* fix(embeddeddolt): commit pending changes before push (#5433) Every Pull variant (Pull, PullWithStrategy, PullRemoteWithStrategy, PullRemote, PullFrom) auto-commits pending working-set changes first (GH#2474, so a crashed command's leftovers don't make the merge refuse to start), but no Push variant (Push, ForcePush, PushRemote, PushTo) did. A write that landed in the working set but was never explicitly committed -- auto-commit configured off, or a session that ended right after a write with no intervening commit -- silently pushed nothing new while `bd dolt push` still printed "Push complete." and exited 0. This is the half of #5433's silent-stall report the maintainer traced and confirmed in code (issue comment, quoting exact line numbers this matches on current main): "every Pull variant calls CommitPending before pulling... no Push variant does... push never flushes the working set, then prints 'Push complete.' on nil error with no remote-ref or working-set check. Strongest hook for a narrow push-honesty PR." Reporter confirmed: "Yes, please open the push-honesty PR. That is the surface that actually cost us." Scoped to exactly this half -- the confirmed, traced, mechanical gap matching an already-established pattern (copy Pull's own guard to Push). Did NOT attempt the harder half discussed in the same thread: verifying the remote ref actually advanced and reporting honestly when it didn't. That needs a deliberate design call the maintainer themselves treated as separate (distinguishing a legitimate no-op push, already in sync, from one that had something to send but transferred nothing) and risks false positives if done carelessly -- a bigger change than this issue's narrowed scope asked for. Flagging in the relay rather than guessing at that design. Also confirmed the server-mode backend (internal/storage/dolt) has the identical asymmetry -- Push/ForcePush/PushRemote all route through pushToRemote, which likewise never calls CommitPending -- but the issue's repro and the maintainer's trace are both embedded-mode specific, so left server-mode untouched pending its own verification. Does not claim to fix the mysterious full stall itself (bd dolt commit also returning "Nothing to commit" while HEAD stays frozen, even when called directly) -- the reporter and maintainer both independently concluded that has no known deterministic trigger. This closes one real, confirmed contributing gap: push not flushing pending changes the way pull always has. RED: added TestPushCommitsPendingChangesFirst, table-driven over all four Push variants. Creates an issue (a genuinely pending, uncommitted change -- verified via GetCurrentCommit before/after CreateIssue, so the test would fail loudly if CreateIssue ever started auto-committing on its own), then calls each Push variant and asserts HEAD moved. Push/ForcePush/PushRemote route through the swappable vcPush/ vcForcePush vars (remote_peer_auth_test.go's existing captureRemoteEntryPoints seam) so the test stops with errStopRemote before any network IO; PushTo calls versioncontrolops.Push directly with no equivalent seam, so it actually attempts the push and fails on "remote not found" in this remote-less test store -- still proves CommitPending ran first, since that failure only happens after this fix's guard clause. All four failed pre-fix (HEAD did not move). GREEN: all four pass after the fix. Verified: full remote-verb/peer-auth suite (TestWithPeerAuth and its variants, TestRemoteVerbsPresentStoredPeerCredentials, TestRemoteVerbPlainRemote*) still green -- confirms this fix doesn't disturb credential resolution or the swappable-seam test infrastructure it shares. go vet clean, go build ./... clean across the repo. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_018c3vLFNVgPebSpCJrd8xTD * chore(bazel): enumerate push_commit_pending_test.go in embeddeddolt go_test srcs Generated by `bazel run //:gazelle`. The PR predates base's addition of internal/storage/embeddeddolt/BUILD.bazel, so its new test file was not in the explicit go_test srcs list; bazel would have silently run zero of its cases. Base parity at the rebase base was 118/118 enumerated. * fix(autopush): record post-push HEAD, and cover the clean-store push path Maintainer review fixup for #5795 (adopt-pr review iteration 1). Push now commits pending changes before pushing (GH#5433), so HEAD can move across the call. maybeAutoPush read HEAD only before the push and recorded that hash as LastCommit, which is now the parent of what was actually pushed, so the next tick saw a mismatch and pushed once redundantly before re-latching. Route the recorded hash through pushedCommitForState, which re-reads HEAD after a successful push and falls back to the pre-push value if the re-read fails (the push itself already succeeded, so dropping LastCommit would lose change detection entirely). Also: - Add a clean-store arm to TestPushCommitsPendingChangesFirst. The guard relies on commitAll(tolerateEmpty=true) mapping Dolt's "nothing to commit" to (false, nil) for every push against an already-committed store, and nothing in this package asserted it. Flipping that flag reds the new arm alone, with the four pending-change arms still green. - Correct the `bd dolt remote reset-data` confirmation, which promised the remote is "rebuilt from local HEAD" while the push now adds a commit created after the operator types y. Not taken here, tracked separately: extending auto-push change detection to also fire on a dirty working set (that gate predates this PR, so skipping it is an incompleteness rather than a regression this PR introduces), and the PR-body disclosure that push now commits on the auto-push and `bd sync` cadences. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 5 <noreply@anthropic.com> Co-authored-by: Gas City Adopt-PR <adopt-pr@gascity.local>
224 lines
7.2 KiB
Go
224 lines
7.2 KiB
Go
package main
|
|
|
|
import (
|
|
"context"
|
|
"encoding/json"
|
|
"fmt"
|
|
"os"
|
|
"path/filepath"
|
|
"time"
|
|
|
|
"github.com/steveyegge/beads/internal/beads"
|
|
"github.com/steveyegge/beads/internal/config"
|
|
"github.com/steveyegge/beads/internal/debug"
|
|
"github.com/steveyegge/beads/internal/storage"
|
|
)
|
|
|
|
// pushState tracks auto-push state in a local file (.beads/push-state.json)
|
|
// instead of the Dolt metadata table, to avoid merge conflicts on multi-machine
|
|
// setups (GH#2466).
