Rename sync-incremental to sync, fix sync elapsed time, TUI and build polish#2
Merged
Conversation
- Fix CLIProgress showing ~2562047h elapsed during incremental sync because startTime was never initialized (OnStart not called) - Always do full cache rebuild on TUI launch when changes detected, since incremental rebuilds can produce incorrect results - Quote $GITHUB_SHA in release workflow - Hide "unknown" version in TUI title bar Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
The shorter name is easier to type for the most common operation. sync-incremental is preserved as a hidden alias for backward compatibility. Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
wesm
changed the title
- Rename file to match the new command name - OnStart unconditionally resets startTime for reuse scenarios - OnLatestDate initializes startTime when called before OnStart - Add CLIProgress unit tests for callback ordering edge cases Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
When OnStart is never called (incremental sync path), lastPrint was zero so the throttle check always passed, and elapsed was near-zero producing absurd rates like 120000000.0/s. Initialize lastPrint alongside startTime in OnProgress/OnLatestDate fallbacks, and guard rate calculation against sub-second elapsed time. Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Runs make test and make lint on macOS (matching the primary dev environment and CGO/DuckDB build requirements) for all PRs and pushes to main. Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Newer versions enable stricter checks by default, causing spurious lint failures. Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Brew doesn't support versioned formulae for golangci-lint. Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
The pre-built binary was compiled with Go 1.24 which refuses to lint Go 1.25 code. Building from source with go install uses the CI's Go 1.25 toolchain. Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Check error returns flagged by errcheck - Remove unused headerStyle and hasPrerelease - Suppress intentionally ignored errors with _ = Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
jesserobbins
added a commit
to jesserobbins/msgvault-1
that referenced
this pull request
Apr 23, 2026
…ilure Two resume-path corrections: - A fbmessengerCheckpoint saved mid-way through the first thread has ThreadIndex == 0, which the prior resume guard (`if prior.ThreadIndex > 0`) treated as "no progress yet" and ignored. Functionally this didn't lose data (source_message_id dedup made retry safe and the outer loop already starts at threadIdx=0) but summary.WasResumed stayed false, the "resuming" log didn't fire, and cumulative counters from the prior run didn't carry forward — so the summary undercounted and the CLI didn't tell the user a resume was happening. Gate removed; any well-formed checkpoint with a matching RootDir is now resumable. Added TestImportDYI_ResumeFromFirstThreadCheckpoint. - The outer thread loop previously advanced the per-thread checkpoint to threadIdx+1 even after a hard error on that thread, so a resumed run would skip it. A transient failure (DB lock, I/O blip) should retry on the next run; source_message_id dedup keeps retry safe for any messages already written. On hard error we now log, flag HardErrors=true, and `continue` without advancing the checkpoint. A persistent error will surface via HardErrors=true across runs and the user can rerun with --no-resume if they want to force past it. Addresses roborev combined review (6a28ac0) Medium findings kenn-io#2 and kenn-io#3. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
sternryan
referenced
this pull request
in sternryan/msgvault
May 6, 2026
- TestSourcesDB_OpenAndLookup: fresh URL scores 1.0, ingested URL scores 0.0 - TestSourcesDB_ReadOnly: INSERT against opened handle errors out Confirms criterion #2 (URL gold) end-to-end with mattn/go-sqlite3 mode=ro on a real file-backed sources.db. The five lexical scorers (urlgold, curiosity, recurrence, decision, expert) were implemented in Task 1 scorer.go alongside Composite — Task 2b only needs the sources.db integration test (per plan §14-02-02b).
webgress
added a commit
to webgress/msgvault
that referenced
this pull request
May 22, 2026
External review caught three SQL shapes in internal/query/sqlite.go (which backs both SQLite and PostgreSQL via the QueryDialect interface) that PG rejects at parse time, breaking TUI startup and API aggregate / deletion-staging endpoints when the store is PG. HIGH kenn-io#1 — buildAggregateSQL emitted `FROM ( <inner SELECT> )` with no alias on the outer derived table. PG requires every derived table in FROM to be aliased ("syntax error at or near ')'"). Added `AS agg`; SQLite tolerates the explicit alias unchanged. HIGH kenn-io#2 — GetGmailIDsByFilter used SELECT DISTINCT … ORDER BY m.sent_at DESC, m.id DESC. PG rejects DISTINCT when the ORDER BY expressions are not in the SELECT list, and the multiplying joins to message_recipients / message_labels were why DISTINCT was there in the first place. Converted every multiplicative filter (sender, sender name, recipient, recipient name, domain, label) into an EXISTS subquery — per the SQL guidance in CLAUDE.md — so the join graph no longer multiplies, DISTINCT is unnecessary, and the ORDER BY stands on its own. The sources scope (1:1 with messages) stays as a JOIN. MEDIUM kenn-io#3 — ListMessages's ORDER BY referenced `m.size_estimate` and `m.subject` raw while the SELECT exposed them through `COALESCE(m.size_estimate, 0)` and `COALESCE(m.subject, '')`. Under SELECT DISTINCT, PG insists the ORDER BY expressions textually match something in the SELECT list. Switched the size and subject sorts to the COALESCE expressions; the date sort already matched. Test - internal/query/pg_compat_test.go exercises Aggregate (ViewSenders), GetGmailIDsByFilter with a Label filter (the previously multiplying join, now EXISTS) and verifies no duplicate IDs come back, and ListMessages with every supported sort field. Uses testutil.NewTestStore, so MSGVAULT_TEST_DB=postgres://… runs the same suite against PG and catches regressions of these three bug shapes.
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
1 participant
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.
Summary
sync-incrementalcommand tosyncfor brevity (sync-incrementalkept as hidden alias)Test plan
msgvault sync you@gmail.comworksmsgvault sync-incremental you@gmail.comstill works (alias)msgvault --helpshowssyncnotsync-incremental🤖 Generated with Claude Code