Skip to content

chore(nix): update flake.lock - #1

Open
github-actions[bot] wants to merge 2 commits into
mainfrom
update_flake_lock_action
Open

chore(nix): update flake.lock#1
github-actions[bot] wants to merge 2 commits into
mainfrom
update_flake_lock_action

Conversation

@github-actions

@github-actions github-actions Bot commented Mar 1, 2026

Copy link
Copy Markdown

Automated changes by the update-flake-lock GitHub Action.

Flake lock file updates:

• Updated input 'nixpkgs':
    'github:NixOS/nixpkgs/a4bf066' (2026-04-25)
  → 'github:NixOS/nixpkgs/687f05a' (2026-05-18)

Running GitHub Actions on this PR

GitHub Actions will not run workflows on pull requests which are opened by a GitHub Action.

To run GitHub Actions workflows on this PR, close and re-open this pull request.

@github-actions
github-actions Bot force-pushed the update_flake_lock_action branch from 19c379e to 82c4714 Compare March 8, 2026 01:00
@jla415
jla415 force-pushed the main branch 2 times, most recently from 7f5deee to 415ee4d Compare March 13, 2026 01:07
@github-actions
github-actions Bot force-pushed the update_flake_lock_action branch from 82c4714 to f8dc919 Compare March 15, 2026 01:08
@jla415
jla415 force-pushed the main branch 2 times, most recently from 2bf3e15 to 9f3603b Compare March 15, 2026 23:28
@github-actions
github-actions Bot force-pushed the update_flake_lock_action branch from f8dc919 to 6a58929 Compare March 22, 2026 01:03
@github-actions
github-actions Bot force-pushed the update_flake_lock_action branch from 6a58929 to c4bb760 Compare March 29, 2026 01:10
@github-actions
github-actions Bot force-pushed the update_flake_lock_action branch from c4bb760 to a5d7f91 Compare April 5, 2026 01:11
@github-actions
github-actions Bot force-pushed the update_flake_lock_action branch from a5d7f91 to 1c22600 Compare April 12, 2026 01:15
@github-actions
github-actions Bot force-pushed the update_flake_lock_action branch from 1c22600 to 9b85e38 Compare April 19, 2026 01:17
@github-actions
github-actions Bot force-pushed the update_flake_lock_action branch from 9b85e38 to 125323e Compare April 26, 2026 01:20
@github-actions
github-actions Bot force-pushed the update_flake_lock_action branch from 125323e to 9856c2b Compare May 3, 2026 01:27
@github-actions
github-actions Bot force-pushed the update_flake_lock_action branch from 9856c2b to d1ec986 Compare May 10, 2026 01:31
@github-actions
github-actions Bot force-pushed the update_flake_lock_action branch from d1ec986 to a39f268 Compare May 17, 2026 01:47
Flake lock file updates:

• Updated input 'nixpkgs':
    'github:NixOS/nixpkgs/a4bf066' (2026-04-25)
  → 'github:NixOS/nixpkgs/687f05a' (2026-05-18)
@github-actions
github-actions Bot force-pushed the update_flake_lock_action branch from a39f268 to bba86e5 Compare May 24, 2026 01:54
jla415 pushed a commit that referenced this pull request Aug 11, 2026
…r issues (wy-jpd3.6)

Federation ask #1, the flagship (doc/beads-upstream-asks.md section 1 in the
wyvern rig). beads stamps issues.updated_at on EVERY mutation, so ANY two
same-issue edits on two replicas between syncs collide on that cell even when
the semantic fields are disjoint — machine A adds a comment while machine B
adds a label, and the issues row conflicts on nothing but the timestamp both
bumped. The conflict rate is therefore far higher than the semantic-conflict
rate, and the row-level LWW resolver could only take the safe half of it: it
declined whenever both sides had moved updated_at past the merge base, because
taking one side's whole row would silently drop the other side's edits.

Replace it with a FIELD-level three-way merge (automerge.go):

  - a cell only one side changed relative to the merge base keeps that side's
    value, so disjoint edits both survive;
  - a cell both sides changed to different values is settled last-write-wins
    by updated_at (the ask's rule for status/assignee/updated_at, applied
    uniformly);
  - updated_at therefore merges to max(ours, theirs);
  - add/add, delete/modify, and a contested cell whose two sides carry equal
    or unparseable updated_at values still go to the operator.

Resolution uses dolt's manual-resolution path — write the merged values over
our working-set row, then DELETE the conflict row — because
DOLT_CONFLICTS_RESOLVE is table-level and cannot express a per-cell merge.

Also admit the ask's companion classes to the auto-resolver: labels by
set-union and comments/events by append-only union. All three are keyed such
that different additions are disjoint rows dolt already unions, so a conflict
can only be the same row on both sides; when its columns agree it is resolved
by keeping it, and a row missing on one side (a deletion racing an insert) or
diverging columns in a supposedly immutable row still go to the operator.

Tests: the merge rules are pure and unit-pinned (disjoint fields both survive,
contested cell LWW both directions, equal/unparseable timestamps decline,
add/add and delete/modify decline, NULL vs '' distinct, []byte vs string
equal, column classification). The dolt-backed integration tests are updated
for the new semantics and a disjoint-field case added.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant