9be20b4274a0324eebdabdc2bb5dd79aa90f783f
668 Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
 Mattsson
|
9be20b4274 | test(bookkeeping): use hotel account in VAT fixture (#1587) | ||
|
Mattsson
|
3829b6add3 |
fix(ai): complete plain-key self-hosting path (#1584)
* feat(ai): resolve the Claude backend from the environment Tier 1 of #1406: a self-hosted deployment can now run every AI feature on a plain ANTHROPIC_API_KEY, with no AWS account. Hosted behaviour is unchanged. lib/ai/provider.ts resolves the backend once, from the environment: AI_PROVIDER explicit override, bedrock|anthropic AWS static key pair Bedrock ANTHROPIC_API_KEY the direct Anthropic API nothing set Bedrock, so the AWS credential provider chain (instance profile, IRSA) still resolves Bedrock deliberately wins when both credential sets are present. EU residency in eu-north-1 is a BFL/GDPR posture rather than a default, so adding an Anthropic key for an experiment must not silently move production inference out of the region. AI_PROVIDER is the way to say you meant it. Model ids are written bare in code and prefixed to eu.anthropic.* only for Bedrock, which needs the cross-region inference profile for on-demand throughput. An operator override that already carries a prefix passes through untouched, so BEDROCK_MODEL_ID and friends keep working as written. Converted call sites: the agent composer, invoice-inbox extraction, the document-extraction model label, and both receipt-hunt clients. The last two are not named in the issue, which predates receipt-hunt landing in main. @anthropic-ai/sdk is declared at 0.95.0, the version @anthropic-ai/bedrock-sdk 0.29.1 already pulled in transitively, so the lockfile dedupes to one copy with no new download. scripts/smoke-bedrock.ts becomes scripts/smoke-ai.ts and grows two steps. Unit tests can only prove which provider and model id get resolved; they cannot prove the resulting request is one the backend accepts. The script now sends real traffic over all three shapes the app uses: a plain create, a streamed turn carrying adaptive thinking, an effort level, an hour-long cache breakpoint and a tool, and document extraction end to end when given a file. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Signed-off-by: Bjorn Bergenheim <29535152+bjornbergenheim@users.noreply.github.com> * docs(self-hosting): document the AI smoke test The script added alongside the provider split is what closes the #1406 acceptance criterion ("document extraction and the assistant both work"), so a self-hoster needs to know it exists. Covers both invocations and states that it exits non-zero, which is what makes it usable as a post-deploy check. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Signed-off-by: Bjorn Bergenheim <29535152+bjornbergenheim@users.noreply.github.com> * test(ai): split the smoke test's thinking probe from its tool probe The combined probe could not falsify what it claimed to. It asked a question that needs a tool call, so the tool was used and adaptive thinking correctly declined to reason about it: the zero thinking-block count that came back was uninformative rather than a signal. 2a keeps the tool and drops thinking. 2b asks a question with several dependent steps (reverse charge, then a partial deduction, then the affected boxes) so that a model honouring the parameter must reason, and reports the thinking text length as well as the block count, since display:"summarized" can yield blocks with empty text. The cached system prompt is also padded past the 1024-token minimum cacheable prefix. Below that the API caches nothing and reports no error, so the old probe's cache counters read zero whether or not caching worked. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Signed-off-by: Bjorn Bergenheim <29535152+bjornbergenheim@users.noreply.github.com> * fix(document-extraction): stop requiring AWS_REGION in the manifest The extension now needs one of two credential sets, AWS static keys or ANTHROPIC_API_KEY, and the manifest schema cannot express "one of". Since requiredEnvVars only drives a build-time warning and never gates anything, listing AWS_REGION told every self-hoster running the direct API to set a variable that has no effect for them. The description was also still promising Sonnet 4.6 via Bedrock specifically, which is no longer what the extension does. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Signed-off-by: Bjorn Bergenheim <29535152+bjornbergenheim@users.noreply.github.com> * fix(ai): read documentKind defensively in the smoke test The field arrived with the receipt-aware extraction work, so referencing it directly stops the script compiling against any checkout from before that landed. tsconfig includes **/*.ts and next.config does not disable type checking, so on such a checkout this failed the production build rather than just the script: caught while preparing a test branch for a self-hosted instance that had not synced yet. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Signed-off-by: Bjorn Bergenheim <29535152+bjornbergenheim@users.noreply.github.com> * fix(deps): restore the nested @swc/helpers entry in the lockfile Declaring @anthropic-ai/sdk with `npm install --package-lock-only` also pruned node_modules/next-intl/node_modules/@swc/helpers@0.5.23, an optional peer entry the local npm 11 considers redundant and the image's npm 10.9.8 does not. The result passed every local check and failed `npm ci` inside the Docker build, which is the only place the lockfile is actually enforced. The lockfile is now the previous one plus the single root dependency line, verified with `npm ci --dry-run`. @anthropic-ai/sdk needed nothing else: it was already in the tree as a transitive dependency of @anthropic-ai/bedrock-sdk. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Signed-off-by: Bjorn Bergenheim <29535152+bjornbergenheim@users.noreply.github.com> * Update DECISIONS.md Co-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com> * Update Docker documentation for AI provider credentials Clarify the role of credentials in AI provider selection and document extraction requirements. * Update SELF-HOSTING.md with smoke-ai script details Clarify usage of smoke-ai script for credential checks and document extraction. * Improve error handling and logging in smoke-ai script * fix(ai): complete plain-key self-hosting path Signed-off-by: Emil <emilmattsson14@gmail.com> --------- Signed-off-by: Bjorn Bergenheim <29535152+bjornbergenheim@users.noreply.github.com> Signed-off-by: Emil <emilmattsson14@gmail.com> Co-authored-by: Bjorn Bergenheim <29535152+bjornbergenheim@users.noreply.github.com> Co-authored-by: Claude Opus 5 <noreply@anthropic.com> Co-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com> |
||
|
Mattsson
|
0c3864cae5 |
fix(payroll): normalize KU10 organisation numbers (#1583)
Normalize KU10 employer identities to Skatteverket's 12-digit schema format, validate the structural XSD contract, correct FK201's XML element name, and add focused sourced tests. Resolves #1410. |
||
|
Jakob Wennberg
|
9c891ee72d |
fix(import): SEB CSV imports survive BOMs and bad format choices (#1565)
* fix(import): handle BOMs at the byte level in decodeFileContent Inspect leading bytes before decoding: EF BB BF strips the UTF-8 BOM and decodes the remainder (falling back to Windows-1252 for the remainder only, so the fallback can no longer produce a literal mojibake prefix), and FF FE / FE FF decode as UTF-16LE/BE. stripBOM additionally strips a literal mojibake BOM prefix for string paths pre-decoded elsewhere. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(import): make an explicit SEB choice at least as good as auto-detect Three changes for the SEB bank CSV report: - parseBankFile: when an explicit format parses 0 transactions, fall back to auto-detection; a different format that parses rows is returned with a prepended info issue naming both formats. A working explicit parse is never overridden, and explicit generic_csv (the manual mapping escape hatch) is exempt. - SEB profile: sniff the header delimiter (';' vs ',') and split with the quote-aware parseCSVLine; accept a bare Datum date column as a lowest priority tier in parse only, never in detect. Its user-reachable issue strings are now Swedish. - Import page: when a parse yields 0 transactions, show the parser's real issues instead of only the generic no-transactions hint. The v1 agent route now decodes through the shared decodeFileContent and stamps external ids, import_source, and the stored file format from the format the parse result actually carries, so fallback imports dedup identically to auto-detected ones. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * docs(api-v1): the bank import route also decodes UTF-16 CodeRabbit on #1565: decodeFileContent gained UTF-16LE/BE BOM support but the route overview and the registered endpoint description still listed only UTF-8 / Windows-1252. Skill regenerated (apiskill:generate). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
0d3ba5268d |
fix(transactions): close the booking duplicate guard's blind spots (#1573)
