Skip to content

Android app icon generation - #5693

Open
pythonpoet wants to merge 22 commits into
DioxusLabs:mainfrom
pythonpoet:main
Open

Android app icon generation#5693
pythonpoet wants to merge 22 commits into
DioxusLabs:mainfrom
pythonpoet:main

Conversation

@pythonpoet

@pythonpoet pythonpoet commented Jul 21, 2026

Copy link
Copy Markdown

Android app icon generation still doesn't work.

This code is from @fontlos from PR #4063

Don't know why it didnt got merged already, however, I can confirm its working and android devs would appreciate this feature.

@pythonpoet
pythonpoet requested a review from a team as a code owner July 21, 2026 13:02
@nicoburns nicoburns changed the title Androind app icon generation (fontlos) Android app icon generation (fontlos) Jul 23, 2026
@nicoburns nicoburns changed the title Android app icon generation (fontlos) Android app icon generation Jul 23, 2026
hardliner66 and others added 17 commits August 8, 2026 18:25
…Labs#5704)

`set_server_url` stored into a `OnceLock` and unwrapped the result, so a
second call panicked. That makes the base URL a permanent property of the
process, which a client that talks to more than one backend cannot work
with: the backend has to be chosen before `launch`, since `launch` claims
the slot itself when it is still empty, and it can never change afterwards.

Store it in an `RwLock` instead so the most recent value wins. `fetch_inner`
already reads the URL as each request is built, so a change applies to the
next server function call with no restart. `get_server_url` still returns
`""` when unset, so `launch`'s "set the devserver default if empty" check
is unaffected.

This also removes a latent panic in `launch`: the native block
(`cfg(any(desktop, mobile, native))`) and the web base-path block both call
`set_server_url`, and the second does so unconditionally, so a binary
compiled with both feature sets panicked inside the framework whenever
`dioxus_cli_config::base_path()` returned `Some`.

The neighbouring `REQUEST_HEADERS` is already a `Mutex` with a re-settable
`set_request_headers`, so this brings the two client-configuration globals
into line.
* fix(cli): prevent duplicate SSG output on rebuild

* remove tests for should_build_sequentially (just an or operation)

* fix formatting

---------

Co-authored-by: sanmaxdev <sanmaxdev@users.noreply.github.com>
Co-authored-by: Evan Almloff <evanalmloff@gmail.com>
…s#5619)

* Support copying resource dirs for the bundler

Recursively copies files and directories using a custom method `copy_entries`.

Fix DioxusLabs#5504

* Ensure destination directory is created in copy_resources

This is necessary to avoid the `NotFound` error.

* use the existing copy_dir_recursive

---------

Co-authored-by: Evan Almloff <evanalmloff@gmail.com>
Co-authored-by: cyphercodes <cyphercodes@users.noreply.github.com>
…Labs#5641)

The rust_callback native function gated compilation behind
target_pointer_width = "64" with a compile_error!, which broke building
dioxus for armv7-linux-androideabi (and any other 32-bit Android target)
starting with DioxusLabs#4842.

The handler pointer is transmitted across the JNI boundary as two jlong
halves. The existing reconstruction already produces a pointer-width
usize: on 64-bit both halves are significant, and on 32-bit the pointer
fits in the low half (high half zero) so the as usize cast drops the
unused high bits and yields the original pointer. Removing the gate lets
the same code compile and run correctly on both widths.

Closes DioxusLabs#5637
…xusLabs#5711)

Co-authored-by: Staging-Devin AI <166158716+staging-devin-ai-integration[bot]@users.noreply.github.com>
…kage name (DioxusLabs#5719)

Signed-off-by: Nico Burns <nico@nicoburns.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.