Skip to content

Bring 0.10.0a pre-release changes back to main for final release - #7524

Merged
masenf merged 30 commits into
mainfrom
r/pre-2026.10.06-37579583012
Oct 8, 2026
Merged

masenf merged 30 commits into
mainfrom
r/pre-2026.10.06-37579583012

Conversation

@masenf

@masenf masenf commented Oct 8, 2026 •

Copy link
Copy Markdown
Collaborator

View guided diff Turn on auto-fix

github-actions Bot and others added 28 commits October 5, 2026 19:28
…lex-components-code@0.10.0a1, reflex-components-gridjs@0.10.0a1, reflex-components-markdown@0.10.0a1, reflex-components-moment@0.10.0a1, reflex-components-plotly@0.10.0a1, reflex-components-radix@0.10.0a1, reflex-components-recharts@0.10.0a1, reflex-docgen@0.10.0a1 (new-prerelease-minor)
Since #7312 made Field the descriptor for state vars, reading a backend
var on the state class returns the Field instead of the default value.
Code that passed that default to a component now fails with
ChildrenTypeError or the LiteralVar TypeError. Add a Breaking Changes
entry to the 0.10.0a1 section with the migration via default_value().

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01R98baxdcSdivcfDYrqxqF6
* Prepare Python 3.11 floors and directional prerelease phases

* Simplify directional lockstep planning

A publish-last member now needs only the early members of its group,
tagged at its own version: it plans from its own baseline, reuses a
sibling tag that already exists, and otherwise plans the sibling at the
same version. Early members release on their own. Two publish-last
members no longer need each other, so pin-exact groups with several
cannot deadlock on each other's tags.

Planning computes targets first and checks the dependency pins of what
it planned, holding back blocked packages and the packages that need
them; an explicit selection that cannot release fails as before. The
exact pin is checked against the declared requirement at plan time as
well as at build time, so a floor raised past a reusable sibling tag
stops the release before any changelog is bumped. prepare-publish
requires every early member's tag before a publish-last member builds.

Sibling floors use the .dev0 of each sibling's next version, and the
docs, template comments and news fragments say the same thing in fewer
words.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MZxGmMwCvKGcUJu6TK99jZ

* Hold back blocked packages before planning their versions

The planner skips a package with an unsatisfiable pin instead of
computing its next version, so a release action that does not apply to
a held dependent no longer stops its sibling's release. Blocked pins are
gathered once for the selection and its lockstep partners; a dependent
whose missing partner is blocked is held back with it, and a partner
reused from an existing tag is not consulted at all.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MZxGmMwCvKGcUJu6TK99jZ

* Delete packages/reflex-hosting-cli/news/+sdk-minor-compatibility.misc.md

* Say which tags the publish-last leg waits for

The workflow comment described the late leg as waiting on every package
in the batch; it runs after the first leg, but the gate is the tags of
the siblings each late member pins.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MZxGmMwCvKGcUJu6TK99jZ

---------

Co-authored-by: Claude <noreply@anthropic.com>
…s-lucide@1.1.0a1, reflex-release@0.2.0a1 (new-prerelease-minor)
…mponents-gridjs@0.10.0a2, reflex-components-moment@0.10.0a2 (continued-prerelease)
…lex-components-react-player@0.10.0a1 (new-prerelease-minor)
…components-radix@0.10.0a2, reflex-components-recharts@0.10.0a2 (continued-prerelease)
reflex-otel raised its floor to Python 3.11 in #7449 without a news
fragment, so the release tooling had nothing to release it with. Add the
same fragment the other packages got, and a news/.gitkeep so the directory
survives materialization.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_011qWgGrSKMqztQtSRWBP7KK
Floor reflex-otel on the 0.10.0a2 alpha already published from this
prerelease train, so its own alpha ships against the train rather than
0.9.x. A prerelease floor is shippable for an alpha release; the final
release lifts it to the first published 0.10 final.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_011qWgGrSKMqztQtSRWBP7KK
…9583012

