Repository navigation
Drop entry-point distribution info when unregistering a plugin - #767
Merged
Merged
Conversation
RonnyPfannschmidt
force-pushed
the
claude/project-thread-p5dhgd
branch
from
October 9, 2026 11:55
c6590f0 to
9af388b
Compare
RonnyPfannschmidt
changed the base branch from
main
to
claude/project-thread-yz2zw2
October 9, 2026 11:55
RonnyPfannschmidt
added this pull request to stack #769
October 9, 2026 12:13
Member
|
The stack is not very clear, it says 2 commits but there are 8? |
RonnyPfannschmidt
force-pushed
the
claude/project-thread-p5dhgd
branch
from
October 9, 2026 12:23
9af388b to
c72b184
Compare
Member
Author
|
pr stacking misstake |
bluetech
approved these changes
Oct 11, 2026
Member
Out of curiosity I clicked this, but it goes to a login wall. I think it would be better to remove these links, we shouldn't add private links to commit messages IMO. |
RonnyPfannschmidt
force-pushed
the
claude/project-thread-p5dhgd
branch
from
October 11, 2026 07:55
c72b184 to
5528401
Compare
Member
Author
|
It works as reference when shared so I kinda like to keep it |
RonnyPfannschmidt
force-pushed
the
claude/project-thread-p5dhgd
branch
from
October 11, 2026 09:18
5528401 to
c72b184
Compare
Member
But it's not shared by default right? IMO commit messages are public-facing messages, so they shouldn't contain cruft that is private to the author only. |
Key the recorded distributions by plugin name instead of keeping a list of (plugin, dist) tuples. unregister() pops the name, so the listing no longer shows unloaded plugins, and reloading an entry point overwrites its slot instead of appending a duplicate. The _name2plugin cleanup becomes an is_blocked() check plus a pop: the old truthiness guard only had to keep a blocked name's marker, and the assert only narrowed the type for mypy. test_unregister_blocked now checks that the block survives. Fixes #763. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
RonnyPfannschmidt
force-pushed
the
claude/project-thread-p5dhgd
branch
from
October 17, 2026 06:04
c72b184 to
50585e0
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Requested by Ronny · project thread
🤖 Written by Claude Opus 5.5 via Claude Code for the pluggy maintainers; I prompted it, it did the work, I read it.
Fixes #763.
Before:
unregister()left the plugin inlist_plugin_distinfo()/list_plugin_distributions()(and so in pytest's "plugins:" header), and reloading the entry point appended a duplicate row.After: both listings only show entry-point plugins that are still registered, and a reload takes the old slot.
How: the private
_plugin_distinfolist of(plugin, dist)tuples becomes_name2dist: dict[str, Distribution], keyed by the registration name like_name2plugin.unregister()pops the name; the listings are built from the two dicts, so no identity map over plugin objects is needed.list_plugin_distinfo()now just wrapslist_plugin_distributions().The
_name2plugincleanup inunregister()becomesif not self.is_blocked(name): pop. The old guard existed only to keep a blocked name's marker, and theassert name is not Nonewas a leftover mypy narrowing from 2019.test_unregister_blockednow asserts the block survives, which it did not check before.🤖 Generated with Claude Code