## Problem
A team that **cancelled** its PAYG subscription kept full subscribed
access:
- **UI didn't reflect cancellation** — the Plan tab still rendered the
subscribed view, never the free/upgrade view.
- **Automation wasn't stopped** — automation / AI / API kept running
without ever falling back to the free-grant gate.
## Root cause
`TeamBillingService.compute` decided `subscribed` as:
```java
boolean subscribed =
subscriptionId != null
|| extOpt.map(PaygTeamExtensions::getStripeCustomerId).filter(s -> !s.isBlank()).isPresent();
```
On cancellation, the `customer.subscription.deleted` webhook calls
`payg_unlink_subscription`, which nulls `payg_subscription_id` but
**deliberately keeps `stripe_customer_id`** (so a future re-subscribe
can reuse the Stripe customer).
`payg_link_subscription` is the **only** writer of
`payg_team_extensions.stripe_customer_id`, and it writes it in the
*same* `UPDATE` as `payg_subscription_id` (on
`customer.subscription.created`). So the customer id is never set before
the subscription id — the "pre-webhook stand-in" the old comment claimed
**cannot happen**. The fallback only ever pinned a team that *ever*
subscribed to `subscribed` forever, because the Stripe customer outlives
the subscription.
Both symptoms are this one flag:
- `PaygWalletController` status → `SUBSCRIBED` vs `FREE`
- `EntitlementService` gate branch → monthly-cap vs free-grant
## Fix
Gate `subscribed` purely on `payg_subscription_id != null`. A cancelled
team now correctly drops to free (UI shows free; billable ops gate on
the one-time grant). This aligns the wallet/entitlement read with the
**meter path** (`JobChargeService.close`), which already gated on
`payg_subscription_id`.
Handles both Stripe cancel modes: "cancel at period end" keeps the sub
`active` (id stays set) until `.deleted` fires at period end → access
through the paid period; immediate cancel fires `.deleted` now → flips
to free now.
**No data migration / backfill** — already-cancelled teams have
`payg_subscription_id = NULL`, so they flip to free as soon as this
ships (within the 30s billing-cache TTL).
## Tests
Adds `TeamBillingServiceTest` — the `subscribed` computation previously
had **no** unit coverage (which is how this shipped). Covers: subscribed
iff subscription id present; **cancelled team (customer id remains,
subscription id null) ≠ subscribed** + free grant survives;
no-subscription/no-customer; no extension row.
`:saas:test` + `:saas:spotlessCheck` green; coverage gates met.
## Not included (optional hardening, can fast-follow)
- Cross-check the synced `stripe.subscriptions.status` to guard a
*missed* `.deleted` webhook leaving `payg_subscription_id` stale.
- Push cache-invalidation from the webhook (currently ≤30s TTL
staleness).
## What
Removes all em dash (`—`) characters from the **user-facing text** on
the Plan page (PAYG section), replacing them with colons, commas, or
restructured punctuation so the copy reads naturally.
## Changes
- `frontend/editor/public/locales/en-GB/translation.toml` — all `payg.*`
strings (this is what actually renders on the page)
- `PaygFree.tsx` — `t()` default fallbacks + the `{" — "}` JSX
benefit-list separators (now `{": "}`)
- `Payg.tsx` — `t()` default fallback for the editor-plan body
## Notes
- The en-dash range separator (`{{start}} – {{end}}`) in the
billing-period string is intentionally **kept** — only em dashes were
targeted.
- JSDoc / code comments containing em dashes were **left unchanged**,
since they aren't rendered text on the page.
<img width="990" height="502" alt="image"
src="https://github.com/user-attachments/assets/13d89b0f-007c-4b4c-b72d-1d912f968bc7"
/>
## Summary
Pay-as-you-go (PAYG) billing for Stirling-PDF SaaS. Manual PDF editing
stays free forever; only **automation, AI, and API** usage is metered.
Every team gets a **one-time lifetime free grant** (default 500 PDFs)
before any billing; past that, a team adds a card and pays per metered
document, with a self-set monthly spending cap.
This branch combines and supersedes the in-flight BE (#6574) and FE
(#6579) work plus the SaaS edge functions (Stirling-PDF-SaaS PR, now on
`v3`), hardened into a single reviewable feature after a pre-merge
dead-code/security review.
## Billing model
- **Always free:** manual / JWT web-tool usage is `BYPASSED` — never
metered, no matter where it's triggered.
- **Billable categories:** `AUTOMATION`, `AI`, `API`.
- **One-time lifetime free grant** (`pricing_policy.free_tier_units`,
default 500): never resets, survives subscribing. It gates unsubscribed
teams (billable API calls hard-stop with a 402 once exhausted) and
decides the free-vs-paid split of every job.
- **Subscribed:** paid documents (beyond the grant) are metered to a
Stripe Billing Meter; an optional monthly spending cap degrades billable
categories when reached.
- **Dedup:** the same file pushed through several steps within a
workflow window counts **once** (lineage join), so API/AI chaining on
one file isn't double-charged.
## What's included
**Database** — Flyway migrations `V11`→`V21` with matching Supabase
twins: pricing policy + per-team sidecar (`payg_team_extensions`:
subscription id, Stripe customer, free-grant counter), append-only
`wallet_ledger`, shadow charges, subscription-state RPCs (`V14`), audit
logs (`V15`), billing category (`V16`), one-time lifetime free grant
(`V19`), launch-grant seed (`V20`), drop of the unused
`wallet_category_summary` view (`V21`).
**Charge pipeline** — `PaygChargeInterceptor` (open/join a process,
split the free grant, write the ledger DEBIT), `JobChargeService`
(consume the grant under a row lock, restore it on a first-step refund,
meter only the paid portion on completion), `StaleJobCloser` fallback
(idempotent close → meter).
**Entitlement** — `EntitlementService` (per-team cached snapshot:
grant-gated for free teams, monthly-cap-gated for subscribed) +
`EntitlementGuard` (401 `SIGNUP_REQUIRED` / 402 `FEATURE_DEGRADED` /
`PAYG_LIMIT_REACHED`).
**Metering** — `PaygMeterReportingService` writes a durable
`payg_meter_event_log` row around every POST to the `meter-payg-units`
edge fn (pending → posted/failed); `PaygMeterReconcileScheduler` retries
unposted events under the same idempotency key inside Stripe's 24h dedup
window.
**Billing facts** — `TeamBillingService` reads the synced `stripe.*`
mirror (subscription window, per-document rate; the unsubscribed-team
estimate resolves the rate by Price `lookup_key = plan:processor`).
**Wallet API** — `PaygWalletController`: `GET /api/v1/payg/wallet`,
`PATCH /api/v1/payg/cap`.
**Frontend** — PAYG Plan page (two-card free layout + subscribed views),
`useWallet`, upgrade modal with lazy-loaded Stripe Embedded Checkout and
a shared `SpendCapControl`, customer-portal link, 402/401 interceptor
toast, en-GB i18n. (Per-member usage shows each teammate's spend; the
activity feed is behind a flag until polished.)
**SaaS edge functions** (`Stirling-PDF-SaaS` `v3`) —
`create-checkout-session`, `create-payg-team-subscription`,
`create-customer-portal-session`, `meter-payg-units`,
`payg-subscription-webhook`, `stripe-sync`, plus the stripe-sync
`migrate` + scoped-`backfill` scripts. All price lookup is DB-driven (no
`STRIPE_PAYG_PRICE_ID_*` env vars).
## Release prerequisites (prod)
1. Apply Flyway migrations (`V11`→`V21`) and the Supabase migration
twins.
2. Stripe Sync Engine: run `stripe-sync:migrate`, then a **scoped**
backfill — `product`, `price`, `customer`, `subscription` only (not
`all`, which rate-limits).
3. Register 2 PAYG webhook endpoints (each its own signing secret):
`stripe-sync` (product/price/customer/subscription `.*`) and
`payg-subscription-webhook` (`customer.subscription.created`/`.deleted`
drive state; `.updated` + `invoice.*` observed). Keep the legacy
`stripe-webhook` only if credits/self-hosted flows still run.
4. Stripe Billing Meter: `event_name = payg_doc_units`, value key
`processed_documents`.
5. Env: `PAYG_METER_ENDPOINT` + `SUPABASE_EDGE_FUNCTION_SECRET`
(backend); the webhook signing secrets (edge fns). The default pricing
policy must point at the PAYG Stripe Price(s); `V20` seeds
`free_tier_units = 500`.
## Testing
- `:saas:test` green, `:saas:spotlessCheck` clean, edge-fn Deno tests
green, FE saas typecheck clean (the remaining errors are pre-existing
`proprietary/*` + `prototypes/*`, untouched here). Cucumber shadow-mode
suite + CI workflow included.
## Pre-merge review
An independent dead-code/security pass came back **clean on security**
(team-derived authz / no IDOR, leader-only cap mutation, no
billing-category downgrade, dev/mock hooks gated to
`import.meta.env.DEV` + `/dev/`, no secrets/injection, fail-open
metering by design). The dead/unwired code it flagged has been removed
in this branch (unenforced sub-cap control, an unused JDBC DAO + its
view, dead methods).
## Follow-ups (tracked, not blocking)
- **Enforce per-member sub-caps** — the control was removed because it
read for display but never gated a request; the per-member usage display
and `cap_units` column are retained for when enforcement is wired.
- **API/AI chaining billing model + `ProcessType` enum** — confirm
same-file dedup covers API chaining; define per-tool AI charging; decide
whether the unused enum values stay.
- **Activity feed** — hidden behind a flag until the meter-event surface
is polished.
---------
Co-authored-by: Reece <[email protected]>
## What this PR does
Bundles the **low-risk polish items** from the [multi-agent review of
#6519](https://github.com/Stirling-Tools/Stirling-PDF/pull/6519). Each
change is independent, mechanical, and ships with focused unit-test
coverage.
The medium-severity items (\`?async=true\` OUTPUT recording,
JSON-consumes endpoint coverage, SpringBootTest harness) are tracked
separately in [\`notes/PAYG_DESIGN.md\` §7.5
PR-S4](https://github.com/Stirling-Tools/Stirling-PDF/blob/payg-s4-hardening/notes/PAYG_DESIGN.md)
— they need design decisions + bigger infrastructure work, so this PR
sticks to the mechanical wins.
Stacked on #6519. When that merges to main, this rebases cleanly — no
code changes.
## Changes
| Area | What | Why |
|---|---|---|
| **\`tool_id\` becomes route pattern** |
\`PaygChargeInterceptor.resolveToolId()\` prefers
\`HandlerMapping.BEST_MATCHING_PATTERN_ATTRIBUTE\` over
\`request.getRequestURI()\`. Truncates to 128 + WARN log +
\`payg.filter.errors\` increment when truncation fires. | Audit rollups
aggregate by endpoint instead of by every individual request's
path-variable / matrix-param variant. Silent truncation now louder. |
| **Direct PDF magic-byte check** | \`PaygOutputExtractor.extract()\`
magic-checks the body even for direct \`application/pdf\` responses. |
Asymmetric with the ZIP-entry path which always magic-checks. A tool
that emits \`application/pdf\` for a JSON / HTML payload would otherwise
pollute \`job_artifact_hash\`. |
| **DESKTOP_APP detection** | \`X-Stirling-Client: desktop\` header →
\`JobSource.DESKTOP_APP\`. | The enum value was unreachable from
\`determineSource()\`; Tauri shell traffic was mis-classified as WEB. No
anti-spoof — V12 step limits are identical for WEB/DESKTOP_APP so the
worst-case abuse value is zero today. |
| **\`max-bytes\` sensible default** | 500 MiB instead of \`null\`
(unbounded). | Covers the largest realistic Stirling responses (full
split-to-ZIP on a 1000-page document) while preventing pathological
cases from tying up the interceptor for minutes. Set to \`null\` to
disable. |
| **\`BufferedOutputStream\` for spill** | Wraps the spill
\`OutputStream\` in 64 KiB \`BufferedOutputStream\`. | Previously every
Tomcat chunk (default 8 KiB) was a separate syscall. Big spilled
responses get a syscall-bound speedup. |
| **Duration timer per phase** | \`payg.filter.duration\` tagged
\`phase=preHandle\` vs \`phase=afterCompletion\`. | Two distinct latency
distributions were blended into one histogram; hard to alert on. |
## Tests
| Test | What it covers |
|---|---|
|
\`PaygOutputExtractorTest.pdfContentType_butBodyMissingPdfMagic_returnsEmpty\`
| New direct-PDF magic-byte gate. |
|
\`PaygChargeInterceptorTest.preHandle_desktopClientHeader_setsJobSourceDesktopApp\`
| New \`X-Stirling-Client: desktop\` → DESKTOP_APP path. |
|
\`PaygChargeInterceptorTest.preHandle_toolId_prefersBestMatchingPattern\`
| Route pattern wins over URI when both are set. |
|
\`PaygChargeInterceptorTest.preHandle_toolId_truncatesAndCountsWhenLongerThan128\`
| Oversized values truncate + increment errors counter. |
Full saas suite green (210 tests), coverage targets met.
