This changes tests so that we fill example data and create a new
pivot table from scratch, and create a new pivot chart using that
pivot table.
This enables that we test different cases without an example
document which creates a pivot table for us.
In addition add two test cases:
- pivot table with one column field and one data field
- pivot table with one row field and one data field (disabled, as
it currently crashes)
Change-Id: I8397a598857953e0b2d7da8dd91d7709a3495b71
Reviewed-on: https://gerrit.libreoffice.org/36625
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
Tested-by: Tomaž Vajngerl <quikee@gmail.com>
This is a squashed commit of the pivot chart implementation.
Some of the changes:
- Add pivot chart specific (pivot table) data provider which
provides the data from a pivot table to the associated chart.
- When inserting a chart and the cursor is in a pivot table,
in that case insert a pivot chart
- Modify the pivot chart when the pivot table changes
- Collect and set the number format for the values
- isDataFromSpreadsheet check for the creation wizard
- In ChartView (and VLegend) check if the data provider is a
pivot chart data provider and get the pivot table field names
to create the buttons on the UI.
- Adds the functionallity to show a filter pop-up (from calc)
when clicking on row / column / page field buttons.
- Remove (X)PopupRequest as we won't need it.
- Add ODF import/export for pivot charts:
+ Added loext:data-pilot-source attribute on chart:chart
which is the internal name of the pivot table with which the
pivot chart is associated with. If the element is present, then
the it means the chart is a pivot chart, else it is a normal
chart
+ Added service to create pivot chart data provider through UNO
+ Add new methods to XPivotChartDataProvider to create value and
label data sequences separately from the data source, which is
needed for pivot chart import
+ When importing defer setting the data provider until a later
time when we know if we are creating a chart od a pivot chart
- Pivot chart ODF round-trip test
- Add table pivot chart supplier API:
This adds the XTablePivotChartSupplier and related interfaces so
we can access, create, delete pivot charts from UNO in a sheet
document. With this we now distinguish between normal charts
and pivot charts. This was mainly needed because we can't extend
the "published" interfaces of TableChartSupplier.
- Added an extensive test, which uses the API to create a new
pivot chart when there was none, and checks that the pivot chart
updates when the pivot table updates.
Change-Id: Ia9ed96fd6b1d342e61c2f7f9fa33a5e03dda21af
Reviewed-on: https://gerrit.libreoffice.org/36023
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
Tested-by: Tomaž Vajngerl <quikee@gmail.com>
Mostly generated using
make check COMPILER_EXTERNAL_TOOL=1 CCACHE_PREFIX=clang-rename-wrapper RENAME_ARGS="-qualified-name=Rectangle -new-name=tools::Rectangle"
Except some modules have their own foo::tools namespace, so there have
to use ::tools::Rectangle. This commit just moves the class from the
global namespace, it does not update pre/postwin.h yet.
Change-Id: I42b2de3c6f769fcf28cfe086f98eb31e42a305f2
Reviewed-on: https://gerrit.libreoffice.org/35923
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
Tested-by: Jenkins <ci@libreoffice.org>
- if the UNO_QUERY_THROW cast throws, this skips all of the disposing
(and thus leaks)
- the whole try-catch-ignore case has been there since early c6b95527
- even in that commit message, no sane justification has been given
- apparently the try-block was Java-style "catch exceptions until it
runs" cargo-cult
- this change is still the more conservative change: probably the whole
try-catch framing should be removed
Change-Id: Ie3a71e48eb056afd8a844e19898dc1555de4297e
Reviewed-on: https://gerrit.libreoffice.org/35690
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Björn Michaelsen <bjoern.michaelsen@canonical.com>
To re-integrate the fixes in revert commit
ab92c84dfe600f320f37b6126cab11385a6fbdd1
we have also to update the binary ods,
to avoid break of "make check".
Change-Id: Ic044d43c6c3c3d9fc3e003c282fac4ba888a0115
Reviewed-on: https://gerrit.libreoffice.org/35685
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
... so admit that the VclPtr that ChartController used to have is really
owned by m_xViewWindow, and get fetch that (stack-scoped only) from
there.
Change-Id: Ib9e2cd20201ebe5019c3607a8c65cb1a9856812f
Reviewed-on: https://gerrit.libreoffice.org/35603
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Björn Michaelsen <bjoern.michaelsen@canonical.com>
On linux it was not a problem. On Windows there were some
random failures, but that might be fixed by using
Liberation Mono font.
Change-Id: Iae2684b302b3380c3a63b1fa80b1d0585454ed56
Reviewed-on: https://gerrit.libreoffice.org/35241
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Tamás Zolnai <tamas.zolnai@collabora.com>
Based on the casts in chart2/source/view/main/ChartView.cxx
and the similarity of naming of values, I conclude that this
enum was intended to abstract over css::text::WritingMode2.
Added a comment to that effect.
Change-Id: I3af8bbe8b6ac8c4a9375f6ccde145b98b9c69a57
Reviewed-on: https://gerrit.libreoffice.org/35164
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
Make sure the default is for kerning to be enable, and honor the setting
to disable it.
This also updates the expectations for chart2_xshape tests for file that
had kerning disabled and was previously ignored.
Change-Id: Icf6bb75e71cae43868fba323ca3c61e4a3e81523
Reviewed-on: https://gerrit.libreoffice.org/34169
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
This is not needed; the problem was in the Fedora 25 epoxy headers:
They got badly broken in an update, and should not be used from C++
code. If you use Fedora 25, make sure to use --without-system-epoxy
for now. See https://bugzilla.redhat.com/show_bug.cgi?id=1431994
This reverts commit cbb2e6b76868d5c67a713b68310b44be59bfab3e.
Change-Id: I55c77ff279f1716004ac77cd3a08c4bb3fdf644b
For some reason this was not detected earlier, but now when the epoxy
headers in f25 were updated, it is.
Change-Id: I7be830153792414b0f913aaaecb2a1336ebf6fce
