7

commit 85d7f57591c4d6377dd64b69d4af05eced57d297
Author: Kohei Yoshida <kohei.yoshida at collabora.com>
Date:   Fri Oct 3 16:28:54 2014 -0400

    GetSymName() not used.
    
    Change-Id: Ia369bf99a5e381a6f1f9c3d8a2499aa1780f04b8

commit df70fda98f82374bd0e8d22790161dc6f3eb0963
Author: Kohei Yoshida <kohei.yoshida at collabora.com>
Date:   Fri Oct 3 16:26:05 2014 -0400

    GetNameAsString() identical to GetName(). Remove this and use GetName().
    
    Change-Id: I26dce2dd11792ee118e78d23d652a5feb0789830

commit c90bcafff39ad959320a726c2bffdaba03a81bb8
Author: Kohei Yoshida <kohei.yoshida at collabora.com>
Date:   Fri Oct 3 16:21:14 2014 -0400

    ( void ) -> ()
    
    Change-Id: I0d3d1d9ab5f7bc270c89a2a98d45ebea3cc37e02

commit ab6389e20d32afa7aca43348989ec0a674ef0190
Author: Matúš Kukan <matus.kukan at collabora.com>
Date:   Fri Oct 3 23:12:56 2014 +0200

    FastSerializer: Also use cache for writing to ForMerge if we are inside mark()
    
    To ensure the correct order of calling ForMerge methods,
    call flush always before touching maMarkStack.
    
    This was the missing piece in optimizing write() methods,
    because of writeBytes() checking each time what to call.
    E.g. for Calc documents we don't use maMarkStack at all.
    So, just transfer the output to proper "ForMerge" when inside mark()
    and allow optimizations.
    This commit makes write() methods almost 1/3 as fast.
    
    Change-Id: I96c13888206c81f87e29b998839f78ea9d5570af

commit 4e6cc67dc0632b79b56e2f9150225f579e3522a9
Author: Matúš Kukan <matus.kukan at collabora.com>
Date:   Fri Oct 3 20:45:04 2014 +0200

    inline TokenMap::getUtf8TokenName
    
    Change-Id: Icd9c6ebc9feb3e7aba28b01729b582a8f49c832a

commit f72680e8307559f682267072786fcae28f119ba7
Author: Matúš Kukan <matus.kukan at collabora.com>
Date:   Fri Oct 3 16:37:26 2014 +0200

    FastSerializer: Simplify OUString write a bit more
    
    Change-Id: Ifa0746d635ec43cdc37867cf94bcc128bb8e2aca

commit 3c62004196470c102b69fa7ec893266b067abdf3
Author: Kohei Yoshida <kohei.yoshida at collabora.com>
Date:   Fri Oct 3 15:16:06 2014 -0400

    Let's consistently spell OpenCL as 'OpenCL'.
    
    Change-Id: I8bdcf0667b326b77d11001efeeefd5247987ebdc

commit f4d84ee7066de295996acfb287cbed8c2021a1fe
Author: Matúš Kukan <matus.kukan at collabora.com>
Date:   Fri Oct 3 12:30:51 2014 +0200

    FastSerializer: Avoid some cycles when dealing with doubles
    
    Would be easier to use OStringBuffer, but we can't get its pData member.
    Also its append(double) is suboptimal (or anything that uses
    rtl_str_valueOfDouble) - should be doing something like this commit.
    
    Change-Id: I8f3140081a574a84f0e60dc85cce1bd2de23cd34

commit 431b92f83174c169001074bb0113f55971f11ffd
Author: Matúš Kukan <matus.kukan at collabora.com>
Date:   Fri Oct 3 11:35:46 2014 +0200

    FastSerializer: Use fixed sized Sequence directly as cache
    
    Well, at least the allocated space is fixed size.
    When passing that to XOutputStream, change the size in a hacky way.
    
    Change-Id: I24fa134286e3086beda25c9a6915549e7c69119a

commit 203917ebf245c33a490181d2e8ed2a0ddbfda24a
Author: Matúš Kukan <matus.kukan at collabora.com>
Date:   Fri Oct 3 10:41:18 2014 +0200

    FastSerializer: Use -1 for unknown string length
    
    Change-Id: I3920caf9d95f20992b7961873f1668468d797e8e

commit 561717025ffec33b5491e35816e3fcdb87b2582d
Author: Matúš Kukan <matus.kukan at collabora.com>
Date:   Fri Oct 3 11:16:39 2014 +0200

    Remove few pointless OUString::number() and one method
    
    Change-Id: I3e9a302a7513eebfeff07402f71fc3dde22e4cc2

commit 2156a9f93ed59f997dd2ae0cf6a9ea9113b95719
Author: Kohei Yoshida <kohei.yoshida at collabora.com>
Date:   Thu Oct 2 15:24:46 2014 -0400

    Make the coding style consistent with the rest of Calc code.
    