* fix(transactions): close booking duplicate guard blind spots G1-G3 The booking-time duplicate guard missed the most common bank-fee twin shapes: - G1: the sibling scan matched on the EXACT date only, so a duplicate import with a drifted date (CSV bokforingsdag vs PSD2 valutadag) was invisible. The scan now uses a +-3 day window with a deterministic ranking where exact-date candidates always outrank drifted ones (force=true re-detection stays bound to the reviewed candidate). - G2: booked-ness required transactions.journal_entry_id, so bulk-booked (transaction_voucher_links) and multi-allocated (invoice_payments / supplier_invoice_payments) siblings read as unbooked. The scan now batch-fetches the anchor rows and resolves the verifikat via getPrimaryJournalEntryId (is_transaction_booked semantics). - G3: the ledger scan excluded every voucher linked to any transaction, so a voucher booked from a date-drifted duplicate row escaped BOTH halves and the booking proceeded with no warning. A voucher whose linking transaction itself matches the target (same ore in the same currency, compatible cash account, date in the window) is now returned as the twin with transaction_id set. All candidate picks keep explicit total-order tiebreakers so a force re-detect returns the same candidate the user reviewed, and the SEK-or-null amount contract is unchanged. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(transactions): offer match/ignore for sibling duplicates and route all 409s into the dialog The duplicate dialog hid its match action for sibling-transaction candidates (canMatch required transaction_id === null), so the user who most needed steering saw only 'Bokfor anda'. manualLink explicitly allows N:1 links, so the match action is now offered for both candidate kinds. Sibling candidates get question-form body copy ('vill du matcha mot verifikatet i stallet?') and an additional 'Ignorera transaktionen' action via the existing POST /api/transactions/[id]/ignore, which is the correct resolution when the row itself is a duplicate import (matching would double-count the bank side, booking the ledger side). Two clients dead-ended the TRANSACTION_BOOK_POSSIBLE_DUPLICATE 409 in a destructive toast with no way forward: - the counterparty-template branch of handleQuickReviewConfirm now sets the shared duplicateWarning state exactly like runCategorize, with the force retry bound to the reviewed candidate's voucher - BankReconciliationView's quick-book now opens the same dialog, with match/ignore refreshing the reconciliation lists New sv/en strings: dialog_duplicate_body_sibling, dialog_duplicate_ignore, dialog_duplicate_ignore_failed. File-level parity tests pin the 409 routing and the dialog affordances. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(transactions): duplicate guard on the bulk-book samlingsverifikation path /api/transactions/bulk-book never called detectBookingDuplicate, so a batch containing an already-booked twin minted a second verifikat with no warning. The route now runs the shared per-tx guard before the RPC, with intra-batch exclusions (the other selected txs are distinct events the user picked, and the link-existing target voucher is the batch's own destination), returning 409 TRANSACTION_BOOK_POSSIBLE_DUPLICATE with the candidate and the flagged tx id. BulkBookDialog routes the 409 into DuplicateBookingDialog for review (view voucher / cancel / book anyway) instead of a dead-end toast; 'Bokfor anda' re-runs the batch with force=true. On force the route re-detects and records each dismissed candidate as BankTransactionDuplicateDismissed in behandlingshistorik (BFNAR 2013:2 kap 8), parity with the /categorize bypass. Detection failures stay fail-open. Note: the MCP RPC twin (gnubok_bulk_book_transactions) bypasses this route and remains unguarded; guarding inside the RPC needs a migration and is out of scope here. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(transactions): gate the duplicate-dialog ignore hint on the action being present The sibling body copy mentioned ignoring the row, but two render sites (the manual booking form and the bulk dialog) show sibling candidates without the ignore action. The guidance now lives in a separate dialog_duplicate_ignore_hint string rendered only when the Ignorera button itself renders, so copy never points at a button that is not there. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
1b829883ae |
feat(reconciliation): promote bulk matching and bridge it from the inbox (#1571)
* feat(reconciliation): accept confidence_threshold on the bank run route Mirror the v1 route: RunReconciliationSchema gains an optional confidence_threshold (0..1) that passes through to runReconciliation as the server-side floor on the apply path. The UI sends 0.85 with a strong-only apply so a pair the fresh re-run scores lower is skipped instead of committed; omitting it keeps the legacy behavior where every selected pair applies. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * feat(reconciliation): promote the bulk match flow and bridge it from the inbox The dry-run preview with pre-ticked strong matches existed but was never found: users matched whole migrations row by row. Three discoverability changes, no engine changes: - Bankavstamning: an attention line above the toolbar while unmatched transactions exist and no preview has run, with Forhandsgranska promoted to the filled variant. When every ticked preview pair is a strong match (>= 0.85) the apply button relabels to 'Matcha X starka traffar' and the apply sends confidence_threshold 0.85; mixed selections keep the plain label and omit the floor so manually ticked weaker pairs still apply. - Autorun bridge: ?autorun=1 on /reports/bank-reconciliation runs the preview once, only after appliedDates is set and not while datesDirty, so it can never cover a different window than the on-screen lists. - Transactions inbox: with >= 5 unbooked bank rows visible, an attention line links to the reconciliation with autorun (static text + count, no probe; the preview is the honest source of how many actually match). The review step stays: autorun lands on the preview table, one click from apply, and the server intersection guard is untouched. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
1eebb75269 |
feat(transactions): move an unbooked transaction to another cash account (#1570)
A bank transaction that ingested under the wrong cash account (or with no
account at all: legacy connections, own-account transfers the backfills
deliberately skipped) surfaces under the primary account's reconciliation
and can never be matched on the account it belongs to, because
cross-account matching is deliberately blocked. There was no first-party
way to fix the binding.
New PATCH /api/transactions/[id]/cash-account moves a movable staging row
(not booked, not invoice/supplier-invoice matched, not anchored via
transaction_voucher_links) to another of the company's cash accounts,
addressed by its BAS 19xx ledger account. Cross-currency moves are
hard-rejected (the row would vanish from every report's currency scope),
and the movable gate is re-asserted atomically in the UPDATE filter
against a concurrent book/auto-match, mirroring the title route. The tvl
check runs as a pre-check query since PostgREST cannot express NOT EXISTS
in an update filter; a tvl row appearing concurrently implies the booking
flow, which sets its own transaction state.
UI: 'Flytta till annat konto' in the transaction inbox row menu (opens a
radio-list dialog of the enabled cash accounts, current one preselected
and disabled) and direct 'Flytta till {name}' items in the bank
reconciliation unmatched-row menu that PATCH and refetch the view.
New structured error codes: TRANSACTION_MOVE_BOOKED,
TRANSACTION_MOVE_UNKNOWN_ACCOUNT, TRANSACTION_MOVE_CURRENCY_MISMATCH.
Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com>
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
|
||
|
Jakob Wennberg
|
e1f13f870a |
feat(import): warn about already-imported rows in the bank-file wizard (#1567)
* fix(transactions): paginate the ingest dedup maps past the 1000-row cap
buildExistingTransactionMaps issued un-paginated selects for the booked and
unbooked dedup maps, so PostgREST silently truncated each at 1000 rows: a
re-import over a wide date range in an active company deduped against a
partial map and inserted everything past the cap as duplicates. Both queries
now go through fetchAllRows with a stable .order('id') for range paging.
Also exports the function and its types for the upcoming read-only duplicate
preview, which must share the exact stored-row universe execute-side ingest
dedups against.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(import): add read-only duplicate preview endpoint for bank files
New POST /api/import/bank-file/check-duplicates (withRouteContext + Zod,
transactions capped at 20000) computes external_ids with the exact
generateExternalId(tx, format, index) derivation execute uses and runs
previewDuplicates: Layer-1 id collisions plus the Layer-2 text bridge with
counting semantics and the currency guard, against the same stored-row maps
ingest builds (buildExistingTransactionMaps). The result is advisory; execute
stays authoritative and mirrors/settlement-account guards are documented
preview/execute differences.
