Rongsong Shen
d26ff8870d
The buglog was one 27K-line JSON file whose fields covered only symptom,
root cause and fix, as free text: 2% of entries named the code version a
bug was found on, 12% the build profile, 36% a way to reproduce, and 58% of
the file was OpenWolf hook auto-detections ("Missing guard clause",
"Function not marked async" on .md files).
- .claude/skills/buglog/bug-record.md: the record standard. Eight
sections (Summary as the title, Description, Environment, Reproduction;
Root cause, Fix, Verification, References once fixed), each with the
content it must contain -- e.g. Environment needs the OS, the libc the
failing binary loads (from its ELF interpreter: glibc 2.42 in the nix
shell vs 2.40 on the host), toolchain, build profile and commit.
- .wolf/bugs/<id>.md: 500 records migrated from buglog.json; every field
verified to round-trip. They carry legacy: true and an explicit "not
recorded" where the old log had no such field; component derived from
file paths, severity from keywords (else unknown), fix_commits from
hashes in status_note/fix that git resolves (91 of 463 fixed).
- .wolf/buglog-archive.json: the 710 hook auto-detections, verbatim.
- .wolf/buglog.json is now an index generated from the records, keeping
the fields OpenWolf's hooks and `openwolf bug search` read.
- scripts/buglog.py rewritten: new (template + Environment collected from
the machine and --binary), check, status (refuses a status the record
doesn't support; fixed needs Root cause, Fix, Verification and
--commit hashes git resolves), note, recur, set, show, list, triage,
env, index. 29 unittest cases in scripts/tests/test_buglog.py.
- scripts/git-hooks/pre-commit runs `check --staged`: incomplete staged
records, a legacy record hand-changed to fixed without the fix-time
sections, a new record claiming legacy: true, or a stale index refuse
the commit.
- scripts/gitea-sync.py reads the records: issue body = the record's
sections plus linked fix commits (origin remote, GITEA_COMMIT_REPO to
override), and a body hash so an issue's text is refreshed when the
record changes.
- bug-1866 rewritten to the full standard as the worked example.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
38 lines
3.3 KiB
Markdown
38 lines
3.3 KiB
Markdown
---
|
|
id: "bug-1792"
|
|
status: "fixed"
|
|
status_updated: "2026-09-24"
|
|
status_note: ""
|
|
component: "controller"
|
|
severity: "unknown"
|
|
timestamp: "2026-09-09T12:40:00.000Z"
|
|
last_seen: "2026-09-19T09:00:00Z"
|
|
occurrences: 1
|
|
files: ["src/controller/testsuite/test_test_cli.c"]
|
|
tags: ["libedit", "readline", "tab-completion", "pty", "ECHO", "test-harness", "xce-test-cli-ux", "task-2.3", "test_test_cli.c", "silent-input-mode"]
|
|
related: []
|
|
legacy: true
|
|
---
|
|
|
|
# bug-1792: test_test_cli.c tab-completion test (xce-test-cli-ux task 2.3): Tab keystroke produced zero effect…
|
|
|
|
## Description
|
|
test_test_cli.c tab-completion test (xce-test-cli-ux task 2.3): Tab keystroke produced zero effect (no bell, no redraw, no completion) when driven through the existing clitest_spawn() pty harness -- 'wo' + Tab + Enter submitted literally "wo" ("wo: No such command for xce test."), never "worker".
|
|
|
|
## Environment
|
|
_Not recorded (migrated from buglog.json)._
|
|
|
|
## Reproduction
|
|
_Not recorded (migrated from buglog.json)._
|
|
|
|
## Root cause
|
|
clitest_spawn() (task 1.7) deliberately clears kernel ECHO on the pty slave BEFORE forking, documented at the time as affecting ONLY the line discipline's own re-echo of injected bytes, 'not readline()/libedit's own prompt/redraw output, which stays intact'. An A/B probe (standalone C program linked against this project's exact pinned libedit, driven both ways via a raw openpty()+fork() harness, controlling for controlling-terminal presence too) showed that assumption does not hold for THIS libedit build: when the slave's ECHO bit already reads off at the moment readline() is invoked, libedit treats the session as a silent/non-interactive (password-style) input mode and suppresses ALL of its own live editing feedback, not just kernel echo -- Tab is accepted as a no-op, the line is only ever consumed whole on Enter, and no bell/redraw/candidate-list ever appears. With kernel ECHO left at its default-on state (libedit reconfigures the tty itself once its own readline() call takes over, taking on echo/redraw duties without any raw kernel double-echo reaching the captured stream), the identical Tab sequence produced the expected bell + in-place completion + ambiguous-match candidate listing. Controlling-terminal setup (setsid()/TIOCSCTTY) was also probed and made no observable difference -- ECHO state was the sole deciding factor.
|
|
|
|
## Fix
|
|
Refactored clitest_spawn() into clitest_spawn_ex(sess, path, argv, disable_echo) with the ECHO-clearing block made conditional on the new parameter; clitest_spawn() itself is preserved as a one-line wrapper calling clitest_spawn_ex(..., true), so every existing (output-only) caller is unaffected. The new xce-test-cli-ux task 2.3 tab-completion test calls clitest_spawn_ex(..., false) directly to keep kernel ECHO on, which restored full interactive completion behavior end-to-end (single-match auto-complete, ambiguous-match double-Tab listing, all verified via the real xce_test binary). Documented the finding in both clitest_spawn_ex()'s own comment and the file's top-of-file API summary so future tasks in this spec (3.5, 4.1-4.4) that add more interactive-feedback-dependent tests to this same file know to use clitest_spawn_ex(..., false) rather than rediscovering this the hard way.
|
|
|
|
## Verification
|
|
_Not recorded (migrated from buglog.json)._
|
|
|
|
## References
|
|
_Not recorded (migrated from buglog.json)._
|