Change-Id: I96ef317e974a9db1713ad3005d0efcecb2bacda5

Is not a good idea after all, says mmeeks. Plus, we should check if
the device actually supports it (with clGetDeviceInfo
(CL_DEVICE_QUEUE_PROPERTIES) before attempting to use it.

This reverts commit a4681be5.

Change-Id: I11896ab2d40bd8abcb11a21ce5df4eb75816013a

and make it statically allocated, no point in dynamically allocating
such a small object

Change-Id: If476bf5dc1e0d535383d16bc49c8d567776f16cd

Change-Id: I288f7ff0327831603eda6e827c8acbae678dfaff

Also extended our spreadsheeet test to search through more columns, open spreadsheets
with macros enabled, and added a test for the the Year(), Month(), Day(), Hour(),
Minute(), and Second() functions comparing Calc's formulas vs StarBasic's runtime functions.

Found-by: villeroy
Patch-by: Damjan Jovanovic

(cherry picked from commit a6849326)

Change-Id: I8f2115c623a1d35db5b7fc8184a9118c3eca6fcd

Introduced by the commit 968f4d72,
problem found by Stephan Bergmann.

Change-Id: If735185c34a0ba69a5cd753ef76032b1b6a4a0cf

Change-Id: Iff5ef2a5b2315b19cf8f0e84a08c776c3c97e0ae

For example, "B2:" or ":B2" were accepted as valid range, with the
missing part set to A1, which resulted in a A1:B2 range.

An omitted sheet specifier always results in a valid relative sheet,
hence that flag is set, so testing flags for non-null before the final
result is available isn't sufficient.

Change-Id: I57d1b7cf71a298e7ac108b0e13aab88208f0f00c

Build ID will show enabled OpenGL with an extra string "-GL"
helping the fix of rendering issues.

Change-Id: Id7bf2db2edb165542bf7a2a253c698c494278a03

Change-Id: I544a67e3706c7d12414cc075118ef2f0f5ddd0f6

Change-Id: Iea0911d182b843c099cca0f2c5ec638034e7268d

Change-Id: I74d55dde824b00e4de13134c5c4eed16b1b5c5df

Change-Id: I49162355ce5a58c6aaadfcfb9f45471e5c2a1424

Change-Id: Ie95c36fe3705e2645a59cac117d99d7b85388ce1

Change-Id: Ie80dfb003118d40741549c41ebcc7eda4819f05b

Change-Id: Ib6662c09feeab5f86ddbbccf5ce0705929194f43

See (short) discussion on the mailing list, "How was it again, is the
DbgDialog useful?".

Change-Id: Ibde1eb13f16edf94f1f7aebd0befd1b0b171d5c4

Change-Id: I18bce66cc0068815a466291a1a72093704d9fdc7

with the variadic variants.

Change-Id: Ib5335ab6e64ee407e3aa8f9558a24af40a308107
Reviewed-on: https://gerrit.libreoffice.org/17981Reviewed-by: Noel Grandin <noelgrandin@gmail.com>
Tested-by: Noel Grandin <noelgrandin@gmail.com>

with the variadic variants.

Change-Id: If88e1d741075e86997c91dc2c59eeebe45f67c1f
Reviewed-on: https://gerrit.libreoffice.org/17980Reviewed-by: Noel Grandin <noelgrandin@gmail.com>
Tested-by: Noel Grandin <noelgrandin@gmail.com>

Change-Id: I21fbfe37f89fa404688cd1b287163801da9f7e0a
Reviewed-on: https://gerrit.libreoffice.org/17974Reviewed-by: Noel Grandin <noelgrandin@gmail.com>
Tested-by: Noel Grandin <noelgrandin@gmail.com>

sberg did not like this macro, let's see if I can find a better
solution.

This reverts commit e51fa143.