    I just ran SlickEdit's 'beautify' command with some follow-up manual
    editing.
    
    Change-Id: I402e782e7d147c4c95ebaa8ffcc40b50a0c565a7

commit 9297396563f3ca187d2baa568278467d003805b1
Author: Matúš Kukan <matus.kukan at collabora.com>
Date:   Thu Oct 2 10:26:00 2014 +0200

    maTokenNames.size() is constant: use it and be a bit faster
    
    Change-Id: I39a6e2badf0c159e87763e2782bc89f0ee6068ec

commit c559af19d9e50f7f4f49ace0294d34dbe86c2d35
Author: Matúš Kukan <matus.kukan at collabora.com>
Date:   Thu Oct 2 12:02:40 2014 +0200

    FastSerializer: Faster write(OUString): add ascii check
    
    Saves about 80m pcycles for 180k calls.
    
    Change-Id: I9c9b3bf5a076df56d1b5b87f0a85ac3404abe8a4

commit a861b10823ee74d41b2f756b7c03618a9b701ab3
Author: Matúš Kukan <matus.kukan at collabora.com>
Date:   Wed Oct 1 18:45:23 2014 +0200

    FastSerializer: Simplify a bit
    
    Change-Id: I7dee908f5441ad34128d688e1008fd6ffad2ab3f

commit 86eb145e7361b7231cae3538156608358367d136
Author: Matúš Kukan <matus.kukan at collabora.com>
Date:   Wed Oct 1 18:24:58 2014 +0200

    Unused includes / using-declarations
    
    Change-Id: I8b43635ab1f77b61469f060c9a092f1494388955

commit 2be261b4c705218628de17a71c1a1611a850f465
Author: Matúš Kukan <matus.kukan at collabora.com>
Date:   Wed Oct 1 17:39:00 2014 +0200

    Use sc::CompileFormulaContext for faster export
    
    It's a cache for maTabNames (sheet names mangled for the current grammar
    for output) which ScCompiler needs and is expensive to get.
    
    Change-Id: I7e2954104427a3c220f0f097be6cd45a66485b56

commit 41ffcc9cb78f2599ded3fd546cd8d66b6e8b5234
Author: Kohei Yoshida <kohei.yoshida at collabora.com>
Date:   Wed Oct 1 20:34:36 2014 -0400

    Be sure to copy the cell text attributes values to and from clip.
    
    Otherwise we'd have to unnecessarily re-calculate the script types again
    which is not cheap...
    
    Change-Id: Ie589fb4a7e5ec9b5ef646dabea4e6bd0c0aca560

commit 696119c87410851e9353570061f9a647fcd7464c
Author: Kohei Yoshida <kohei.yoshida at collabora.com>
Date:   Wed Oct 1 15:18:17 2014 -0400

    Set script type to latin for formula cells with numeric results.
    
    But only when the column contains only standard number format types.
    
    Change-Id: I83982d2e87f9776cf03754beaf183e35675be992

commit 340d1d8212e7c5ca962aba992ab6b8ae29cc0d99
Author: Kohei Yoshida <kohei.yoshida at collabora.com>
Date:   Wed Oct 1 11:12:49 2014 -0400

    Exit early in case the column has no cell notes to copy to destination.
    
    Change-Id: Ifca77ccda7b2065b00ee29f29f377da599929843

commit bb484d5487f924993a6535b26442c50972ec14a3
Author: Kohei Yoshida <kohei.yoshida at collabora.com>
Date:   Tue Sep 30 23:28:41 2014 -0400

    Set latin script to numeric blocks of applicable columns.
    
    Change-Id: Ib81ef144f168fed38100127bd63f43ea5a835a13

commit a35f23e60e8dec2945f58d9b441b6fa1592f86db
Author: Kohei Yoshida <kohei.yoshida at collabora.com>
Date:   Tue Sep 30 22:23:16 2014 -0400

    Check for standard number format rather than just for 'General'.
    
    That's what the ods import filter does.
    
    Change-Id: Ibcd9a80a51785a448594d29a02283cca0ec53e95

commit da7ac33ae8924783d649d1add22eab9131fe2b91
Author: Kohei Yoshida <kohei.yoshida at collabora.com>
Date:   Tue Sep 30 22:00:05 2014 -0400

    Move this useful function to sc::NumFmtUtil.
    
    Change-Id: I7b42a4418408ee2c988c32e7f1adeee3dfe269c8

commit b92f8567d8092fad8f6fcaa92f4b61f5fd1cce7f
Author: Kohei Yoshida <kohei.yoshida at collabora.com>
Date:   Tue Sep 30 21:30:17 2014 -0400

    Try to determine whether or not a column has all 'General' number format
    
    during import.  We'll then use this information to set script type to latin
    for all numeric cells in those columns rather than leaving the script type
    'unknown'.
    
    Change-Id: I69eae1effc32c57290b0265bc6c87e58f51944b1