aboutsummaryrefslogtreecommitdiffstats
AgeCommit message (Collapse)AuthorFilesLines
2020-03-01Fix arrow style in API documentationDavid Robillard1-2/+8
Hiding this broke the indentation for nested lists. This reverts to something like the default Doxygen style.
2020-02-18Cairo Test: Preserve clip rectangle for background fillDavid Robillard1-2/+1
2020-02-18X11: Use ephemeral Cairo surfaces for drawingDavid Robillard1-51/+66
This uses a similar approach to Gtk, where surfaces are created every draw call. It is a bit slower for a single application in most cases, but uses less memory when drawing isn't happening and plays more nicely with other views or applications.
2020-02-18Pass expose event to backend enter and leave functionsDavid Robillard11-66/+65
This is needed for clipping. Unfortunately, the puglEnterContext() and puglLeaveContext() API was not suitable for this, but this shouldn't matter in user code because it is only used for setup, and is slated for removal anyway. Instead, just call the backend functions directly in the implementation.
2020-02-18Remove immediate dispatch of exposed rectsDavid Robillard1-30/+2
This was a hack to support only exposing the rects that were explicitly exposed with puglPostRedisplayRect(), but it caused flaky drawing issues because it circumvented the deferral of exposure until the end of the loop. Instead, simply expand the pending expose to be dispatched later as usual. This means that only the union will be exposed in the end, so more area might be drawn than necessary, but this is probably good enough. If not, we will have to maintain a set of rects and be more clever about combining them.
2020-02-18X11: Use a single Cairo contextDavid Robillard1-22/+20
Cairo has a built-in cache that makes this fast enough to not matter, and it removes some state which is always good.
2020-02-18Fix recursive context entryDavid Robillard1-2/+0
2020-02-15Set waflib as vendored to fix Github language detectionDavid Robillard1-0/+1
2020-02-11X11: Fix XMoveResizeWindow error checkZoë Sparks2-3/+4
In Xlib, a return value of 0 indicates an error (see "Xlib - C Language X Interface", Ch. 1, "Errors"). XMoveResizeWindow accordingly returns 1 on success, but the error check in puglSetFrame used the usual C convention, and so was backwards.
2020-02-02Add tags to Windows builds to exclude Gitlab shared runnersDavid Robillard1-8/+4
Gitlab now has shared Windows runners, which is cool, but unfortunately they don't have Python, so they are useless here and will cause failed builds if they pick up the job.
2020-02-02Only resize backend when necessaryDavid Robillard2-17/+26
This avoids resizing the backend when the window is only moved, which fixes flicker with Cairo where resizing is expensive.
2020-02-02X11: Dispatch exposures from event callbacks in the same iterationDavid Robillard2-1/+16
2020-02-02X11: Avoid calculating expose rect if view is not visibleDavid Robillard1-10/+10
2020-02-02X11: Factor out adding a pending exposureDavid Robillard1-14/+22
2019-12-11GL3 Test: Use instancingDavid Robillard3-60/+88
2019-12-11GL3 Test: Update bindings to OpenGL 3.3David Robillard2-8/+324
2019-12-11GL3 Test: Spread out rectangles moreDavid Robillard1-7/+11
2019-12-11GL3 Test: Use core profile in shadersDavid Robillard2-2/+2
2019-12-11GL3 Test: Use camel case for shader variablesDavid Robillard1-11/+11
2019-12-11GL3 Test: Calculate border color on the GPUDavid Robillard2-22/+14
2019-12-11GL3 Test: Move shaders to separate filesDavid Robillard4-43/+94
2019-12-09Mac: Fix missing NSWindowStyleMask type on 10.11Hanspeter Portner2-3/+9
2019-12-09Fix use of incorrect union elementDavid Robillard1-4/+4
2019-12-09Cleanup: Fix whitespaceDavid Robillard1-1/+1
2019-11-21Cleanup: Fix flake8 warningsDavid Robillard1-7/+7
2019-11-21Check explicitly for GLX and only link against the necessary libraryDavid Robillard1-2/+10
It turns out that on some systems GLX is implemented in libGLX (which brings in only libGLdispatch), while on others it is implemented in libGL.