A dedicated endpoint because the generic_csv path re-parses client-side and
never re-hits /parse. Also removes the dead existing_transaction_count field
from the parse response (a raw date-range count consumed by nothing).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(import): surface duplicate rows in the bank-file import wizard
Overlapping bank imports used to dedup silently: the wizard promised
'Importera N transaktioner', ingest skipped the twins, and the user saw fewer
rows than parsed with zero explanation. The wizard now calls check-duplicates
after a successful parse AND inside handleColumnMappingConfirm (the
generic_csv path never re-hits parse), and:
- BankFilePreviewStep: warning card in the AlertTriangle pattern ('{count}
rader finns redan', skipped automatically) plus a 'Finns redan' badge on
flagged rows in the 50-row table
- BankFileConfirmStep: repeats the summary card (generic path skips preview)
and the CTA counts 'Importera {parsed - duplicates} transaktioner'
- BankFileResultStep: renders result.duplicates when > 0, closing the loop
ingest.ts documents as unrendered
Execute semantics unchanged: all rows are sent, ingest skips; the preview is
advisory and never promises an exact final number. New strings in both
messages/sv.json and messages/en.json next to the import_psd2 anchors.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
---------
Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com>
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
|
||
|
Jakob Wennberg
|
7ccaab7a08 |
fix(agent): contain the floating assistant panel on open and resize (#1575)
A persisted float rect saved at the viewport edge, or on a larger monitor, passes clampFloatRect (which only keeps 48px reachable so a live drag may deliberately hang off an edge) and renders the panel as a 48px sliver on every open. Add containFloatRect, which snaps the whole window inside the viewport, and an AgentSheet effect that validates the persisted rect on sheet mount and on viewport resize and persists the corrected position. The rect is read through a ref so drag commits do not re-trigger containment: parking the window half off-screen still works within a session. The live drag path and clampFloatRect are unchanged, and containFloatRect's output is always a fixpoint of clampFloatRect, so the render clamp stays a no-op. Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
78a581bca1 |
fix(import): guard against CP437-as-CP1252 mojibake entering via pre-decoded SIE text (#1569)
* refactor(arcim-migration): remove the dead gateway SIE export path fetchSIEExport and SIEExportFile have had zero callers since the direct provider clients replaced the Arcim Sync gateway (#181, #718). The path returned SIE as a pre-decoded string, and the gateway's decode of CP437 bytes as windows-1252 is what wrote the 2026-03-17 mojibake into posted entries. Deleting it makes the string-typed SIE fetch impossible to re-wire; a comment marks the grave. The consent lifecycle and entity accessors stay untouched. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(import): warn when SIE text carries CP437-as-CP1252 mojibake The 2026-03-17 migration wrote mojibake ("L"neutbetalning"-style C1 specials) into posted entries because the retired gateway handed the /import-sie handler an already-decoded string: byte-level encoding detection never saw it, and nothing downstream checked. The live bug is gone; this is the tripwire so the signature can never land silently again. - lib/import/sie-artifact-scan.ts: pure scanner over parsed SIE account names and voucher/line descriptions, reusing hasCp1252Artifact from charset-repair; flags at >= 2 hits so a lone legitimate curly quote or apostrophe cannot false-positive a whole file. - arcim-migration /import-sie: warn-never-block; the Swedish warning rides on result.warnings, which the workspace UI already renders, plus a server-side log.warn. - wizard parse route: same scan, surfaced through the existing parse-issue warnings card in the preview, pointing at the first affected line. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * test(bookkeeping): pin the reported gateway mojibake strings Adds the four strings reported from the affected company's journal as reverse_cp437 cases (all reverse losslessly) plus a false-positive guard: space-padded typography must never route into the CP437 reversal. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
c4adc8eb7d |
fix(salary): stop RLS from failing vab/parental absence registration (#1568)
Migration 20260517135000 rewrote the franvaro-specifikationsnummer trigger functions to insert audit rows into salary_absence_franvaro_audit, a table with RLS enabled and zero policies, while leaving the functions SECURITY INVOKER (its comment claimed implicit SECURITY DEFINER, which is false in Postgres). Every vab/parental insert from role authenticated (dashboard absence POST, web /pending approval, in-app Assistenten chat) then failed with 42501, surfaced as a generic 500, and left no diagnosable trace. - New migration 20260813120000: ALTER both trigger functions to SECURITY DEFINER with search_path pinned to public, pg_temp. No RLS policy is added on the audit table: trigger/service-only writes stay the design intent. - mapInsertError: 42501 now maps to the new bilingual DB_PERMISSION_DENIED code instead of INTERNAL_ERROR, and 23514 is split so only the 24h-cap trigger's 'Total tid' message becomes ABSENCE_HOURS_CONFLICT; other CHECK violations map to VALIDATION_ERROR. - commitRegisterAbsence/commitDeleteAbsence: log the underlying PG details and persist the sanitized structured code in result_data.error_code so the next failure is traceable from the op row. - Dashboard absence route: only ABSENCE_HOURS_CONFLICT passes details.message through to the client; every other code shows the registry Swedish message instead of raw Postgres text. - New pg-real regression test locks the authenticated-role parental/vab insert path, the shared per-month specnummer sequence, the audit rows, and idempotent upsert retries. Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Mattsson
|
5769e35869 |
fix(api-v1): propagate underlag when booking via v1 categorize routes (#1564)
The v1 categorize and batch-categorize routes create the journal entry via createTransactionJournalEntry directly and never ran the shared underlag propagation, so a booking made through the API-key surface left the transaction's pinned document unanchored and matched inbox items unstamped: the same "Underlag saknas" gap #1560 closed for the dashboard, /book and bulk-book paths, surviving on this one surface. Both routes now call propagateUnderlagForBookedTransaction after the CAS write succeeds (only when this request owns the booking; skipped on partial success and lost CAS races). Best-effort by contract, same as every other caller: a propagation failure is logged inside the helper and never fails the booking. Also adds the attach-after-bulk-book unit test salvaged from the closed duplicate PR #1559: a document pinned to a bulk-booked transaction (verifikat anchored via transaction_voucher_links) is anchored against the samlingsverifikat when attached after the booking. Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Mattsson
|
d9dddba682 |
fix(transactions): anchor the pinned document to the verifikat on booking (#1560)
A document pinned to a transaction (transactions.document_id) with no unconsumed inbox item was never anchored onto the verifikat when the transaction was booked: document_attachments.journal_entry_id stayed null and every underlag surface reported "Underlag saknas" for a booking that HAS its underlag (attach-before-book via the manual booking dialog, the 2026-08-13 user report). PR #1547 already routed /book, bulk-book and categorize through the shared propagateUnderlagForBookedTransaction helper, but that helper only walked matched inbox items. This adds a pinned-document leg to the helper, so all booking paths anchor the pin in one place: - the pin is read fresh inside the helper (not from the caller's pre-booking snapshot) so a concurrent attach is still anchored - same guard semantics as inbox docs, via the extracted anchorDocumentToJournalEntry: no-op when already anchored to this verifikat, never steal another verifikat's underlag, log-and-continue on failure (the booking is already posted; a re-run repairs the link) - the bulk-book RPC already anchors pins atomically, so the leg no-ops there Route tests cover the three plan cases: pinned doc anchored, matched inbox item stamped, and propagation failure never failing the booking. Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
8f1b1fb5cd |
feat(ui): company monogram in user menu + mobile web touch polish (#1531)
* feat(ui): replace generic building icon with company monogram in user menu The company row and switcher flyout in the sidebar user menu showed lucide Building2 for every company. Render the company's initial in a small rounded square instead (square = company, circle = person), so each company gets a mark of its own. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(mobile): touch behavior polish for the mobile web experience - kill -webkit-tap-highlight-color flash; touch-action: manipulation on interactive elements (no double-tap-to-zoom wait); user-select: none on buttons (long-press no longer enters text selection) - overscroll-behavior-y: contain on html/body: pull-to-refresh no longer hijacks list scrolling, inner scrollers stop chaining to the document (contain, not none, so iOS rubber-banding survives) - 16px font-size floor for form fields on coarse pointers: iOS Safari stops zooming into focused inputs; desktop keeps text-sm - min-h-screen -> min-h-dvh everywhere: correct height under collapsing mobile browser chrome, identical on desktop - active: variants mirror hover: on Button: Tailwind 4 gates hover: behind (hover: hover), so touch devices previously got zero pointer feedback - theme-color now tracks the app: default was a leftover blue #304D83; SSR emits white and ThemeColorSync mirrors the computed --background into the meta tag across dark mode and palette switches Hover-stuck-after-tap and viewport-fit/safe-area were already covered (Tailwind 4 hover gating; existing viewportFit: cover + safe-area utilities). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * docs: log monogram and overscroll decisions Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
