* feat(auth): API keys for programmatic access
Adds the @better-auth/api-key plugin so scripts and CI pipelines can call
the existing endpoints with an x-api-key header instead of a session
cookie. Keys are owned by a user, hashed at rest, prefixed gm_, never
expire unless an expiry is chosen, and are not rate limited. A small
guard plugin refuses key management calls that arrive with a key, so a
leaked key cannot mint or revoke keys.
New API Keys section on the Authentication tab with create, show once,
copy and revoke. Migration 0017 adds the api_keys table with the
validator fixture. docs/API.md documents the header and the calls
automation needs. An e2e spec covers create, use, refuse and revoke over
HTTP. bun.nix regenerated for the new package.
Closes#314
Claude-Session: https://claude.ai/code/session_01Tp9pmi65a8k5jLMQFLf4JX
* fix(e2e): send Origin on cookie-authenticated key management calls
Better Auth rejects a cookie-authenticated POST without an Origin header (403 MISSING_OR_NULL_ORIGIN). Browsers always send one, the Playwright request context does not, so the spec sets it on the create and delete calls. Also asserts the guard's 403 code and documents the Origin requirement for scripts that manage keys with a session.
Claude-Session: https://claude.ai/code/session_01Tp9pmi65a8k5jLMQFLf4JX
Adds a release asset limit next to the release limit. The release limit
still decides how many of the newest releases exist in Gitea at all. The
new asset limit decides how many of those also get their assets uploaded;
older releases inside the release limit are created with their notes and
tag only. Unset means assets for every mirrored release, which is what
happened before, so existing setups do not change. Zero means release
notes only. The limit inherits global to organization to repository
through the existing limit-key machinery, and lowering it never deletes
assets that were already uploaded; it only gates what a sync uploads.
Settings gets a second box next to the release limit, the overrides dialog
gets a matching row for organizations and repositories, and the env loader
reads RELEASE_ASSET_LIMIT. The README explains both limits and gains a
troubleshooting entry for the read-only push error people hit on pull
mirrors.
Closes#311Closes#59
Claude-Session: https://claude.ai/code/session_01Tp9pmi65a8k5jLMQFLf4JX
better-auth, @better-auth/oauth-provider and @better-auth/sso move from
1.6.23 to 1.7.2. This is the first stable line with the oauth-provider fix
for unbound resource indicators (Dependabot alert #55); no 1.6.x release
has it.
1.7 keys accounts by (issuer, accountId) and looks the credential account
up with issuer "local:credential" on sign-in, so migration 0016 adds the
column and backfills existing rows: local accounts get local:credential,
SSO accounts take their provider's issuer, other OAuth accounts get the
local:oauth namespace. It also adds jwks.alg and jwks.crv, the 1.7
oauth-provider columns, and the oauth_resources, oauth_client_resources
and oauth_client_assertions tables.
A database that booted the reverted 1.7.0-rc.4 build (on main for a few
minutes on 2026-08-06, and any dev database from then) already has part of
that schema without a 0016 record, which makes 0016 fail on every boot.
repairStrandedBetterAuth17Schema drops the leftovers before migrate runs,
in the same way the #312 repair handles stranded SSO columns; the
migration validator exercises it.
createOAuthClient no longer takes "type"; the SSO applications route sends
application_type (web or native) and reads it back with the legacy column
as fallback. bun.nix is regenerated. reset-users writes the issuer.
Supersedes #374.
Claude-Session: https://claude.ai/code/session_01Tp9pmi65a8k5jLMQFLf4JX
Adds a Source dropdown to the configuration card (GitHub, GitLab beta,
Gitea/Forgejo) backed by a source provider interface with three adapters.
Discovery, mirroring, recovery and cleanup go through the provider;
repositories record their source (migration 0015); clone credentials are
built per host and a repository from another host is refused at the
mirror sites. Issues, pull requests, releases and star lists stay GitHub
only. The destination card gets a Gitea/Forgejo dropdown, and both hosts
lock once repositories exist, with an explicit confirmation to change.
Closes#375. Helps #371.
Claude-Session: https://claude.ai/code/session_01Tp9pmi65a8k5jLMQFLf4JX
A job that had just started, with in_progress=1 and no checkpoint yet,
matched findInterruptedJobs immediately. Since #297 the middleware runs
that check on every request, so a live job was "resumed" by recovery
while the original process was still working on the same repositories.
The request-level 15 second timeout then released the in-flight latch
without cancelling recovery, and later requests logged misleading
"already in progress" and "completed with some issues" lines.
- Move the liveness rule into interrupted-job-detection.ts, as both a
predicate and the SQL condition used by findInterruptedJobs. A job
with no checkpoint is only interrupted once it is older than the 10
minute checkpoint window (or has no recorded start at all).
- Stamp an initial checkpoint on in-progress jobs at creation.
- Refresh the checkpoint every 2 minutes from processWithResilience so a
single long item cannot make a live job look interrupted. The refresh
only touches rows still in progress.
- Keep the middleware recovery latch held until the recovery promise
settles, not until the request stops waiting, and clear the timeout
timer so it no longer leaves a dangling rejection.
Tests cover the predicate, the SQL against an in-memory SQLite database,
and the wiring into helpers, concurrency, and middleware.
Claude-Session: https://claude.ai/code/session_01Tp9pmi65a8k5jLMQFLf4JX
Follow-up to #362 and #365 for the remaining points raised in #361.
- Add `releaseLimit` to the per-repository and per-organization mirror
overrides (repo > org > global > 10), surface it in the Mirror Options
dialog with gating and validation, and pass the resolved value into the
release mirror. Tags need no equivalent: Gitea's git mirror fetches every
ref and they cost nothing beyond the commits already in the clone.
- Reconcile the Gitea description and topics with GitHub on every sync, not
only at migration time. Read both live in one call, write the refreshed
description back to the local row, and skip writes that would not change
anything. Mirrors created before #224 or whose upstream description
changed since now pick it up on the next sync.
- Rework the mobile repository card: name, Custom badge and menu on one
vertically centred row, secondary badges on their own row below, status
and last sync stacked on the right.
Fork PRs (and Dependabot PRs) never receive repository secrets, so the
Docker Hub login ran with empty credentials and Docker Scout then died
with "user githubactions not entitled to use Docker Scout". Every
external contributor got a red docker check they could not do anything
about (#368, #356, #357, #359).
- Gate the Docker Hub login and both Scout PR steps on the PR coming
from this repo, the same guard the login/push/comment steps already use.
- Add a Trivy scan of the locally loaded image on every PR. It needs no
credentials, so forks get real vulnerability feedback. Output goes to
the job summary because GITHUB_TOKEN is read-only on fork PRs.
- Only upload the Scout SARIF when a file exists, so fork PRs stop
logging "Path does not exist: scout-results.sarif".
- Update the workflows README, which still described a docker-scan.yml
that no longer exists.
Not using pull_request_target on purpose: it would run untrusted PR code
with the Docker Hub token available.
Three fixes for the two problems reported in #366:
1. Auto-trust registered SSO identity provider origins. better-auth
1.6.23 (shipped in v3.21.0) added SSRF hardening to the SSO plugin:
sign-in rejects IdP endpoints whose hostnames resolve to private
addresses unless the origin is in trustedOrigins. Homelab split-DNS
setups (IdP domain resolving to a LAN IP from inside the container)
broke on every sign-in with a 400. Registering a provider is an
explicit operator action, so its issuer and endpoint origins are now
added to trusted origins automatically.
2. Surface SSO sign-in errors in the login form. The auth client
resolves with { data, error } instead of throwing, so server-side
rejections were silently swallowed - the button flipped back from
"Redirecting..." with no feedback and nothing in the logs.
3. Honor deleteFromGitea (CLEANUP_DELETE_FROM_GITEA). It was documented
as "Delete repositories from Gitea" defaulting to false, but cleanup
never read it and always archived/deleted orphans on the Gitea side.
It now gates the Gitea-side operation: when disabled (default),
orphans are only marked archived or removed in gitea-mirror's own
database and the Gitea/Forgejo copies stay untouched.
Verified end to end against a live server: sign-in with a private-IP
IdP returns 400 discovery_private_host on v3.27.1 and a 200 with the
authorization URL on this branch; the login form now shows the server
error as a toast; delete cleanup with the flag off removes only the DB
row while the flag on contacts Gitea.
