_isReady() can be true even if the _docLayer is not
created yet and in this case we dropped the window
messages which should have triggered the input bar
and the sidebar creation.
cypress tests were failing because of this issue sometimes.
Change-Id: Ifda1d66c952c65c420db06bfd61918411426ae93
Reviewed-on: https://gerrit.libreoffice.org/c/online/+/103010
Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
Reviewed-by: Tamás Zolnai <tamas.zolnai@collabora.com>
The hammer library sometimes or random when
consecutively pinch events, the center and scale
has infinity values
Change-Id: Ide6605bcbc0c7b4818fd27b7b44706fb8122d80a
Reviewed-on: https://gerrit.libreoffice.org/c/online/+/100744
Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
Reviewed-by: Andras Timar <andras.timar@collabora.com>
(cherry picked from commit 091eca0ae421ca436ba95ea1e478909f2a18d0a7)
Reviewed-on: https://gerrit.libreoffice.org/c/online/+/103047
Also re-render as soon as we have grid positions.
Change-Id: I57095683e662991badcb3a58832c81ebb3bc460d
Reviewed-on: https://gerrit.libreoffice.org/c/online/+/103043
Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
Reviewed-by: Michael Meeks <michael.meeks@collabora.com>
I reported the issue. Let's have a workaround until,
the bug fix is released.
Change-Id: I7d7719d311e0c0584b7d6e710286f28962d00a96
Reviewed-on: https://gerrit.libreoffice.org/c/online/+/103038
Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
Reviewed-by: Tamás Zolnai <tamas.zolnai@collabora.com>
- Avoid having multiple document-header elements when on notebookbar
- Add classes to the main-nav element when in presence of Readonly
Change-Id: I82cb816fcfa28d790202c96484a7e8c99f3875f6
Reviewed-on: https://gerrit.libreoffice.org/c/online/+/103004
Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
Reviewed-by: Pedro Silva <pedro.silva@collabora.com>
Newly close button does not actually close the document,
but steps back to read only mode. So we need to push it
twice.
Change-Id: Ic38f1e658396d9afe49496cfb75aa68df1c4623f
Reviewed-on: https://gerrit.libreoffice.org/c/online/+/102997
Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
Reviewed-by: Tamás Zolnai <tamas.zolnai@collabora.com>
So that it's more obvious that they don't even read the state of the
object.
Change-Id: I71d39bf3a20989c8fbf3410c2f1d2a98661c49d5
Reviewed-on: https://gerrit.libreoffice.org/c/online/+/102986
Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
It's more effective, since it reruns only the failing test,
not the whole test suite. Also it has a nicer output.
Change-Id: I8a50a20366b76a5cedb4e9383220c53f367c5a68
Reviewed-on: https://gerrit.libreoffice.org/c/online/+/102995
Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
Reviewed-by: Tamás Zolnai <tamas.zolnai@collabora.com>
* Fixes random failures in interactive test runner.
* Introduces test retries, what we did with an own script so far.
Change-Id: Iab5c32e854618a694e7d10d3c92c1c6c32465983
Reviewed-on: https://gerrit.libreoffice.org/c/online/+/102991
Tested-by: Tamás Zolnai <tamas.zolnai@collabora.com>
Reviewed-by: Tamás Zolnai <tamas.zolnai@collabora.com>
Don't re-send un-changed tiles that we can detect easily.
Also avoids some PNG compression / CPU overhead server-side.
Change-Id: Ieca05680d9194e0bfc177b8db338010e5ffafe75
Reviewed-on: https://gerrit.libreoffice.org/c/online/+/102954
Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
Reviewed-by: Jan Holesovsky <kendy@collabora.com>
When the span starts in the middle of the view don't render backwards.
Change-Id: Icc97fef88a65c0ca83167ddb72c03bece9a8e047
Reviewed-on: https://gerrit.libreoffice.org/c/online/+/102951
Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
Reviewed-by: Jan Holesovsky <kendy@collabora.com>
This fixes setup of many cypress tests.
Change-Id: I4eb626050d2d4202104ab01a6aa0b01248ae4eb5
Reviewed-on: https://gerrit.libreoffice.org/c/online/+/102965
Tested-by: Jan Holesovsky <kendy@collabora.com>
Reviewed-by: Jan Holesovsky <kendy@collabora.com>
When loading a spreadsheet document, the server should send a message
to show the cursor appropriately
Change-Id: I520a2b21fab903fc6b17ea612bbe1691ef311dbd
Reviewed-on: https://gerrit.libreoffice.org/c/online/+/102692
Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
Reviewed-by: Henry Castro <hcastro@collabora.com>
Per the rfc (https://tools.ietf.org/html/rfc6455#section-5.5.2):
"An endpoint MAY send a Ping frame any time after the connection
is established and before the connection is closed."
And "Upon receipt of a Ping frame, an endpoint MUST send a Pong
frame in response, unless it already received a Close frame."
Here we allow for pings to come from clients and we respond
to them by pongs, as required by rfc 6455.
Change-Id: I8e285f095526e4b67373ecb3ae1efc9c8717d756
Reviewed-on: https://gerrit.libreoffice.org/c/online/+/102948
Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
Reviewed-by: Andras Timar <andras.timar@collabora.com>
The split panes are not modified that often, no need to have buttons to
set that up in the toolbar.
Change-Id: I9c7920f6dd57de5f507450e15fcde615ad941f1c
It is too buggy, let's instead allow only setting of that via the menu,
similarly to what the desktop LO does.
Change-Id: I219b2535e0cbd5310fe2a3a9ebf2098536d5c175
This is partial revert of 901ecca313.
Handles (Markers) 'down' handler disabled
main TouchGesture handler what caused
taps to be ignored. Eg. double click didn't
enter edit mode for a Calc cell.
Change-Id: Id7097e9ad513e56a71423a6d6a8ac5105c373f77
On mobile, the center of the zooming has to be according to the center
of the pinch-to-zoom, ie. the center has to be taken into account.
Change-Id: I3ba2ea90b7bac9bc1ba27f8068ea7ed6bbb4910d
This avoids display corruption when panning, whereby stale/old
canvas content would continue to be rendered in the 'holes'
where invalid tiles were not rendered.
Change-Id: Ic886c0924c5a930116b1437c8e0cf35726ab76a5
The core of the fix is in _getNewPixelOrigin() where the round() behaves
non-predictably / inconsistently with the rest of the code, causing
random off-by-one error that shows (or not) depending on the window
size.
The biggest problem of this is that this off-by-one is then multiplied
somewhere by the zoom factor, causing the canvas being completely
blurry; but eventually when the user clicked into the sheet, it
'magically' fixed itself.
The rest of the changes (in setZoom()) should actually do the same thing
as the previous code, but using existing methods, instead of computing
the shifts manually.
Change-Id: If0ecb1301b7c1e65cfe8126385ef959c584c5d16
With this, if you increase or decrease the browser zoom, the document
zoom still stays the same.
Before this, when you had eg. 100% document zoom and 150% browser zoom
and try to zoom out, it actually zooms in instead, because the browser's
zoom is added to the mix; and it displays the wrong value in the
dropdown. Even worse, to get the 100% again, you have to choose 80% so
that the correction for the browser zoom is added, resulting in the
100%.
We should keep both the document and browser zoom separately. The
questions is then whether to combine them later for the actual document
rendering; I believe we should not, but even if we should, we cannot do
it directly in the setZoom() method, but instead closer to the painting
itself.
Change-Id: Ib7f3d2ae8b4e6e6086f14e933b215c32326c6be6