38a890c8d1 |
fix(underlag): carry the phone photo that is too big to send, and say why when we cannot (#1550)
* fix(whatsapp-inbox): register the channel question event types Every follow-up question the WhatsApp intake asks has been failing its processing_history append in production: ChannelQuestionAsked, ChannelQuestionAnswered and ChannelQuestionExpired were never added to the processing_event_types catalog the event_type FK points at. appendQuestionHistory() catches and logs that failure by design, so the reply to the sender still goes out and nothing looked broken from the outside. What was lost is the durable record of the exchange, which is part of how the underlag was obtained (BFNAR 2013:2 kap 8). Catalog rows only: aggregate_type 'System' already passes the CHECK. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(underlag): say why an upload failed, and get out of an expired session A user reported that none of the three ways to add a receipt from a phone worked, all of them answering "Uppladdning misslyckades. Nagot gick fel, forsok igen" immediately. Production told us nothing: every upload request that reached the route in the same 24 hours returned 200. Both halves of that are the same bug. The workspace read failures as `throw new Error(json.error)`, which loses a body that is not JSON (the res.json() call throws first) and stringifies the structured envelope to "[object Object]", so anything the route did not answer with a plain string arrived as the generic fallback. The middleware 401 for an expired cookie session is exactly that envelope shape, and a phone tab left open is exactly where the session expires unnoticed: the controller's timers are throttled in the background, so the request the user just made is what finds out. Now the response is resolved where it fails, through the house helper that already knows the status map, and an expired session is announced on the session-timeout BroadcastChannel so the controller signs out and routes to /login the same way it does for an expired heartbeat. Failed uploads also post metadata (status, size, mime type, resolved reason) to /api/log, the one API path exempt from the timeout gate, so a request answered before the route runs stops being invisible. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(underlag): carry the phone photo that is too big to send The reported failure was not the account and not the session: hosted rejects any request body over 4.5 MB itself, before the function runs. Measured against production, 4.4 MB reaches the route and 4.6 MB comes back as a plain-text FUNCTION_PAYLOAD_TOO_LARGE. Nothing invokes the function, so nothing lands in the logs, which is why one user's failing uploads were invisible while every upload that arrived returned 200. An iPhone photo in "Most Compatible" mode is 4-12 MB, so whether it worked depended on whose phone took the picture. Meanwhile the route advertises a 10 MB limit it can never be handed. Photos are now re-encoded in the browser when they exceed what the platform will carry: 2400px on the long edge at JPEG q0.85, stepping the quality down only if that is not enough. That keeps the small print on a receipt legible, which is what BFL 7 kap asks of an archived underlag ("varaktigt läsbart skick", a faithful reproduction), and a refusal is not. What cannot be shrunk (a PDF, or HEIC where the browser will not decode it) is refused before the upload starts, naming its actual size and the limit rather than failing in transit. 413 joins the HTTP status map so a rejection we cannot pre-empt still says what happened: the platform's body is plain text, so the status is the only thing there is to translate. Self-hosted Docker has no proxy in front of the app, so none of this applies there and the route's own MAX_FILE_SIZE keeps governing. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Mattsson
|
98612fb0ac |
fix(providers): stop requesting unapproved Fortnox scopes that broke every connect (#1549)
PR #1541 added archive and connectfile to the Fortnox DEFAULT_SCOPES for the voucher attachment import, but the registered Fortnox app does not have those scopes approved in the Fortnox Developer Portal. Fortnox rejects the authorize request with invalid_scope before login, which broke every Fortnox connect in production within minutes of the deploy (verified in Vercel runtime logs). Remove the two scopes from the connect request; the attachment import logic from #1541 stays fully intact and already degrades gracefully: a 403 becomes PROVIDER_DOCUMENT_SCOPES_REQUIRED with a reconnect follow-up card. Re-add the scopes once the portal registration has them approved. Also add charset=utf-8 to the OAuth callback HTML responses: without it browsers render the Swedish error text as Latin-1 mojibake. Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Mattsson
|
b9bf60234d |
feat(transactions): filter /transactions by rakenskapsar and kvartal (#1545)
* feat(transactions): filter /transactions by rakenskapsar and kvartal User request: booking a specific period (including brutet rakenskapsar, e.g. July-June) meant scrolling past every other year's transactions. - New FyPicker chip in the toolbar scopes both the inbox and history views to a fiscal year; quarter chips (Q1-Q4, fiscal-year aligned) appear once a year is selected. Clicking the active quarter widens back to the year. - Bounds are pushed into the Supabase queries (window, pending backlog, badge count, load-more) so pagination and the Att bokfora count stay consistent with the visible list; skattekonto rows are bounded client-side. - Scope persists under a page-local localStorage key, deliberately separate from the shared report scope so a year picked on a report page never silently hides pending inbox rows. - lib/transactions/period-filter.ts derives quarter bounds from fiscal period dates (handles brutet, shortened and extended years); unit tested. - FyPicker gains an optional storageKeyPrefix prop; default unchanged. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(transactions): never hide pending rows behind the period filter Swedish accounting review on PR #1545: scoping the pending-backlog fetch and badge count to the period made unbooked rows outside the selected year vanish from the inbox worklist (BFL 5 kap: pending affarshandelser must stay visible until booked). - Pending-backlog fetch and the DB pending count are unscoped again; only the history window pages server-side within the period. - The inbox applies the period client-side over the complete backlog; the tab badge counts pending rows inside the scope. - When pending rows (bank or skattekonto) fall outside the scope, the footer says how many and offers Visa alla, which clears the filter and its persisted value. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(transactions): guard list fetches against stale cross-scope responses CodeRabbit on PR #1545: - fetchTransactions/loadMoreTransactions now carry a fetch generation; a response applies only if no newer fetch (scope change, realtime refresh, load-more) started meanwhile, so a slow pre-filter request can no longer overwrite the active period scope's window, paging offsets, or loading skeleton. - FyPicker restore effect includes storageKeyPrefix in its deps. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(transactions): label quarter chips as fiscal-year quarters Swedish accounting review note on PR #1545: Q1-Q4 follow the company's rakenskapsar, which on a brutet rakenskapsar differs from the calendar quarters that momsdeklaration periods use. Say so in the group's aria-label and hover title so the chips are not mistaken for VAT periods. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Mattsson
|
27383fd02b |
fix(inbox): surface full BAS catalog in BookDirectlyDialog account picker (#1543)
* fix(inbox): surface full BAS catalog in BookDirectlyDialog account picker The account combobox in the inbox book-directly flow was fed only the company's active chart, so typing a prefix like 65 showed just the two activated 65xx accounts and a search for 6540 found nothing, which reads as the account not existing. Pass the cached BAS catalogue (same pattern as JournalEntryForm) so every standard account is searchable; picking a not-yet-activated account flows through the existing ActivateAccountsDialog rail at booking. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(bookkeeping): log BAS catalog fetch failures Compliance swarm finding (SOC 2 CC7.2): loadBasCatalog swallowed fetch errors silently, leaving catalog-load failures unobservable. Log inside the client's catch, which is the only place the error actually surfaces: callers' own .catch handlers are unreachable since the shared promise already resolves to an empty list on failure. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * test(bookkeeping): unit-cover loadBasCatalog fetch, fallback, and retry CodeRabbit finding on PR #1543: the catalog client had no focused coverage. Tests assert the success path with promise caching, empty-list fallback with logging on non-OK responses, missing data field handling, and cache clearing after a failure so the next call refetches. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Mattsson
