Skip to content

v0.2.0 — progress timeline + stall detection

Choose a tag to compare

View all tags
@leeguooooo leeguooooo released this 09 Jun 01:50
· 13 commits to main since this release
v0.2.0
e7b26b4

Generation can take 2-3 min for detailed images, and slow runs used to look frozen or crash with a raw traceback. This release makes the wait legible and robust.

Fixes

  • Premature read timeout: urlopen()'s 30s connect timeout was silently governing every read, so a >30s quiet gap mid-stream (measured ~66s of silence during generating) killed a live generation with a raw TimeoutError traceback. The per-read idle window is now configurable and bounded by the total budget.
  • Network errors (TimeoutError/ConnectionError/URLError) now surface as a clean one-line message, never a traceback.

Features

  • Progress timeline on stderr, each line stamped with elapsed seconds: queued -> generating -> receiving image -> saved. Decoupled from --quiet, so agents capturing stdout via $() still see progress on stderr.
  • --stall-timeout (default 120s): detects a hung backend and reports which phase it stalled in.
  • --no-progress to silence the timeline.
  • Default --timeout 180 -> 300s (covers 2-3 min images).

Update

npx skills add leeguooooo/chatgpt-imagegen -g
Assets 2
Loading