## What's NOT in this PR (deliberately)
- **\`?async=true\` OUTPUT recording.** The JobExecutorService returns a
synchronous \`JobResponse{jobId}\` body before the async tool actually
runs; \`afterCompletion\` fires too early. Needs a design decision:
short-circuit PAYG when \`async=true\` OR hook into \`TaskManager\`
completion. Tracked in PR-S4 design doc.
- **JSON-consumes endpoint coverage.** The
\`MultipartHttpServletRequest\` cast skips endpoints with \`consumes =
APPLICATION_JSON_VALUE\` (e.g.
\`ConvertPdfJsonController.exportPartialPdf\`). Fix is either extract a
request-body hash for JSON or add a CI lint forbidding non-multipart
\`@AutoJobPostMapping\`. Design discussion needed.
- **SpringBootTest harness for filter + interceptor wiring.** Saas
module doesn't have one yet. Separate work — PR-S3 takes a different
approach (docker-compose + Behave); a SpringBootTest layer would be
additive in-process coverage.
These are tracked in \`notes/PAYG_DESIGN.md §7.5\` so they don't slip.
## Tracked in
\`notes/PAYG_DESIGN.md\` §7.5 PR-S4.
## Summary
Two Flyway migrations + matching JPA entity updates. **Part 1 of 2** in
the Stripe/Supabase wire-up (PR-SB-1 in
`payg-stripe-supabase-plan.html`); the companion SaaS PR carries the
twin Supabase migrations + new edge functions.
### V14 — payg_subscription_state.sql
- `payg_team_extensions.payg_subscription_id` — the single switch that
decides whether a team is billed. NULL = free-tier or block; NOT NULL =
post Stripe meter events.
- `pricing_policy.free_tier_units_per_cycle` — per-policy free allowance
before a card is required. Default 0.
- `payg_link_subscription(team_id, customer_id, sub_id)` RPC —
idempotent.
- `payg_unlink_subscription(team_id, reason)` RPC — called on
`subscription.deleted`.
- AFTER-INSERT trigger on `teams` so every new signup gets a
`payg_team_extensions` sidecar row from creation.
- Backfill for existing teams without a sidecar row.
- RLS: SELECT permissive (any team member), UPDATE restricted to LEADER.
Service-role bypasses (backend reads + day-1 migration writes).
### V15 — payg_audit_logs.sql
- `payg_meter_event_log` — backend audit of every Stripe meter event
POST attempt (idempotency-key UNIQUE; index on unposted rows for nightly
reconcile).
- `payg_subscription_change_log` — written by V14 RPCs on every
link/unlink.
### Entity updates
- `PaygTeamExtensions.paygSubscriptionId` — read-only field; RPC
functions are the only writers.
- `PricingPolicy.freeTierUnitsPerCycle` — read by upcoming
`PaygTeamUsageService` (PR-SB-4).
### Behaviour change
**None yet.** The columns + functions sit unused until PR-SB-4 wires
`PaygMeterReportingService` and the free-tier gate into
`JobChargeService`. This PR is pure schema + JPA wiring.
## Test plan
- [x] `./gradlew :saas:test` — BUILD SUCCESSFUL
- [x] Manual schema review: column types, FK directions, RLS scope
- [ ] Apply against v3-Supabase via `supabase db push` (after companion
SaaS PR merges)
- [ ] Smoke-test the trigger: `INSERT INTO teams(...)` → assert
`payg_team_extensions` row appears
- [ ] Smoke-test RPCs: SQL-only test of `payg_link_subscription` +
`payg_unlink_subscription` produces expected row + audit entries
## References
- `notes/PAYG_DESIGN.md` (revision note 2026-06-03)
- `payg-stripe-supabase-plan.html` §3.1 (RPC functions), §3.5 (RLS),
§3.10 (audit-log tables)
## What this PR is
End-to-end cucumber coverage for the PAYG shadow charging engine (the
filter + interceptor stack from #6519), wired into CI via a new
`docker-compose-tests-saas.yml` workflow that runs only on PAYG-touching
PRs.
Stacked on #6519.
## Automated scenarios (run by `docker-compose-tests-saas.yml`)
See
[`testing/cucumber/features/payg/shadow_charges.feature`](../tree/payg-s3-cucumber/testing/cucumber/features/payg/shadow_charges.feature):
| Scenario | Validates |
|---|---|
| First tool call writes a CHARGED row | Filter + interceptor fire
end-to-end |
| Lineage join — second call on output | `JobService.joinOrOpen`
matching; no new shadow row |
| 4xx leaves the row CHARGED | "Customer paid for the attempt" semantics
|
| ZIP-returning tool records per-PDF OUTPUT | `PaygOutputExtractor`
unpacks + records signatures |
| Multi-file input writes a single shadow row | Multi-input group sizing
|
| `X-Stirling-Automation` sets PIPELINE source | Header → `JobSource`
detection |
All 6 run locally via `./testing/test-payg.sh` and will run on CI for
any PR that touches `app/saas/**`, the PAYG cucumber features, the saas
compose stack, or the workflow itself.
## Manual-only scenarios — documented in design doc, not in this suite
Two parts of the shadow engine are deliberately not automated; the
engine paths are unit-tested in
`PaygChargeInterceptorTest.afterCompletion_5xx_opened_*`, and the manual
procedures (which require a temporary throw endpoint or a container
restart with a flag flipped) live in [`notes/PAYG_DESIGN.md` §7.5.2
"PAYG cucumber: manual-only
scenarios"](../tree/payg-s3-cucumber/notes/PAYG_DESIGN.md).
- **5xx first-step failure → REFUNDED + CLOSED.** No reliably-5xx-ing
endpoint exists; manual procedure adds a throw endpoint, runs, asserts,
removes.
- **Kill-switch (`PAYG_FILTER_ENABLED=false`).** Needs a container
restart mid-suite; manual procedure tears down, flips env, brings up,
asserts zero shadow rows.
If either gets a hot-reload path (test-only throw endpoint shipped
behind a profile gate, or admin endpoint for the kill switch), automate
it in a follow-up and drop the manual procedure.
## CI workflow
`.github/workflows/docker-compose-tests-saas.yml` (new) —
self-contained, not wired into `build.yml`'s `files-changed` matrix so
the saas-cucumber job fails and succeeds independently. Triggers only on
PAYG-relevant paths. No JaCoCo coverage in v1 (saas compose doesn't have
the coverage override; can add later).
## Test infrastructure (recap)
- **`testing/compose/docker-compose-saas.yml`** — Stirling-PDF backend
with `STIRLING_FLAVOR=saas` + Postgres holding the `stirling_pdf`
schema. Supabase JWT auto-config disabled; API-key auth via
`SECURITY_CUSTOMGLOBALAPIKEY` is the live path the cucumber tests
exercise.
- **`testing/compose/payg/saas-init.sql`** + **`saas-seed.sql`** —
schema bootstrap + idempotent seed (team / user / wallet_policy).
- **`testing/cucumber/features/payg/shadow_charges.feature`** — the 6
scenarios above.
- **`testing/cucumber/features/steps/payg_step_definitions.py`** — step
defs using `requests` (HTTP) + `psycopg` (direct DB inspection). Direct
DB reads are deliberate — we want to see the filter's side effects, not
relay them through another API layer.
- **`testing/test-payg.sh`** — companion runner to `testing/test.sh`.
Brings up the saas compose, waits for health, seeds, runs behave, tears
down.
- **`behave.ini`** excludes `features/payg` from the default behave run
(the saas-cucumber CI job invokes it explicitly).
## Why a separate harness from `testing/test.sh`
The existing `test.sh` covers the proprietary-flavour stack (no PAYG
tables, no saas profile). Coupling two CI matrices that fail and succeed
independently into one script is asking for trouble. Keep the
saas-cucumber job focused on its own concerns; once the harness is
mature, the wider team can decide whether to merge them.
## Tracked in
`notes/PAYG_DESIGN.md` §7.5 (PR-S3) + §7.5.2 (manual scenarios).
## What this PR does
Wires the **B-1 shadow charging engine** into real HTTP request flow.
After this lands, flipping an internal team to ``PAYG_SHADOW`` via SQL
begins populating ``payg_shadow_charge`` automatically — with **zero
impact** on the legacy credit deduction path.
**This is the load-bearing PR for shadow mode.** Without it, B-1's
engine sits idle — nothing in the codebase calls
``JobChargeService.openProcess()`` from a real HTTP request.
Stacks on top of #6477 (PR B-1).
## Components
| Class | Role |
|---|---|
| ``PaygResponseBodyWrapperFilter`` | Servlet filter, installs tee'ing
response wrapper. Defers wrapper close to ``AsyncListener`` for
``DeferredResult`` / ``CompletableFuture`` controllers so the lifetime
spans the async window. |
| ``PaygResponseBodyWrapper`` | ``HttpServletResponseWrapper`` —
in-memory ``ByteArrayOutputStream`` up to 10 MiB; spills to ``TempFile``
above. ``materialisedPath()`` always returns a uniform ``Path``
interface. |
| ``PaygChargeInterceptor`` | ``AsyncHandlerInterceptor`` mirroring
``UnifiedCreditInterceptor`` shape. ``preHandle`` gates on
``@AutoJobPostMapping``, materialises multipart inputs, calls
``JobChargeService.openProcess``. ``afterCompletion`` branches on HTTP
status. |
| ``PaygOutputExtractor`` | Pulls PDFs out of the response body. Direct
``application/pdf`` returns body verbatim; ``application/zip`` iterates
entries and keeps each ``.pdf`` entry whose first bytes match the
``%PDF-`` magic. |
| ``PaygWebMvcConfig`` | Registers filter at end of Spring filter chain
(after security); interceptor after ``UnifiedCreditInterceptor``. |
| ``PaygFilterProperties`` | ``payg.filter.enabled`` master switch +
in-memory threshold + optional max-bytes ceiling. |
## Status branching in afterCompletion
| HTTP status | Action |
|---|---|
| **2xx** | Append OK step; extract PDFs from response;
``JobService.recordOutput`` per PDF |
| **4xx** | Append FAILED step with ``errorCode``. No refund — customer
paid for the attempt. No OUTPUT recording. |
| **5xx + OPENED** (first-step) |
``JobChargeService.markFirstStepFailed`` → shadow row flipped to
``REFUNDED``, process CLOSED. Refund counter incremented. |
| **5xx + JOINED** (mid-chain) | ``JobChargeService.decrementStepCount``
— step slot returned without resetting ``lastStepAt`` (workflow window
stays active for retry). |
## New ``JobChargeService`` methods
- **``markFirstStepFailed(jobId, reason)``** — flips shadow row to
``REFUNDED`` with ``refundedAt`` + ``refundReason``, closes the process.
Idempotent. Mimics the eventual Stripe
``meter_event_adjustment(cancel)`` flow that real-mode will invoke at
the same callsite. **Refund implies close** so a same-input retry can't
lineage-join into a refunded chain for free work.
- **``decrementStepCount(jobId)``** — defensive floor at 1; never drives
count negative.
