Age | Commit message (Collapse) | Author | Files | Lines | |
---|---|---|---|---|---|
2020-04-08 | Remove deprecated backend headers | David Robillard | 2 | -46/+0 | |
2020-04-08 | Mac: Use high-resolution backing surfaces | David Robillard | 3 | -43/+111 | |
2020-04-06 | Implement puglSetTransientFor() for Mac and Windows | David Robillard | 4 | -1/+41 | |
2020-04-03 | Rewrite C++ bindings | David Robillard | 9 | -70/+919 | |
2020-04-03 | Mac: Use NSMakeRect over CGRectMake | David Robillard | 1 | -1/+1 | |
Using CGRectMake here apparently doesn't work on older versions of MacOS. | |||||
2020-04-03 | Remove glew.h compatibility header | David Robillard | 1 | -31/+0 | |
Not sure what the thinking behind this was, glew.h isn't available on MacOS anyway. | |||||
2020-04-02 | Make puglGetVisible() take a const view argument | David Robillard | 2 | -2/+2 | |
2020-04-02 | Make puglSetString() safe to call with equal source and destination | David Robillard | 1 | -3/+5 | |
This is sometimes used by puglSetWindowTitle() using the existing title. | |||||
2020-04-02 | Add missing include | David Robillard | 1 | -0/+1 | |
2020-04-01 | Add types and documentation for all event types | David Robillard | 1 | -6/+72 | |
These are not necessary since they are just aliases for PuglEventAny, but provide a place to put the documentation, and can make code clearer where a specific event type is known. | |||||
2020-04-01 | Consistently use "view" terminology in API | David Robillard | 7 | -30/+41 | |
The sloppy use of "window" caused quite a bit of confusion, since views only correspond to top-level windows in some cases, and on MacOS, a non-top-level view is not a "window" at all. | |||||
2020-04-01 | Cleanup: Fix mismatched file comments | David Robillard | 1 | -1/+1 | |
2020-04-01 | Add PUGL_BAD_PARAMETER status | David Robillard | 2 | -1/+4 | |
2020-04-01 | Replace puglShowWindow() with puglRealize() | David Robillard | 4 | -44/+81 | |
2020-03-31 | Mac: Fix various warnings | David Robillard | 3 | -25/+30 | |
2020-03-31 | X11: Don't wait for timeouts less than a millisecond | David Robillard | 1 | -1/+1 | |
2020-03-28 | Tolerate puglPostRedisplay() in configure handlers | David Robillard | 1 | -10/+11 | |
2020-03-18 | Reorder event struct definitions to match PuglEventType | David Robillard | 1 | -27/+27 | |
2020-03-17 | Use a more logical order for PuglEventType | David Robillard | 1 | -5/+5 | |
2020-03-17 | Use clearer names for pointer events | David Robillard | 4 | -25/+34 | |
These old "notify" names are a smell from X11 which is a bit strange and inconsistent here, since nearly everything is a "notification" of sorts. I think the new names here are much more clear since they are consistent with the keyboard focus events. | |||||
2020-03-16 | Cleanup: Add missing includes | David Robillard | 1 | -0/+2 | |
2020-03-16 | Simplify puglRequestAttention() | David Robillard | 3 | -37/+7 | |
Now that timers are exposed, applications can repeatedly nag for attention themselves if they really want to. | |||||
2020-03-16 | Add timer events | David Robillard | 6 | -2/+276 | |
2020-03-16 | Cleanup: Remove unused include | David Robillard | 1 | -1/+0 | |
2020-03-16 | MacOS: Fix semantics of puglUpdate() with negative timeout | David Robillard | 1 | -0/+6 | |
2020-03-16 | MacOS: Move initial configure to just before map | David Robillard | 1 | -11/+10 | |
2020-03-16 | X11: Dispatch a configure event before the map for child windows | David Robillard | 1 | -0/+9 | |
It does not seem to be guaranteed that child windows receive a ConfigureNotify at all. Work around this by explicitly dispatching one first, so child views receive events consistently with how top level views do. | |||||
2020-03-16 | Separate cached configuration from frame | David Robillard | 2 | -6/+3 | |
This was a bad idea and a never-ending source of problems. The frame represents what the "current" frame is from a Pugl perspective, but with the asynchronicity of X11 and other issues this cant be used to filter configure events. Instead, simply cache the last configure event that was sent and compare with that. | |||||