2019-11-21Rename "backend" headersDavid Robillard20-165/+234
Working on Vulkan clarified what has always been slightly smelly about the design and organization here: not everything that is API specific is really in a "backend" (a PuglBackend). The concrete example is puglGetProcAddress(), which only makes sense for GL and is actually implemented in the "backend" files. Arguably puglGetContext() is also such a thing. So, rename the headers so they can be the place where API-specific things go in general, which happens to include a backend most of the time. The stub is a bit of an exception to this, but whatever. The includes look tidier this way. In place of the old headers are compatibility stubs that just emit a warning and include the new version, which will be maintained for a while.
2019-11-21Use consistent include guardsDavid Robillard5-15/+15
2019-11-21X11: Remove libGL dependency from GL backendDavid Robillard2-4/+1
This was the only use of actual GL in the backend, and I'm relatively sure this flush is pointless anyway. If, for some bizarre reason, anyone really wants to run without double buffers, they are always free to flush themselves.
2019-11-21Clean up configuration codeDavid Robillard1-48/+58
2019-11-20Mac: Move puglGetProcAddress() to mac_gl.mDavid Robillard2-17/+17
2019-11-20Mac: Fix unused parameter warningDavid Robillard1-1/+1
2019-11-18Cleanup: Remove unnecessary includesDavid Robillard4-6/+0
2019-11-18Move puglGetProcAddress to pugl_gl_backend.hDavid Robillard2-11/+11
This establishes a general pattern for backend-specific APIs, so that pugl.h doesn't become a mess. The name of these headers, and the definition of "backend", is a little fuzzy here, but it was before in reality anyway.
2019-11-18Add test utility function for logging errorsDavid Robillard5-16/+23
2019-11-18Ignore clang-tidy warnings in system headersDavid Robillard1-1/+1
2019-11-18Add error checking option to test programsDavid Robillard3-1/+8
2019-11-17X11: Don't select on server socket if events are already queuedosch1-1/+3
2019-11-17X11: Support pasting (almost) arbitrarily long textDavid Robillard1-1/+1
This is a bit ridiculous, but using a "probably big enough" value seems to be what everyone does. This particular value is the one used by Gtk.
2019-11-17X11: Don't send expose events to invisible windowsDavid Robillard1-1/+3
2019-11-17X11: Track visibility via system eventsDavid Robillard1-3/+8
This ensure that visibility is properly updated when a window is made visible or invisible by the user or window system. Unfortunately it does not seem to propagate automatically to child windows, but this is at least better than the previous situation, and good enough for most cases.
2019-11-17Use typedefs for flag typesDavid Robillard1-72/+82
This doesn't change anything about type-safety (C can't do that), but at least makes the intent of things clearer.
2019-11-17Cleanup: Use consistent formatting for field documentationDavid Robillard1-141/+141
2019-11-09Add clang-tidy file and update lint targetDavid Robillard2-20/+27
2019-11-09Fix various clang-tidy warningsDavid Robillard5-18/+20
2019-11-09Make pugl_gl3_test deterministicDavid Robillard1-16/+15
2019-11-09Add puglStrerror() and improve test program error reportingDavid Robillard5-8/+41
2019-11-03Remove redisplay flag and add puglPostRedisplayRect()David Robillard8-36/+84
2019-11-03X11: Only merge expose events if they intersectDavid Robillard1-13/+45
Towards actual fine-grained exposure. This uses the previous behaviour if expose events intersect to reduce redundant drawing, but dispatches expose events immediately (after flushing and pending configure events) if they don't to support partial redraws.
2019-11-03Add puglGetNativeWorld()David Robillard4-0/+28
generated by cgit v1.2.1 (git 2.18.0) at 2024-11-23 15:34:57 +0000