## Schema
- Backend: ``V13__payg_shadow_charge_status.sql`` adds ``status``
(``CHARGED`` | ``REFUNDED``) + ``refunded_at`` + ``refund_reason``.
``DEFAULT 'CHARGED'`` so existing B-1 rows stay correct without
backfill.
- Supabase: matching migration in
[Stirling-PDF-SaaS#payg-shadow-charge-status](https://github.com/Stirling-Tools/Stirling-PDF-SaaS/tree/payg-shadow-charge-status)
## Fail-open semantics in shadow
Any ``RuntimeException`` in ``preHandle`` / ``afterCompletion`` is
logged at WARN, increments ``payg.filter.errors``, and lets the
customer's tool call proceed unbilled. This **reverses to fail-closed**
when ``wallet_policy.engine = PAYG`` (real charging) — that reversal
lives inside ``JobChargeService`` and ships with the cap evaluator PR
(PR-C1 in PAYG_DESIGN.md).
## Observability
Micrometer metrics:
- ``payg.filter.errors`` Counter — internal failures (preHandle +
afterCompletion). Alert source.
- ``payg.filter.calls`` Counter, tagged ``disposition`` (``OPENED`` |
``JOINED`` | ``SHORT_CIRCUIT``)
- ``payg.filter.refunds`` Counter — first-step 5xx refunds
- ``payg.filter.duration`` Timer — preHandle + afterCompletion
wall-clock per request
## Test coverage (38 tests across 4 classes)
- **PaygResponseBodyWrapperTest** (12 tests) — in-memory, spill,
threshold crossing mid-chunk, writer vs outputStream exclusivity,
``resetBuffer`` with and without spill, close idempotency, single-byte
writes across threshold.
- **PaygOutputExtractorTest** (7 tests) — direct PDF, parametrised
content type, ZIP with mixed entries + magic-byte gate, corrupt ZIP
fail-open, empty ZIP.
- **PaygChargeInterceptorTest** (13 tests) — all preHandle
short-circuits, OPENED disposition stash, fail-open on chargeService
exception, 2xx recordOutputs path, 5xx OPENED → markFirstStepFailed, 5xx
JOINED → decrementStepCount, 4xx FAILED step append, max-bytes ceiling
skip, PIPELINE header detection.
- **JobChargeServiceTest extended** (+6 tests) — markFirstStepFailed
happy path, idempotency, missing-shadow-row case, long-reason trim;
decrementStepCount happy path, floor-at-1 defence, missing-job no-op.
## What's NOT in this PR (deliberate)
- **No SpringBootTest layer.** The saas module doesn't have bootstrap
test infrastructure (Supabase JWT config + H2 schema harness).
Integration confidence comes from the manual staging deploy + SQL-flip
of an internal team. Bootstrap-test infra is a focused follow-up if
needed.
- **No saas-mode Behave / docker-compose.** Per design §17 — deferred.
Existing ``testing/cucumber/`` infrastructure doesn't yet have a
saas-profile compose target; that's its own PR when warranted.
- **No CreditService wire-in** (per design §13 decision). Per-row
comparison data moves to the reconciliation report PR (PR-S2).
``legacy_credits_charged`` + ``diff_pct`` columns stay at 0 in shadow
rows.
- **No reconciliation report endpoint.** Direct SQL queries against
``payg_shadow_charge`` cover the data-access need until patterns emerge.
## Rollback levers
| Symptom | Lever |
|---|---|
| Some / all tool calls breaking due to filter |
``payg.filter.enabled=false`` + restart (~20s) |
| Shadow rows look wrong for a specific team | ``UPDATE wallet_policy
SET engine = 'LEGACY' WHERE team_id = ?`` |
| Mass shadow weirdness | ``UPDATE wallet_policy SET engine = 'LEGACY'``
|
| Memory exhaustion from response tee | Lower
``payg.filter.response.in-memory-threshold-bytes`` |
## Test plan
- [ ] CI green (build + tests)
- [ ] Aikido / Snyk / SonarCloud clean
- [ ] Manual: deploy to staging
- [ ] Manual: flip one internal team via ``UPDATE wallet_policy SET
engine = 'PAYG_SHADOW' WHERE team_id = ?``
- [ ] Manual: hit ``/api/v1/security/add-password`` with that team's
JWT; verify a ``payg_shadow_charge`` row appears with
``status='CHARGED'``
- [ ] Manual: trigger a 503 (e.g. via temporary backend kill
mid-request); verify the resulting row is ``status='REFUNDED'`` + the
process is ``CLOSED``
- [ ] Manual: hit ``/api/v1/general/split`` with a multi-page PDF;
verify one OUTPUT signature per inner PDF appears in
``job_artifact_hash``
- [ ] Manual: chain ``add-password`` → ``compress`` on the output;
verify the second call JOINS the first process (no new shadow row) and
the inner output OUTPUT signature is what drove the lineage join
## Stacks on / references
- Stacks on: #6477 (B-1 — shadow charging engine)
- Schema mirror: Stirling-PDF-SaaS#payg-shadow-charge-status branch
- Design doc: ``notes/PAYG_FILTER_DESIGN.md`` (all 19 decisions DECIDED)
> 📌 **Stacked on
[#6464](https://github.com/Stirling-Tools/Stirling-PDF/pull/6464)**
(lineage primitives, still in review). #6469 has merged so its commits
are no longer in this PR's diff. Once #6464 merges, a final rebase
collapses the lineage-primitives commits out of this diff too — leaving
only the B-1 work.
## What this is
Process tracking + shadow charging engine. Bundles PR-I7 service half
with the non-filter piece of PR-I7a so the pieces ship together — none
of them is useful in isolation.
**Review focus:** the new files in:
- \`app/saas/src/main/java/stirling/software/saas/payg/job/\`
(\`JobService\`, \`JobContext\`, \`JoinOrOpenResult\`,
\`StaleJobCloser\`)
- \`app/saas/src/main/java/stirling/software/saas/payg/charge/\`
(\`JobChargeService\`, \`ChargeContext\`, \`ChargeOutcome\`,
\`JobInput\`)
-
\`app/saas/src/main/java/stirling/software/saas/payg/lineage/LineagePruneScheduler.java\`
- their tests
The 8 files inherited from #6464 (lineage primitives) are unchanged from
there — they ride along in this diff until #6464 lands.
The remaining work for shadow-in-staging is the ingress/egress filter
that wires controllers into this engine — that's PR B-2.
## Scope
### \`JobService\` — persistence + lineage policy
- **\`joinOrOpen\`** — the multi-input "any-match-joins, newest wins"
rule. Hash every input via the lineage detector; if any matches an open
process in the workflow window, attach to the one with the freshest
\`lastStepAt\`. Step-limit overflow on the matched job spawns a fresh
process; the new job's input signatures are still recorded so
\`mostRecentMatchWins\` routes future calls forward.
- **\`recordOutput\`** — post-tool-success path. Records OUTPUT
signatures so the next call that takes this file as input
lineage-matches into the same process.
- **\`appendStep\`** — audit-trail step row written after a tool
completes.
- **\`close\`** — idempotent; safe to call from multiple paths
(explicit, FE on-unload, scheduler). Returns the same row on re-close,
no state mutation.
- **\`findStale\` / \`closeStale\`** — workflow-window-based stale
closure used by the scheduler.
### \`JobChargeService\` — the orchestrator (shadow variant)
\`openProcess\` resolves the effective policy via
\`PricingPolicyService\` (now in main via #6469), derives the step-limit
for the current \`JobSource\` (with a defensive fallback if the policy
is missing an entry), delegates to \`JobService.joinOrOpen\`, and on
OPENED runs the \`DocumentClassifier\` + writes a \`payg_shadow_charge\`
row. Applies the policy-level \`minChargeUnits\` floor per design § 3.4.
Shadow variant only — never debits the ledger, never posts a Stripe
meter event. The real-charging follow-up reuses the same orchestration
and swaps the side-effect. \`legacyCreditsCharged\` on the shadow row
stays \`0\` until the legacy \`CreditService\` is wired in (PR B-2),
where the comparison becomes meaningful.
### Schedulers (both plain \`@Scheduled\`)
- **\`StaleJobCloser\`** — fixed-rate 60 s. Closes \`OPEN\` jobs idle
past the workflow window. API users never have to call close explicitly
— this is the safety net.
- **\`LineagePruneScheduler\`** — hourly cron, retention 1 h. Deletes
\`job_artifact_hash\` rows older than the retention window.
- **No \`@SchedulerLock\` / no \`shedlock\` table** — consistent with
the 5 existing unguarded \`@Scheduled\` tasks in \`:saas\`
(\`CreditResetScheduler\` and friends, none of which are guarded today).
Cluster-correctness across all 7 saas schedulers is tracked in design §
9 as a separate focused cleanup. Underlying operations are idempotent —
duplicate firings on multi-pod would be wasted DB load, not data
corruption.
### Records (call-shape glue for PR B-2's filter)
- \`JobContext\` / \`JoinOrOpenResult\` — input/output for
\`JobService\`.
- \`ChargeContext\` / \`ChargeOutcome\` — input/output for
\`JobChargeService\`.
- \`JobInput\` — paired \`(MultipartFile, materialised Path)\` so the
upcoming ingress filter can pass both views without re-materialising.
## Tests
**26 new, all green.**
- 14 × \`JobServiceTest\` — no-match → opens new, single-match → joins
existing, multi-input any-match-joins, multi-match newest-wins (older
job never even looked up), step-limit hit spawns fresh job (and original
\`stepCount\` is NOT mutated), empty inputs reject, stale-signature
handling, recordOutput delegation, close idempotency, closeStale,
appendStep persistence.
- 7 × \`JobChargeServiceTest\` — JOINED skips classifier + shadow write
entirely, OPENED writes shadow row + classifies (single + multi file
paths), \`minChargeUnits\` floor applied, step-limit resolved
per-\`JobSource\` from policy, missing source entry falls back to
conservative default of 10.
- 2 × \`StaleJobCloserTest\`, 3 × \`LineagePruneSchedulerTest\` —
scheduler-wiring smoke + constructor-validation tests.
\`ENABLE_SAAS=true ./gradlew :saas:test\` — BUILD SUCCESSFUL.
## What's not in this PR (lands in PR B-2)
- **Tool ingress/egress servlet filter.** The highest-risk piece —
materialises the request body into a \`JobInput\`, calls
\`JobChargeService.openProcess\` from every \`@AutoJobPostMapping\`,
records OUTPUT after success. Edge cases to validate: multipart parts,
async controllers, streaming responses, errored 5xx paths, very large
files. Design decisions for the filter are being worked through in
\`notes/PAYG_FILTER_DESIGN.md\` before any code is written.
- **Wire shadow path into legacy \`CreditService\`.** Every legacy debit
writes a comparison row carrying both the PAYG would-be units and the
legacy actual credits, populating \`diffPct\`.
## Design doc
\`notes/PAYG_DESIGN.md\` — PR-I7 + PR-I7a status updated to reflect this
bundle. § 9 carries the cluster-correctness deferral note alongside the
existing LISTEN/NOTIFY trade-off note. § 7.5.1 readiness summary shows
the path now needs just **1 more PR** (the filter half + CreditService
wire-in, both bundled into PR B-2).
## What this is
Three orthogonal interfaces — each with one production impl — for
detecting whether an incoming tool call should join an existing process
via content-hash lineage. Groundwork for PR-I7a: nothing in this PR
calls the detector yet; the ingress/egress filter that wires it into
every controller lands separately.