(cherry picked from commit e3cd97ce0b2e153b8b6812cc66f2e7d33504a7da)

Conflicts:
	vcl/opengl/program.cxx

Change-Id: I86ae50bd8d2148e86a6c8fc0821d3e2bc29a0cd4

Change-Id: I3d5eae99427ca68a5121ad43e9f4e7df03ddc23e
Reviewed-on: https://gerrit.libreoffice.org/18005Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Tested-by: Markus Mohrhard <markus.mohrhard@googlemail.com>

Change-Id: Ia9ae1f8583948620b6f496afcadd292671cdc906
Reviewed-on: https://gerrit.libreoffice.org/18004Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Tested-by: Markus Mohrhard <markus.mohrhard@googlemail.com>

Change-Id: Id407614c75eb04fb920fcf75d368b070323a0b79
Reviewed-on: https://gerrit.libreoffice.org/18003Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Tested-by: Markus Mohrhard <markus.mohrhard@googlemail.com>

Any string in one of given matrices should result in IllegalParameter
error.

Change-Id: Ia08d879ec109e04021373c34e83f1873be6815c2
Reviewed-on: https://gerrit.libreoffice.org/18002Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Tested-by: Markus Mohrhard <markus.mohrhard@googlemail.com>

Strings should be omitted by the algorithm, so such test cases are
important (were missed before).

Change-Id: I2e77b2087e398bf27ac71ba4da0b26c66cc08311
Reviewed-on: https://gerrit.libreoffice.org/18001Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Tested-by: Markus Mohrhard <markus.mohrhard@googlemail.com>

Same as in ScStatisticsInputOutputDialog.

Change-Id: I0e3eb06bc86cf77c405c54f312340c7b2551c1ec

Same as in ScSamplingDialog.

Change-Id: Iff709ec2284673efa8824a9b8cfc037e4d6d567a

Same as in ScRandomNumberGeneratorDialog.

Change-Id: I1bc9296bfc8b1b2b8f3fc20183e2c626f94dee09

... and disable/enable Apply/Ok buttons for in/valid ranges.

Change-Id: I52b5eae2f34e3fca7c0761b14124477a0d160b01

The string support certainly isn't complete or correct
anyway. Partially revert c3383aaf.

Change-Id: Ica86f39daf864a1a62d92f8d8300d75d020c0ee0

If an OpenGL zone takes >2s to make progress, disable OpenGL.
If an OpenGL zone takes >5s to make progress, abort the app.

Change-Id: I776c06a3f8ba460ff9842a9130c21f9ee2147eee
Reviewed-on: https://gerrit.libreoffice.org/17986Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Michael Meeks <michael.meeks@collabora.com>
Tested-by: Michael Meeks <michael.meeks@collabora.com>

Change-Id: I937ec900044bbc6027ff8d4ae37f2f275dde974f

Instead of a bare NaN.

Change-Id: I170c540478315eedd23f5851d54c30175d21ef96

This reverts commit af9751a7.

just shove the unneeded pam into dead space

Change-Id: Id0a3d550cd9d54394888e6b2d62e21eaded24a7c

so do that in the ctor and drop the arg from a bunch of places

Change-Id: Iaad71e9854b38d8632df057dfe9c96468d1d0098

The commit c3383aaf was missing an !
operator in two places where it checks whether a cell is a numeric or
string one, resulting in it always using the string value, which for
NULL (the case I was looking at, only numeric cells in the array)
meant zero was returned.

As such I am not sure if it is entirely correct to do the check
whether a cell is a numeric or string value in the order the generated
OpenCL code does here (and all over the place perhaps). The
documentation in <formula/vectortoken.hxx> says:

 * Single unit of vector reference consists of two physical arrays.
 *
 * If the whole data array consists of only numeric values, mpStringArray
 * will be NULL, and NaN values in the numeric array represent empty
 * cells.
 *
 * If the whole data array consists of only string values, mpNumericArray
 * will be NULL, and NULL values in the string array represent empty
 * cells.
 *
 * If the data array consists of numeric and string values, then both
 * mpNumericArray and mpStringArray will be non-NULL, and a string cell will
 * be represented by a non-NULL pointer value in the string array.  If the
 * string value is NULL, check the corresponding value in the numeric array.
 * If the value in the numeric array is NaN, it's an empty cell, otherwise
 * it's a numeric cell.

Note how that implies one should first check whether the value in the
string array is NULL or not, and only if it is NULL, look at the vale
in the numeric array. The code in the generated OpenCL VLOOKUP
implementation does it backwards. Scary. But probably equivalent for
the subset of cases we actually handle in OpenCL, which (I think) are
those where no string cells are involved.

More bug fixes for the OpenCL VLOOKUP will follow.

Change-Id: Id567c245a0700267584be6032320863a4a66df83