Currently every user needs to manually adjust for system insets.
This is rather verbose and most don't need to deviate from this behavior.
To disable the automatic adjustment, use:
e.Insets = system.Insets{}
ctx := layout.NewContext(ops, e)
Signed-off-by: Egon Elbre <egonelbre@gmail.com>
Commit 94d242d broke the widget.Image's Scale field so
that it no longer had any effect on the actual size of
the displayed image. This commit fixes that, as well as
adding tests to confirm that the widget.Image type
scales appropriately with DPI changes and its own Scale
field.
Signed-off-by: Chris Waldon <christopher.waldon.dev@gmail.com>
Use dxc (DirectXShaderCompiler) for compiling, which is newer than fxc
and doesn't not fail compilation with dynamically uniform flows with
barriers.
Currently requires -directcompute to enable generating the shaders.
Signed-off-by: Egon Elbre <egonelbre@gmail.com>
That change makes possible to provide custom PKCS#12 (JKS/PFX) using `-signkey` and the password with `-signpass`.
By default the gogio will use the `debug.keystore`, if no key is provided.
Signed-off-by: Inkeliz <inkeliz@inkeliz.com>
Previously that patch, gogio unzip the `link.apk` (generated by AAPT2) to an temporary folder and then compress it again to a new `app.ap_` file.
For some unknown reason, that unzip-then-zip doesn't work. The resources are included but is corrupted in somehow. That PR aims to fix that by avoid the extraction to an temporary folder.
Signed-off-by: Inkeliz <inkeliz@inkeliz.com>
The Wayland protocol implicitly dup(2)s the pipe write end descriptor passed to
wl_data_offer_receive. As long as we also have an open descriptor for the write
end, the pipe will not close and signal the completion of the clipboard read.
This change explicitly and immediately closes our write descriptor. Before this
change, reading the Wayland clipboard worked with some delay because the Go
garbage collector closed the write end of the transfer pipe after some time.
Signed-off-by: Elias Naur <mail@eliasnaur.com>
We're about to change the last stage of the compute pipeline to only
accept images, not sampled textures. This change prepares materials
for pixel-aligned image copying by pre-rendering images to a texture,
applying transforms.
Signed-off-by: Elias Naur <mail@eliasnaur.com>
The webassembly port changed to use transparent black to allow content
below Gio to show through. However, the end-to-end tests expect a white
background.
Signed-off-by: Elias Naur <mail@eliasnaur.com>
EGL_PLATFORM=surfaceless allows rendering tests to complete in headless
mode. However, the headless chrome tests fail with that setting. Change
the linux builder to only set the variable when needed.
Enable verbose output of tests while here. We can keep an eye on skipped
tests that way.
Signed-off-by: Elias Naur <mail@eliasnaur.com>
A recent change to the native toolchain or system on the builder infrastructure
made the arguments for the apple cross compiler needed always, not just in
CGO_CFLAGS and CGO_LDFLAGS.
Remove them altogether and rely on a recent change to the giouiorg repository
for including them in the clang-ios and clang-macos wrappers.
Signed-off-by: Elias Naur <mail@eliasnaur.com>
The order of subtraction when calculating f.pos from value was wrong,
so setting a minimum value for a Float never really worked, although
min = 0 worked as intended which is why this probably went unnoticed.
Signed-off-by: vsariola <5684185+vsariola@users.noreply.github.com>
The README is deliberately light; we want users to use the gioui.org site for
resources, not the site that happens to host the project Git repository.
Signed-off-by: Elias Naur <mail@eliasnaur.com>
The iOS GL implementation doesn't optimize BufferSubData of the entire
buffer, leading to GPU stalls. The workaround is to explicitly clear the
buffer before replacing it.
Signed-off-by: Elias Naur <mail@eliasnaur.com>
Add support to Router so that the cursor can be changed with CursorNameOp without any mouse movement.
Enter and Leave events are also delivered as areas change.
Signed-off-by: pierre <pierre.curto@gmail.com>
Cursor and Enter/Leave events should follow the same rules, specifically:
when pointerInfo.pressed is true, Enter/Leave processing ignores
pointer.Areas not hit during the pointer.Press event.
Signed-off-by: pierre <pierre.curto@gmail.com>
Post a dedicated message upon Window.SetCursor calls.
Make sure that the cursor is only changed if the cursor is in the window.
Signed-off-by: pierre <pierre.curto@gmail.com>
Broken by gioui.org/commit/2dce8a0155ae192f54f74747d660b9f311e9e364,
found while debugging the dormant OpenGL backend for Windows.
Signed-off-by: Elias Naur <mail@eliasnaur.com>
If you created an Editor and immediately SetCaret, it panicked because
e.lines was nil and it looked at e.lines[0].
- Add e.makeValid at the top of SetCaret.
- Add a test case for this situation.
Signed-off-by: Larry Clapp <larry@theclapp.org>
Chris Waldon