The per-repo overrides from #361 were reachable only from the desktop
repositories table. The mobile card layout had no dropdown at all, and
the organizations page had the inverse gap: the item existed in the
mobile card menu but not the desktop one.
Mobile repo cards get a three-dot menu in the header opening the same
MirrorOverridesDialog, plus the Custom badge shown when a repository
overrides something. The organizations desktop dropdown gains the same
Mirror Options item the mobile variant already had.
Four older articles directed readers to UI elements by names that no
longer exist. Verified against src/components/config/:
"Content & Data" section -> "Mirror Content"
"Mirror metadata" toggle -> "Repository metadata"
"Git LFS (Large File Storage)" -> "Git LFS objects"
"Mirror releases" -> "Releases & tags"
"Preserve structure" strategy -> "Preserve Structure"
The new articles from #363 already used current names or generic
wording, so only the older four needed changes.
* feat(www): generated sitemap, article metadata, and de-orphaned comparison page
The hand-written sitemap listed one URL of 24 and was dated January 2025.
astro.config now sets `site` and uses @astrojs/sitemap, so every built
page lands in sitemap-index.xml on each build. robots.txt points at the
new index and drops the duplicate user-agent block.
Use-case layouts gain the OG/Twitter tags the docs layout already had,
and all three article layouts now emit TechArticle JSON-LD.
Nothing on the site linked to /comparison/github-backup-tools/. It now
has a header nav item, a footer link, and a homepage CTA next to the
use-cases button. The footer Documentation link points at the site docs
instead of the GitHub tree, and footer internal links no longer open in
new tabs.
* feat(www): five new articles and an interlinking pass
New guides, each grounded in shipped behavior:
- Migrate from GitHub to Gitea (parallel-run migration playbook)
- Mirror Git LFS repositories (why one bad LFS repo fails its whole
migration, and the per-repo override from v3.27.0 that fixes it)
- GitHub API rate limits (ETag conditional requests making 304s free)
- Mirror issues and pull requests (what transfers, the PR-to-issue
conversion, why processing is sequential)
- GitHub backup on Unraid (based on the RayLabsHQ CA template)
All five are registered in the use-cases index; appended after the
existing entries so the homepage's featured three stay unchanged.
Interlinking pass across the existing articles: contextual inline
links where the text already referenced a topic, plus a short related
guides section per article. The comparison page's next steps now cover
the new guides, and the docs quickstart bridges to the use-case section.
Every article now carries 4 to 9 internal links; a build-time check
found 0 broken links across 440.
* feat: per-repository and per-organization mirror option overrides
Closes the gap reported in #361: a repository whose LFS fetch fails could
not be mirrored at all, because LFS was a single global switch. Gitea runs
the LFS fetch inside its own migration, so a failure aborts the whole
migration with nothing to salvage. The only fix available to us is to stop
asking for LFS on that repository.
Mirror options now resolve across three tiers, per flag, most specific
first: repository override, then organization override, then the global
config. NULL at a tier means inherit, so overriding LFS on one repo leaves
its other options alone.
Adds resolveMirrorOptions(), which replaces 24 scattered reads of
config.giteaConfig.* across three near-identical blocks: the two mirror
paths in gitea.ts and the sync path in gitea-enhanced.ts. The third was
easy to miss and mattered: without it, overrides would have been honored
on first mirror and silently ignored on every scheduled sync afterwards.
The starredCodeOnly clamp is folded into the resolver and deliberately
outranks explicit overrides, preserving the existing behavior that starred
repos mirror code only.
Editing is on the objects themselves, via the existing three-dot menus on
the Repositories and Organizations pages, not in Configuration, which keeps
holding the global defaults. A repositories filter and a row badge make it
possible to find which repos deviate.
Malformed override JSON degrades to "inherit" rather than throwing, so a
bad value can never break a mirror run.
* fix(ui): disable mirror toggles that cannot take effect, and explain why
A toggle the runtime will ignore should not look editable. Three cases
were doing exactly that, so they now share one mechanism:
getMirrorOverrideGating() returns a reason string per flag, and the dialog
disables the control and prints that reason underneath.
Starred clamp. When a repo is starred and starredCodeOnly is set, the
resolver forces every metadata flag off regardless of the override. The
dialog previously let the user set them anyway. The clamped set now lives
in STARRED_CLAMPED_KEYS, shared by the resolver and the gating helper so
the flags the UI disables cannot drift from the ones the runtime clamps.
LFS is deliberately excluded from that set, since turning LFS off per
repository is the point of #361, and a test pins that.
Labels. shouldMirrorLabels is `mirrorLabels && !mirrorIssues` in both
mirror paths, so labels cannot take effect while issues are mirrored. That
gate reacts live to the in-progress edit rather than only the saved state.
Inherit hint. For a repository the hint now reflects global -> org instead
of global alone, via a new name-keyed GET
/api/organizations/mirror-overrides that reuses
loadOrganizationMirrorOverrides. Personal repos skip the fetch, a failed
fetch degrades to the global values rather than blocking the dialog, and
the hint is suppressed while in flight so it is never briefly wrong.
Widens the UI to mirrorLabels and mirrorMilestones. mirrorMetadata stays
out: no mirror path reads it, so a per-object override would resolve
correctly and then do nothing. See the report for that finding.
useGiteaConfig additionally returns advancedOptions so callers can read
starredCodeOnly without a second request.
* fix(ui): read inherited mirror flags from mirrorOptions, not giteaConfig
/api/config does not return the mirror flags on giteaConfig. On the way
out, mapDbToUiConfig reshapes them into a separate mirrorOptions object
using different names (mirrorLFS, not lfs) and nesting the metadata flags
under metadataComponents with short names (issues, not mirrorIssues).
The dialog was written against the DB shape, so every lookup returned
undefined. Coerced to false, that is indistinguishable from a real "off",
which produced two symptoms: every inherit hint read "currently off"
regardless of the actual global config, and the labels gate never fired
because the effective issues value it keys on came from the same dead
source.
Adds mirrorOptionsToFlags() as the single conversion point, matching the
derivation in mapUiToDbConfig exactly, including that mirrorMetadata is a
master switch over the metadata components while lfs and mirrorReleases
sit outside it. Both dialogs now go through it, and useGiteaConfig returns
mirrorOptions alongside advancedOptions.
Server-side mirroring was never affected: the resolver reads config
straight from the DB, where the flags really do live on giteaConfig.
The existing tests could not catch this because they build a synthetic
config already in DB shape, so a client/server mismatch is invisible to
them. The new tests push flags through the real config-mapper in both
directions and assert the derived flags equal what mapUiToDbConfig would
persist, so they fail if that mapping changes again. One test pins the
root cause directly: that giteaConfig in the API payload carries no flags.
* fix(ui): surface the mirror-options filter on desktop, and on organizations
The overrides filter only existed inside the mobile filter drawer, so on
desktop the Repositories page showed a "Custom" badge on overridden rows
with no way to filter to them. Being able to answer "which repos deviate
from my defaults" on this page is the reason the Configuration-page
listing was dropped, so desktop could not do the one job it was given.
Adds the control to the desktop filter row next to status and sort, using
the bare Select-with-placeholder style those use rather than the drawer's
labelled markup. The mobile control is unchanged.
The Organizations page had a wider version of the same gap: it renders the
"Custom options" badge but never had this filter on either layout, and
OrganizationsList did not filter on it at all. Added the predicate plus
the control in both its layouts, so the two pages behave the same.
Also folds hasOverrides into activeFilterCount on both pages and into the
Organizations clear-all reset. Without that the mobile filter badge
undercounted an active overrides filter, and clearing filters left it set.
* fix(ui): stop double-applying the mirrorMetadata switch when reading config
mirrorOptionsToFlags ANDed each metadata component with mirrorMetadata on
the way in. That is a write-path rule: mapUiToDbConfig already applies it
when persisting, so a config saved through the settings UI has it baked
into the stored flag. Applying it again on read double-applies it.
The read path does not need it. mapDbToUiConfig puts the raw stored values
into metadataComponents, so metadataComponents.issues is exactly
giteaConfig.mirrorIssues, which is the field gitea.ts and
gitea-enhanced.ts read. Mapping the components straight through is 1:1
with runtime behavior.
Double-applying was invisible while the stored state was self-consistent,
since false && false is still false. It diverged when the state did not
come from the UI write path, which env vars allow: MIRROR_METADATA=false
with MIRROR_ISSUES=true stores mirrorMetadata:false, mirrorIssues:true.
The runtime mirrors issues; the dialog reported off. Measured against a
config in that state, five flags misreported.
I previously described this as an unrecoverable loss in the API shape.
That was wrong. Both values reach the client separately and unmerged, so
the payload carries full information and the loss was one we introduced.
Rewrites the test that asserted the derived flags match what
mapUiToDbConfig would persist. Comparing against the write derivation is
what pinned the bug in place. It now asserts the property that matters,
that flags derived from the API payload equal the stored flags the runtime
reads, plus a case checking agreement with resolveMirrorOptions on an
inconsistent config. Both fail if the AND returns.
* docs: move canonical documentation to the website
The website now hosts the full documentation at /docs with a proper docs
layout: sidebar navigation, on-page table of contents, mobile nav, theme
support, canonical and OG meta, and overflow-safe code blocks and tables.
Ten pages, all rewritten from the current code rather than copied from
the old in-app docs: quickstart, deployment (Docker, Helm, Nix, LXC,
bare metal), configuration, environment variable reference,
notifications (all four providers including webhook payload signing),
authentication (including header auth), force-push protection,
architecture, advanced, and custom CA certificates.
This fixes every inaccuracy found in the docs audit: the wrong
raylabs/gitea-mirror image name, JWT_SECRET presented as the live auth
secret instead of BETTER_AUTH_SECRET, the missing auth env vars, both
wrong DATABASE_URL defaults, the contradicting starred-org default, and
health endpoint fields the API deliberately does not return.
The in-app /docs pages are retired: a stub redirects old bookmarks to
the website, and the sidebar and 404 links point there directly. The
markdown files under docs/ stay as the versioned offline reference;
NOTIFICATIONS.md now covers Gotify and Webhook. README links the docs
site, mentions notifications, and drops stale version markers. The app
viewport meta gains initial-scale=1.
* docs: mark new-repo notification as unimplemented, bump helm appVersion to 3.24.0
* feat(ui): redesign dashboard and configuration screens
New settings design language built from the design/giteamirror.pen file:
cards with icon headers and status footers, header-level enable switches,
toggle switches instead of checkboxes, uppercase section titles, selection
tiles with icon chips and a check on the active option, segmented controls,
and an indigo accent. Implemented via shared primitives in
src/components/config/settings-ui.tsx and applied across:
- Automation: header switches, schedule card, one-line auto-mirror copy
with info tooltip, full-width Repository Cleanup card with Skip/Archive/
Delete tiles and dry run row
- Notifications: segmented provider picker (ntfy/Apprise/Gotify/Webhook),
events card with per-event switches
- Connections: GitHub/Gitea connection cards with token creation guide and
field helpers, Repository Selection and Mirror Content cards covering
every mirror option, Organization Structure card with strategy tiles,
destructive update protection tiles (BETA label removed)
- Authentication: sign-in methods status card, identity providers restyle
- Dashboard: flatter stat cards, icon panel headers, indigo view-all links
All existing state handling, autosave and API behavior is unchanged.
Light mode keeps working via theme tokens. README and website screenshots
regenerated, docs references to renamed cards updated.
* fix(ui): design polish pass from local review
- Recent Activity rows get status icon circles (check, sync, sparkles, alert)
- Connections tab restructured: connection cards share a stretched grid row
so GitHub and Gitea stay equal height; Mirror Content moved to the right
column; forms split into placeable cards via a part prop
- Token guide panel: link moved to header as icon, larger text; redundant
card footers removed (scopes line, test-connection hint)
- Repository Selection gains a footer note; retention explanation moved
below the selector
- Authentication tab matches the design: side-by-side cards, row dividers,
disabled state-reflecting switches with info hints, footers; SSO dialog
restyled (segmented protocol tabs, field labels, indigo primary)
- Import GitHub Data button is the indigo primary; disabled state is muted
- Time format menu redesigned (locale pill, live examples, live clock in
the trigger); theme switcher moved to sidebar as icon segmented control,
system preference now persists correctly; legacy ModeToggle removed
- Automation timezone pill no longer shows stored legacy UTC as a choice
- Config tab bar wraps 2x2 on narrow screens instead of overflowing
- README, website and PR screenshots regenerated
better-auth 1.6.23 (fixes GHSA high stored XSS via javascript: redirect_uri), esbuild >=0.28.1 override for www (GHSA low), Astro 7 / @astrojs/node 11 / @astrojs/react 6 / @astrojs/mdx 7 / lucide-react 1.x (inline GitHub icon replaces removed brand icon) and all in-range updates on both the app and the website. The remaining @better-auth/oauth-provider medium advisory is patched upstream only in 1.7-rc and will be picked up when 1.7 stable lands.
Timestamps now follow the browser locale by default (previously hardcoded en-US/12-hour), with a clock toggle in the header for Auto / 12-hour / 24-hour, persisted in localStorage.
Fixes#342
Adds stuck-status recovery: repositories (and orgs) stranded in an in-flight status by a crash/restart are reset to failed with an explanation, on container start and every scheduler tick, guarded by the existing 2h liveness window.
Fixes#339
Config saves now preserve every field the settings form doesn't expose (mirror interval and other env-only options) instead of resetting them to defaults.
Fixes#338
Three related fixes for the "repos keep getting archived and then fail to
sync with HTTP 405" report:
1. Orphan cleanup no longer archives on bulk-list absence alone.
Repos added via the "+" Add Repository dialog (foreign owner, not
starred) can never appear in the authenticated bulk fetches, so every
cleanup cycle deterministically flagged them as orphaned and archived
them. identifyOrphanedRepositories() now runs a targeted per-repo
confirmation (starred check or repos.get) and only treats a clean 404
as gone; any other outcome fails safe.
2. The archived-* rename is persisted. archiveGiteaRepo() now returns
the actual post-rename name and the cleanup service records it in
mirroredLocation, so the DB no longer points at a name that only
301-redirects.
3. Sync self-heals repos renamed in Gitea/Forgejo. Requests to a
renamed repo get a 301; fetch follows it, downgrading POST to GET,
which lands on the POST-only mirror-sync endpoint as a 405. The sync
candidate loop now adopts the canonical owner/name from the GET
response body before POSTing, tries an archived-{name} fallback for
archived repos (guarded by an original_url source match), and keeps
archived repos archived: no mirror-interval PATCH, status stays
'archived' per the documented Manual Sync contract.
Also hardens mirrorGitHubReleasesToGitea to derive GitHub coordinates
from fullName so Gitea-side names can never leak into GitHub API calls.
Verified end-to-end on Forgejo 15.0.3 (rootless): pre-fix reproduces the
exact 405; post-fix the stale-name sync succeeds (GET stale -> 301, GET
canonical -> 200, POST canonical mirror-sync -> 200), archived repos keep
interval "0s" with no PATCH issued, and non-archived renamed repos heal
and get the configured interval applied.
* fix(releases): send Gitea release title as `name`, not `title` (#334)
Gitea/Forgejo expose the release title through the JSON field `name`
(the API Go struct is `Title string `+"`"+`json:"name"`+"`"+`). The release
create and update payloads sent `title:` instead, which Gitea silently
ignores, so every mirrored release landed with a blank title.
Verified live against Gitea 1.24.7: a POST/PATCH with `title` yields
`name: ""`; the same call with `name` sets the title correctly. The
update path also self-heals previously-mirrored releases whose names were
left blank, since the existing-vs-expected name comparison already drives
a PATCH.
Adds gita-release-name.test.ts, which drives the real
mirrorGitHubReleasesToGitea create/update paths with a mocked fetch and
asserts the payload carries `name` (and never `title`).
* fix(issues): reconcile labels on issue/PR update via the labels sub-resource (#334 sibling)
Gitea/Forgejo's `EditIssueOption` has no `labels` field (only
`CreateIssueOption` does), so a `labels` key in a `PATCH .../issues/{index}`
body is silently dropped — the same silent-ignore class as the release
`title` vs `name` bug. The issue and PR-as-issue update paths sent `labels`
in the PATCH body, so label changes never propagated onto already-mirrored
issues (and a deadlock-orphaned issue recovered via PATCH never got its
labels).
Fix: add `reconcileGiteaIssueLabels`, which replaces the label set via
`PUT .../issues/{index}/labels` (idempotent — adds new, removes deleted).
Call it on the two issue update paths and the two PR-issue update paths,
and drop the dead `labels` key from those PATCH bodies. Labels on freshly
created issues still come from CreateIssueOption on the POST.
Verified live against Gitea 1.24.7 (PATCH ignores labels; PUT applies them)
and end-to-end (a drifted mirrored issue reconciled from no-labels to its
GitHub label set). Adds gitea-issue-labels.test.ts driving the real
mirrorGitRepoIssuesToGitea update path; the test carries a self-contained
http-client mock so it is immune to another suite's global module mock.
* test: make #334 regression tests deterministic via pure payload builders
The prior tests drove the real mirror functions with a global `fetch` mock.
That is order/version-fragile: another suite installs a process-global
`mock.module("@/lib/http-client")`, and bun 1.3.13 (CI) runs test files
concurrently, so `globalThis.fetch` races across files and
`isRepoPresentInGitea` (raw fetch) intermittently sees the wrong mock —
green locally on bun 1.3.6, red in CI.
Extract the payload construction into pure, exported builders and assert on
those instead (the repo's existing `classify*` pattern): buildGiteaReleasePayload
(create+update send `name`, never `title`), buildGiteaIssueEditPayload (edit
body never carries `labels`), buildGiteaIssueLabelsPayload (labels sub-resource
body). Behavior is unchanged — the builders return the exact same objects the
call sites built inline — and the fixes remain verified live on Gitea 1.24.7.
Release creation failed on some Gitea/Forgejo instances with
"HTTP 404: The target couldn't be found", so no release (and therefore no
assets) was ever created — re-syncing never recovered.
Root cause: the create payload always sent `target: target_commitish`
(e.g. "main"). When the release's git tag is not yet present in the Gitea
mirror — which happens when Gitea's own git mirror clone lags behind the
metadata sync — Gitea tries to *create* the tag from `target`; if that ref
can't be resolved it returns a generic 404 ("The target couldn't be
found"), and if it can, it would create a brand-new tag at the wrong commit.
Reproduced the reporter's exact stack (Forgejo 15 rootless + read_only +
cap_drop ALL + postgres, plus Gitea 1.20-1.26 and Forgejo 1.21-15): a
healthy repo always succeeds — the 404 only occurs when the tag is absent
at create time.
Fix:
- Before creating a release, verify the git tag already exists in Gitea.
If it isn't synced yet, skip it (logged) and let a later sync create it
once the mirror has the tag — never create a tag via `target`.
- Drop the `target` field from both the create and update payloads. For a
mirror the tag is synced from upstream, so Gitea attaches the release to
the existing tag; `target` is unnecessary and is what triggers the 404.
- Surface skipped-missing-tag releases in the summary log for diagnosability.
Verified end-to-end against the real mirror function on a Forgejo instance:
a release whose tag exists is created with its assets; a release whose tag
was removed is skipped cleanly (no 404, no bogus tag) and picked up once the
tag is present.
Release assets were uploaded only on the create path of
mirrorGitHubReleasesToGitea(). When a Gitea release already existed, the
update path PATCHed the changelog/title and `continue`d without ever
touching assets. So any release whose assets were not fully uploaded on
that single create-path run — first sync interrupted, a transient
download/upload failure, large multi-MB assets, etc. — stayed permanently
asset-less, and re-syncing always hit the update path and could never
recover it. Asset failures were also swallowed to console.error, so the
job still reported success (the "no errors in the logs" in #331).
Reproduced on a real Forgejo pull-mirror with shauninman/MinUI: a GitHub
release with two ~35-40MB binaries became a Gitea release with 0 assets
(just Forgejo's auto-generated source archive), and re-syncing left it at
0 while logging "Updating existing release".
Fix:
- Add reconcileReleaseAssets(), an idempotent reconciler run on BOTH the
create and update paths. It compares Gitea's existing attachments to
GitHub's by name+size: skips matches, uploads missing ones, replaces
size-mismatched copies. Existing broken releases self-heal on next sync.
- Extract the pure decision into classifyAssetsForReconciliation() for
unit testing (the absence of asset tests is why this slipped past #310).
- Surface asset upload counts in the summary and emit a visible warning
when any fail, instead of silently swallowing them.
- Add 5 unit tests covering the broken state, partial backfill,
idempotency, size-mismatch replacement, and the no-assets case.
Verified end-to-end against the real function on a Forgejo pull-mirror:
0 -> 2 assets backfilled, already-present assets skipped (no re-download),
second run uploads 0.
The "Name collision strategy" dropdown (starredDuplicateStrategy) never
persisted: the field was absent from both directions of the UI<->DB config
mapper. On save, mapUiToDbConfig dropped it before the DB write; on load,
mapDbToUiConfig never read it, so the UI reset to the "suffix" (repo-owner)
default. Mirror logic in gitea.ts then read undefined and also defaulted to
suffix — so repos really were created with that pattern regardless of the
user's choice. It has been broken since the field was introduced.
- Map starredDuplicateStrategy in mapUiToDbConfig and mapDbToUiConfig
- Add STARRED_DUPLICATE_STRATEGY env var for parity (reporter could not
work around it via compose because no env var existed) + docs
- Round-trip tests covering save, load, and the missing-field default
Repository discovery requested the `organization_member` affiliation
unconditionally, so repos from every org a user belongs to were imported —
even orgs they never explicitly added. `skipPersonalRepos` only dropped
user-owned repos and left org repos unfiltered, which surprised users who
expected "only mirror org repos" to mean "only the orgs I chose" (reported
on #304).
Wire up the previously-dormant `includeOrganizations` config field as an
opt-in allowlist: when non-empty, only repos owned by the listed
organizations are imported. Empty = all org repos (backward-compatible).
Owned and collaborator repos are never restricted, so it composes cleanly
with `skipPersonalRepos`.
- Filter org repos by the allowlist in getGithubRepositories
- Add includeAllOrgsOverride so the cleanup service bypasses the allowlist
and never false-orphans a previously-mirrored repo from an org the user
later removes from the list
- UI control under Filtering & Behavior; INCLUDE_ORGANIZATIONS env var
- Case-insensitive dedup/trim in the UI<->DB mapper round-trip
- 7 unit tests covering the filter, composition, and the cleanup override
- Add `skipPersonalRepos: z.boolean().default(false)` to githubConfigSchema
- Filter out user-owned repos in getGithubRepositories when flag is true
- Wire ONLY_MIRROR_ORGS env var to skipPersonalRepos in env-config-loader
- Add checkbox UI in GitHubMirrorSettings Filtering & Behavior section
- Round-trip skipPersonalRepos through config-mapper (UI ↔ DB)
- Add skipPersonalRepos to AdvancedOptions TypeScript type
- Mark include/exclude arrays in configSchema as unused/reserved
- Update ENVIRONMENT_VARIABLES.md to document ONLY_MIRROR_ORGS effect
* fix(scheduler): make enabled flag authoritative for auto-start
checkAutoStartConfiguration() and performInitialAutoStart() previously
used `scheduleEnabled || hasMirrorInterval`, allowing a configured
GITEA_MIRROR_INTERVAL to trigger boot-time auto-start even after the
user disabled scheduling via the UI toggle.
env-config-loader already writes scheduleConfig.enabled=true when
GITEA_MIRROR_INTERVAL is set at container startup, so the interval is
a timing detail, not an enable signal. The documented env-var contract
is preserved: GITEA_MIRROR_INTERVAL at boot → env-config-loader sets
enabled=true → auto-start fires. But a later UI disable now sticks.
Add a focused unit test for the gate logic.
* fix(backup): always derive clone URL from user-configured Gitea URL
The pre-sync backup preferred repoInfo.clone_url, which reflects
Gitea's ROOT_URL setting. In Tailscale MagicDNS deployments (and any
setup where ROOT_URL is an external address), this URL is unreachable
from the app itself, causing bundle backup to fail.
Always build the clone URL as:
${config.giteaConfig.url.trimEnd('/')}/${owner}/${repo}.git
This matches the URL the app already uses for all other Gitea API
calls and is guaranteed reachable.
* feat(jobs): cancel-pending endpoint + fix misleading Delete All copy
Add POST /api/job/cancel-pending that sets the current user's
repositories with status "imported" or "failed" to "ignored",
preventing the scheduler from re-queuing them. In-flight "mirroring"
rows are left alone. Returns the count and logs one activity entry.
Fix the "Delete All Activities" dialog to clearly state it only clears
the history log and does not stop pending work. Rename button/title to
"Clear History" so intent is unambiguous.
Add a "Stop Pending Mirrors" button (StopCircle icon, amber) in both
mobile and desktop activity log toolbars, with a confirmation dialog
explaining repos are set to Ignored and can be re-enabled from the
Repositories page.
* fix(sync): actionable 405 error for non-pull-mirror repos
Gitea returns HTTP 405 with an empty body when the target repository is
no longer a pull mirror — e.g. the mirror was auto-disabled by Gitea or
the repository lost its mirror state after a manual edit.
Previously this fell through to the generic error handler which stored
the raw HttpError message (often empty) giving the user no guidance.
Now a 405 response is caught alongside the existing 400 handler and
sets the repository to "failed" with an actionable error message:
"Gitea reports this repository is not a pull mirror (HTTP 405).
In Gitea check Settings → Mirror Settings; if the mirror section is
missing, delete the repository in Gitea and re-mirror it from
gitea-mirror."
The same message is written to the activity log for visibility in the
dashboard.
Root cause (Theory A): the `needsRecreation` check compared GitHub
published_at-based expected indices against Gitea's API order. Gitea mirror
repos sort releases by tag-commit date, which can permanently disagree with
published_at order (e.g. unaconfig_dart v0.1.0 published after v0.1.1 but
tagged before). This made `currentExpectedIdx < nextExpectedIdx` evaluate
true on every sync, triggering delete-all-and-recreate forever — spamming
Gitea's activity feed with "released X" events (#310).
Fix: replace the destructive order-check machinery with set-based
reconciliation via `classifyReleasesForReconciliation`. Releases are
created when missing in Gitea and skipped (or PATCH-updated if content
drifted) when already present. No deletions are ever triggered by ordering.
Retain the existing release-limit trimming (retention cleanup) unchanged.
Also removes the 1-second per-release delay that was only needed for the
creation-order dance, significantly speeding up initial mirrors.
Adds unit tests covering: normal ordered repos, the unaconfig_dart inversion
fixture, missing→create, present→skip, and edge cases.
Starred (and other) repos duplicated on every re-mirror (starred/Repo,
Repo-owner, Repo-owner-1, ...) because the existence check only asked
"does a repo with this name exist?" and never "is the existing repo a
mirror of THIS same source?". The repo's own prior mirror counted as a
collision, so generateUniqueRepoName bumped to the next suffix each run,
repointing mirroredLocation at the newest copy. Under a single re-call,
3 concurrent/retried jobs each computed a DIFFERENT suffixed name, so the
location-based in-flight guard never matched and the race produced extra
copies.
Fix (source-identity aware):
- New shared helper src/lib/utils/mirror-source-match.ts:
- normalizeCloneUrl / cloneUrlsMatch: credential-, .git-, slash- and
host-case-insensitive clone URL comparison.
- isMirrorOfSource: a Gitea repo is "ours" only if it is a mirror AND
its original_url matches this repo's source.
- findExistingMirror: resolves an existing same-source mirror via the
recorded mirroredLocation first (survives strategy changes — #309),
then the base candidate name.
- classifyCandidateName: pure available/reusable/taken decision.
- gitea-enhanced: export GiteaRepoInfo and add original_url (Gitea's
recorded migration source) for source matching.
- Both create paths (mirrorGithubRepoToGitea, mirrorGitHubRepoToGiteaOrg):
run findExistingMirror BEFORE name generation; on a hit, reuse that
location and route into the existing "already mirrored" handling rather
than calling generateUniqueRepoName. Names now converge under
concurrency so the in-flight guard becomes effective.
- generateUniqueRepoName is now source-aware: an occupied name held by a
mirror of the SAME source is reused (no suffix); suffixing only happens
on a genuine different-source collision, preserving #95/#236 behavior.
The per-user DB claim check is retained so two users mirroring the same
source into a shared org stay separated.
- Phantom-fork guard (#309): the existingRepoInfo.mirror branches now
verify same-source before marking "mirrored"; on mismatch they fall
through to unique-name generation and create a separate mirror.
- Scheduler: a `failed` repo whose mirroredLocation still resolves to a
live same-source mirror is routed to syncGiteaRepo instead of re-create,
breaking the failed-metadata re-create loop cheaply.
- Remove dead src/lib/starred-repos-handler.ts (zero importers across all
git history); its correct base-name/.mirror reuse logic now lives in the
shared helper.
Tests: src/lib/utils/mirror-source-match.test.ts (30 cases) covers URL
normalization, reuse at base name, reuse via mirroredLocation across a
strategy change, genuine different-source collision (suffix), phantom
fork, stale mirroredLocation fallback, per-user DB-claim separation, and
the suffix-vs-reuse classification. Full suite: 319 pass, 0 fail.
Migration 0013 runs as a single transaction that rebuilds `organizations`
and then `ALTER TABLE sso_providers ADD saml_config` / `ADD domain_verified`.
On instances where those columns were already created outside Drizzle (declared
in schema.ts and added via db:push / an SSO-register round-trip on an
intermediate build), the ADD throws "duplicate column name: saml_config". That
rolls back the entire 0013 transaction, so 0013 is never recorded in
`__drizzle_migrations` and is retried — failing identically — on every boot,
crash-looping the server.
Add a pre-migrate repair (mirroring the existing repairFailedMigrations() for
the 0009 case): when 0013 is unrecorded but the columns already exist, preserve
any real SAML provider config, drop the stranded columns so the canonical 0013
runs in full (organizations rebuild included), then restore the preserved
values once the columns are re-added. No-op on fresh installs, clean upgrades,
and already-migrated databases.
This lets affected instances recover automatically on the next boot after
upgrading — no manual SQLite surgery required.
- src/lib/db/migration-repairs.ts: repairDuplicateSsoColumns + restoreSsoDataAfter0013
- src/lib/db/index.ts: wire both around migrate()
- scripts/validate-migrations.ts: cover the broken-upgrade + data-preservation path
Resolves#306. SSO sign-in via OIDC (Authentik / Keycloak / etc.) now links the
SSO identity to an existing email/password admin instead of bouncing to /login
with `?error=UNKNOWN`. Account-linking is gated on the operator-supplied
**Domain** field — cross-domain claims from a compromised IdP are refused.
Also bundles the deprecated `oidcProvider` → `@better-auth/oauth-provider`
migration. **Operators using the OAuth-provider feature must rotate registered
client secrets after upgrade** (legacy plaintext → hashed storage; see the
0012 migration notes).
Verified end-to-end on the pr-307 image against a real Authentik instance:
SSO login lands on the dashboard, `accounts` table gets both `credential` and
`authentik` rows for the same user. See PR description for full details.
Header / forward authentication has been end-to-end broken since the
v3 rewrite. The middleware populated `context.locals.user` from
trusted upstream headers (Authentik / Authelia / oauth2-proxy /
Caddy), but never minted a Better Auth session, and never set a
cookie. Server-rendered pages saw the user, but the React SPA's
`/api/auth/get-session` call hit Better Auth's handler — which only
reads its session cookie — and got `null`. The auth guard then
redirected to `/login`, even though the upstream proxy had already
authenticated the user.
Reported on issue #29 by @lanrat with a clean repro on v3.16.1.
Fix: add a small Better Auth plugin (`header-auth`) that exposes
`POST /api/auth/sign-in/header`. The endpoint validates the trusted
headers via `authenticateWithHeaders`, creates a real session row via
`internalAdapter.createSession`, and attaches the `Set-Cookie` via
`setSessionCookie` — the same pattern the magic-link, anonymous, and
phone-number plugins use after their respective verification steps.
The Astro middleware now calls this endpoint when no cookie session
exists and header auth is enabled, forwards the `Set-Cookie` onto the
outbound response, and populates `context.locals` from the minted
session. After the first request the browser has the cookie; every
subsequent request takes the normal cookie-auth fast path and the
bridge doesn't fire.
Fail-open everywhere: any endpoint failure (header auth disabled,
auth rejected, DB blip, malformed response) returns null from the
bridge and the request proceeds as anonymous. A broken header-auth
configuration must never lock everyone out of the cookie-auth path.
Tests:
- `auth-header.test.ts` — unit tests for `extractUserFromHeaders` and
`isHeaderAuthEnabled`, including lanrat's reported config shape
(same header for username and email).
- `auth-header-plugin.test.ts` — locks down plugin id, endpoint key,
path, and method so an accidental rename can't silently break the
middleware bridge.
- `auth-header-bridge.test.ts` — covers the cookie-extraction logic
and the fail-open paths (non-2xx, thrown error, malformed JSON,
missing fields, no Set-Cookie attached).
Stacks on top of #301 (better-auth 1.6.11 bump).
Refs: #29
Updates `better-auth` and `@better-auth/sso` from 1.5.5 to 1.6.11 to
pick up the patch fixes that have landed since (OAuth state CSRF
verification, scrypt non-blocking password hashing, account cookie
comparison fix, session freshness alignment, etc.). All 270 local
tests pass against the new version.
The only behavioral surface to watch:
- `freshAge` now aligns with session `createdAt` instead of
`updatedAt`. Not applicable here — we don't gate anything on
session freshness.
- 1.6.2 adds OAuth state-parameter CSRF verification. Applies to
OIDC/SAML SSO flows; no code changes needed.
- 1.6 emits a deprecation warning for `oidc-provider` in favor of
`@better-auth/oauth-provider`. The plugin still works in 1.6.x;
migrating it is a separate cleanup.
Prep work for an upcoming header-auth fix (issue #29 follow-up).
Same hardening principle as #293 for GitHub Actions: pin to an immutable
identifier so a future tag move can't silently change what we build against.
- Bumps oven/bun from 1.3.13 to 1.3.14 (released 2026-05-13)
- Pins to multi-arch digest sha256:9dba1a1b...db6f rather than just the tag
- Applies to both base and runner stages
Adds an account menu under the avatar in the header with Change password and
Change email actions, each opening a small dialog. Calls Better Auth's existing
change-password / change-email endpoints — no new API routes.
- Enables `user.changeEmail` in Better Auth with `updateEmailWithoutVerification`
since the app runs with email verification disabled and no email sender is
wired up
- Hides Change password for SSO-only users (no `credential` provider account),
fails open if the listAccounts probe errors
- Resolves discussion #291 ("Change user password/email")
Tags are mutable. A compromised maintainer (or a maintainer's compromised
machine) can force-move v-tags to point at malicious commits, and any workflow
using `@vN` picks up the malicious code on its next run — see the recent
`actions-cool/issues-helper` / `maintain-one-comment` incident exfiltrating
credentials from `Runner.Worker` memory.
This commit pins every third-party action in the two workflows that handle
secrets (GHCR push, Docker Hub login, Scout token) to immutable 40-char SHAs,
with a trailing comment naming the release version for readability. SHAs are
the latest released tag at time of pin.
The two DeterminateSystems actions were on `@main` — a *branch* ref that moves
on every push, materially worse than a tag — and are now pinned to the latest
release SHAs (v22 / v13).
First-party `actions/*` and `github/codeql-action` are left on tags for now;
they're a separate, lower-risk follow-up.
Removed 5 overrides whose constraints have since been picked up naturally
by the transitive dep graph. Verified by removing each and confirming the
resolved version (and the dep tree) is identical to what the override
produced:
- defu ^6.1.7 → still resolves 6.1.7
- fast-xml-parser ^5.5.6 → still resolves 5.5.6
- node-forge ^1.3.3 → package not in tree at all (override was dead)
- rollup >=4.59.0 → still resolves 4.59.0
- svgo ^4.0.1 → still resolves 4.0.1
Kept overrides that are still doing real work:
- @esbuild-kit/esm-loader → npm:tsx@^4.21.0 — deliberate replacement shim
- @xmldom/xmldom ^0.8.13, devalue ^5.8.1, fast-uri ^3.1.2,
fast-xml-builder ^1.1.7, kysely ^0.28.17 — active CVE pins (#289)
- lodash ^4.18.1 — pins to the newer 4.18.x line over the legacy 4.17.x
that transitive deps still pull
- picomatch ^4.0.4 — without it, picomatch@2.3.2 is added as a duplicate
copy via a transitive that asks for 2.x
Future drift would be caught by Dependabot + the weekly Docker Scout
scan; the overrides above remain because they currently affect the tree.
Patches 9 Docker Scout HIGH alerts surfaced by the weekly image scan:
- @xmldom/xmldom 0.8.12 → 0.8.13 (CVE-2026-41672/3/4/5)
- devalue 5.6.4 → 5.8.1 (CVE-2026-42570)
- kysely 0.28.16 → 0.28.17 (CVE-2026-44635)
- fast-uri (new override) → ^3.1.2 (CVE-2026-6321, CVE-2026-6322)
- fast-xml-builder (new override) → ^1.1.7 (CVE-2026-44665)
All five resolve to fixed versions after `bun install`. Tests and astro
build pass locally.
Remaining open Docker Scout alerts (git-lfs Go stdlib, gnutls28, nghttp2)
are base-image or upstream-binary issues, not addressable via npm.
GitHub's listForAuthenticatedUser defaults to returning every repo the
user has access to (owner + collaborator + organization_member), which
imports a lot of noise for users who only want their own repos.
Adds an `includeCollaboratorRepos` toggle, defaulting to true to preserve
existing behavior. When disabled, the affiliation filter scopes the API
call to "owner" only.
The cleanup service overrides the filter to always include collaborator
repos when computing the "what's still on GitHub" list. Without this,
toggling the option off would mark previously-mirrored collab repos as
orphaned and archive/delete them from Gitea.
Wired through the schema, both UI<->DB mappers, the env-config loader
(with new INCLUDE_COLLABORATOR_REPOS env var), and the settings UI.
The fix in v3.15.8 made scheduleConfig.autoMirror an independent trigger
in the scheduler, but it remained reachable only via the AUTO_MIRROR_REPOS
env var. This adds a UI checkbox under the Automatic Syncing section so
the option can be toggled per-config without touching the environment.
The toggle is conditional on scheduling being enabled (since auto-mirror
without a scheduler is meaningless) and is independent of the existing
"Auto-mirror new starred repositories" toggle in GitHub settings. Together
they cover the full owned/starred matrix that the scheduler already
supports.
Plumbing: config-mapper.ts now round-trips autoMirror through the UI/DB
boundary, and ScheduleConfig in types/config.ts gets the matching field.
No schema or migration change — autoMirror was already in the zod schema.
The "Auto-mirror new starred repositories" checkbox in the GitHub settings
was a filter layered on top of scheduleConfig.autoMirror, which itself is
only settable via the AUTO_MIRROR_REPOS env var (no UI). So users who
checked the box saw their starred repos auto-imported but never mirrored.
Treat autoMirror and autoMirrorStarred as independent triggers in the
scheduler: autoMirror covers owned (and self-starred) repos, autoMirrorStarred
covers repos starred from other owners. Either flag on its own is enough
to enter the auto-mirror phase, and the filter scopes the work accordingly.
Also normalize the owner comparison to lowercase since GitHub usernames are
case-insensitive — previously a self-starred repo whose stored owner casing
differed from the configured owner would be misclassified as a third-party
star.
Behavior change worth flagging in release notes: anyone who currently has
the starred checkbox on (broken state) will start getting starred repos
mirrored on upgrade. AUTO_MIRROR_REPOS=true users see no change.
* fix: hoist migrateSucceeded above try so catch can update DB on failure (fixes#268)
`let migrateSucceeded` was declared inside the try block of
mirrorGithubRepoToGitea and mirrorGitHubRepoToGiteaOrg, but the catch
block referenced it. Block-scoping made it invisible to catch, so any
error inside the try (network timeout, transient 5xx, etc.) crashed the
catch with `ReferenceError: migrateSucceeded is not defined` before
reaching the DB update that marks the repo "failed". Result: repos
stuck in "mirroring" forever with no entry in the activity log.
Hoisting the declaration above the try restores the intended behavior:
catch updates the repo to failed, clears mirroredLocation when migrate
hadn't succeeded, writes a failed activity-log entry, and re-throws
with the original error message preserved.
TypeScript was flagging this as "Cannot find name 'migrateSucceeded'"
but esbuild stripped the types during build, so the bug shipped.
