Skip to content
Open
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
28 changes: 28 additions & 0 deletions memory/upstream-sync.md
Original file line number Diff line number Diff line change
Expand Up @@ -342,3 +342,31 @@ having been open ~40 days; the fixes are wanted downstream regardless.
cursor now covers them only when they merge.
- Supersedes the clone caveat above: the local clone is fetched to `da08db1`
and GitHub API access works from this environment.

## 2026-11-17: cursor already at upstream HEAD; nothing to sync

- Verification (GitHub API, authenticated `gh api`):
`compare/da08db16714b608d7f5db248c2b2fcc30a756660...main` returns
`status: identical`, `ahead_by: 0`, `behind_by: 0`, `total_commits: 0`.
Upstream `pushed_at` is `2026-10-29T09:19:44Z`, the default branch `main`'s
HEAD is `da08db167` itself (`Merge pull request #2073 ... changelog-2.10.0`),
and the newest tag `2.10.0` points at the same commit. No merged upstream
commit exists after the cursor.
- `.sync-upstream.json` unchanged: `last_synced_commit` stays `da08db1`,
`sync_count` stays `8`, `last_synced_at` untouched (nothing was analyzed or
ported, so the timestamp would only imply work that did not happen).
- Downstream state: the 2.10.0 ports are on `main` as `7efc234` (#262), which
carries all three ports plus the cursor advance - the earlier "branch layout
deviation" note is now superseded by an actual merge. The remote branches
`sync/ant-design-x-2066`, `sync/ant-design-x-2070`, `sync/ant-design-x-2071`
and `chore/sync-upstream-da08db1` are history-only leftovers (their content
is on `main`; they are not ancestors because `#262` was a squash, so
`merge-base --is-ancestor` reports them unmerged). Do not re-port from them.
- Per the standing policy (2026-10-19) no open upstream PRs were enumerated and
no tracking issues were filed; the empty range ends the sync.
- Targeted re-check of the six upstream PRs already recorded in memory
(#2055, #2057, #2058, #2065, #2067, #2068 - not an enumeration): all still
`state=open`, `merged_at: null`. #2057 (`9ff58c0fe5`) and #2058
(`dbc5120865`) have **new head SHAs** since the 2026-10-18/19 entries, so
upstream reworked them; if either ever merges, diff against the merge commit,
not the head recorded here.
Loading