|
|
type pushState struct {
|
|
LastPush string `json:"last_push"` // RFC3339 timestamp
|
|
LastCommit string `json:"last_commit"` // Dolt commit hash
|
|
}
|
|
|
|
type autoPushTarget interface {
|
|
GetCurrentCommit(ctx context.Context) (string, error)
|
|
Push(ctx context.Context) error
|
|
}
|
|
|
|
func pushStatePath() (string, error) {
|
|
beadsDir := beads.FindBeadsDir()
|
|
if beadsDir == "" {
|
|
return "", fmt.Errorf("%s", activeWorkspaceNotFoundError())
|
|
}
|
|
return filepath.Join(beadsDir, "push-state.json"), nil
|
|
}
|
|
|
|
func loadPushState() (*pushState, error) {
|
|
path, err := pushStatePath()
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
data, err := os.ReadFile(path) //nolint:gosec // path is constructed internally
|
|
if os.IsNotExist(err) {
|
|
return nil, nil
|
|
}
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
var ps pushState
|
|
if err := json.Unmarshal(data, &ps); err != nil {
|
|
return nil, err
|
|
}
|
|
return &ps, nil
|
|
}
|
|
|
|
func savePushState(ps *pushState) error {
|
|
path, err := pushStatePath()
|
|
if err != nil {
|
|
return err
|
|
}
|
|
data, err := json.MarshalIndent(ps, "", " ")
|
|
if err != nil {
|
|
return err
|
|
}
|
|
return atomicWriteFile(path, data)
|
|
}
|
|
|
|
// autoPushTimeout bounds the st.Push() call that shells out to git fetch,
|
|
// which blocks indefinitely when the remote is unreachable (GH#3370).
|
|
const autoPushTimeout = 30 * time.Second
|
|
|
|
// isDoltAutoPushEnabled returns whether auto-push to Dolt remote should run.
|
|
// Returns true only when explicitly opted in via dolt.auto-push=true in
|
|
// .beads/config.yaml (or BD_DOLT_AUTO_PUSH=true env var).
|
|
//
|
|
// Previously the default was to auto-enable when an "origin" remote exists.
|
|
// That is unsafe at any concurrency above one writer: git+ssh remotes have no
|
|
// chunk-level upload atomicity, so concurrent dolt pushes race on the remote
|
|
// manifest and can leave it referencing chunks that were never uploaded. Any
|
|
// subsequent fetch/clone/push propagates the dangling reference. Set
|
|
// dolt.auto-push=true to restore the old behavior on single-writer setups.
|
|
func isDoltAutoPushEnabled(_ context.Context) bool {
|
|
return config.GetBool("dolt.auto-push")
|
|
}
|
|
|
|
// pushWithContext is a caller-side guard. Push implementations should honor
|
|
// ctx directly, but this keeps auto-push from blocking forever if one does not.
|
|
func pushWithContext(ctx context.Context, target autoPushTarget) error {
|
|
errCh := make(chan error, 1)
|
|
go func() {
|
|
errCh <- target.Push(ctx)
|
|
}()
|
|
|
|
select {
|
|
case err := <-errCh:
|
|
return err
|
|
case <-ctx.Done():
|
|
return ctx.Err()
|
|
}
|
|
}
|
|
|
|
// pushedCommitForState returns the commit to record as LastCommit after a
|
|
// successful auto-push. Push commits pending changes before pushing (GH#5433),
|
|
// so HEAD can move during the push call; reusing the pre-push value would leave
|
|
// LastCommit pointing at the parent of what was actually pushed, and the next
|
|
// tick would see a mismatch and push once redundantly before re-latching.
|
|
//
|
|
// A failed re-read is not a push failure -- the push already succeeded -- so
|
|
// fall back to the pre-push commit rather than leaving LastCommit unset and
|
|
// losing change detection entirely.
|
|
func pushedCommitForState(ctx context.Context, target autoPushTarget, prePush string) string {
|
|
c, err := target.GetCurrentCommit(ctx)
|
|
if err != nil {
|
|
debug.Logf("dolt auto-push: failed to re-read commit after push: %v\n", err)
|
|
return prePush
|
|
}
|
|
return c
|
|
}
|
|
|
|
// maybeAutoPush pushes to the Dolt remote if enabled and the debounce interval has passed.
