fix(windows): add CREATE_BREAKAWAY_FROM_JOB to windows_hide_flags()

[ms-alan]

Closes #55604

Summary

windows_hide_flags() was returning only _CREATE_NO_WINDOW, missing _CREATE_BREAKAWAY_FROM_JOB which is already used by windows_detach_flags() for the same reason: Electron and Tauri desktop apps wrap their subprocesses in job objects, and without the breakaway flag a child process (e.g. cmd.exe windows flash) can be killed when the parent Electron/Tauri process exits.

Root cause

CREATE_BREAKAWAY_FROM_JOB (0x01000000) already exists at line 110 of _subprocess_compat.py and is used by windows_detach_flags(). It was simply missing from windows_hide_flags().

Fix

Added _CREATE_BREAKAWAY_FROM_JOB to windows_hide_flags():

Note: DETACHED_PROCESS is intentionally not added — windows_hide_flags() is for short-lived synchronous children where stdout capture is needed, and DETACHED_PROCESS would sever stdio.

Related

• [Bug]: Subprocess calls flash black cmd.exe windows on Windows #53053 — "Subprocess calls flash black cmd.exe windows on Windows"
• Windows Desktop GUI: CREATE_NO_WINDOW insufficient — cmd windows flash on every terminal() call #53273 — "Windows Desktop GUI: CREATE_NO_WINDOW insufficient"
• [Bug]: The Hermes desktop client on Windows keeps flickering black command prompt windows nonstop, making the program completely inoperable. This is a critical blocking bug! #53342 — "Hermes desktop client keeps flickering black command prompt windows"

[alt-glitch]

This was generated by AI during triage.

Duplicate of #55613 — identical core change (return _CREATE_NO_WINDOW | _CREATE_BREAKAWAY_FROM_JOB in windows_hide_flags(), hermes_cli/_subprocess_compat.py) fixing the same issue #55604. #55613 is canonical: triaged earlier and includes a regression test (test_windows_hide_flags_includes_breakaway_from_job); this PR is the one-line change with only a longer comment, no test. Related to the other identical OPEN twins #55667 and #55692, and to issue #55604.

[teknium1]

Thanks for tracing the existing breakaway constant and keeping DETACHED_PROCESS out of the synchronous path.

Problems

• hermes_cli/_subprocess_compat.py:138-144 documents that CREATE_BREAKAWAY_FROM_JOB can cause ERROR_ACCESS_DENIED / PermissionError when the parent job rejects breakaway. This change adds that bit to every windows_hide_flags() consumer, but generic callers such as tools/environments/base.py:145-153 pass the result directly to subprocess.Popen without the documented retry-without-breakaway path.
• The PR has no regression test for the changed helper behavior. The linked canonical duplicate, #55613, includes a helper-level test.

Suggested changes

• Scope breakaway to spawns that actually need to outlive the parent job, or provide and test a safe fallback for hide-only callers before changing the shared helper.
• Add a focused simulated-Windows regression test for the final behavior.

Automated hermes-sweeper review.

[teknium1]

CREATE_BREAKAWAY_FROM_JOB can make CreateProcess fail when the parent job disallows breakaway; current main documents this at hermes_cli/_subprocess_compat.py:138-144. Because this helper is used by synchronous callers without an OSError retry (for example tools/environments/base.py:145-153), please avoid applying this bit globally unless the fallback is designed and covered.

[teknium1]

Closing after the class-level fix in PR #70205 (merged as 0dbf639) — the premise here doesn't hold against Win32 semantics, and the underlying symptom is addressed at a different layer.

Two problems with adding CREATE_BREAKAWAY_FROM_JOB to windows_hide_flags():

1. Job objects don't affect console/window creation. CREATE_NO_WINDOW is not "silently ignored inside a job object" — jobs govern process lifetime and limits, not console allocation. The flag that actually neutralizes CREATE_NO_WINDOW is DETACHED_PROCESS (documented in the Process Creation Flags spec), and the real cause of the Electron/desktop flashes was the parent being console-less (GUI-subsystem pythonw.exe / DETACHED daemons), so every console-subsystem child had to allocate its own visible console. That's fixed at the root: commit aa2ae36 launches the desktop backend as hidden-console python, and fix(windows): hidden-console daemons — extend the parent-console flash fix to every detached spawn path #70205 extends the same to all detached daemon paths.

2. Breakaway on every short-lived spawn is harmful. windows_hide_flags() feeds ~40 synchronous capture-output call sites. In a job that doesn't set JOB_OBJECT_LIMIT_BREAKAWAY_OK, CREATE_BREAKAWAY_FROM_JOB makes CreateProcess fail with ERROR_ACCESS_DENIED — and none of those sites have a retry, so probes and helpers would start crashing in restrictive Windows Terminal/container/kiosk configs. Short-lived children being torn down with the parent's job is also the desired behavior for helpers.

If a flash persists at a specific spawn site on current main, that's worth a fresh issue with the process name from the flash — it would point at a leg the parent-console fix doesn't cover. Thanks for digging into the Windows spawn behavior.