Built to be modular along three axes. Swapping any of them should not
require changes elsewhere:
| Axis | Interface | V1 impl | Plausible future impl |
|---|---|---|---|
| Hash algorithm | `LineageSignatureExtractor` |
`ByteHashSignatureExtractor` (SHA-256) | `PdfMetadataSignatureExtractor`
(PDF `/ID`, content-stream hash) |
| Storage backend | `JobLineageStore` | `JpaJobLineageStore` |
`RedisJobLineageStore` |
| Matching policy | `HashLineageDetector` | `DefaultHashLineageDetector`
| strategy-driven variant (any-match-joins for multi-input — lives in
JobService) |
## Interfaces
### `LineageSignatureExtractor` — what counts as a fingerprint
```java
public interface LineageSignatureExtractor {
Set<LineageSignature> extract(Path file) throws IOException;
String name();
}
```
File-based (not stream-based) so a future PDF-aware extractor can open
the same file via jpdfium / PDFBox and pull `/ID[0]` or a content-stream
hash. Multiple extractors compose at the detector layer — Spring
auto-wires all `LineageSignatureExtractor` beans, the detector unions
their results.
Production impl: **`ByteHashSignatureExtractor`** — SHA-256 over the
file via 64 KiB-buffered `DigestInputStream`. Hardware-accelerated by
the JVM (Intel SHA-NI, ARM SHA).
### `JobLineageStore` — where signatures live
```java
public interface JobLineageStore {
void record(UUID jobId, Set<LineageSignature> signatures, ArtifactKind kind);
Optional<LineageMatch> findOpenJobForSignatures(Long userId, Set<LineageSignature> candidates, Duration window);
int pruneOlderThan(Instant cutoff);
}
```
Knows nothing about storage technology. Production impl
**`JpaJobLineageStore`** runs a single joined query against
`job_artifact_hash` ⋈ `processing_job` — status + window filtering
happen at the database. The query is bounded by `Limit.of(1)` on the hot
path so a job set sharing a popular signature doesn't materialise
unwanted rows. A future `RedisJobLineageStore` (or write-through hybrid)
is a drop-in.
### `HashLineageDetector` — the high-level API
```java
public interface HashLineageDetector {
Optional<LineageMatch> detect(Long userId, Path inputFile) throws IOException;
void record(UUID jobId, Path file, ArtifactKind kind) throws IOException;
}
```
**`DefaultHashLineageDetector`** delegates extraction to every
registered `LineageSignatureExtractor`, storage to the configured
`JobLineageStore`, and reads `payg.lineage.workflow-window` (default
`PT5M`) from config. When a single extractor throws (e.g. a future
PDF-aware extractor against a malformed PDF), the other extractors still
contribute — failures don't block the byte-hash from landing.
## Profile gating
All three `@Component` beans (`JpaJobLineageStore`,
`ByteHashSignatureExtractor`, `DefaultHashLineageDetector`) are
`@Profile("saas")` — consistent with every other `:saas` bean. Without
this guard the JPA store would fail to wire against its profile-gated
repository in non-saas profiles that pull `:saas` onto the classpath.
## Tests
Run entirely in-memory; no database required.
- **`LineageSignatureTest`** — storage-key encoding round-trips, rejects
malformed `"type:value"` keys.
- **`ByteHashSignatureExtractorTest`** — identical bytes → identical
sigs; empty file hashes to the well-known SHA-256-of-empty constant; 10
MiB file streams without OOM.
- **`DefaultHashLineageDetectorTest`** — same-user / within-window /
status=OPEN filtering, multi-signature matching (one extractor sees
`pdf-id` and matches even when bytes differ), most-recent-job-wins,
record+detect round-trip, extractor-throwing-doesn't-break-others.
**`InMemoryJobLineageStore`** (in test sources) implements the same
`JobLineageStore` interface as the JPA impl, plus a `registerJob` hook
for tests to model job state. Same contract — proves the abstraction is
portable. When the Redis impl lands it gets the same contract tests.
## What's not in this PR (deliberate)
- The tool ingress/egress filter that wires the detector into every
controller — separate, focused review.
- `JobChargeService.openProcess()` — uses the detector, part of the
charging machinery, separate PR.
- Prune scheduler that calls `pruneOlderThan` — small follow-up
alongside the `shedlock` foundational table.
- PDF-aware extractor (`PdfMetadataSignatureExtractor`) — to be added
when we measure how often byte-hash-only misses real workflows.
- Multi-input "any-match-joins" lineage policy — that's a `JobService`
decision (PR-I7), not a primitive.
## Self-review pass applied
An independent code-review on this PR caught:
- **HIGH:** Missing `@Profile("saas")` on the three `@Component` beans →
fixed.
- **MEDIUM:** `pruneOlderThan` missing `@Transactional` (its
`@Modifying` query would have thrown
`InvalidDataAccessApiUsageException`) → fixed.
- **MEDIUM:** `findOpenJobsForSignatures` fetching the whole match set
just to `get(0)` → now takes `Limit`, JPA store passes `Limit.of(1)` on
the hot path.
- **LOW:** `InMemoryJobLineageStore` used both `synchronized` methods
and `ConcurrentHashMap` → dropped the redundant `ConcurrentHashMap`.
Deferred: project-wide UTC unification (`LocalDateTime.now()`
system-zone is the established convention; flipping one file mid-stack
caused a real test failure — proper fix needs its own audit).
## Rollback
Straight `git revert`. No callers yet; deleting these classes wouldn't
break anything.
---
## Checklist
- [x] Tests pass: `ENABLE_SAAS=true ./gradlew :saas:test`
- [x] No new warnings
- [x] Self-review performed (HIGH + MEDIUM findings addressed)
## What this is
PR-I1 service half from `notes/PAYG_DESIGN.md`. Built on top of the data
model from #6460 — answers "what pricing policy applies to this team
right now?" with a fast cache and an admin write surface.
## Scope
| Piece | Where |
|---|---|
| `PricingPolicyService` — `getEffectivePolicy(teamId)` with 30s
Caffeine cache + admin write paths |
`app/saas/.../payg/policy/PricingPolicyService.java` |
| `PolicyChangedEvent` — published after admin writes for in-process
cache invalidation | `app/saas/.../payg/policy/PolicyChangedEvent.java`
|
| Admin REST — list / get / create / set-default / set team override /
get effective |
`app/saas/.../payg/policy/admin/PricingPolicyAdminController.java` +
DTOs |
| `PricingPolicyRepository.clearDefaultFlag()` — atomic clear for
set-default | repository update |
| `SaasJpaConfigScanTest` — drift guard against the JPA scan paths going
stale (carried over from the #6460 review concern) | new test |
| V12 default-policy seed (`v1-initial`, 25 pages/unit, 5 MiB/unit,
per-`JobSource` step limits) | `V12__seed_default_payg_policy.sql` |
## Lookup precedence
1. `PaygTeamExtensions.pricingPolicyId` set → return that policy
2. Else return the `pricing_policy` row with `is_default = TRUE`
3. Override row points at a deleted policy → log warn, fall back to
default (safety net for racing deletes)
4. No default → `IllegalStateException` (V12 seed guarantees one exists)
## Cache behaviour
- 30s `expireAfterWrite` Caffeine, max 10k entries, keyed by `teamId`.
- **Single correctness model: the TTL.** Cross-instance propagation is
at-most-30-seconds. The writer instance sees its own change immediately
via the `PolicyChangedEvent` after-commit publish. Other instances pick
it up on the next TTL expiry.
- Admin reads use `getEffectivePolicyUncached` so admins always see
their own write straight back.
**Why no LISTEN/NOTIFY runner.** An earlier cut of this PR included a
Postgres `LISTEN policy_changed` runner so cross-instance propagation
was instant. Dropped — admin policy changes are events-per-week and the
30s TTL is already the correctness floor; the listener was ~250 lines of
nontrivial code (raw JDBC outside HikariCP, daemon thread, reconnect
loop, lock-protected connection lifecycle) for a use case that isn't on
the hot path. Trade-off is documented in `notes/PAYG_DESIGN.md` §9 with
three concrete triggers that would justify reintroducing it (aggressive
cap enforcement, Redis landing for other reasons, real-time admin UI).
## Writes — transactional, fire `PolicyChangedEvent` after commit
- `create(draft)` — rejects pre-set `policy_id` or `is_default=true`
(promotion must go through `setDefault` so the partial unique idx is
freed first).
- `setDefault(id)` — atomically clears the existing default via
`clearDefaultFlag()` then flips the new row. Idempotent: silent no-op if
the row is already default.
- `setTeamOverride(teamId, policyId | null)` — validates the policy
exists before save; `null` clears the override.
`publishOnCommit` uses `TransactionSynchronizationManager.afterCommit`
so listeners never see pre-commit state. Outside a transaction (test
paths) falls through to immediate publish.
## Admin REST surface — `/api/v1/admin/payg/...`
All endpoints `@PreAuthorize("hasRole('ADMIN')")`:
- `GET /policies` — list all
- `GET /policies/{id}` — read one
- `POST /policies` — create new (non-default)
- `POST /policies/{id}/set-default` — atomic promote
- `PUT /teams/{teamId}/policy-override` — set or clear per-team override
- `GET /teams/{teamId}/effective-policy` — cache-bypassing live read
Validation errors → 400, unknown rows → 404.
## Counterpart Supabase PR
[`Stirling-PDF-SaaS#298`](https://github.com/Stirling-Tools/Stirling-PDF-SaaS/pull/298)
— seeds the same V1 default policy on the Supabase side via
`20260528000002_payg_seed_default_policy.sql`.
## Tests
- 17 × `PricingPolicyServiceTest` — lookup precedence, cache hit/miss,
invalidation on event, mutation paths publishing event, error cases.
- 14 × `PricingPolicyAdminControllerTest` — every endpoint's happy path
+ error mapping, DTO defensive-copy invariant.
- 2 × `SaasJpaConfigScanTest` — reflection-based guard that
`payg.repository` is in `@EnableJpaRepositories` and `payg` is in
`@EntityScan`. Without this, new sub-packages can silently fail to wire
at runtime — same class of bug that the #6460 review caught.
Full `:saas:test` BUILD SUCCESSFUL.
## Design doc
`notes/PAYG_DESIGN.md` §7.4 PR-I1 — completes the service half (the
schema half landed in #6460). §9 carries the 30s-TTL trade-off note.
# Description of Changes
Two layers — the `DocumentClassifier` utility plus the full data model
for the new billing engine. Nothing wires the entities into application
behaviour yet; services and controllers land in follow-up PRs.
**Companion PR:**
[Stirling-PDF-SaaS#296](https://github.com/Stirling-Tools/Stirling-PDF-SaaS/pull/296)
— Supabase migration for the v3 dev branch, schema-equivalent to the
Flyway migration in this PR.
## 1. DocumentClassifier (under `payg.docs`)
`DocumentClassifier` computes the doc-unit cost of an uploaded file (or
multi-file input) under a `PricingPolicy`. PDFs read page count via
`stirling.software.jpdfium.PdfDocument`; non-PDFs are bytes-only.
Formula: `max(ceil(pages / docPagesPerUnit), ceil(bytes /
docBytesPerUnit))` clamped to `[1, fileUnitCap]`. Multi-file is the sum
of raw per-file units capped at `fileUnitCap × file_count`.
Two floors, by design: the classifier returns `docUnits` with an
absolute `1` floor for non-empty input; the policy-level
`minChargeUnits` is intentionally applied later, at process-open time in
`JobChargeService`, per design § 3.4 (`unitsForProcess =
max(policy.min_charge_units, docUnits)`). Documented in the interface +
impl javadoc.
Upload bytes are materialised through
`TempFileManager.createManagedTempFile` so jpdfium gets a `Path`; the
temp file auto-deletes on close.
Twelve tests, all in-memory fixtures generated with PDFBox at test time
— no committed binary blobs.
## 2. PAYG data model (under `payg.*`)
JPA entities, repositories, and a Flyway migration covering the full
schema in §6 of the design.
**Enums** (`payg.model`):
`JobSource`, `ProcessType`, `JobStatus`, `JobStepStatus`,
`ArtifactKind`, `LedgerEntryType`, `LedgerBucket`, `ReferenceType`,
`EntitlementState`, `FeatureSet`, `FeatureGate`, `WalletEngine`,
`CapPeriod`, `AutoGroupStrategy`.
**Entities + repositories:**
| Entity | Table | Notes |
|---|---|---|
| `PricingPolicy` | `pricing_policy` | Promoted from a record.