* test: replace integration test with structural source check (#268)
The behavioral version of this regression test passed locally but
failed in CI because of mock.module pollution between files: bun's
mock.module is process-wide, so my mock for @/lib/gitea-enhanced
leaked into gitea-enhanced.test.ts (its real-module assertions saw
my null-returning mocks), and gitea-enhanced.test.ts's own
@/lib/http-client mock could supersede mine depending on file
discovery order, causing my mirrorGithubRepoToGitea call to not
throw at all in CI.
Replace with a structural assertion that reads gitea.ts and verifies
`let migrateSucceeded` is declared before the outermost try in both
mirrorGithubRepoToGitea and mirrorGitHubRepoToGiteaOrg. Verified the
new test fails on the pre-fix source with a clear error message
pointing to issue #268, and passes on the fixed source.
* fix: honor GH_API_URL across all Octokit call sites
Six Octokit call sites constructed `new Octokit(...)` directly instead of
going through `createGitHubClient()`, so `GH_API_URL` (and the
`GITHUB_API_URL` fallback) only applied to the handful of flows that used
the helper. For GHES / GHEC-with-data-residency users this surfaced most
visibly as the "Test Connection" button hitting `api.github.com/user`
and failing with 401 even when `GH_API_URL` was set correctly (#269).
Route everything through `createGitHubClient()`:
- src/pages/api/github/test-connection.ts (the reported failure)
- src/pages/api/sync/repository.ts (public-repo sync)
- src/lib/gitea-enhanced.ts (force-push detection + metadata octokit)
- src/lib/scheduler-service.ts (auto-discovery, auto-mirror, auto-start)
- src/tests/test-metadata-mirroring.ts (dev harness, for consistency)
Side benefit: scheduler + sync paths now also get throttling, rate-limit
tracking, and the standard User-Agent, which they were missing.
`createGitHubClient`'s `token` parameter is made optional so the
public-repo sync path (`new Octokit()` with no auth) can keep working.
Fixes#269
* fix: address review findings
- scheduler: pass config.githubConfig?.owner (the real DB field) instead
of ?.username, which doesn't exist on the DB row and was silently
resolving to undefined — matches every other DB-reading call site.
- sync/repository.ts: revert to bare Octokit for the unauthenticated
public-repo lookup to preserve fast-fail on the 60 req/hr limit.
Still reads GH_API_URL / GITHUB_API_URL inline so GHES / GHEC
data-residency users benefit. The throttling plugin's retry-with-
backoff is wrong UX for a one-shot button click.
- github.ts: revert createGitHubClient token back to required (no
remaining callers pass undefined after the above).
- gitea-enhanced.ts: make the leftover Octokit import type-only.
- test-connection.test.ts: replace mid-test mock.module re-call with a
mutable stub reference — safer against ESM live-binding semantics.
- Add metadata state guards for issues/PRs in syncGiteaRepoEnhanced to
prevent re-mirroring on every sync (fixes#262)
- Fix httpPut → httpPatch for Gitea release updates (HTTP 405 error)
- Add paginated release retention cleanup to enforce releaseLimit by
deleting the oldest excess releases (fixes#264)
- Always send auth credentials during migration for all repos, not just
private ones, preventing "terminal prompts disabled" on Forgejo (fixes#263)
- Add missing service: "git" to org migration payload
- Make buildGithubSourceAuthPayload return {} instead of throwing when
token is missing, allowing unconditional use
- Preserve existing backup settings in env-config-loader so UI-configured
backup strategy and retention values survive restart (fixes#267)
Closes#262, #263, #264, #267
* feat: add custom sync start time scheduling
* Updated UI
* docs: add updated issue 240 UI screenshot
* fix: improve schedule UI with client-side next run calc and timezone handling
- Compute next scheduled run client-side via useMemo to avoid permanent
"Calculating..." state when server hasn't set nextRun yet
- Default to browser timezone when enabling syncing (not UTC)
- Show actual saved timezone in badge, use it consistently in all handlers
- Match time input background to select trigger in dark mode
- Add clock icon to time picker with hidden native indicator
* feat: add notification system with Ntfy.sh and Apprise providers (#231)
Add push notification support for mirror job events with two providers:
- Ntfy.sh: direct HTTP POST to ntfy topics with priority/tag support
- Apprise API: aggregator gateway supporting 100+ notification services
Includes database migration (0010), settings UI tab, test endpoint,
auto-save integration, token encryption, and comprehensive tests.
Notifications are fire-and-forget and never block the mirror flow.
* fix: address review findings for notification system
- Fix silent catch in GET handler that returned ciphertext to UI,
causing double-encryption on next save. Now clears token to ""
on decryption failure instead.
- Add Zod schema validation to test notification endpoint, following
project API route pattern guidelines.
- Mark notifyOnNewRepo toggle as "coming soon" with disabled state,
since the backend doesn't yet emit new_repo events. The schema
and type support is in place for when it's implemented.
* fix notification gating and config validation
* trim sync notification details
* fix: improve reverse proxy support for subdomain deployments (#63)
- Add X-Accel-Buffering: no header to SSE endpoint to prevent Nginx
from buffering the event stream
- Auto-detect trusted origin from Host/X-Forwarded-* request headers
so the app works behind a proxy without manual env var configuration
- Add prominent reverse proxy documentation to advanced docs page
explaining BETTER_AUTH_URL, PUBLIC_BETTER_AUTH_URL, and
BETTER_AUTH_TRUSTED_ORIGINS are mandatory for proxy deployments
- Add reverse proxy env var comments and entries to both
docker-compose.yml and docker-compose.alt.yml
- Add dedicated reverse proxy configuration section to .env.example
* fix: address review findings for reverse proxy origin detection
- Fix x-forwarded-proto multi-value handling: take first value only
and validate it is "http" or "https" before using
- Update comment to accurately describe auto-detection scope: helps
with per-request CSRF checks but not callback URL validation
- Restore startup logging of static trusted origins for debugging
* fix: handle multi-value x-forwarded-host in chained proxy setups
x-forwarded-host can be comma-separated (e.g. "proxy1.example.com,
proxy2.example.com") in chained proxy setups. Take only the first
value, matching the same handling already applied to x-forwarded-proto.
* test: add unit tests for reverse proxy origin detection
Extract resolveTrustedOrigins into a testable exported function and
add 11 tests covering:
- Default localhost origins
- BETTER_AUTH_URL and BETTER_AUTH_TRUSTED_ORIGINS env vars
- Invalid URL handling
- Auto-detection from x-forwarded-host + x-forwarded-proto
- Multi-value header handling (chained proxy setups)
- Invalid proto rejection (only http/https allowed)
- Deduplication
- Fallback to host header when x-forwarded-host absent
* fix: prevent starred repo name collisions during concurrent mirroring (#95)
When multiple starred repos share the same short name (e.g. alice/dotfiles
and bob/dotfiles), concurrent batch mirroring could cause 409 Conflict
errors because generateUniqueRepoName only checked Gitea via HTTP, missing
repos that were claimed in the local DB but not yet created remotely.
Three fixes:
- Add DB-level check in generateUniqueRepoName so it queries the local
repositories table for existing mirroredLocation claims, preventing two
concurrent jobs from picking the same target name.
- Clear mirroredLocation on failed mirror so a failed repo doesn't falsely
hold a location that was never successfully created, which would block
retries and confuse the uniqueness check.
- Extract isMirroredLocationClaimedInDb helper for the DB lookup, using
ne() to exclude the current repo's own record from the collision check.
* fix: address review findings for starred repo name collision fix
- Make generateUniqueRepoName immediately claim name by writing
mirroredLocation to DB, closing the TOCTOU race window between
name selection and the later status="mirroring" DB update
- Add fullName validation guard (must contain "/")
- Make isMirroredLocationClaimedInDb fail-closed (return true on
DB error) to be conservative about preventing collisions
- Scope mirroredLocation clear on failure to starred repos only,
preserving it for non-starred repos that may have partially
created in Gitea and need the location for recovery
* fix: address P1/P2 review findings for starred repo name collision
P1a: Remove early name claiming from generateUniqueRepoName to prevent
stale claims on early return paths. The function now only checks
availability — the actual claim happens at the status="mirroring" DB
write (after both idempotency checks), which is protected by a new
unique partial index.