|
8d56219c31 |
fix(inbox): booked items no longer strand in Att gora as matched-forever (#1547)
* fix(inbox): booked items no longer strand in Att gora as matched-forever A matched inbox item only left the active inbox when created_journal_entry_id was stamped, and only categorizeTransactionCore stamped it. Booking the matched transaction through any other path (the /book dialog route, bulk-book, link-to-existing-voucher) or matching a receipt to an already-booked transaction (receipt hunt approvals, attach-document, match-transaction) left the item "linked" forever, pointing at a transaction that had already left the transactions work list. Todays hunt fix (#1524) turned this July-old gap into a visible flood of stuck items. Two-part fix, because stamps alone cannot cover the reported case: created_journal_entry_id is UNIQUE (20260515090000), so on a bulk-book samlingsverifikat only one of N matched items can ever carry it. Write side: lib/transactions/inbox-underlag.ts is the shared implementation all paths now call. It links matched items' documents to the anchoring verifikat (BFL 5 kap 6-7 kap: underlag on the verifikation) and stamps created_journal_entry_id best-effort (CAS on null, unique_violation tolerated). Wired into categorize-core (replacing its inline block), /book, bulk-book, linkTransactionToJournalEntry, both attach paths (REST + pending-operation), and the inbox match-transaction handler. The attach paths and the doc-conflict guard also resolve bulk-booked transactions through transaction_voucher_links, which they previously treated as unbooked. Read side: GET /items (and /items/:id) enrich matched-but-unstamped items with matched_transaction_journal_entry_id, and the workspace derives "booked" from it. This is what clears the stuck rows already in prod without a status backfill, and what covers the N-1 samlingsverifikat items the UNIQUE constraint refuses to stamp. Bulk-book selection filters exclude such items so "Bokfor valda" no longer offers 409 fodder. scripts/backfill-inbox-booked-underlag.ts (dry-run by default) repairs the historical document->verifikat links the old paths never made. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(inbox): stamp only settled underlag, and give the backfill behandlingshistorik Both from the Swedish accounting compliance review. The consumed-stamp is now conditional on the underlag actually referencing a verifikat: stamping over a failed document link hid the item from the .is('created_journal_entry_id', null) query forever, leaving a posted verifikation without its underlag reference (BFL 5 kap 6-7 kap) and nothing left to surface or repair it. A failed link now leaves the item unstamped so re-runs and the backfill can finish the job; a document preserved on another verifikat still counts as settled. The backfill script now appends an InboxUnderlagBackfilled event per repaired transaction to processing_history (BFNAR 2013:2 kap 8): a mass repair touching underlag-to-verifikat linkage leaves a changelog trail distinguishing it from the original booking action. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * refactor(inbox): backfill writes behandlingshistorik through the shared appender From the Swedish accounting compliance review round 2: a hand-rolled processing_history insert in the backfill script could drift from the shared row shape and skip the PII validation. appendProcessingHistory now delegates to appendProcessingHistoryWithClient, which takes a caller-supplied service-role client, so standalone scripts write behandlingshistorik through the exact same code path as the app (BFNAR 2013:2 kap 8: one reconcilable change log across writers). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(inbox): leave the item unstamped when its document belongs to another verifikat Swedish accounting review round 3: refusing to steal the document was right, but stamping the item consumed anyway hid the fact that the transaction's own verifikat ended up with no underlag reference from it (BFL 5 kap 6-7 kap). The anchored-elsewhere case now leaves created_journal_entry_id null so the mismatch keeps surfacing for reconciliation, same posture as a failed link. Also documents in the backfill script header why its writes cannot land in locked periods: linkToJournalEntry's UPDATE is guarded by the enforce_period_lock DB trigger, which fires for service-role writes too. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Mattsson
|
a97b0023d4 |
feat: import Fortnox voucher attachments (#1541)
* feat: import Fortnox voucher attachments * fix: show Fortnox document import follow-up * fix: harden optional Fortnox document import * test: pin optional Fortnox import flow * fix: use browser timer handle type * fix: avoid serializing OAuth resume state |
||
|
Mattsson
|
45d7f1be4e |
feat(mileage): surface Körjournal in the nav behind a settings toggle (#1540)
* feat(mileage): surface Körjournal in the nav behind a settings toggle The /mileage page shipped hidden: the route works but no nav row points at it. Add company_settings.mileage_enabled (mirroring dimensions_enabled) with a switch in Fönster -> Bokföring, and show the Arbeta nav row when the toggle is on OR the company already has mileage_trips rows, the same hybrid gate as webshop orders, so trips created via API/MCP can never become invisible underlag. UI visibility only, never load-bearing for correctness. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(migrations): move mileage_enabled migration after already-applied 20260812153208 origin/main merged in 20260812153208 which prod has already applied; a new file sorting before it risks an out-of-order db push abort. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
67febd5097 |
feat(ui): prev/next record navigation on detail pages (#1530)
* feat(ui): prev/next record navigation on detail pages Customer feedback: stepping between invoices in a reskontra (56 -> 57 -> 58) required going back to the list for every record. List pages now write their full ordered id array to sessionStorage when a row is opened (Accounted:list-context:<scope>:<companyId>), and the detail pages for kundfakturor, leverantorsfakturor, and verifikat show a compact prev/next pager (chevrons + 'n av m') next to the back control. ArrowLeft/ArrowRight step too, except while typing in a text field or while a dialog is open. Navigation uses router.replace so 'tillbaka' returns to the list in one step. Deep links and new tabs have no context: the pager hides and pages behave as before. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(pager): overlay-aware arrow guard, notes-draft safety, context on Visa detaljer Review fixes on the detail-record pager: - The keyboard guard matched any mounted [role=dialog], so the agent sheet (which stays mounted display:none once opened) killed arrow paging for the rest of the tab session, while open dropdown menus did not block at all. The guard now mirrors the AgentSheet Esc selector (data-state="open" variants incl. alertdialog and radix menu/select/listbox content) and also yields while focus sits inside a dialog/menu/listbox container. Extracted as pure functions in lib/hooks/detail-pager-guards.ts so the rules are testable in the node test environment. - Arrow keys could unmount the verifikat page and destroy an unsaved notes draft once the textarea lost focus. useDetailPager and DetailPager now take a keyboard flag, and the verifikat page disables keyboard paging while editingNotes is active; the chevron buttons stay live. - The expanded-row Visa detaljer link in JournalEntryList navigated without writing the list context, producing stale pager snapshots; it now calls rememberListContext like the voucher link. - ListContext.listPath was written and strictly validated but never consumed: removed from the interface, all writers, and the read validation. Reads stay tolerant of extra properties so contexts stored by older builds still parse. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * refactor(pager): quiet wayfinding strip instead of buttons in the title cluster The pager sat between the back arrow and the H1, which read as a toolbar of three boxed buttons and made the title jump horizontally per record. All three detail pages now share the verifikat page's pattern: a muted back text-link on the left and the pager right-aligned on the same quiet row. The pager itself drops to 16px glyphs, muted ink, and hides when the list context holds a single record. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
bffa57a565 |
feat(invoices): bulk Bokfor, per-view filter counts, review-queue draft CTA (#1533)