`stepLimits` is `Map<JobSource, Integer>` persisted via normalised child
table `pricing_policy_step_limit`. `stripePriceIds` is `Set<String>`
persisted via `pricing_policy_stripe_price` — currency comes from
`stripe.prices` via Sync Engine, not stored locally. |
| `ProcessingJob` | `processing_job` | UUID PK. Tracks lineage window
via `step_count` and `last_step_at`. |
| `ProcessingJobStep` | `processing_job_step` | Per-tool-call audit. |
| `JobArtifactHash` | `job_artifact_hash` | Composite key `(job_id,
content_hash, kind)`. `content_hash VARCHAR(128)` so multiple signature
schemes coexist as `"type:value"` storage keys. Lineage detector queries
this. |
| `WalletLedgerEntry` | `wallet_ledger` | Append-only, signed
`amount_units`. Two unique indexes kill double-posting. |
| `WalletPolicy` | `wallet_policy` | Per-team engine + cap + degradation
rules + lineage strategy. No `@Version` — admin-only writes (documented
in javadoc). |
| `WalletEntitlementSnapshot` | `wallet_entitlement_snapshot` |
Composite key `(team_id, user_id)`; `user_id = 0` is the team-wide
sentinel. No `@Version` — full-row recompute via
`EntitlementService.recompute` (documented in javadoc). |
| `PaygShadowCharge` | `payg_shadow_charge` | Per-job diff while in
`PAYG_SHADOW` engine mode. |
| `PaygTeamExtensions` | `payg_team_extensions` | Sidecar 1:1 with
`teams` carrying `pricing_policy_id` (per-team override) +
`stripe_customer_id`. Sidecar pattern (mirrors `saas_team_extensions`)
so OSS Hibernate ddl-auto never sees PAYG columns on `teams`. |
**Column adds:**
- `team_memberships.cap_units` (optional per-member sub-cap)
**Width split (intentional, documented in V11):** per-row deltas
(`wallet_ledger.amount_units`, `processing_job.charged_units`) are
`INTEGER` because no single charge realistically approaches 2B units.
Cap and period-rollup columns (`team_memberships.cap_units`,
`wallet_policy.cap_units`,
`wallet_entitlement_snapshot.period_spend_units / period_cap_units`) are
`BIGINT` because they accumulate across a billing period and admins may
legitimately set headroom-cap values into the millions.
**JPA wiring:** `SaasJpaConfig` was updated to include
`stirling.software.saas.payg.repository` in
`@EnableJpaRepositories.basePackages` and `stirling.software.saas.payg`
in `@EntityScan` (covers `payg.policy` / `payg.job` / `payg.wallet` /
`payg.entitlement` / `payg.shadow` recursively). New
`SaasJpaConfigScanTest` reads the annotations reflectively and asserts
every expected package is wired — catches the next time someone adds a
new sub-package without updating the scan paths.
**Migration:** `V11__saas_payg_model.sql` (purely additive).
Schema-equivalent to the Supabase migration in the companion PR —
including the `VARCHAR(128) content_hash` width that's needed for the
multi-signature-scheme storage encoding the lineage layer uses.
## 3. Smoke tests
`PaygEntitiesSmokeTest` exercises each entity via the no-arg ctor JPA
requires, plus getter/setter round-trips and composite-key equality —
catches Lombok/annotation regressions without needing a database.
Real-DB integration coverage lands alongside the services that consume
each entity.
## Why this is safe to land now
- All schema changes are additive — no existing rows modified, no
columns dropped.
- The entities are not yet referenced from any production code path;
they exist for the next PRs to build on.
- The v3 Supabase dev branch picks up the schema via the companion PR;
the main repo's Flyway migration applies the same shape when an instance
boots against a freshly-migrated v3 database.
## Open decisions made
- **Step-limits keyed by `JobSource`** rather than by `ProcessType`.
Captures the "self-hosted gets a different knob" framing in earlier
feedback. Trivially overridable per pricing policy version.
- **Step limits + Stripe price IDs normalised into child tables** rather
than JSONB on `pricing_policy` (per Connor's review on #296). Typed
columns, queryable directly, no JSON parsing.
- **Currency dropped from `pricing_policy_stripe_price`** — it lives on
`stripe.prices.currency` and is resolved via Sync Engine. App is
currency-blind.
## Rollback
Straight `git revert` on this PR. The Supabase migration in #296 is
additive and can be left in place safely — the running app ignores
tables it doesn't reference.
---
## Checklist
- [x] I have read the [Contribution
Guidelines](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/CONTRIBUTING.md)
- [x] I have read the [Stirling-PDF Developer
Guide](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/DeveloperGuide.md)
- [x] I have performed a self-review of my own code
- [x] My changes generate no new warnings
- [x] I have run `task check` (via `./gradlew :saas:test` with
`ENABLE_SAAS=true`) — passes
# Description of Changes
Two narrowly-scoped hardening changes to the credits engine.
## 1. CreditService — move Stripe meter call to `afterCommit`
The Stripe metered-usage call sits inside the surrounding
`@Transactional`, holding the `user_credits` row lock for the duration
of an HTTP round-trip to Supabase. Under load this starves concurrent
debits; a transient Stripe blip rolls back a (correct) free-credit
consumption and forces the caller to retry.
The Stripe call now runs in a `TransactionSynchronization.afterCommit`
hook — DB commits first, Stripe fires immediately after. If Stripe fails
after commit, we log + increment a new `credits.stripe_report.failures`
counter; the idempotency key is stable, so a manual replay recovers
without double-charging.
Applied to both `consumeCreditBySupabaseId` and
`consumeCreditWithWaterfall`.
**Dead-code removed:**
- Unreachable UUID fallback for MDC `requestId` — `CorrelationIdFilter`
already guarantees the key on every request.
- The `"Unable to report usage to Stripe"` `RuntimeException` and its
catch block — the afterCommit refactor eliminates the throw path.
- `StripeRollbackOnFailureTest` — pinned the rollback-on-Stripe-fail
behaviour this refactor replaces.
## 2. `@AutoJobPostMapping` — build-time lint for `resourceWeight`
`UnifiedCreditInterceptor` multiplies `resourceWeight` into the per-call
charge. An endpoint that falls through to the annotation default
produces a charge derived from a value nobody chose.
- Annotation default flipped from `1` to `Integer.MIN_VALUE` (sentinel).
Both runtime readers (`UnifiedCreditInterceptor`, `AutoJobAspect`)
already clamp into `[1, 100]` so behaviour is unchanged.
- New `AutoJobPostMappingWeightTest` scans the classpath and fails the
build if any method leaves the sentinel.
- Initial run caught 11 endpoints relying on the default. Explicit
weights now declared, chosen by comparing to peer endpoints:
- `EditTextController` — LARGE
- `EmailController#sendEmailWithAttachment` — SMALL
- `ConvertPDFToMarkdown` — MEDIUM
- `AttachmentController` (extract/list/rename/delete) — SMALL × 4
- `ConvertImgPDFController` (cbr/cbz ↔ pdf) — MEDIUM × 2, LARGE × 2
## Tests
- `StripeUsageIdempotencyKeyTest` — pins the `(supabaseId, overage,
requestId)` idempotency key shape so Stripe always dedupes a retry.
- `StripeAfterCommitOrderingTest` — pins that `afterCommit` fires after
commit and NOT on rollback.
- `AutoJobPostMappingWeightTest` — the lint itself, plus a self-check
that the classpath scan finds at least 10 `@AutoJobPostMapping` methods
(guards against the lint passing vacuously).
Build verified: `ENABLE_SAAS=true ./gradlew :stirling-pdf:test
:saas:test`.
---
## Checklist
### General
- [x] I have read the [Contribution
Guidelines](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/CONTRIBUTING.md)
- [x] I have read the [Stirling-PDF Developer
Guide](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/DeveloperGuide.md)
(if applicable)
- [ ] I have read the [How to add new languages to
Stirling-PDF](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/devGuide/HowToAddNewLanguage.md)
(if applicable) — no translation changes
- [x] I have performed a self-review of my own code
- [x] My changes generate no new warnings
### Documentation
- [ ] I have updated relevant docs on [Stirling-PDF's doc
repo](https://github.com/Stirling-Tools/Stirling-Tools.github.io/blob/main/docs/)
— internal-billing change, no public docs impact
- [ ] I have read the section [Add New Translation
Tags](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/devGuide/HowToAddNewLanguage.md#add-new-translation-tags)
— N/A
### Translations (if applicable)
- [ ] Not applicable
### UI Changes (if applicable)
- [ ] Not applicable
### Testing (if applicable)
- [x] I have run `task check` (via `./gradlew :stirling-pdf:test
:saas:test` with `ENABLE_SAAS=true`) — passes
- [x] I have tested my changes locally
# Description of Changes
## What & why
Customers using ADFS (or any generic OIDC provider that doesn't emit
`email`) hit `Attribute value for 'email' cannot be null` during OAuth2
login with no visibility into what claims the provider actually sent.
The only available remedy was guessing at
`security.oauth2.useAsUsername` until something worked.
This PR adds a new opt-in `security.oauth2.debugLogging` flag (default
`false`). When enabled, `CustomOAuth2UserService` logs:
- All ID token claims (sorted, with values)
- All UserInfo endpoint claims (if any)
- The merged attribute key set Spring exposes to `getAttribute()`
- The value the configured `useAsUsername` actually resolved to
- A **`Hint:`** line listing the claim keys present in the token that
map to a valid `UsernameAttribute` enum value — i.e. exactly what the
operator could put in `useAsUsername` to make login work
Logged at `INFO` on the success path and `ERROR` on failure (inside the
existing `catch (IllegalArgumentException)` block that throws
`OAuth2AuthenticationException`). The block is wrapped with a `[OAUTH2
DEBUG] ... [/OAUTH2 DEBUG]` banner and ends with a PII warning so
operators don't leave it on in production.
Default off → zero observable change for anyone not actively
troubleshooting.
## Files changed
| File | Why |
|---|---|
| `app/common/.../ApplicationProperties.java` | New `debugLogging` field
on the `OAUTH2` config class with javadoc warning about PII |
| `app/core/src/main/resources/settings.yml.template` | Documents
`oauth2.debugLogging` so it appears on next startup |
| `app/proprietary/.../security/service/CustomOAuth2UserService.java` |
Emits the claim dump + suggestion hint when the flag is on |
|
`app/proprietary/.../security/service/CustomOAuth2UserServiceDebugLoggingTest.java`
(new) | Unit test: mocks the OIDC delegate, asserts off-path is silent
and on-path emits the dump with the right Hint contents |
## End-to-end verification
Ran the bundled `testing/compose/docker-compose-keycloak-oauth.yml`
Keycloak realm, configured `security.oauth2.useAsUsername: mail`
(Keycloak emits `email`, not `mail`) and `provider: demarest` (matches
the original customer bug report). Triggered the OAuth flow at
`http://localhost:8080/oauth2/authorization/demarest` and confirmed:
- The ERROR-level dump fires with the full 19-claim ID token decoded
- `-- Value at 'mail' : <NULL — this is why login fails>` correctly
identifies the missing claim
- `-- Hint:` correctly suggests `[email, family_name, given_name,
preferred_username]` (the four keys present that map to valid
`UsernameAttribute` values)
- Auth still fails with the original `OAuth2AuthenticationException` —
no change to control flow, just added diagnostic logging
Unit test (`CustomOAuth2UserServiceDebugLoggingTest`) covers both
branches.
## Reviewer notes
- **No new public APIs.** The flag is config-only; no servlet endpoints
exposed.
- **PII is logged when the flag is on.** This is the whole point —
operators need to see the claims to fix their config — but it's gated,
defaults off, and the dump self-documents with a `WARNING: ... Set
security.oauth2.debugLogging=false once troubleshooting is complete.`
footer.
- **Why log everything, not just sub/email?** Because the operator
doesn't know in advance which claim they actually want. ADFS uses `upn`
in some configs and `preferred_username` in others; Azure AD uses `oid`;
the customer here had neither. Dumping the full set is the only way to
make the diagnostic self-service.