2020-03-16 | Factor out dispatching configure events in the drawing context | David Robillard | 4 | -9/+25 | |
The updates here need to happen whenever a configure is dispatched, even outside puglDispatchEvent(). This removes the last remaining direct calls to the event callback so the common implementation can always do the right thing. | |||||
2020-03-16 | Only send update events when the view is visible | David Robillard | 3 | -3/+11 | |
2020-03-15 | Cleanup: Fix documentation links | David Robillard | 1 | -2/+2 | |
2020-03-15 | Unify event loop functions as puglUpdate() | David Robillard | 5 | -71/+151 | |
The previous separation between polling and dispatching was a lie, especially on MacOS where it is impossible to only poll for events without dispatching anything. Providing such an API is misleading, and problematic in various other ways. So, merge them into a single puglUpdate() function which can do the right thing on all platforms. This also adds the behaviour of actually processing all events in the given time interval, which is almost always what clients actually want to do when using a positive timeout (naively doing this before caused terrible input lag). | |||||
2020-03-15 | X11: Factor out flushing pending exposures | David Robillard | 1 | -16/+22 | |
2020-03-15 | Add type and flags to world | David Robillard | 6 | -8/+42 | |
Unfortunately this is an API break, but there's no reasonable way to deprecate the old function and this is required for things to work correctly. The type will be used in following commits to tick the main loop and dispatch events correctly for either case. | |||||
2020-03-15 | Cleanup: Add puglDispatchSimpleEvent() internal utility | David Robillard | 5 | -14/+21 | |
2020-03-15 | Cleanup: Fix misleading variable name | David Robillard | 1 | -2/+2 | |
2020-03-15 | Cleanup: Remove redundant PUGL_API declarations | David Robillard | 3 | -3/+3 | |
2020-03-15 | X11: Simplify implementation slightly | David Robillard | 1 | -6/+2 | |
These continue statements are not necessary since the events will not be translated anyway. | |||||
2020-03-15 | X11: Read from server if necessary in puglDispatchEvents() | David Robillard | 1 | -3/+3 | |
This reduces input lag. The previous approach was an over-optimization: what's important here is to keep the iteration bounded by not continually flushing, but we do want to read everything to be as up to date as possible. | |||||
2020-03-15 | X11: Don't use CAIRO_OPERATOR_SOURCE for blitting | David Robillard | 1 | -1/+0 | |
This causes artifacts in some scenarios. Although CAIRO_OPERATOR_SOURCE should be faster, the default operator works better in the absence of real double buffering. | |||||
2020-03-15 | X11: Only enter drawing context if an expose is pending | David Robillard | 1 | -2/+2 | |
2020-03-15 | X11: Update frame immediately on ConfigureNotify | David Robillard | 1 | -0/+4 | |
This fixes some jittery display issues since the frame can be used before the deferred configure actually gets dispatched. | |||||
2020-03-15 | X11: Factor out selection event handling | David Robillard | 1 | -40/+69 | |
2020-03-15 | Move configure shortcut and frame update to common implementation | David Robillard | 3 | -36/+23 | |
2020-03-15 | Remove backend resize method | David Robillard | 12 | -43/+6 | |
2020-03-15 | Windows: Use ephemeral Cairo surface for drawing | David Robillard | 1 | -38/+37 | |
2020-03-15 | Mac: Remove unnecessary resize handler in GL backend | David Robillard | 1 | -12/+2 | |
2020-03-14 | Add logging API | David Robillard | 7 | -18/+104 | |
2020-03-13 | Strengthen warnings | David Robillard | 2 | -10/+12 | |
2020-03-09 | Deprecate puglEnterContext() and puglLeaveContext() | David Robillard | 2 | -29/+37 | |
These are prone to abuse, and have caused confusion with people who try to use them like in other libraries that support explicit drawing in the main loop. The drawing parameter was also wrong, and these were already just compatibility veneers since the internal context API needs to be more expressive. So, now that PUGL_CREATE and PUGL_DESTROY exist, they can be deprecated to force clients to draw only at the correct time. |