Skip to content

slack: the comment on _PUBLISHED_NOT_SERVED dates only the entries the 2026-10-20 run measured - #1445

Open
tony1233200 wants to merge 1 commit into
brekkylab:mainfrom
tony1233200:slack-comment-dates-1428
Open

tony1233200 wants to merge 1 commit into
brekkylab:mainfrom
tony1233200:slack-comment-dates-1428

Conversation

@tony1233200

Copy link
Copy Markdown
Contributor

Closes #1428.

What changed

The comment on _PUBLISHED_NOT_SERVED in backlot/routers/slack.py now dates only the entries with no note of their own to the run of 2026-10-20, and says an entry the baseline gained later carries its own dated note there, so the comment stays true when the next scheduled run adds one. The admin.conversations.kick note in backlot/fidelity/baseline/slack.json now also records the GET measurement. The docstring of test_slack_answers_each_published_method_no_route_serves points at the comment for the run and at each later entry's note for its own date.

Why this is what the real API does

The baseline's own notes date 19 entries after 2026-10-20: 3 on 2026-10-28 (admin.users.deactivate, admin.users.access.pause, admin.users.access.resume), 11 on 2026-10-29 (the four admin.slackbot.skills.* and seven slackbot.skills.* methods), 1 on 2026-11-05 (admin.conversations.kick), 2 on 2026-11-10 (admin.analytics.slackbot.skills, admin.analytics.slackbot.tools), and 2 on 2026-11-19 (admin.analytics.members.activity, admin.analytics.members.metadata). The GET measurement added to the admin.conversations.kick note is the one reported in the issue: measured against slack.com on 2026-11-19, GET /api/admin.conversations.kick?limit=1 answers HTTP 200 {"ok":false,"error":"not_authed"} with no token and {"ok":false,"error":"invalid_auth"} with a bogus one, which is what Backlot answers and what the test asserts. No served answer changes; only the prose now matches the measurements already recorded.

Verification

  • Tests — pytest tests/test_slack.py on a .[dev] install: 422 passed.

  • Optional surfaces that ran rather than skipped — none beyond .[dev]: the optional extras were not installed, so the s3/sdk/mcp/llamaindex/integrations surfaces skipped, as the template notes.

  • Lint — ruff check . && ruff format --check . on the changed files: all checks passed, both files already formatted. python scripts/prose_check.py: no findings.

  • A test fails without this change

  • A new endpoint serving corpus content is ACL-scoped, proved for both an admin and a scoped token

  • Comments state what was measured, not the history of the fix

@nuriyoo nuriyoo left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks for this. On this head the baseline's notes date the 19 entries the description lists, and each of those notes records a GET and a POST with no token and with a bogus one, the admin.conversations.kick note now included, so the test's claim holds for them. Two things in the prose:

  1. The comment on _PUBLISHED_NOT_SERVED (inline). "An entry the baseline gained later carries its own dated note there, which is the date its measurement was made on, so this stays true when the next scheduled run adds one" describes the comment and when entries arrived rather than what was measured, and "which" reads as attaching to the note. The suggestion keeps the measurement sentence as it was on main and ends it with the two dates: 2026-10-20 for an entry with no note, and the note's date for one with a note.
  2. The test docstring (inline). With the comment dating both kinds of entry, the pointer can go back to one clause; as written it repeats the comment's split.

With the suggestions applied, tests/test_slack.py passes.

Comment thread backlot/routers/slack.py
Comment on lines +983 to +991
#: reference index lists that no route here serves, by name. The entries with no note of their
#: own were measured on 2026-10-20 with no token and with a bogus one, by `GET` and by `POST`,
#: each answering the credential's refusal (`not_authed`, `invalid_auth`) but those
#: :data:`_OTHER_ANSWERS` and :data:`_ANSWERED_BY_ARGUMENTS` hold. An entry the baseline gained
#: later carries its own dated note there, which is the date its measurement was made on, so
#: this stays true when the next scheduled run adds one. Ten that Slack's archived OpenAPI
#: document lists and the index does not, `oauth.token`, three `workflows` ones and six
#: `apps.permissions` ones, real answered `unknown_method` for, as for a name it never had,
#: whatever the credential, one that resolves included.

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Item 1: one sentence for the measurement, and the date beside each kind of entry.

Suggested change
#: reference index lists that no route here serves, by name. The entries with no note of their
#: own were measured on 2026-10-20 with no token and with a bogus one, by `GET` and by `POST`,
#: each answering the credential's refusal (`not_authed`, `invalid_auth`) but those
#: :data:`_OTHER_ANSWERS` and :data:`_ANSWERED_BY_ARGUMENTS` hold. An entry the baseline gained
#: later carries its own dated note there, which is the date its measurement was made on, so
#: this stays true when the next scheduled run adds one. Ten that Slack's archived OpenAPI
#: document lists and the index does not, `oauth.token`, three `workflows` ones and six
#: `apps.permissions` ones, real answered `unknown_method` for, as for a name it never had,
#: whatever the credential, one that resolves included.
#: reference index lists that no route here serves, by name. Measured with no token and with a
#: bogus one, by `GET` and by `POST`, each answered the credential's refusal (`not_authed`,
#: `invalid_auth`) but those :data:`_OTHER_ANSWERS` and :data:`_ANSWERED_BY_ARGUMENTS` hold: on
#: 2026-10-20 for an entry with no note, and on the date its note gives for an entry with one.
#: Ten that Slack's archived OpenAPI document lists and the index does not, `oauth.token`, three
#: `workflows` ones and six `apps.permissions` ones, real answered `unknown_method` for, as for a
#: name it never had, whatever the credential, one that resolves included.

Comment thread tests/test_slack.py
Comment on lines 4187 to +4189
those ``slack._ANSWERED_BY_ARGUMENTS`` leaves to the router (``slack._PUBLISHED_NOT_SERVED``
dates the measurement): with no credential and a bogus one, for a `GET` and a `POST`."""
dates the run that measured the entries with no note of their own, and a later entry's note
dates its own): with no credential and a bogus one, for a `GET` and a `POST`."""

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Item 2: the comment dates both kinds of entry, so the pointer can say so in one clause.

Suggested change
those ``slack._ANSWERED_BY_ARGUMENTS`` leaves to the router (``slack._PUBLISHED_NOT_SERVED``
dates the measurement): with no credential and a bogus one, for a `GET` and a `POST`."""
dates the run that measured the entries with no note of their own, and a later entry's note
dates its own): with no credential and a bogus one, for a `GET` and a `POST`."""
those ``slack._ANSWERED_BY_ARGUMENTS`` leaves to the router (``slack._PUBLISHED_NOT_SERVED``
dates the measurements): with no credential and a bogus one, for a `GET` and a `POST`."""

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

slack: the comment on _PUBLISHED_NOT_SERVED dates every listed method's measurement to 2026-10-20, where 17 entries were measured on later dates

2 participants