- **Out of scope for this PR (follow-ups):**
- The `UsernameAttribute` enum doesn't include `upn` / `unique_name`
(common ADFS claims). If the customer's token only has `upn`, the Hint
will be empty even though the operator can see `upn` in the dump. Worth
a separate PR to extend the enum.
- The known-provider validator in `Provider.java` (rejects e.g.
`useAsUsername: mail` for `provider: keycloak` at startup) bypasses our
diagnostic for those provider names. ADFS customers using `provider:
<name>` fall into the `default` branch so are not affected — but it's a
sharp edge worth documenting.
---
## Checklist
### General
- [x] I have read the [Contribution
Guidelines](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/CONTRIBUTING.md)
- [x] I have read the [Stirling-PDF Developer
Guide](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/DeveloperGuide.md)
(if applicable)
- [ ] I have read the [How to add new languages to
Stirling-PDF](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/devGuide/HowToAddNewLanguage.md)
(if applicable) — N/A, backend-only change
- [x] I have performed a self-review of my own code
- [x] My changes generate no new warnings
### Documentation
- [ ] Doc-repo update (if functionality has heavily changed) —
diagnostic flag is self-documenting via the `settings.yml.template`
comment and the in-log warning; happy to add a doc-repo entry if
reviewers want one
- [ ] Translation tags — N/A
### UI Changes (if applicable)
- [ ] N/A — backend-only
### Testing (if applicable)
- [x] Unit test added (`CustomOAuth2UserServiceDebugLoggingTest`)
covering on/off paths and Hint correctness
- [x] End-to-end verified locally against bundled Keycloak compose with
intentionally misconfigured `useAsUsername`
- [x] Full `:proprietary:test` suite passes
## Summary
Adds a new AI specialist that finds **textual contradictions** across
one or more PDFs — conflicting claims, recommendations, points of view,
contested facts — built entirely in Python on top of the new
`DocumentService` + `ChunkedReasoner` stack from #6314. Replaces the
closed#6304, which was started before #6314 landed and therefore
over-engineered (Java orchestrator, two-round handshake, resume
artifact, discriminated-union lift).
Two commits:
1. **`refactor(engine): extract ChunkedMapper[T] from ChunkedReasoner`**
— pure refactor, public API of ChunkedReasoner unchanged. New
`ChunkedMapper[T: BaseModel]` is a generic parallel-chunk primitive
(slicing, semaphore, time-bounded extraction, cancellation drain,
progress events) that's now a peer to ChunkedReasoner rather than locked
inside it. The compression loop stays on ChunkedReasoner where it
belongs.
2. **`feat(ai): add Contradiction Agent on ChunkedMapper`** — the agent
itself, plus integrations into `PdfReviewAgent` and `PdfQuestionAgent`.
## Architecture
- **Python-only.** No Java code. No `AgentToolId.CONTRADICTION_AGENT`.
No dedicated HTTP endpoint. No resume artifact, no discriminated-union
lift in `contracts/common.py`. Detector runs inside the Python engine
and the Python engine alone.
- **Review path** (`PdfReviewAgent`): a new
`ContradictionIntentClassifier` fires on contradiction-flavoured
prompts; agent runs detection synchronously and emits a single
`EditPlanResponse(steps=[ADD_COMMENTS])`. Single-turn flow — no resume.
- **Question path** (`PdfQuestionAgent`): a new
`ContradictionCapability` joins `RagCapability` and
`WholeDocReaderCapability` in the smart-model toolset, exposing
`find_contradictions(query)`. The smart model picks it from the toolset
alongside `search_knowledge` and `read_full_document`.
## Inside `ContradictionDetector.detect()`
1. `DocumentService.read_pages(file_id)` → ordered `list[Page]`.
2. `ChunkedMapper[_ExtractedClaims].map_pages(...)` — char-budgeted
multi-page slicing; each slice runs the claim-extractor LLM in parallel
under a semaphore.
3. Page-traceability: the extractor returns `_ExtractedClaim.page`
(which `[Page N]` marker the claim came from). The wrapper validates
`page ∈ chunk.pages`; if not, mechanical fallback searches the chunk's
page text for the verbatim quote and reassigns. If still no match, drop
the claim.
4. `Claim.anchor_quality: Literal[\"verbatim\", \"paraphrased\"]` is set
by a substring check against the declared page's text. Verbatim quotes
feed `anchor_text` for snap-to-quote add-comments placement; paraphrased
ones fall back to margin geometry.
5. Subject canonicalisation: ONE fast-model LLM call collapses synonyms
across the document. Fails open to lexical bucketing.
6. Pre-filters: drop identical-quote pairs; drop same-page same-polarity
paraphrases.
7. Per-bucket pair detection in parallel (separate semaphore, cap 5).
Buckets > 12 claims chunk into windows of 12 with overlap 2; pairs
deduped across overlapping windows by frozen `(i, j)` index pair.
8. Summary fast-model call with fallback string on error.
## Prompt-injection hardening
Every prompt that interpolates user-supplied or PDF-extracted text wraps
content in `<user_message>` / `<verdict>` / `<content>` tags with an
explicit SECURITY preamble instructing the model to treat tagged content
as data only.
## Limitations
- **Combined math + contradiction intent**: when both intent classifiers
fire on the same prompt, contradiction takes precedence and the math
intent is silently dropped. Documented in the Review module docstring
and pinned by
`test_review_integration.py::test_contradiction_precedence_over_math`.
- **Cross-window contradiction reach**: within a subject bucket, pairs
more than ~10 claim indices apart in the same chunked window may be
missed by the overlap-2 strategy. Documented in `test_detector.py`.
Acceptable for v1.
## Settings (engine/src/stirling/config/settings.py)
```python
contradiction_detect_concurrency = 5 # per-bucket detector semaphore
contradiction_bucket_chunk_size = 12 # max claims per detector call
contradiction_bucket_chunk_overlap = 2 # overlap for >threshold buckets
```
`chars_per_slice` and extraction concurrency are reused from the
existing `chunked_reasoner_*` settings.
## Test plan
- [x] `uv run pytest tests/ -v` — **245/245 pass** (210 pre-existing +
35 new)
- [x] `uv run ruff check src/ tests/` — clean
- [x] `uv run pyright src/stirling/agents/contradiction/
src/stirling/contracts/contradiction.py` — 0 errors
- [x] `./gradlew :proprietary:test` — green; no Java was touched, but
verified untouched
- [x] Page-traceability tests cover: valid page kept, hallucinated page
dropped, mechanical-reassign on misattribution, anchor-quality verbatim
vs paraphrased
- [x] Review integration: ADD_COMMENTS plan with two paired CommentSpecs
per contradiction; NeedIngestResponse precheck; precedence vs math
intent pinned
- [x] Question integration: all three capabilities wired into
smart-model toolset; `find_contradictions` returns formatted report text
- [x] ChunkedMapper standalone: slicing, multi-chunk ordering, worker
failures, timeouts, cancellation drain, semaphore saturation
- [x] ChunkedReasoner regression: all pre-existing tests pass unchanged
after the internal split
## Relationship to closed#6304#6304 was closed in favour of this PR. The closed PR predated #6314 and
modelled the agent as a Java-orchestrated two-round examine/deliberate
flow with its own HTTP endpoint and a discriminated-union resume
artifact. With #6314 making full ordered page text available to the
engine via `DocumentService.read_pages`, none of that is needed. Net
effect: drop ~600 lines of Java, drop the two-round handshake, drop the
`ToolReportArtifact` lift, while ending up with a more scalable agent
(chunk-based instead of page-based extraction; tested to
ChunkedReasoner-equivalent scale).
## Summary
`#6145` (port picker) and `#6244` (gradle unification) combined to break
`task dev` / `task dev:all` on Windows. Two independent regressions,
both addressed here.
### 1. `find-free-port.ps1` panic (`#6145`)
```
$ task --dry dev
The argument 'scriptsfind-free-port.ps1' to the -File parameter does not exist.
panic: ended up with a non-nil exitStatus.err but a zero exitStatus.code
```
- **Backslash stripped.** `Taskfile.yml` had
`scripts\find-free-port.ps1`. go-task pipes the `sh:` block through
mvdan/sh, which treats `\` as a POSIX escape and silently drops it,
leaving `scriptsfind-free-port.ps1`. PowerShell can't find the file,
exits non-zero, mvdan/sh panics on the inconsistent exit status.
Switched to a forward slash.
- **`-Preferred 8080,5173` is brittle.** Relies on PowerShell parsing
the comma-list into `[int[]]`. Dropped the named flag; switched the
script's `param` to `[Parameter(ValueFromRemainingArguments =
$true)][int[]]$Preferred`; pass each port as its own positional token
(matching how the bash variant is called).
### 2. `bash gradlew` can't find Java on Windows (`#6244`)
```
[backend:dev] ERROR: JAVA_HOME is not set and no 'java' command could be found in your PATH.
```
`#6244` unified all gradle invocations under `bash gradlew` on the
assumption that Git-Bash inherits Windows-side env. It doesn't (and
neither does WSL bash, which can also shadow `bash` on a developer's
PATH). The standard Adoptium/Temurin installer sets `JAVA_HOME` in the
Windows env only, so `bash gradlew` fails before Spring Boot even loads.
Restored the per-platform branch for every backend task: `cmd /c
".\gradlew.bat ..."` on Windows, `./gradlew ...` on Linux/macOS. Two
Windows-specific gotchas to be aware of for future edits:
- mvdan/sh strips `\` outside of double quotes, so `cmd /c
.\gradlew.bat` ends up as `.gradlew.bat`. The entire payload must be
wrapped in double quotes (`cmd /c "..."`).
- Modern Windows excludes cwd from cmd.exe's search path, so the leading
`.\` is required — bare `cmd /c gradlew.bat` errors with "is not
recognized" even from the repo root.
## Test plan
- [x] `task --dry dev` / `task --dry dev:all` on Windows resolve ports
cleanly and dispatch the inner tasks
- [x] `powershell -NoProfile -File scripts/find-free-port.ps1 8080 5173
5001` prints three ports
- [x] `task backend:dev PORT=8081` on Windows starts Spring Boot in ~9s;
`/api/v1/info/status` returns `{"status":"UP"}`
- [x] `task dev:all` on Windows brings up all three services healthy:
backend `/api/v1/info/status` 200, engine `/health` 200, frontend `/`
200
- [ ] Linux / macOS path unaffected (only the Windows branches changed
in behaviour)
## Summary
The frontend was rerendering excessively across many interactions —
typing, clicking tools, opening modals, toggling the sidebar — because
**multiple compounding ref-instability cascades defeated `memo()` checks
in hot paths**. This PR fixes the cascades structurally.
Seven focused commits, low-to-high blast radius:
1. `perf(contexts): memoize BannerContext provider value`
2. `perf(contexts): memoize CommentAuthor and ActiveDocument provider
values`
3. `perf(contexts): memoize AppConfigContext provider value`
4. `perf(useToolManagement): stop spreading tool entries to keep refs
stable` — **root-cause fix**
5. `refactor(ToolPicker): hoist module-scope styles and helpers`
6. `feat(ToolWorkflowContext): add ref-stable Actions and Data subset
contexts` — additive
7. `perf(tools): migrate hot consumers to slim contexts and wrap in
memo()`
(Plus `style: apply prettier formatting` for CI.)
## What was wrong
Whenever something high-up in the tree caused a render, a chain of
unstable references propagated downward and forced every `ToolButton` to
re-execute its full body (hooks, derived computations, hook
subscriptions to other contexts). The chain:
- **4 unstable Context providers** (`Banner`, `CommentAuthor`,
`ActiveDocument`, `AppConfig`) were passing fresh `value={{ … }}`
objects on every render. Every consumer rerendered on every ancestor
render.
- **`useToolManagement.toolRegistry`** spread `{...baseTool, name,
description}` — a no-op spread that manufactured a new tool object
identity on every memo recompute.
- **The big `ToolWorkflowContext`** (25+ fields including
`state.searchQuery`) rebuilt its entire value on every
keystroke/click/toggle, forcing every `useToolWorkflow()` consumer (~36
files) to rerender.