* feat(invoices): bulk Bokfor, per-view filter counts, review-queue draft CTA Customer feedback: MCP-created invoices land in Granskning and then sit as unnumbered drafts that each need individual issuance, and the list filter gives no signal about where the work is. - New POST /api/invoices/bulk-book: drafts get an F-number + mark-sent semantics (no email) and book inline when the company books at issue; sent/overdue unbooked invoices get the deferred /book semantics. Sequential loop keeps voucher numbers ordered; per-item Swedish errors. - Extracted the shared cores into lib/invoices/issue-and-book-invoice.ts and lib/invoices/book-invoice-deferred.ts, now used by the per-id mark-sent and book routes AND the bulk loop, so they cannot drift. Per-id route behavior unchanged (existing route tests untouched, green). - Invoice list: multi-select with hover-reveal checkboxes (supplier-invoices shape), bulkbar with mode-aware action label, ConfirmationDialog with a draft/sent breakdown, one aggregate toast. Kontantmetoden hides selection entirely. - ContextPicker: count annotations on every status view via the one shared predicate (counts always match rows), active view written back to the URL (?status=) for shareable views. No seg/chip row: founder-locked pattern. - Granskning: after a bulk approve that committed create_invoice ops, the summary toast links to /invoices?status=draft to finish with bulk Bokfor. Verified: npm run lint clean, npm test 13845 passed, npm run check:guards passed. New tests: bulk-book route (11), issueAndBookInvoice (7), bookInvoiceDeferred (7). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(invoices): bulk-book review findings, deferred drafts, dupes, URL params - Deferred-booking companies (accrual + defer_invoice_booking): a draft in bulk-book no longer gets silently ISSUED (F-number consumed, marked sent, invoice.sent emitted) while reporting status 'booked' with a null journal_entry_id. The draft branch now requires booksInvoicesOnIssue(); otherwise the item fails per-row with the new INVOICE_BOOK_DEFERRED_DRAFT code (Swedish + English) before the invoice is touched. - Duplicate ids in one request no longer double-book: the second iteration read the stale pre-loop snapshot, passed the already-booked check, and minted a voucher the CAS claim then cancelled (cancelled verifikat + gap explanation per duplicate). Ids are deduped before the loop. - Bulkbar: the select-all link is hidden when the current view has no selectable rows; "Markera alla (0)" only wiped the existing selection. - Invoice dialog open/close handlers (new invoice, self-billed, ROT/RUT payout) rewrite only their own query keys instead of hardcoding '/invoices', so the ?status= view write-back survives them. - /pending: the "Bokfor utkasten" toast CTA is suppressed for kontantmetod and deferred-booking companies where the invoice list offers no draft bulk Bokfor (dead end); the neutral hint sentence stays. Tests: deferred-draft rejection (asserts issueAndBookInvoice never called, sent invoice in the same batch still books) and duplicate-id dedupe (exactly one booking call); both fail without the route fix. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
7cf0e34434 |
feat(supplier-invoices): sarskild loneskatt (SLP) pair on pension premium lines (#1534)
* feat(supplier-invoices): sarskild loneskatt (SLP) pair on pension premium lines Booking a tjanstepension invoice (e.g. Avanza) needs the buyer's own SLP beyond the payable: debit 7533 / credit 2514 at 24.26% of the premium (SLF 1991:687). The item-based debit-only form could not express the self-balancing pair, so users had to hand-edit the verifikat. - new leaf module lib/bookkeeping/slp-lines.ts: SLP_RATE (single source, re-exported by the bokslut calculator), isSlpPensionAccount (741x), generateSlpLines (7533 D / 2514 K, nets to zero) - migration adds supplier_invoice_items.apply_slp boolean default false - registration, cash, and privately-paid generators inject the pair for flagged 741x items, mirroring the reverse-charge injection; the balance guarantees keep 2440/1930/2893 at exactly the invoice total; the credit note generator reverses the pair (7533 K / 2514 D) - privately-paid balance guarantee now subtracts existing credits so the SLP 2514 leg never inflates the owner account - schema field apply_slp + guards in all create paths (main route, inbox convert, v1 REST, pending-operations executor): 400 SI_CREATE_SLP_INVALID_ACCOUNT on non-741x accounts, 400 SI_CREATE_SLP_ACCRUAL combined with periodisering - form: advisory hint on unflagged 741x rows with one-click opt-in and a quiet confirmation line when applied; totals box untouched (the invoice total stays the payable); AB review preview injects the same pair via the same generator for parity - year-end double-count guard: calculateSarskildLoneskatt subtracts SLP already posted to 7533 during the year (floored at zero) so bokslut never provisions flagged premiums twice Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * chore(api-skill): regenerate suppliers reference for apply_slp The apiskill:check CI gate requires the generated accounted-api skill to stay in sync with the endpoint registry after the apply_slp addition. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(slp): carry apply_slp through v1 routes, MCP staging, preview and credit reversal Review findings on the SLP PR: - v1 credit route: SI_FULL_COLUMNS now projects items.apply_slp, so createSupplierCreditNoteEntry sees the flag and reverses the 7533/2514 pair booked at registration (it previously stood forever and the year-end netting under-provisioned). The flag is also copied onto the created credit-note items for parity with the web credit route. - v1 mark-paid: the items sub-select now includes apply_slp, so a kontantmetoden payment via v1 books the cash entry WITH the SLP pair, matching the web mark-paid. - v1 GET ?expand=items: SI_ITEM_COLUMNS includes apply_slp so the flag is readable back through the public API. - credit-note SLP base is abs of the SIGNED sum of flagged line_totals, not per-item abs: a mixed-sign flagged original (+10000/-2000) booked SLP on 8000 at registration and now reverses exactly that, not 12000. The expense-bucket per-item abs convention is untouched. - kontantmetod bank-match preview appends the same generateSlpLines pair the POST books, so the approved lines equal the committed lines. - MCP gnubok_create_supplier_invoice_from_inbox: line_overrides accepts apply_slp (optional boolean), plumbs it into the staged operation's items, and rejects non-741x resolved accounts at staging time with the bilingual SI_CREATE_SLP_INVALID_ACCOUNT texts. - DECISIONS.md: five entries for today's decisions. Every behavioral fix has a test verified to fail without it. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Mattsson
|
11995b1b0c |
feat(auth): make automatic logout an opt-in per-user setting (#1536)
* feat(auth): make automatic logout an opt-in per-user setting Session timeouts (30 min idle / 12 h absolute on hosted) now apply only to users who enable "Automatic logout" in Settings > Security. Default is off: sessions live for the full Supabase refresh-token lifetime, the behavior from before the 2026-07 session hardening. - user_preferences.auto_logout (migration, default false), toggled via the extended /api/user/preferences route - The opt-in is snapshotted into the signed timeout cookie at mint, so enforcement stays DB-read-free per request; the preferences route clears the cookie on change so a toggle takes effect immediately - Pre-toggle cookies are authentic-but-stale: re-minted preserving their timers, never routed down the tamper path, so the rollout does not log anyone out - NEXT_PUBLIC_SESSION_TIMEOUT_FORCE_ALL=true enforces timeouts for every user regardless of preference (emergency lever, also plumbed through the Docker image); self-hosted stays disabled by default Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(auth): resolve PR #1536 review findings - Replace the spread upsert in /api/user/preferences with one literal payload per field: the phantom-column schema guard cannot resolve spread payloads (Unit tests 3/4 ceiling failure) - Map the preferences 500 through getErrorMessage so the user-facing text is Swedish (CodeRabbit) - fetchAutoLogoutPreference now returns null on a FAILED read instead of a fail-open false: callers skip minting so an unknown preference is never persisted into the year-long signed cookie, and the next request retries; failures log at error level, distinct from the normal opt-out path (compliance swarm GDPR Art.32(1)(b) / ISO A.8.5) Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(auth): write multi-field preference updates as one atomic upsert A request carrying both hide_assistant_fab and auto_logout previously issued two sequential writes, so a failure of the second returned 500 after half the request had persisted (CodeRabbit, PR #1536). One literal upsert per accepted field combination keeps the write atomic and stays resolvable for the phantom-column schema guard. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
845add4573 |
feat(documents): dedupe intake channels on content, not just provenance (#1528)
* feat(documents): dedupe intake channels on content, not just provenance Every ingestion path already computes and stores sha256_hash, but only WhatsApp ever read it back: the manual upload, Resend inbound, and mail hunt deduped on provenance keys alone (or not at all), so the same receipt forwarded to two inboxes, re-hunted by a sweep, or uploaded twice became a second archived document and a second inbox item. With the hunt live and three channels feeding one inbox, that is an unbounded duplicate generator (flows plan, prerequisite PR 1). uploadDocument gains an opt-in dedupeByContent flag: before storing, it looks for a current-version document in the same company with the same SHA-256 and returns it (marked deduplicated) instead of archiving a copy. Opt-in because archival callers must store what they produced even when bytes repeat; the SELECT-then-insert race is accepted exactly as in the WhatsApp intake precedent. uploadAndExtract turns the flag on for every inbox channel. On a hit it adopts the oldest inbox item for that document, so callers always receive a real inbox_item_id, and only files a new item (against the EXISTING document) when the content entered the archive outside the inbox. The mail hunt skips outright: its provenance key catches the same message re-hunted, the content check catches the same receipt arriving through another inbox. WhatsApp keeps its own pre-check, which also drives the duplicate reply to the sender. No migration: the hash column and its index have existed since the original archive schema. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(documents): review round: fail closed, adopt-or-file in the hunt, audit trail CodeRabbit: both dedupe lookups failed OPEN, so a transient DB error would silently archive the duplicate the feature exists to prevent; both now throw before anything is stored, and a regression test locks it. The ingest test also asserts the dedupeByContent flag in the production call, so removing the flag fails the suite. Swedish compliance review, both findings real: (1) the mail hunt's unconditional skip could swallow a receipt whose content matches a document that never passed the inbox (a manually attached copy), leaving an affärshändelse without underlag routing (BFL 5 kap): the hunt now mirrors the funnel's adopt-or-file semantics, skipping only when an inbox item already carries the document and otherwise filing an item against the EXISTING document. (2) The skip decision now lands in behandlingshistorik as DocumentDuplicateSkipped (BFNAR 2013:2 kap 8), not just the app log. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(receipt-hunt): keep the audit payload pseudonymous; lock the skip trail in tests Review round 2. The DocumentDuplicateSkipped payload carried the mailbox address, violating the processing-history contract (pseudonymous IDs only, never emails); the digit-shaped PII validator would not have caught it, which is exactly why the contract must hold at the call site. Which mailbox first delivered the receipt is already on the existing item's channel_context. Tests now assert the audit event lands with the right identifiers and no address, and that a history outage still skips rather than filing a duplicate. Not changed: a duplicate-lookup error still soft-fails the attachment (warn + continue). Aborting the candidate would contradict this function's documented contract (one bad message never costs the night's hunt); fail-closed holds either way, and the next sweep retries since no item was filed. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Mattsson
|
c35b2547fb |
feat(webshop-orders): Orders page with per-store, per-payment-method booking (#1525)
* feat(webshop-orders): schema, types and error codes for the orders surface webshop_orders (order/refund rows, financial-freeze trigger, member select/update RLS, no DELETE) + webshop_store_settings (per-store payment method -> account map), source_type 'webshop_order', multi-store index drop, customer_country, and a one-time woo cursor reset so the switch-over backfills and cross-marks existing feed rows. Tables classified in the full-archive export; pg-real coverage for RLS, freeze and CHECK. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * feat(webshop-orders): core service (ingest, booking lines) upsertWebshopOrders: two-phase order/refund upsert with FX enrichment, legacy-feed cross-marking, frozen-row protection and field-wise jsonb comparisons (Postgres does not preserve object key order). Booking-line builder: per-rate VAT split with SIGNED buckets (discounts book as revenue reductions), refund mirroring, 3740 residual, per-store account prefill, and advisory export/EU + OSS warnings. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * feat(webshop-orders): API routes for list, booking, invoicing and mapping Booking is draft -> atomic claim -> commit (conditional link-back closes the concurrent double-book race; a lost claim cancels the voucher-free draft). Legacy-feed guard honors transactions.is_ignored on both the book and create-invoice paths. Invoice conversion reuses buildInvoiceWriteData for an unnumbered draft with dominant-rate fallback and drift-safe unit prices. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * feat(webshop-orders): Orders page, booking/invoice dialogs and gated nav /orders lists per-store orders with status tabs (server-side filters), exception chips and one action per row. Booking dialog prefills from the per-store payment-method mapping with an opt-in remember; invoice dialog converts to a draft kundfaktura. The Order nav item renders only for companies with an active WooCommerce connection or existing order rows (Shopify deliberately excluded until its sync writes webshop_orders). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * feat(woocommerce): switch the order sync to webshop_orders, multi-store The sync maps rich wc/v3 payloads (billing, line/shipping/fee taxes, refund allocations with parent-prorated VAT fallback) and upserts order rows instead of transactions-inbox rows; already-imported feed rows stay bookable and get cross-marked. Multi-store: several active connections per company, per-store panel cards with the account-mapping editor. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * docs(webshop-orders): decision log entries and ratchet baseline Baseline moves DOWN only: naive-ore-round 638 -> 637 via roundOre adoption; hand-rolled invariants stay at 115 (ACCOUNT_NUMBER_RE imported, not inlined). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(webshop-orders): resolve PR #1525 review findings and CI failures Review batch (Superagent, CodeRabbit, Swedish compliance review): - Mutual-exclusion claims: booking guards invoice_id, invoice link-back guards journal_entry_id AND treats zero matched rows as the conflict it is (409 + rollback), closing both TOCTOU races. - Freeze v2 migration (20260812124858): the link columns themselves are protected: invoice links immutable, journal links clearable only while the entry is still a draft (the booking rollback path). - Scraped orgnr no longer auto-written to customers.org_number; rate fallback applies only on single-VAT-bucket orders; refunds get their own WEBSHOP_ORDER_REFUND_NOT_CONVERTIBLE code; VAT advisories outrank the invoice-mode hint in the booking dialog. - Ingest compares every synced field (billing corrections no longer drop as unchanged); sync guards absent refunds arrays; /sync aggregates per-store results; panel disables all cards while a request runs; orders page separates load failure from empty; account field explains itself. CI: regenerated skills/accounted-api; pg tests restructured for transaction-abort/rollback semantics + freeze-link coverage; unresolvable- expression ceiling 375 -> 378 with documented reason (partial-update payloads in ingest, shapes covered by unit tests). Declined: CodeRabbit docstring-coverage advisory (house style: comments only where the code cannot say it). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
555a2a20ae |
feat(inbox): Underlag rebuilt to answer what is missing, where to get it, and how it would be booked (#1524)
* fix(mail): stop Gmail refusing the search, and stop calling that "hittade inget" Pressing Leta produced mails=25, documents=0 on a real two-mailbox run. Nothing was found because nothing was searched: every request came back 429 "Too many concurrent requests for user". Two bugs, and the second is the one that matters. The search fanned out with Promise.all over every message id at once, one Gmail request per message, per connection. Gmail enforces a per-user concurrency ceiling as well as a daily quota, and this sailed past it long before any volume worth worrying about. It now runs through a pool of five per connection, which is comfortably under and still finishes a page of results in a couple of round trips. The catch turned each refusal into an empty array, with a comment saying one mailbox's failure must not become the company's. Right instinct, wrong consequence: an empty array is also what an empty mailbox returns, and the manual hunt loop stops on fetched === 0 because that is its signal for "the mailboxes hold nothing more for what is open". So a rate-limited search told the user their receipts do not exist, and stopped looking. searchFailureCount() now separates "could not look" from "nothing there". The run route reports it, and the loop treats a pass with failures as failed rather than finished, so pressing again is the obvious next move instead of a pointless one. This is the failure this feature exists to catch, happening inside the feature: silence that reads as an answer. Restoring the unbounded fan-out fails one test; removing the failure counter fails three. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat(inbox): segment filter as a dropdown, not three rows of pills Five filters wrapped to three lines in a 280px column. The counts are what people actually read, so they stay on the trigger and inside the menu rather than being traded away for the space. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat(inbox): one chip for where underlag come from Three routes in, and the page never said so: the forwarding address sat inline in the header, the mailboxes lived only in Instaellningar, and WhatsApp was invisible here entirely. They are behind one chip now. Which mailbox and when it was last read is what people look up when something seems wrong, not what they read every visit, so it opens rather than occupying the header. A mailbox that has stopped working is the exception, so it surfaces on the chip itself rather than waiting to be found one click in. That silence is the failure this feature exists to catch. Configuration stays in Instaellningar; this only reports. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat(inbox): the kontering first, the evidence folded Reading order was backwards. Nine extracted values came first and the one thing to approve came last, so every matched item meant scrolling past the evidence to reach the decision. The proposed kontering is now the first thing in the rail. The fields fold behind a summary that carries how many of the twelve the extraction actually filled, so a thin extraction is visible without opening it. They stay open when nothing is matched: with no proposal above them the fields are all there is, and folding the only content on the pane would be a hiding place rather than a hierarchy. The counted list is the same one hasAnyExtractedField checks, so the summary cannot claim a field the 'is anything here' test does not count. