Skip to main content
AgeCommit message (Collapse)AuthorFilesLines
2018-12-20Bug 542865: [GTK4] Snippets with any children of Shell crash uponEric Williams1-0/+3
gtk_widget_get_surface_allocation() Don't call gtk_widget_set_has_surface() if the value is TRUE and the widget is SwtFixed. Change-Id: Icd39da60b7c9b68f63e37e5310ffb7ed6770256b Signed-off-by: Eric Williams <>
2018-12-20Bug 542865: [GTK4] Snippets with any children of Shell crash uponEric Williams5-3/+51
gtk_widget_get_surface_allocation() GTK4 no longer supports "own" surfaces for widgets, so this patch removes this logic from the GTK4 SwtFixed implementation. Some adjustments have to made in Control, as we do gtk_widget_get_has_surface_or_window() checks to determine the paintHandle. In GTK4 this is broken so always use the surface of the fixedHandle (or clientHandle in the case of Group). Change-Id: I09c0c096407ab67ac10edfcd21a3c9927cb5f833 Signed-off-by: Eric Williams <>
2018-12-19Bug 542926 - [GTK3] Combo has incorrect initial sizeXi Yan1-0/+28
Using gtk_widget_size_allocate (only works for GtkContainer on GTK3) for Combo#setInitialBounds does not assign the correct size. Use gtk_widget_set_allocation instead to set the correct initial size for Combo. Change-Id: I2a31e1d91dc0724f350ce32c0eb9866ab010d80b Signed-off-by: Xi Yan <>
2018-12-19Bug 542867: [GTK3] Combo pulldown text cutoff on LinuxEric Williams1-0/+8
Enable fit-model on a READ_ONLY Combo's cell renderer when computing size, as it will use the largest item's width to compute the proper size. The fit-model setting can be toggled back to false after performing the compute size operation. Change-Id: I7e5b057e7d3fdd51181cf6b059b03b67dff262d6 Signed-off-by: Eric Williams <>
2018-12-19Bug 541451 - [GTK3] Replace deprecatedXi Yan2-6/+10
gdk_window_set_background_pattern() Removing gdk_window_set_background_pattern for GTK3.22+ Tested with ControlExample, GraphicsExample, snippet with bug 218935, bug 475784, Snippet154, 174, 281, 48. Change-Id: Id269e8babaf6473e9be25c04f55bca2253df8d55 Signed-off-by: Xi Yan <>
2018-12-17Bug 541635 - [Wayland] Random behaviour when moving editor tabsXi Yan2-2/+1
Remove public non-API field. There is no need to set mouseDown flag in DragEnd as when source.dragEnd is called, we set the mouseDown flag when we send the MouseUp signal. Change-Id: Ifa828bebc759ffafc656b7b414ae090b308533f6 Signed-off-by: Xi Yan <>
2018-12-14Bug 535978: [GTK3] Bad Scrolling behaviour of tables with CheckboxEric Williams4-4/+152
The issue at hand in this bug is that Tree/Table editing widgets are children of the fixedHandle, which sits above the GtkTreeView in the widget hierarchy. This means that they are siblings of the tree/table, and thus will continue to draw even on top of the tree/table headers. The fix is to calculate the location and size of the header bars and keep track of them in Composite. When editing widgets are drawn using Composite.propagateDraw(), we check to see if the widgets would be about to be drawn over a tree/table header. If they are, we raise/lower the GdkWindow belonging to the editing widgets, depending on whether or not it needs to be drawn or not. Tested on GTK3.24 using the snippet attached to this ticket and also the snippet from bug 531928. No AllNonBrowser JUnit tests fail, and the update sites editor in a child Eclipse seems to be working as intended. Change-Id: I66d90c438a3d44afdd6bf6112deef5af680cd201 Signed-off-by: Eric Williams <>
2018-12-14Bug 366471 - Improve OSX scroll performance with drawText cachingPhillip Webb1-28/+138
Add caching to GC.drawText to prevent multiple calls with the same single character from needed to perform expensive layout operations. Any call to GC.drawText() that is rendering a single character on a transparent background is now a cache candidate. The cache is limited to 20 items and is specifically designed to work well with classes such as WhitespaceCharacterPainter which frequently call drawText() but operate with a relatively small set of characters. Change-Id: I0ac4e3478e212ba5e410c5a06a6827279143f7be Also-by: Karsten Thoms <> Signed-off-by: Phillip Webb <>