Reconcile fragments main changed after the train materialized them: carry
the #7513 wording of the #7312 breaking entry and the #7516 wording of the
#7461 entry into the a2/a3 changelogs, and drop the #7495 patch-restore and
storage-default entries that #7516 backed out.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01XWWGUbPYwC5ELRyhCjm5ym
…9583012

Reconcile fragments main changed after the train materialized them:
carry #7519's `set_default` wording of the #7516 breaking entries into
the reflex and reflex-base a4 changelogs, and keep both materialized
fragments deleted.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SjCrbdguMkfe4bhbaREiwo
Collapse each package's pre-release sections under its newest pre-release
heading, group the larger categories under subheadings, and drop entries
that only fixed regressions introduced and resolved within the 0.10
pre-releases (boot hydration storage, cross-tab storage echoes, mangled-name
guard, supervised output draining) or that are superseded or internal
(reflex-base 0.9.12 floors, sibling version floors, BUN_OPTIONS, routes.json
constant, unreleased reflex-workflow logging). Merge related entries
(reflex component CLI removal, BackendVarFormatError, buffered uploads) and
split the reflex-hosting-cli output changes into a list.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0163uzvnGNDwtVmqGKqwSmWC
@masenf
masenf requested review from a team and Alek99 as code owners October 8, 2026 20:24
@masenf masenf added skip-changelog For doc/internal changes on deck PRs lined up to review / merge next labels Oct 8, 2026
@masenf masenf added this to the v0.10.x milestone Oct 8, 2026

@cubic-dev-ai cubic-dev-ai Bot 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.

All reported issues were addressed across 185 files

Reply with feedback, questions, or to request a fix.

View guided diff | Turn on auto-fix | Re-trigger cubic

Comment thread packages/reflex-base/CHANGELOG.md Resolved
Comment thread packages/reflex-otel/CHANGELOG.md Resolved
Comment thread packages/hatch-reflex-pyi/CHANGELOG.md Resolved
Comment thread CHANGELOG.md Resolved
FarhanAliRaza
FarhanAliRaza previously approved these changes Oct 8, 2026
@codspeed

codspeed Bot commented Oct 8, 2026 •

Copy link
Copy Markdown

Merging this PR will degrade performance by 4.11%

❌ 1 regressed benchmark
✅ 149 untouched benchmarks
⏩ 18 skipped benchmarks1

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

Benchmark BASE HEAD Efficiency
❌ test_state_access[get_handler] 72.5 ms 75.7 ms -4.11%

Tip

Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.


Comparing r/pre-2026.10.06-37579583012 (f05eeb8) with main (85da007)2

Open in CodSpeed

Footnotes

  1. 18 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

  2. No successful run was found on main (f05eeb8) during the generation of this report, so 85da007 was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩

@greptile-apps

greptile-apps Bot commented Oct 8, 2026 •

Copy link
Copy Markdown
Contributor

RetriggerConfidence Score: 4/5

[High impact] Release workflow and changelog consolidation for version 0.10.0.

The PR still needs the release-planning fix before merging, because a blocked group can stop unrelated releases.

Summary

Brings the 0.10 prerelease notes and dependency floors back to main, and updates how related packages release.

  • Lets early members of a publish-last group release separately.
  • Checks exact dependency pins before planning and building.
  • The latest change restores the cross-tab rx.LocalStorage fix to the reflex-base notes.
  • No new findings were accepted.

Reviews (2) · Last reviewed commit: "Restore the sync=True LocalStorage cross..." · Reviewed by Greptile

Comment thread CHANGELOG.md Resolved
…elease notes

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0163uzvnGNDwtVmqGKqwSmWC
@masenf
masenf merged commit 624e3e7 into main Oct 8, 2026
142 of 144 checks passed
@masenf
masenf deleted the r/pre-2026.10.06-37579583012 branch October 8, 2026 20:44
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

on deck PRs lined up to review / merge next skip-changelog For doc/internal changes

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants