feat(web-ui): migrate drifted buttons to canonical Button + add raw-<… - #616
Merged
Conversation
…button> lint gate (byte5ai#290)
# Conflicts: # web-ui/app/_components/ChatTabs.tsx # web-ui/app/_components/StreamToasts.tsx
Weegy
added a commit
to sneumannb5/omadia
that referenced
this pull request
Aug 12, 2026
Merging current main into this branch surfaced three things the branch's last CI run (2026-08-03) could not have caught: - The raw-<button> lint gate added by byte5ai#616 on 2026-08-09 makes CopyIdButton a hard eslint error. It is a bespoke inline data chip whose own label is the key id in the mono-data register, so it takes the rule's documented escape hatch with a reason, matching the SlotEditor/graph precedent. - The `reloadSeqRef` out-of-order guard was asserted by no test: deleting the sequence check left all 15 tests green. Added a regression test that resolves the mount fetch after a post-create reload and asserts the stale list is discarded — it fails without the guard and passes with it. - Dropped the internal issue numbers from the admin card description (en + de). They mean nothing to an operator, and the catalog had only one other such reference in 3565 keys. Also adds the CHANGELOG entry under [Unreleased] that the orphaned twin of this work (byte5ai#551, merged into an already-squashed base and never reaching main) carried.
Weegy
added a commit
that referenced
this pull request
Aug 12, 2026
* feat(web-ui): admin UI for channel API keys (#567) * fix(web-ui): unblock api-keys admin UI on current main Merging current main into this branch surfaced three things the branch's last CI run (2026-08-03) could not have caught: - The raw-<button> lint gate added by #616 on 2026-08-09 makes CopyIdButton a hard eslint error. It is a bespoke inline data chip whose own label is the key id in the mono-data register, so it takes the rule's documented escape hatch with a reason, matching the SlotEditor/graph precedent. - The `reloadSeqRef` out-of-order guard was asserted by no test: deleting the sequence check left all 15 tests green. Added a regression test that resolves the mount fetch after a post-create reload and asserts the stale list is discarded — it fails without the guard and passes with it. - Dropped the internal issue numbers from the admin card description (en + de). They mean nothing to an operator, and the catalog had only one other such reference in 3565 keys. Also adds the CHANGELOG entry under [Unreleased] that the orphaned twin of this work (#551, merged into an already-squashed base and never reaching main) carried. --------- Co-authored-by: Weegy <mwege@byte5.de>
Weegy
added a commit
to sneumannb5/omadia
that referenced
this pull request
Aug 12, 2026
… guard Two things surfaced only after merging current main: - PR byte5ai#616's `no-restricted-syntax` gate for raw <button> landed on 2026-08-09, five days after this branch's last green CI run, and rejects six buttons on the datasets page. All six are the rule's own listed non-candidates (a bare text link in a table cell, a close link, and two pager pairs), so they get the documented escape-hatch comment with a reason, matching the idiom already used in admin/duplicates, admin/mcp and admin/providers. - The detailSeq generation guard was asserted by nothing: deleting it left the suite 11/11 green. The guard's real job is not "do not repaint a closed panel" (`selected` already covers that) but "a row page from the dataset you just closed must not land in the next dataset's panel" — and that mismatch is silent, because rows render as `selected.columns x row[col.name]", so foreign rows show up as blank cells rather than visibly wrong text. The new test pins it on the row footer instead, and goes red when the guard is removed.
This was referenced Aug 13, 2026
Closed
Weegy
added a commit
that referenced
this pull request
Aug 13, 2026
main's no-restricted-syntax rule (the Lume button ratchet, #616) landed after this branch was written and rejects the raw <button> the banner used for its dismiss action. Migrated to <Button variant="secondary">, matching ConfirmDialog's precedent for a cancel-style affordance; the uppercase/tracking classes are kept so it still reads as a pair with the adjacent settings link. web-ui lint: 0 errors (was 1).