|
|
// Called from PersistentPostRun after auto-commit and auto-backup.
|
|
func maybeAutoPush(ctx context.Context) {
|
|
if isSandboxMode() {
|
|
debug.Logf("dolt auto-push: skipped (sandbox mode)\n")
|
|
return
|
|
}
|
|
if isDoltLocalOnly() {
|
|
debug.Logf("dolt auto-push: skipped (dolt.local-only=true)\n")
|
|
return
|
|
}
|
|
if !isDoltAutoPushEnabled(ctx) {
|
|
return
|
|
}
|
|
|
|
st := getStore()
|
|
if st == nil {
|
|
return
|
|
}
|
|
if lm, ok := storage.UnwrapStore(st).(storage.LifecycleManager); ok && lm.IsClosed() {
|
|
return
|
|
}
|
|
|
|
// Load local push state (file-based, not in Dolt metadata table).
|
|
// This avoids merge conflicts on multi-machine setups (GH#2466).
|
|
ps, err := loadPushState()
|
|
if err != nil {
|
|
debug.Logf("dolt auto-push: failed to load push state: %v\n", err)
|
|
return
|
|
}
|
|
|
|
// Debounce: skip if we pushed recently
|
|
interval := config.GetDuration("dolt.auto-push-interval")
|
|
if interval == 0 {
|
|
interval = 5 * time.Minute
|
|
}
|
|
|
|
if ps != nil && ps.LastPush != "" {
|
|
lastPush, err := time.Parse(time.RFC3339, ps.LastPush)
|
|
if err == nil && time.Since(lastPush) < interval {
|
|
debug.Logf("dolt auto-push: throttled (last push %s ago, interval %s)\n",
|
|
time.Since(lastPush).Round(time.Second), interval)
|
|
return
|
|
}
|
|
}
|
|
|
|
// Change detection: skip if nothing changed since last push
|
|
currentCommit, err := st.GetCurrentCommit(ctx)
|
|
if err != nil {
|
|
debug.Logf("dolt auto-push: failed to get current commit: %v\n", err)
|
|
return
|
|
}
|
|
if ps != nil && currentCommit == ps.LastCommit && ps.LastCommit != "" {
|
|
debug.Logf("dolt auto-push: no changes since last push\n")
|
|
return
|
|
}
|
|
|
|
// Push with a bounded timeout so an unreachable remote doesn't block
|
|
// the CLI indefinitely (GH#3370). The timeout is configurable via
|
|
// dolt.auto-push-timeout (default 30s).
|
|
pushTimeout := config.GetDuration("dolt.auto-push-timeout")
|
|
if pushTimeout == 0 {
|
|
pushTimeout = autoPushTimeout
|
|
}
|
|
pushCtx, pushCancel := context.WithTimeout(ctx, pushTimeout)
|
|
defer pushCancel()
|
|
|
|
debug.Logf("dolt auto-push: pushing to origin (timeout %s)...\n", pushTimeout)
|
|
if err := pushWithContext(pushCtx, st); err != nil {
|
|
if !isQuiet() && !jsonOutput {
|
|
if pushCtx.Err() == context.DeadlineExceeded {
|
|
fmt.Fprintf(os.Stderr, "Warning: dolt auto-push timed out after %s (remote may be unreachable)\n", pushTimeout)
|
|
} else {
|
|
fmt.Fprintf(os.Stderr, "Warning: dolt auto-push failed: %v\n", err)
|
|
}
|
|
if isAncestorPKMismatchErr(err) {
|
|
printAncestorPKMismatchGuidance(err)
|
|
} else if isDivergedHistoryErr(err) {
|
|
printDivergedHistoryGuidance("push")
|
|
}
|
|
}
|
|
debug.Logf("dolt auto-push: push error: %v\n", err)
|
|
// Throttle retries after failure so a hanging remote doesn't make every
|
|
// subsequent bd command pay the push timeout. We record the attempt
|
|
// timestamp but NOT a new LastCommit, so when the remote recovers the
|
|
// change-detection check (currentCommit != LastCommit) still triggers.
|
|
if ps == nil {
|
|
ps = &pushState{}
|
|
}
|
|
ps.LastPush = time.Now().UTC().Format(time.RFC3339)
|
|
if saveErr := savePushState(ps); saveErr != nil {
|
|
debug.Logf("dolt auto-push: failed to save push state after error: %v\n", saveErr)
|
|
}
|
|
return
|
|
}
|
|
|
|
// Record last push time and commit to local file.
|
|
now := time.Now().UTC().Format(time.RFC3339)
|
|
if err := savePushState(&pushState{LastPush: now, LastCommit: pushedCommitForState(ctx, st, currentCommit)}); err != nil {
|
|
debug.Logf("dolt auto-push: failed to save push state: %v\n", err)
|
|
}
|
|
|
|
debug.Logf("dolt auto-push: pushed successfully\n")
|
|
}
|