- **`useToolNavigation`** transitively subscribed every `ToolButton` to
the full workflow context.
- **`ToolButton` & `ToolPicker`** weren't `memo()`-wrapped, so nothing
checked.
- **`ToolPanel`** passed inline `onSelect={(id) =>
handleToolSelect(...)}` — fresh ref every render, defeats child
memoization.
- **`ToolPicker`** allocated inline styles / `[]` / `toTitleCase` inside
the function body — churned `useToolSections`'s internal memo.
## Interaction matrix — what improves
The PR fixes the underlying ref-stability problem; the same fix benefits
*every* interaction that previously triggered the cascade:
| Interaction | Before | After |
|---|---|---|
| **Typing in tool search** | All visible buttons rerender per keystroke
| Only buttons whose matched-text changes rerender |
| **Clicking a tool** | All 36 `useToolWorkflow()` consumers rerender |
Only previously-selected and newly-selected buttons rerender (via
`isSelected` prop) |
| **Toggling sidebar / panel mode / reader mode** | Every tool button
rerenders | Tool components stay still (slim context doesn't see UI
state) |
| **Switching workbench / navigation** | `handleToolSelect` identity
changes → cascades through `onSelect` props | Ref-stabilized in Actions
context. Identity stable. Children's memo bails |
| **Modal/dialog open/close** | AppConfig churns → every `useAppConfig`
consumer rerenders (ToolButton reads `premiumEnabled`) | AppConfig
memoized; consumers rerender only when config changes |
| **Banner show/hide** | BannerProvider value churns → every consumer
rerenders on any ancestor render | Memoized; AppLayout rerenders only
when banner content changes |
| **Any state update high in the tree** | Compounding cascade defeats
memo everywhere | Stable subscriptions; memo bails out |
## Evidence
Per-keystroke prop instability on `ToolButton` (cleanest measurable
signal, captured via custom memo comparators logging which prop refs
differ):
| | `tool` ref diffs | `onSelect` ref diffs | `matchedSynonym` value
diffs | Total |
|---|---|---|---|---|
| Before | 18 | 18 | 6 | **42** |
| After | 0 | 0 | 6 | **6 (all legitimate)** |
→ **86% reduction** in spurious per-keystroke prop instability. The 6
remaining matched-synonym diffs are correct (different substring
highlighted per keystroke).
Context value rebuild counts during a keystroke (verified with
instrumented `useMemo` factories): `useToolWorkflowData=0`,
`useToolWorkflowActions=0`, `AppConfigContext=0`.
The same stabilization applies to click/toggle/modal interactions — they
were all driven by the same cascading invalidations.
## Honest caveat on render-count metrics
`React.Profiler` counts and function-body execution counts in **dev
mode** came back identical before vs after (StrictMode + concurrent
rendering + Mantine internal commits dominate the numbers). The PR's
value is measured against the **prop-stability signal** above, not
Profiler counts. Production builds — where StrictMode doesn't
double-render and Mantine internals aren't constantly committing — will
show memo bail out properly.
## Risk × benefit
| # | Commit | Risk | Benefit |
|---|--------|------|---------|
| 1 | BannerContext memo | ⬛ Trivial | 🟦 Small |
| 2 | CommentAuthor + ActiveDocument memo | ⬛ Trivial | 🟦 Small |
| 3 | AppConfig memo | ⬛ Trivial | 🟦 Moderate (wide consumer base) |
| 4 | useToolManagement spread removal | ⬛ Trivial | 🟥 **High (root
cause)** |
| 5 | ToolPicker hoist | ⬛ Trivial | 🟦 Small |
| 6 | ToolWorkflowContext split | 🟧 Low-Med | 🟥 **High (foundation)** |
| 7 | Hot consumer migration + memo | 🟧 Low-Med | 🟥 **High
(actualization)** |
Commit 6 introduces an invariant: ref-stabilized callbacks in the
Actions context must only be invoked from event handlers (post-commit),
never during render. All current call sites comply.
## Test plan
- [x] `npx playwright test --project=stubbed` — 145 / 6 skipped / 0
failed before and after.
- [x] Targeted regression: `main-dashboard`, `tool-search`, `navigation`
— 11/11 passing.
- [x] CI passing on commits (one infrastructure flake on
`docker-compose-tests` — "No space left on device" — unrelated;
rerunning).
- [ ] Manual sanity check in a dev build after merge.
## What this enables
The same Actions + Data subset-context pattern can be applied to
`FileContext`, `NavigationContext`, and other big contexts. The
foundation is in place.
## Add Taskfile for unified dev workflow
### Summary
- Introduces [Taskfile](https://taskfile.dev/) as the single CLI entry
point for all development workflows across backend, frontend, engine,
Docker, and desktop
- ~80 tasks organized into 6 namespaces: `backend:`, `frontend:`,
`engine:`, `docker:`, `desktop:`, plus root-level composites
- All CI workflows migrated to use Task
- Deletes `engine/Makefile` and `scripts/build-tauri-jlink.{sh,bat}` —
replaced by Task equivalents
- Removes redundant npm scripts (`dev`, `build`, `prep`, `lint`, `test`,
`typecheck:all`) from `package.json`
- Smart dependency caching: `sources`/`status`/`generates`
fingerprinting, CI-aware `npm ci` vs `npm install`, `run: once` for
parallel dep deduplication
### What this does NOT do
- Does not replace Gradle, npm, or Docker — Taskfile is a thin
orchestration wrapper
- Does not change application code or behavior
### Install
```
npm install -g @go-task/cli # or: brew install go-task, winget install Task.Task
```
### Quick start
```
task --list # discover all tasks
task install # install all deps
task dev # start backend + frontend
task dev:all # also start AI engine
task test # run all tests
task check # quick quality gate (local dev)
task check:all # full CI quality gate
```
### Test plan
- [ ] Install `task` CLI and run `task --list` — verify all tasks
display
- [ ] Run `task install` — verify frontend + engine deps install
- [ ] Run `task dev` — verify backend + frontend start, Ctrl+C exits
cleanly
- [ ] Run `task frontend:check` — verify typecheck + lint + test pass
- [ ] Run `task desktop:dev` — verify jlink builds are cached on second
run
- [ ] Verify CI passes on all workflows
---------
Co-authored-by: James Brunton <[email protected]>
## Summary
- Added "Alpha" badge to the File Storage & Sharing nav item in the
settings sidebar
- Added "Alpha" badge to the File Storage & Sharing page title
- Removed the old inline "(Alpha)" text from the Enable Group Signing
label
- Restructured all toggle cards so the switch is anchored to the right
of each row
- Tightened spacing between cards for a more compact layout
- Extended `ConfigNavItem` interface with optional `badge` and
`badgeColor` fields for reuse elsewhere
<img width="1696" height="1057" alt="image"
src="https://github.com/user-attachments/assets/77ac8276-ed65-4cae-8470-65de8f56dd74"
/>
## Summary
This PR adds full desktop (Tauri) support for the shared signing feature
when connected to a self-hosted server, and fixes several bugs
discovered during that work.
### Feature gating
Shared signing, file sharing, and share links are proprietary server
features that require an authenticated self-hosted session. Previously
these were read directly from `config` with no awareness of connection
mode or auth state, meaning the UI could appear in SaaS/local mode or
when logged out.
- Introduce `useGroupSigningEnabled` and `useSharingEnabled` hooks with
core implementations (web behaviour unchanged) and desktop overrides
that require `selfhosted` mode + an active authenticated session
- Extract shared subscription logic into `useSelfHostedAuth` (connection
mode + auth state + config refetch)
- `QuickAccessBar` now derives all three flags from the hooks instead of
raw config
### Config timing fix
When a user logs in via the SetupWizard, the `jwt-available` event fires
a config fetch *before* the mode is switched to `selfhosted`. This meant
the config was fetched from the local bundled backend (port ~59567)
which has no knowledge of `storageGroupSigningEnabled`, causing the
group signing button to stay hidden until a full page refresh.
`useSelfHostedAuth` detects the mode transition and triggers a fresh
config fetch at the correct moment, after the self-hosted URL is active.
### Bug fixes
**`SignPopout.tsx`** — Manually setting `Content-Type:
multipart/form-data` on two `FormData` POST requests stripped the
auto-generated boundary, causing a `400 bad multipart` from the server.
Removed the explicit headers so Axios sets them correctly.
**`tauriHttpClient.ts`** — `response.json()` was called before
`response.ok` was checked. A plain-text error body from the server (e.g.
`"Cannot sign..."`) caused a `SyntaxError` that fell into the network
error catch block and was reported as `ERR_NETWORK`, hiding the real
failure. The fix checks `response.ok` first, reads error bodies as text,
and handles empty 200 bodies (returning `null` instead of throwing).
---
## Testing
### Prerequisites
- Desktop app running in self-hosted mode pointed at a local
Stirling-PDF instance (`http://localhost:8080`)
- The self-hosted instance has group signing and storage enabled in
settings
- At least two user accounts on the self-hosted instance
### 1. Feature gating — group signing button
| Step | Expected |
|---|---|
| Open the desktop app in **local mode** (no server configured) | Group
signing button absent from QuickAccessBar |
| Switch to self-hosted mode but **do not log in** | Group signing
button absent |
| Log in to the self-hosted server | Group signing button appears
without requiring a page refresh |
| Log out | Group signing button disappears immediately |
| Log back in | Group signing button reappears without a page refresh |
### 2. Feature gating — file sharing
Repeat the same steps above, verifying the share and share-link buttons
in the file manager follow the same visibility rules.
### 3. Create a signing session
1. Log in, open the group signing panel from QuickAccessBar
2. Select a PDF, add a participant, configure signature defaults and
submit
3. Verify the session is created successfully (no `400 bad multipart`
error)
### 4. Participant signing
1. As the invited participant, open the signing request from
QuickAccessBar
2. Upload or draw a signature and submit
3. Verify signing completes successfully (no `ERR_NETWORK` error)
### 5. Error surfacing
1. Attempt an action that the server rejects (e.g. sign a document with
an invalid certificate)
2. Verify the actual server error message is shown rather than a generic
network error
## Fix Playwright E2E tests and expand CI to run full suite
### Problem
The full Playwright suite was broken in two ways:
1. **`ConvertE2E.spec.ts` crashed at import time** —
`conversionEndpointDiscovery.ts` imported a React hook at the top level,
which pulled in the entire component tree. That chain eventually
required `material-symbols-icons.json` (a generated file that didn't
exist), crashing module resolution before any tests ran.
2. **CI only ran cert validation tests** — both `build.yml` and
`nightly.yml` hardcoded `src/core/tests/certValidation` as the test
path, silently ignoring everything else.
### Changes
**`ConvertE2E.spec.ts` — complete rewrite**
The old tests were useless in practice: all 9 dynamic conversion tests
were permanently skipped unless a real Spring Boot backend was running
(they called a live `/api/v1/config/endpoints-enabled` endpoint at
module load time). Replaced with 4 focused tests that use `page.route()`
mocking — no backend required, same pattern as
`CertificateValidationE2E`.
New tests cover:
- Convert button absent before a format pair is selected
- Successful PDF→PNG conversion shows a download button (mocked API
response)
- API error surfaces as an error notification
- Convert button appears and is enabled after selecting valid formats
**`conversionEndpointDiscovery.ts` — deleted**
Only existed to support the old tests. The `useConversionEndpoints`
React hook it exported was never imported anywhere else.
**`ReviewToolStep.tsx`**
Added `data-testid="download-result-button"` to the download button —
required for the happy-path test assertion.
**CI workflows (`build.yml`, `nightly.yml`)**
- Added a `Generate icons` step before Playwright runs (`node
scripts/generate-icons.js`) — the icon JSON is generated by `npm run
dev` locally but skipped by `npm ci` in CI
- Removed the `src/core/tests/certValidation` path filter so the full
suite runs
## PR: Certificate Pre-Validation for Document Signing
### Problem
When a participant uploaded a certificate to sign a document, there was
no validation at submission time. If the certificate had the wrong
password, was expired, or was incompatible with the signing algorithm,
the error only surfaced during **finalization** — potentially days
later, after all other participants had signed. At that point the
session is stuck with no way to recover.