2018-12-13Bug 541635 - [Wayland] Random behaviour when moving editor tabsXi Yan3-2/+26
1) Block in dragDetect until mouse is released or drag is detected while using gtk_motion_event for drag detection. See Bug 515396. This was done using extra mouseDown and dragBegun flags to keep track of the mouse or drag state. 2) When editor tabs are being dragged, CTabFolder receives button_pressed event a *single* motion_notify event when drag begins. Then the Tracker rectangle is initialize and all motion_notify event and button_released event is received by the Tracker which is not part of Control. This causes the issue that even when mouse is released, dragDetect still assumes the drag has not ended. So the cursor can still move a tab even after releasing the mouse. The fix is to extend the DND blocking logic and update that mouse was released in Tracker. Change-Id: I71ac7edf3f2c51349426004af514037ff537c0aa Signed-off-by: Xi Yan <>
2018-12-13Bug 540672 - [GTK4] Adapt to GtkStyleContext API changesXi Yan9-73/+294
Change-Id: Ia3b7acd7d34e35c1c3d392728e43c6d9573bffb1 Signed-off-by: Xi Yan <>
2018-12-12Bug 541381: [GTK4] Replace gtk_container_get_border_width()Eric Williams1-2/+2
Fix up certain margin calls to prevent stack overflow. Change-Id: I6ba117b27961127ee8a283a7da0cafd587640e18 Signed-off-by: Eric Williams <>
2018-12-12Bug 541381 - [GTK4] Replace gtk_container_get_border_width()Xi Yan11-21/+330
Replaced with gtk_widget_get_margin_*. Change-Id: I38a78ea6eb8cc5c8724d07fd8462fd8d8b3b0e88 Signed-off-by: Xi Yan <>
2018-12-12Bug 540841: [GTK4] Rename GdkWindow to GdkSurfaceEric Williams2-4/+4
Fix typo that prevents GTK4 bindings compilation. Change-Id: Ie754dea389350563dfef186dfd9051e0adee4b68 Signed-off-by: Eric Williams <>
2018-12-12Bug 542712 - 32-bit compile checks still run for Gerrit jobsAlexander Kurtakov1-8/+0
Stop calling the 32 bit compiles in check_compilation_all_platforms. As the target is used only for gerrit builds it would not affect people building on 32 bit. Although they may encounter missing 64to32 pointer comments in the code . Change-Id: I8d9e62ce68725bcdf3b541672ebe11a2e1037d5e Signed-off-by: Alexander Kurtakov <>
2018-12-12Run tests with surefire 2.20.1.Alexander Kurtakov1-1/+1
Change-Id: I90ea31ce5e396ecc7c7ebae79f781b3811054f06 Signed-off-by: Alexander Kurtakov <>
2018-12-11Bug 540841: [GTK4] Rename GdkWindow to GdkSurfaceEric Williams28-344/+956
Rename GdkWindow to GdkSurface, part 2 (4.11). Change-Id: I7bcb17d4c88b9f22cf75fc2faa4f0de7a6129386 Signed-off-by: Eric Williams <>
2018-12-11Bug 541862: [GTK3] Increase minimum GTK3 version to 3.10Eric Williams21-271/+38
Remove old 3.8 and below code. Change-Id: Ib3cb77939a7146396c36e982a93e135b18318855 Signed-off-by: Eric Williams <>
2018-12-11Bug 539761 - [GTK3.24] Replace deprecated g_type_class_add_private()Xi Yan1-18/+20
Replace g_type_class_add_private with G_ADD_PRIVATE. Replace G_TYPE_INSTANCE_GET_PRIVATE with *_get_instance_private Change-Id: Iac418463335d28d21750f0a97074dc90d872c3d6 Signed-off-by: Xi Yan <>
2018-12-11Bug 532074 - [Wayland] Arrow menu from fast view pops up in wrongXi Yan1-2/+20
location setLocation method is limited on Wayland since we cannot use gdk_popup_at_rect if GdkWindow of the shell is not mapped. When the Problems View is a fast view on Eclipse start, the drop down menu takes PartRenderingEngine's limbo shell (see PartRenderingEngine#safeCreateGui) as the parent shell which is off the screen. The limbo shell's GdkWindow cannot be used on Wayland to popup the menu at a relative location. In this case, we open the menu at the pointer to avoid it being positioned in the wrong location. Change-Id: Ifa120a0e68e0d012b09c1f4806dc959bb447b8f2 Signed-off-by: Xi Yan <>
2018-12-11Revert "Revert "Bug 541854 - [Wayland] Ctrl+1 popup issues""Xi Yan2-73/+57
This reverts commit c88920365b95e77e6369a9a1626007d4d24fe97f. Change-Id: I0e831ecec4754a7edfeb1c4957f88418b5463454
2018-12-11Bug 541378: [GTK4] Implement GtkWindow "close-request" signalEric Williams5-12/+39