Weegy
added a commit
that referenced
this pull request
Aug 13, 2026
…surfaced (#471) * feat(deploy): one-click Render blueprint + one-command Fly.io deploy render.yaml backs a 'Deploy to Render' button: middleware + admin UI from the published GHCR images, managed Postgres 17 (pgvector), a persistent /data disk, and a generated VAULT_KEY — the /setup wizard collects the LLM key at runtime, so a deploy needs zero pasted secrets. MIDDLEWARE_URL reaches the web-ui via fromService/envVarKey RENDER_EXTERNAL_URL. Fly has no blueprint-style button, so fly/deploy.sh is the one-command equivalent: three apps (pgvector/pgvector:pg17 private-only, middleware with /data volume, web-ui), volumes, and generated secrets. Fly's own Postgres offerings don't fit (postgres-flex ships without pgvector, Managed Postgres gates the vector extension behind a dashboard toggle), hence the same pgvector image the compose stack uses. Smoke-tested end to end on a live Fly org. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(web-ui): resolve MIDDLEWARE_URL at request time — route handlers replace baked rewrites Next evaluates rewrites() at build time and freezes the destinations into routes-manifest.json. The published web-ui image was built with ARG MIDDLEWARE_URL=http://middleware:8080, so every browser call through /bot-api proxied to a hostname that only exists on the compose network — on Fly/Render/any other host the proxy died with ENOTFOUND and every operator page showed 'GET … failed: 500'. Runtime MIDDLEWARE_URL was silently ignored (only server-side fetches honored it, which is why page shells still rendered). /bot-api/* and /p/* are now catch-all route handlers backed by app/_lib/middlewareProxy.ts, resolving MIDDLEWARE_URL per request: streaming both directions (SSE, plugin-ZIP uploads), multiple Set-Cookie preserved, redirects passed through, hop-by-hop headers stripped. One image now runs anywhere; the compose default still works because compose sets the same value as runtime env. Consequences swept: Dockerfile build ARG removed (with a warning against reintroducing it), CI workflows no longer pass the inert build-arg, the compose build overlay drops it, and desktop-apps.yml stops pretending to bake the kernel URL — the desktop supervisor already injects it at spawn (this also fixes the latent wrong-port proxy when the kernel doesn't get its preferred port). Verified: 6 regression tests (core assertion: same handler, changed env, changed target), plus a live Fly deployment where /bot-api/v1/auth/providers went 500 → 200 and login started working. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * feat(web-ui): first-run readiness banner; fix 17 broken t.rich placeholders; unbreak next dev Three first-run findings from smoke-testing the one-click deploy: 1. Readiness banner. A fresh install has no LLM key, so the orchestrator never publishes chatAgent@1/orchestratorRegistry@1 and every operator surface answers 503 multi_orchestrator_unavailable — the UI showed raw 'GET … failed: 503' strings with no hint at the cause. RuntimeReadinessBanner (mounted next to SessionWatcher) probes one operator route, and on the structured 503 shows a card naming the cause with a link to /admin/settings. Re-probes on focus + heartbeat while visible, so it clears itself the moment the key is saved. 2. t.rich placeholder syntax (regression from #447). 17 placeholders across 12 message keys used ICU argument syntax ({toolName}) while the component passes chunk functions — next-intl splices the raw function into the render output and React throws ('Functions cannot be passed to Client Components'), hard-crashing the routines page, login no-providers state, onboarding modal, builder preview, and four admin pages whenever those paths render. The locale-parity check couldn't catch it because both locales were consistently wrong. All converted to tag syntax (<toolName></toolName>) in en+de; audited to zero via a t.rich/message cross-check sweep. 3. next dev was 500ing on every page: Tailwind v4 scans comments as class candidates, and two test-file comments containing a literal bracket-class with var(...) compiled into invalid CSS (dev-only; prod builds tolerate it). Reworded the comments. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(web-ui): use the canonical Lume Button in RuntimeReadinessBanner main's no-restricted-syntax rule (the Lume button ratchet, #616) landed after this branch was written and rejects the raw <button> the banner used for its dismiss action. Migrated to <Button variant="secondary">, matching ConfirmDialog's precedent for a cancel-style affordance; the uppercase/tracking classes are kept so it still reads as a pair with the adjacent settings link. web-ui lint: 0 errors (was 1). --------- Co-authored-by: Claude Fable 5 <noreply@anthropic.com> Co-authored-by: Marcel Wege <mwege@byte5.de>
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
2 participants
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.
What
Closes #290. Migrate the 27 post-#291 rectangular text CTAs that drifted in across web-ui/app to the canonical
<Button>(§4.2 variants, §6.4 press feel, §7.3 busy), and add an error-level ESLintno-restricted-syntaxgate so raw<button>can't silently drift back.Why
Post-merge features kept adding raw
<button>(200 at the drift peak) — the material layer preserved the look but not the press feel, and per-call className soup re-diverged from the spec variants. The lint gate turns the recurring manual audit into a permanent one; every deliberate non-candidate (tabs/toggles, icon-only chrome, text links, selection tiles, warning-outline/§10 and--highlightactions, global-error.tsx) carries an inline disable + reason (185 total). One solidbg-dangerrollback normalized to the transparentdangervariant per §7.5.Test plan
npx tsc --noEmit— 0 errorsnpm run lint— 0 errors (new gate green; only pre-existing warnings)npm run i18n:check— OK, 3339 keys (no strings/keys added)npx vitest run— 552/552npm run build— greenRisk / blast radius
web-ui only. No schema, public API, env-var, or CI/release changes. Migrations change some at-rest visuals by design (ink/
--bg-inversefills → accent; solidbg-danger→ transparent danger per §7.5) — verified in both themes. Large file count (93) is inherent to the error-level gate touching every raw-button site.Need help on this PR? Tag
@codesmith-botwith what you need. Autofix is disabled.