Shell Completion

Shell Completion

English · 繁體中文

taiga completion bash
taiga completion zsh
taiga completion fish
taiga completion powershell

Write the output into the completion directory for your shell to enable it; exact paths are in INSTALL.md.

Dynamic candidates

Completion covers more than subcommands and flags: it queries Taiga for real data — projects, work item refs, statuses, members, sprints, and issue metadata. Queries have a 2 second timeout and fall back quietly to static completion when they expire, so your shell never hangs.

The metadata cache

Successful lookups are cached with 0600 permissions:

• Entries under 5 minutes old are considered fresh and used directly.
• Older entries trigger a fresh query. If that query fails, data up to 24 hours old is used instead (stale-on-error), keeping completion usable offline or during a server outage.
• Entries older than 24 hours are discarded and pruned on the next write.

Cache keys hash the profile, API URL, project, and data kind together, so different sites and projects never contaminate each other. No token is stored in the cache, and a corrupted cache file only disables completion — it never affects your shell or any other command.