Use close-request on GTK4 to release/dispose of Widgets when the user clicks the "X" button of a Shell. Change-Id: Ic8dc4b2f53b63d7faf7069accdcb6703e93fcdc8 Signed-off-by: Eric Williams <>
2018-12-11Bug 542475 - [GTK3] DateTime date decreases when focusing outXi Yan2-7/+63
Same issue as bug 538648 with the SWT.MEDIUM style on US locale. Modified previous patch to account for all cases where Month as string is the first field in DateTime spinner. When using SWT.SHORT | SWT.DATE on US locale, increment arrow increases date by 2 and decrement arrow doesn't work. This is because new_value from calendar have month between 0-11 while the adj_value have month betwene 1-12. Fix is to shift adj_value by 1 offset so that we get the correct arrow direction. This doesn't happen if DATE is the first value (i.e. using Canada locale) Change-Id: I52c5ef02e936a89d66566032aff437d165ceb289 Signed-off-by: Xi Yan <>
2018-12-11Bump versions for 4.11 stream.Alexander Kurtakov5-6/+6
Change-Id: I36ed391a09e9e09fe55aaa7acd1df6d701a2397f Signed-off-by: Alexander Kurtakov <>
2018-12-11Bug 541819 - Configure SWT build scripts for 4.11Sravan Kumar Lakkimsetti1-3/+3
Change-Id: I2b30a35dc48c1291fa8190149dadfef35a339be6 Signed-off-by: Sravan Kumar Lakkimsetti <>
2018-12-11Bug 541819 - Configure SWT build scripts for 4.11Sravan Kumar Lakkimsetti2-3/+3
Change-Id: I2284d3a9c38bee4596bba23a0278746efbd0d6dc Signed-off-by: Sravan Kumar Lakkimsetti <>
2018-12-08Bug 541819 - Configure SWT build scripts for 4.11Sravan Kumar Lakkimsetti1-4/+4
Change-Id: I9b1daf09b1b175e7866e80558e709d50124e0a04 Signed-off-by: Sravan Kumar Lakkimsetti <>
2018-12-07Bug 541822 - POM and product version change for 4.11 releaseAlexander Kurtakov18-19/+18
Change-Id: I8454243e87078b718102ce58702c08fe8f9f163c Signed-off-by: Alexander Kurtakov <>
2018-12-06Revert "Bug 541854 - [Wayland] Ctrl+1 popup issues"Xi Yan2-57/+73
This reverts commit 3615a6bf1b910747186515b2526504f91dbe4dae. Change-Id: I16fc9137a5380e7e3f7851fe5710b8a22cc68ee8
2018-12-06Bug 541854 - [Wayland] Ctrl+1 popup issuesXi Yan2-73/+57
Attempting to address all popup related issue (see bug 542104). 1) Patch to bug 515773 ungrabbed focus when shell is hidden in setVisible(false) and used showing/hiding hack to avoid being stuck with no focus, which introduced some of the related issues. Rewrote logic to ungrab focus after shell is disposed. - Removed showing/hiding hack that uses callback to show shell in bringToTop(). - Addressed different cases for single popup and popups attached to a parent popup. - Added additional guard to ungrab the grabbed focus.. 2) Patch to bug 540166 uses callback connected to gtk destroy signal which causes crash in Flatpak due to invalid child handle passed to the callback. Workaround is to move the logic to SWT level by keeping track of the child popup shell of a parent popup shell, and dispose the child popup whenever the parent popup shell is disposed. Known limitation: - For the auto-completion, if you switch focus to the details page using Tab, clicking on an item on the list page will close the auto-completion popup instead of highlighting the list item. This is because we ungrab the focus whenever user clicks outside the focused popup window (in this case its the details page) to avoid being stuck without focus. This means we can only switch between list item using the mouse when the details page is not in focus (i.e. hit Esc first to bring focus back to the list item before clicking list item). - However, we should not fix this with the show/hide hack because with the show/hide hack, hitting Esc does not work when focus is switched to the details page using Tab. See bug 542104 comment #2 for test cases. Change-Id: I4ba90afdf9664e6858aa3c28cf51825f85171a76 Signed-off-by: Xi Yan <>
2018-12-04Bug 542101: [GTK3] Increase minimum GTK3 version to 3.8.0Eric Williams1-1/+1
Bump minimum version to 3.8.0 to avoid JVM symbol lookup errors for an a11y macro which was not public before 3.8.0. Change-Id: Ia9cbca30070907ab9904fdf75c195fa76ff29488 Signed-off-by: Eric Williams <>

Back to the top