Skip to content

fix: Bedework and Zimbra conformance — status line, tz links, task start - #869

Merged
tobixen merged 8 commits into
masterfrom
fix/status-without-reason
Nov 15, 2026
Merged

tobixen merged 8 commits into
masterfrom
fix/status-without-reason

Conversation

@tobixen

@tobixen tobixen commented Nov 15, 2026 •

Copy link
Copy Markdown
Member

Disclaimer: AI-generated code and text below the line, generated by Claude Opus 5.5, from the prompts:

... continue working on getting all the conformance tests passing [in calendaring-client]
[add] a tester probe and new caldav hint [for the Zimbra behaviour, dropping a task's DTSTART without a DUE]
Go with [a caldav-server-tester probe, a caldav hint, and caldav rewriting a tz link name to its canonical zone before saving] for bedework5


Three fixes for server quirks met while getting calendaring-client's conformance tests passing, bundled into one PR since a CI run takes an hour:

  • A status line without a reason phrase (HTTP/1.1 404, as Bedework 5 sends in a multiget) is now read by its parsed status code, so a missing object is reported missing, not raised as ResponseError.
  • New hint save-load.icalendar.tzid-link: Bedework 5 refuses a TZID that is a tz database link name (Europe/Oslo) with 403 unknownTimezone. There save() sends the zone the link points to instead; save(raw_data=...) is left alone.
  • New hint save-load.todo.dtstart-without-due: Zimbra drops a task's DTSTART when the task has no DUE.

The commits carry the details. The matching caldav-server-tester probes come in a PR of their own.

🤖 Generated with Claude Code

Comment thread tests/test_caldav_unit.py Fixed
Comment thread caldav/lib/vcal.py Fixed
@tobixen tobixen changed the title fix: a status line may leave out the reason phrase fix: Bedework and Zimbra conformance — status line, tz links, task start Nov 15, 2026
Bedework 5 answers a calendar-multiget naming a missing object with
"HTTP/1.1 404": no reason phrase, nor the space before it that RFC 9112
section 4 keeps even when the phrase is empty.  The status checks
looked for " 404 " with spaces on both sides, so this raised
ResponseError where the object should have been reported missing.
They now compare the parsed status code, as all_statuses_ok() does.

Prompt: ... continue working on getting all the conformance tests passing [in calendaring-client; this fixes calendaring's test_get_missing_raises on Bedework 5]

Assisted-By: Claude Opus 5.5 <noreply@anthropic.com>
Reviewed-by: Tobias Brox <tobias@redpill-linpro.com>
Zimbra drops a task's DTSTART where the task has no DUE, though RFC
5545 allows either alone; with a DUE it keeps both.  The new hint
save-load.todo.dtstart-without-due records it, unsupported in the
Zimbra 8.8 profile and so in 10.1's (both probed), so a client can tell
it will lose the start.

Prompt: ... continue working on getting all the conformance tests passing [in calendaring-client]
Followup-Prompt: [add] a tester probe and new caldav hint [for the Zimbra behaviour, dropping a task's DTSTART without a DUE]

Assisted-By: Claude Opus 5.5 <noreply@anthropic.com>
Reviewed-by: Tobias Brox <tobias@redpill-linpro.com>
Bedework 5 refuses an object whose TZID is a tz database link name,
such as Europe/Oslo, with 403 "unknownTimezone", VTIMEZONE or not, so
a Norwegian user cannot store an event in their own zone there.  The
new hint save-load.icalendar.tzid-link is ungraceful for it, and
save() then sends each link TZID as the zone it points to, from
tzdata.zi.  Times since 1970 stay where they were; save(raw_data=...)
is left alone.

Prompt: ... continue working on getting all the conformance tests passing [in calendaring-client]
Followup-Prompt: Go with [a caldav-server-tester probe, a caldav hint, and caldav rewriting a tz link name to its canonical zone before saving] for bedework5

Assisted-By: Claude Opus 5.5 <noreply@anthropic.com>
Reviewed-by: Tobias Brox <tobias@redpill-linpro.com>
Code scanning flagged two nits on the tz link change: an
`except ModuleNotFoundError: pass` with no explanation, and
caldav.lib.vcal imported both with `import` and `from ... import`
in the unit tests.  Say why the missing tzdata package is fine, and
import the module the `from` way like the neighbouring tests.

