Audit codebase and implement comprehensive improvements

[Laviesss]

This change conducts an overall audit across all components in the monorepo, fixing bugs in bounding box calculations, process execution, thread management, shell scripts, and build configurations while adding unit tests for the core logic.

---

PR created automatically by Jules for task 10188766728484556702 started by @Laviesss

[google-labs-jules]

👋 Jules, reporting for duty! I'm here to lend a hand with this pull request.

When you start a review, I'll add a 👀 emoji to each comment to let you know I've read it. I'll focus on feedback directed at me and will do my best to stay out of conversations between you and other bots or reviewers to keep the noise down.

I'll push a commit with your requested changes shortly after. Please note there might be a delay between these steps, but rest assured I'm on the job!

For more direct control, you can switch me to Reactive Mode. When this mode is on, I will only act on comments where you specifically mention me with @jules. You can find this option in the Pull Request section of your global Jules UI settings. You can always switch back!

New to Jules? Learn more at jules.google/docs.

---

For security, I will only act on instructions from the user who triggered this task.

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: f1d2614e61

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Keep circle hit tests in screen space

When a user configures a type=1 circular block or exclude zone on a non-square display, the overlay still renders it with radius x2 * width, but this shared contains check compares normalized x/y deltas directly, which makes the effective vertical radius x2 * height. Because both the Xposed hook and overlay services now call region.contains, tall phones will block or whitelist touches well outside the drawn circle; keep the circle hit test in pixel/screen space or pass the display dimensions into this calculation.

Useful? React with 👍 / 👎.

[Laviesss]

Review complete — summary in the review body; two line comments with suggested fixes. Net positive: mergeable after addressing the two comments (or filing follow-up issues) and adding a CHANGELOG note for the circle-geometry change.

[Laviesss]

The timeout here is effectively dead code: reader.readText() on the line above blocks until the process closes stdout (or EOF), so process.waitFor(10, SECONDS) is only reached after the process already exited. A wedged adb (device offline, server hung) still blocks the PC tool indefinitely — the exact case this timeout is meant to protect against.

Suggested fix — bound the read instead of the wait:

val output = try {
    // read up to the timeout, then kill
    val future = CompletableFuture.supplyAsync { reader.readText() }
    future.get(10, TimeUnit.SECONDS)
} catch (_: TimeoutException) {
    process.destroyForcibly()
    ""
} catch (e: Exception) {
    process.destroyForcibly()
    println("Error running command ${cmd.firstOrNull()}: ${e.message}")
    ""
}

(executor must be a daemon thread so the pool doesn't outlive the JVM).

[Laviesss]

FOUND is set inside the get_regions | while ... pipeline, which runs in a subshell — the assignment never propagates to the parent shell. Combined with the removed [ "$FOUND" -eq 0 ] guard (old line ~1297), remove_region with a nonexistent ID now silently succeeds: the region isn't found, nothing is removed, and mv still rewrites the config file with no error reported.

Also note the echo "Removed region..." fires before the file is actually rewritten — if the region exists this is fine, but the pre-delete message plus the missing error path is the regression here.

Suggested fix:

FOUND=0
REGIONS=$(get_regions)          # capture once, no pipe subshell
printf '%s\n' "$REGIONS" | while IFS= read -r line; do
    [ -z "$line" ] && continue
    COUNT=$((COUNT + 1))
    if [ "$COUNT" -eq "$target_id" ]; then
        FOUND=1
        echo "Removed region [$target_id]: $line"
    else
        echo "$line" >> "$TEMP_FILE"
    fi
done
if [ "$FOUND" -eq 0 ]; then
    rm -f "$TEMP_FILE"
    echo "Error: Region $target_id not found"
    return 1
fi

(The same subshell pattern also makes COUNT in show_status dead in the parent — harmless there since TOTAL is computed separately, but worth knowing.)

[google-labs-jules]

Good catch! I have refactored remove_region in module/system/bin/inputblocker to avoid subshell pipeline execution when processing regions line-by-line. FOUND is now tracked in the main shell environment, and if FOUND is 0, an error message is printed and execution terminates with exit status 1 without modifying the config file.