Additionally, `buildKeystore` in the finalization service only
recognised `"P12"` as a cert type, causing a `400 Invalid certificate
type: PKCS12` error when the **owner** signed using the standard
`PKCS12` identifier.
---
### What this PR does
#### Backend — Certificate pre-validation service
Adds `CertificateSubmissionValidator`, which validates a keystore before
it is stored by:
1. Loading the keystore with the provided password (catches wrong
password / corrupt file)
2. Checking the certificate's validity dates (catches expired and
not-yet-valid certs)
3. Test-signing a blank PDF using the same `PdfSigningService` code path
as finalization (catches algorithm incompatibilities)
This runs on both the participant submission endpoint
(`WorkflowParticipantController`) and the owner signing endpoint
(`SigningSessionController`), so both flows are protected.
#### Backend — Bug fix
`SigningFinalizationService.buildKeystore` now accepts `"PKCS12"` and
`"PFX"` as aliases for `"P12"`, consistent with how the validator
already handles them. This fixes a `400` error when the owner signed
using the `PKCS12` cert type.
#### Frontend — Real-time validation feedback
`ParticipantView` gains a debounced validation call (600ms) triggered
whenever the cert file or password changes. The UI shows:
- A spinner while validating
- Green "Certificate valid until [date] · [subject name]" on success
- Red error message on failure (wrong password, expired, not yet valid)
- The submit button is disabled while validation is in flight
#### Tests — Three layers
| Layer | File | Coverage |
|---|---|---|
| Service unit | `CertificateSubmissionValidatorTest` | 11 tests — valid
P12/JKS, wrong password, corrupt bytes, expired, not-yet-valid, signing
failure, cert type aliases |
| Controller unit | `WorkflowParticipantValidateCertificateTest` | 4
tests — valid cert, invalid cert, missing file, invalid token |
| Controller integration | `CertificateValidationIntegrationTest` | 6
tests — real `.p12`/`.jks` files through the full controller → validator
stack |
| Frontend E2E | `CertificateValidationE2E.spec.ts` | 7 Playwright tests
— all feedback states, button behaviour, SERVER type bypass |
#### CI
- **PR**: Playwright runs on chromium when frontend files change (~2-3
min)
- **Nightly / on-demand**: All three browsers (chromium, firefox,
webkit) at 2 AM UTC, also manually triggerable via `workflow_dispatch`
Tested:
* Can sign in on saas -> can run local tools with or without credits->
can run saas only tools (if credits) -> can't run saas only tools
without credits
* Can sign in self-hosted -> can run all tools on remote if available ->
can run local when self-hosted unavailable
Clouds show on saas tools when connected
Tools are disabled when connected to self-hosted but cannot find server.
You also get banner
#cantwaitforplaywritetests
* Adds a fallback mechanism so the desktop app routes tool operations to
the local bundled backend when the user's self-hosted Stirling-PDF
server goes offline, and disables tools in the UI that aren't supported
locally.
* `selfHostedServerMonitor.ts` independently polls the self-hosted
server every 15s and exposes which tool endpoints are unavailable when
it goes offline
* `operationRouter.ts` intercepts operations destined for the
self-hosted server and reroutes them to the local bundled backend when
the monitor reports it offline
* `useSelfHostedToolAvailability.ts` feeds the offline tool set into
useToolManagement, disabling affected tools in the UI with a
selfHostedOffline reason and banner warning
- `SelfHostedOfflineBanner `is a dismissable (session-only) gray bar
shown at the top of the UI when in self-hosted mode and the server goes
offline. It shows:
PostHog is now initialized with persistence: 'memory' so no cookies are
written on first load. Consent is handled in a PostHogConsentSync
component that switches to localStorage+cookie persistence only when the
user accepts, using the official @posthog/react package (cherry-picked
from 14aaf64)
---------
Co-authored-by: James Brunton <[email protected]>
Makes sure settings step collapses in results step
Makes sure result step doesn't always reset even in development for
baseTool
Make sure result step doesn't reset for convert
Custom processors can now return consume all inputs flag. This allows to
have many inputs to single output consumption
Fixed multi call conversion logic
2025-12-03 17:39:49 +00:00
ConnorYohGitHubCopilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com>
Removed current plan section from static plan to match connected version
stripe publishable key not required to show plans or checkout in hosted
version
lazy load plans when needed not on load
* Added ability to add seats to enterprise
* first logged in date on people page
* Remove Premium config section
* Cleanup add seat flow
* Shrink numbers in plan
* Make editing text a server feature in the highlights
* default to dollar pricing
* clear checkout logic when crash
* Recongnise location and find pricing
* Payment successful page
---------
Co-authored-by: Connor Yoh <[email protected]>
Created PrivateContent Component to be used as wrapper.
This way all tools that need to obscure contents can update this
wrapper.
---------
Co-authored-by: Connor Yoh <[email protected]>
Co-authored-by: James Brunton <[email protected]>
# Description of Changes
Please provide a summary of the changes, including:
## Add PDF File Association Support for Tauri App
### 🎯 **Features Added**
- PDF file association configuration in Tauri
- Command line argument detection for opened files
- Automatic file loading when app is launched via "Open with"
- Cross-platform support (Windows/macOS)
### 🔧 **Technical Changes**
- Added `fileAssociations` in `tauri.conf.json` for PDF files
- New `get_opened_file` Tauri command to detect file arguments
- `fileOpenService` with Tauri fs plugin integration
- `useOpenedFile` hook for React integration
- Improved backend health logging during startup (reduced noise)
### 🧪 **Testing**
See
* https://v2.tauri.app/start/prerequisites/
*
[DesktopApplicationDevelopmentGuide.md](DesktopApplicationDevelopmentGuide.md)
```bash
# Test file association during development:
cd frontend
npm install
cargo tauri dev --no-watch -- -- "path/to/file.pdf"
```
For production testing:
1. Build: npm run tauri build
2. Install the built app
3. Right-click PDF → "Open with" → Stirling-PDF
🚀 User Experience
- Users can now double-click PDF files to open them directly in
Stirling-PDF
- Files automatically load in the viewer when opened via file
association
- Seamless integration with OS file handling
---
## Checklist
### General
- [ ] I have read the [Contribution
Guidelines](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/CONTRIBUTING.md)
- [ ] I have read the [Stirling-PDF Developer
Guide](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/DeveloperGuide.md)
(if applicable)
- [ ] I have read the [How to add new languages to
Stirling-PDF](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/HowToAddNewLanguage.md)
(if applicable)
- [ ] I have performed a self-review of my own code
- [ ] My changes generate no new warnings
### Documentation
- [ ] I have updated relevant docs on [Stirling-PDF's doc
repo](https://github.com/Stirling-Tools/Stirling-Tools.github.io/blob/main/docs/)
(if functionality has heavily changed)
- [ ] I have read the section [Add New Translation
Tags](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/HowToAddNewLanguage.md#add-new-translation-tags)
(for new translation tags only)
### UI Changes (if applicable)
- [ ] Screenshots or videos demonstrating the UI changes are attached
(e.g., as comments or direct attachments in the PR)
### Testing (if applicable)
- [ ] I have tested my changes locally. Refer to the [Testing
Guide](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/DeveloperGuide.md#6-testing)
for more details.
---------
Co-authored-by: Connor Yoh <[email protected]>
Co-authored-by: James Brunton <[email protected]>
Co-authored-by: James Brunton <[email protected]>
Google drive oss. Shouldn't have any effect on pr deployment.
Mainly the removal of the old integration via backend.
I have added the picker service and lazy loading of the required google
dependency scripts when the necessary environment variables have been
implemented.
---------
Co-authored-by: Connor Yoh <[email protected]>
Co-authored-by: James Brunton <[email protected]>
* Top Controls only show when files > 0
* Moved content down so top controls don't obscure
* Viewer background set to match workbench and shadow around pages added
so that page boundaries are visible
* unsaved-changes modal rework
---------
Co-authored-by: Connor Yoh <[email protected]>
All implemented tools now support automation bar Sign. Sign will need
custom automation UI support
---------
Co-authored-by: Connor Yoh <[email protected]>
Co-authored-by: Reece Browne <[email protected]>
Produced PDFs go into recent files
Undo button added to review state
Undo causes undoConsume which replaces result files with source files.
Removes result files from recent files too
---------
Co-authored-by: Connor Yoh <[email protected]>
Changes fallback pdfa extension pdf when a filename isnt returned for
convert.
Also fixes ui bug with footer and pop ups
---------
Co-authored-by: Connor Yoh <[email protected]>
Added post hog project - always enabled
Added scarf pixel - Always enabled
Reworked Url navigation
Forward and back now works without reloading page
---------
Co-authored-by: Connor Yoh <[email protected]>
- [x] Cleanup Automation output name garbage
- [x] Remove Cross button on first two tools
- [x] Automation creation name title to make clearer to the user
- [x] Colours for dark mode on automation tool settings are bad
- [x] Fix tool names not using correct translated ones
- [x] suggested Automation Password needs adding to description
- [x] Allow different filetypes in automation
- [x] Custom Icons for automation
- [x] split Tool wasn't working with merge to single pdf
---------
Co-authored-by: Connor Yoh <[email protected]>
Co-authored-by: James Brunton <[email protected]>
- Added optional title for tool workflow - Not added to any tool. Just
there for when we need it
- Added add files button to files step
- renamed Local files button in filemanager to Upload Files
-
---------
Co-authored-by: Connor Yoh <[email protected]>
Co-authored-by: James Brunton <[email protected]>
* automate feature
* Moved all providers to app level to simplify homepage
* Circular dependency fixes
* You will see that now toolRegistry gets a tool config and a tool
settings object. These enable automate to run the tools using as much
static code as possible.
---------
Co-authored-by: Connor Yoh <[email protected]>
Component Extraction & Context Refactor - Summary
🔧 What We Did
- Extracted HomePage's 286-line monolithic component into focused parts
- Created ToolPanel (105 lines) for tool selection UI
- Created Workbench (203 lines) for view management
- Created ToolWorkflowContext (220 lines) for centralized state
- Reduced HomePage to 60 lines of provider setup
- Eliminated all prop drilling - components use contexts directly
🏆 Why This is Good
- Maintainability: Each component has single purpose, easy
debugging/development
- Architecture: Clean separation of concerns, future features easier to
add
- Code Quality: 105% more lines but organized/purposeful vs tangled
spaghetti code
---------
Co-authored-by: Connor Yoh <[email protected]>
Co-authored-by: James Brunton <[email protected]>
Overview
Replaced scattered file inputs with a unified modal-based upload system.
Users now upload files via a global Files button with intelligent
tool-aware filtering.
Key Changes
🔄 New Upload Flow
- Before: Direct file inputs throughout the UI
- After: Single Files button → Modal → Tool filters files automatically
🎯 Smart File Filtering
- Modal shows only supported file types based on selected tool
- Visual indicators for unsupported files (grayed out + badges)
- Automatic duplicate detection
✨ Enhanced UX
- Files button shows active state when modal is open
- Consistent upload experience across all tools
- Professional modal workflow
Architecture
New Components
FilesModalProvider → FileUploadModal → Tool-aware filtering
Button System Redesign
type: 'navigation' | 'modal' | 'action'
// Only navigation buttons stay active
// Modal buttons show active when modal open
Files Changed
- ✅ QuickAccessBar.tsx - Added Files button
- ✅ FileUploadModal.tsx - New tool-aware modal
- ✅ HomePage.tsx - Integrated modal system
- ✅ ConvertE2E.spec.ts - Updated tests for modal workflow
Benefits
- Unified UX: One place to upload files
- Smart Filtering: Only see relevant file types
- Better Architecture: Clean separation of concerns
- Improved Testing: Reliable test automation
Migration: File uploads now go through Files button → modal instead of
direct inputs. All existing functionality preserved.
---------
Co-authored-by: Connor Yoh <[email protected]>
ConnorYohandGitHub