Prompt: Please fix the comment on the empty except on #869 [#869], and fix consistent import style

Assisted-By: Claude Opus 5.5 <noreply@anthropic.com>
Reviewed-by: Tobias Brox <tobias@redpill-linpro.com>
Explain the NLnet GenAI policy situation: the project is committed to
version 1.1, while version 2.0 only allows read-only use, and the
policy will have to be revised if funding continues.  Reword the
"read this first" rules around contributors adding value, update the
disclosure section beyond 3.3.1, and fix a batch of typos.  The text
is mostly the maintainer's own; the AI did QA and language fixes.

Prompt: Can we piggy-back my uncommitted AI-POLICY-change into the same PR?  Please do a QA on my changes. [moved to #869 instead of the contributor PR #858]
Prompts (comments made inline in the document while reviewing it):
  Please fix up the findings.
  https://nlnet.nl/foundation/policies/generativeAI/2.0/
  https://nlnet.nl/foundation/policies/generativeAI/1.1/
  The project has been receiving funding since before the AI-policy was created, and we're currently committed to the version 1.1 of their AI-policy.
  Ok [to moving the NLnet paragraph into its own section]
  We're a bit in a twilight zone, as NLNet wants the 2.0-policy to be adapted.
Followup-Prompt: I've amended a bit more (including re-adding that discarding NLnet funding may be considered), do the QA again, and push it into my PR.  Do a review-and-merge of #858 also [#858]

Assisted-By: Claude Opus 5.5 <noreply@anthropic.com>
Reviewed-by: Tobias Brox <tobias@redpill-linpro.com>
Two fixtures carried a real personal mail address, copied from server
responses.  Replace it with testuser@example.com (and the organizer's
with noreply@example.com); the tests do not depend on the value.

Prompt: please replace my email address with an @example.com-address

Assisted-By: Claude Opus 5.5 <noreply@anthropic.com>
Reviewed-by: Tobias Brox <tobias@redpill-linpro.com>
Follow-up to #858.  A
component name with trailing whitespace ("BEGIN:VEVENT ") left the raw
UID shortcut and made RawDataState.get_uid() return None; strip it.
Also cover the rule that an earlier UID line the cheap regex skips
(parameters, lowercase) sends the lookup to the full parser.

Prompt: [fix issues found during the review process]
Followup-Prompt: [merge the contributor's PR as is and do the review fixes in a commit of our own on top]

Assisted-By: Claude Opus 5.5 <noreply@anthropic.com>
Reviewed-by: Tobias Brox <tobias@redpill-linpro.com>
UTC, GMT and Zulu are tz database links to Etc/UTC and Etc/GMT;
rewriting TZID=UTC gave a TZID on a UTC (Z) time, which RFC 5545
forbids.  A VTIMEZONE without a TZID made save() raise KeyError on
servers where the link rewrite is active; it is now skipped.

Prompt: [fix issues found during the review process]

Assisted-By: Claude Opus 5.5 <noreply@anthropic.com>
Reviewed-by: Tobias Brox <tobias@redpill-linpro.com>
@tobixen
tobixen force-pushed the fix/status-without-reason branch from 2ca70da to 848f2f1 Compare November 15, 2026 16:26
@tobixen
tobixen merged commit e60d8d3 into master Nov 15, 2026
17 checks passed
tobixen added a commit to python-caldav/caldav-server-tester that referenced this pull request Nov 16, 2026
#62

Two new probes, matching the hints in
python-caldav/caldav#869:
save-load.icalendar.tzid-link (Bedework 5 refuses a tz link name as
TZID) and save-load.todo.dtstart-without-due (Zimbra drops a task's
DTSTART without a DUE).  The tester carries the feature definitions
until a caldav release has them.

Commits:
- 15ae655 feat: probe a TZID that is a tz link name
- 79c489e feat: probe a task start without a due
- 54522b9 test: keep a system pytest-asyncio out of the run

Assisted-By: Claude Opus 5.5
Reviewed-by: Tobias Brox <tobias@redpill-linpro.com>
@tobixen
tobixen deleted the fix/status-without-reason branch November 17, 2026 13:02
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.

1 participant