P1b: Add unique partial index on (userId, mirroredLocation) WHERE
mirroredLocation != '' via migration 0010. This enforces atomicity at
the DB level: if two concurrent workers try to claim the same name,
the second gets a constraint violation rather than silently colliding.
P2: Only clear mirroredLocation on failure if the Gitea migrate call
itself failed (migrateSucceeded flag). If migrate succeeded but
metadata mirroring failed, preserve the location since the repo
physically exists in Gitea and we need it for recovery/retry.
* fix: prevent excessive disk usage from repo backups (#234)
Legacy configs with backupBeforeSync: true but no explicit backupStrategy
silently resolved to "always", creating full git bundles on every sync
cycle. This caused repo-backups to grow to 17GB+ for users with many
repositories.
Changes:
- Fix resolveBackupStrategy to map backupBeforeSync: true → "on-force-push"
instead of "always", so legacy configs only backup when force-push is detected
- Fix config mapper to always set backupStrategy explicitly ("on-force-push")
preventing the backward-compat fallback from triggering
- Lower default backupRetentionCount from 20 to 5 bundles per repo
- Add time-based retention (backupRetentionDays, default 30 days) alongside
count-based retention, with safety net to always keep at least 1 bundle
- Add "high disk usage" warning on "Always Backup" UI option
- Update docs and tests to reflect new defaults and behavior
* fix: preserve legacy backupBeforeSync:false on UI round-trip and expose retention days
P1: mapDbToUiConfig now checks backupBeforeSync === false before
defaulting backupStrategy, preventing legacy "disabled" configs from
silently becoming "on-force-push" after any auto-save round-trip.
P3: Added "Snapshot retention days" input field to the backup settings
UI, matching the documented setting in FORCE_PUSH_PROTECTION.md.
SQLite rejects ALTER TABLE ADD COLUMN with expression defaults like
DEFAULT (unixepoch()), which Drizzle-kit generated for the imported_at
column. This broke upgrades from v3.12.x to v3.13.0 (#228, #229).
Changes:
- Rewrite migration 0009 using table-recreation pattern (CREATE, INSERT
SELECT, DROP, RENAME) instead of ALTER TABLE
- Add migration validation script with SQLite-specific lint rules that
catch known invalid patterns before they ship
- Add upgrade-path testing with seeded data and verification fixtures
- Add runtime repair for users whose migration record may be stale
- Add explicit migration validation step to CI workflow
Fixes#228Fixes#229
* feat: add autoMirrorStarred toggle for selective starred repo mirroring (#205)
Add `githubConfig.autoMirrorStarred` (default: false) to control whether
starred repos are included in automatic mirroring operations. Manual
per-repo actions always work regardless of this toggle.
Bug fixes:
- Cleanup service no longer orphans starred repos when includeStarred is
disabled (prevents data loss)
- First-boot auto-start now gates initial mirror behind autoMirror config
(previously mirrored everything unconditionally)
- "Mirror All" button now respects autoMirrorStarred setting
- Bulk mirror and getAvailableActions now include pending-approval status
Changes span schema, config mapping, env loader, scheduler, cleanup
service, UI settings toggle, and repository components.
* fix: log activity when repos are auto-imported during scheduled sync
Auto-discovered repositories (including newly starred ones) were inserted
into the database without creating activity log entries, so they appeared
in the dashboard but not in the activity log.
* ci: set 10-minute timeout on all CI jobs
* feat: smart force-push protection with backup strategies (#187)
Replace blunt `backupBeforeSync` boolean with `backupStrategy` enum
offering four modes: disabled, always, on-force-push (default), and
block-on-force-push. This dramatically reduces backup storage for large
mirror collections by only creating snapshots when force-pushes are
actually detected.
Detection works by comparing branch SHAs between Gitea and GitHub APIs
before each sync — no git cloning required. Fail-open design ensures
detection errors never block sync.
Key changes:
- Add force-push detection module (branch SHA comparison via APIs)
- Add backup strategy resolver with backward-compat migration
- Add pending-approval repo status with approve/dismiss UI + API
- Add block-on-force-push mode requiring manual approval
- Fix checkAncestry to only treat 404 as confirmed force-push
(transient errors skip branch instead of false-positive blocking)
- Fix approve-sync to bypass detection gate (skipForcePushDetection)
- Fix backup execution to not be hard-gated by deprecated flag
- Persist backupStrategy through config-mapper round-trip
* fix: resolve four bugs in smart force-push protection
P0: Approve flow re-blocks itself — approve-sync now calls
syncGiteaRepoEnhanced with skipForcePushDetection: true so the
detection+block gate is bypassed on approved syncs.
P1: backupStrategy not persisted — added to both directions of the
config-mapper. Don't inject a default in the mapper; let
resolveBackupStrategy handle fallback so legacy backupBeforeSync
still works for E2E tests and existing configs.
P1: Backup hard-gated by deprecated backupBeforeSync — added force
flag to createPreSyncBundleBackup; strategy-driven callers and
approve-sync pass force: true to bypass the legacy guard.
P1: checkAncestry false positives — now only returns false for
404/422 (confirmed force-push). Transient errors (rate limits, 500s)
are rethrown so detectForcePush skips that branch (fail-open).
* test(e2e): migrate backup tests from backupBeforeSync to backupStrategy
Update E2E tests to use the new backupStrategy enum ("always",
"disabled") instead of the deprecated backupBeforeSync boolean.
* docs: add backup strategy UI screenshot
* refactor(ui): move Destructive Update Protection to GitHub config tab
Relocates the backup strategy section from GiteaConfigForm to
GitHubConfigForm since it protects against GitHub-side force-pushes.
Adds ShieldAlert icon to match other section header patterns.
* docs: add force-push protection documentation and Beta badge
Add docs/FORCE_PUSH_PROTECTION.md covering detection mechanism,
backup strategies, API usage, and troubleshooting. Link it from
README features list and support section. Mark the feature as Beta
in the UI with an outline badge.
* fix(ui): match Beta badge style to Git LFS badge
* fix(nix): ensure absolute bundle path in pre-sync backup (#203)
Use path.resolve() instead of conditional path.isAbsolute() check to
guarantee bundlePath is always absolute before passing to git -C. On
NixOS, relative paths were interpreted relative to the temp mirror
clone directory, causing "No such file or directory" errors.
Closes#203
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* fix(nix): ensure absolute bundle path in pre-sync backup (#203)
Use path.resolve() instead of conditional path.isAbsolute() check to
guarantee bundlePath is always absolute before passing to git -C. On
NixOS, relative paths were interpreted relative to the temp mirror
clone directory, causing "No such file or directory" errors.
Extract resolveBackupPaths() for testability. Bump version to 3.10.1.
Closes#203
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* ci: drop macos matrix and only run nix build on main/tags
- Remove macos-latest from Nix CI matrix (ubuntu-only)
- Only run `nix build` on main branch and version tags, skip on PRs
- `nix flake check` still runs on all PRs for validation
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
* feat: add target organization field to Add Repository dialog
Allow users to specify a destination Gitea organization when adding a
single repository, instead of relying solely on the default mirror
strategy. The field is optional — when left empty, the existing strategy
logic applies as before.
Closes#200
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* docs: add screenshot of target organization field in Add Repository dialog
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
* fix(nix): enable sandboxed builds with bun2nix
The Nix package was broken on Linux because `bun install` requires
network access, which is blocked by Nix sandboxing (enabled by default
on Linux).
This switches to bun2nix for dependency management:
- Add bun2nix flake input to pre-fetch all npm dependencies
- Generate bun.nix lockfile for reproducible dependency resolution
- Copy bun cache to writable location during build to avoid EACCES
errors from bunx writing to the read-only Nix store
- Add nanoid as an explicit dependency (was imported directly but only
available as a transitive dep, which breaks with isolated linker)
- Update CI workflow to perform a full sandboxed build
- Add bun2nix to devShell for easy lockfile regeneration
Closes#197
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* fix(nix): create writable workdir for database access
The app uses process.cwd()/data for the database path, but when running
from the Nix store the cwd is read-only. Create a writable working
directory with symlinks to app files and a real data directory.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
The runner stage was missing git, causing pre-sync backups to fail with
"Executable not found in $PATH: git". The backup feature (enabled by
default) shells out to git for clone --mirror and bundle create.
Closes#196
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
ARUNAVO RAY