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat(inbox): one dialog that changes the whole verifikat The rail offered three overlapping ways to alter a booking and none said what it covered: an Aendra beside the date, an Aendra kontering at the bottom, and a menu entry that did what the primary button already did. This is the one control, and its scope is the whole verifikat: date, series, description, every line. It opens pre-filled with the proposal when there is one and empty when there is not, so there is no separate book-manually path to pick between. A dialog rather than an inline editor: a 340px rail cannot hold an account picker, two money columns and a delete control per row without clipping something, and the document has to stay readable while the numbers change. Checking a momssats against the paper is the reason to open it at all. TransactionBookingDialog already has this shape for the same reason. The form is JournalEntryForm unchanged. It carries the series picker, per line descriptions, dimensions, currency, the balance check and the confirm step, and it posts through the sanctioned route. Extending BookDirectlyDialog was the alternative and is not viable: three effects seed its lines and fight anything injected, and its FormLine has no room for line text, dimensions or tax codes. Nothing posts without the form's own review step, so a proposal stays a draft the user commits. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(inbox): show every unreceipted purchase, and fold the mailboxes Three things. The 100 kr floor was hiding 52 of one real company's 119 unreceipted purchases: the page reported 67 and looked tidier for it. The floor was copied from the receipt hunt, where it earns its place because every candidate costs a mail search and a model read. This list costs a query, and bokforingslagen wants an underlag for the 45 kr purchase exactly as much as for the 4 500 kr one. The hunt keeps its floor; the page has none. Mailboxes fold. When it was last searched is what you look up when a mailbox seems to have gone quiet, not what you read on the way past. The address stays on the row, and a connection that needs reconnecting still says so without opening. Dropped the line telling people to go to Instaellningar. The panel reports where underlag come from; sending them elsewhere was the seam this work set out to close. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat(inbox): split the portal purchases out, and say what a run found Four things from looking at the real page beside the artifact. Hamta fran portal is its own list again. Twelve of one company's 119 unreceipted purchases have a supplier whose invoices sit behind a login, and that is a different job from the other 107: go there and fetch it, versus ask somebody. Collapsing them into one list with a badge buried the twelve you can settle now among the hundred you cannot. A run now says what it did. Pressing Leta and being told nothing is why the feature read as broken even on the runs where it worked: three underlag landed and the page looked identical afterwards. WhatsApp folds like the mailboxes and shows its number, which is the fact worth having. Describing the channel to someone who already connected it was not. The forwarding address lost its subtitle, and WhatsApp rows carry the brand mark. Emailed documents keep the generic one: nothing records which mailbox fetched them, so claiming a provider would be a guess. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(inbox): the WhatsApp number, three wrong portals, and somewhere to drop the file The WhatsApp row read the response in snake_case while the route answers camelCase, so a linked number rendered as a dash and a verified link read as unverified. Reading phoneMasked and verifiedAt fixes both. Anthropic, Vercel and Supabase are out of the portal directory. All three email their invoices to European customers, so listing them told somebody to go and log in for a document already sitting in their inbox: worse than saying nothing, because it sends them away from the answer. The directory's bar is 'does not send the invoice', not 'also has a portal'. The poll it was seeded from asked which portals people log into, and people answered with where an invoice can also be found. The same objection may reach further down the list. A purchase with no underlag now offers somewhere to put one. Telling somebody a document is missing without a place to drop it is half an answer, and the drop zone carries the amount and the date so the right file goes to the right purchase. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(portal): the links were never opened, and two of them were wrong The directory shipped with eighteen hand-written paths and none had been clicked. The file said so in its own header and shipped regardless, which is how a founder came to land on a 404 opening Google Workspace. A sweep of every URL found GitHub broken as well. Google Workspace now points at the console root rather than a deep billing path: admin.google.com refuses automated requests, so no deeper path can be verified from here, and a link that lands one click short beats one that lands on an error page. GitHub points at the path that actually answers. Trygg Hansa is removed because neither candidate URL could be reached at all, and an unverifiable link is exactly the promise this file kept warning about. scripts/check-portal-urls.mts sweeps them, so the next wrong URL is found by a script rather than by somebody who trusted the link. A 404 fails it; a host that refuses automation reports as unreachable and does not, because failing on those would train people to ignore the output. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(inbox): the drop zone now actually attaches the file to the purchase It did not. The generic upload sends only the file, so a document dropped while a purchase was selected landed in the inbox unmatched, while the pane showed that purchase's amount and date directly under the drop zone. The copy promised a link the code never made, and the user was left to match by hand what they had already told us. Uploading from a selected purchase now matches the new item to that transaction through the endpoint that already exists, and a file dropped anywhere on the page while a purchase is selected counts as that purchase's receipt rather than a loose upload. When the match fails the document is still safely filed, so it says so plainly instead of claiming a link that is not there. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(inbox): book the underlag against its transaction, and stop claiming links Two blockers found by review, both on the path that writes to the ledger. "Granska och bokför" never sent transaction_id. JournalEntryForm serialises a fixed set of keys and that is not one of them, and BookInboxItemDirectlySchema is a non-strict z.object, so the source_id carrying it was silently stripped. The verifikat posted standalone, the bank transaction stayed unbooked, and matched_transaction_id was overwritten with null: the match somebody had already made, undone, while the rail said Bokförd over all of it. Fixed in three places because one was not enough. JournalEntryForm takes an extraBody passthrough, the dialog sends transaction_id through it, and the route now falls back to the item's existing match rather than null, so a caller that merely forgets the field cannot undo work. Removing that fallback fails the new test. The hunt banner said "kopplades till ett köp" about pending_operations rows. The hunt stages proposals for approval and books nothing, so the number was real and the word was wrong: a user would read it, believe three purchases were done, and leave. It now says how many förslag await granskning, and links there. Booking also left the rail in its pre-booking state, still offering to post, so the same underlag could be submitted twice. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(inbox): no marker on a healthy state, no false empty state, no dropped files Three from review. The sources chip painted a sage dot whenever every mailbox was fine. Convention 12 rules semantic colour out of chrome, and convention 5 rules out a marker on a normal state: a chip every company sees always is a chip that says nothing. What is left is the exception, which is worth an ochre word and an icon. The pre-existing sage on matched rows is untouched; it is not this branch's to change. The empty state asserted "Varje köp har sitt underlag" while the trigger directly above it still showed the unsearched count. Type a term under Att göra, switch to Saknar underlag, and the page told you every purchase was covered while the button beside it read 50. It now says what is true: no matches for that term. A drop of several files onto a selected purchase kept the first and discarded the rest in silence, so a receipt scanned as two images left the purchase looking resolved with half its paperwork gone. They cannot all be one purchase's underlag, so the extras are filed in the inbox and the toast says how many. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(inbox): the hunt banner now says a press is not the last word A press fetches a bounded number of receipts, so an empty result usually means not yet rather than nothing there. The banner said 'Inget matchade något köp' and stopped, which reads as final and sends people away from a mailbox that still holds their receipts. It now says how many purchases are left to search for, and to press again. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(inbox): a count not a score, an honest failure, full-opacity borders '5 av 12' read as a bad extraction even when a kvitto had given up everything a kvitto has: half those twelve fields only exist on an invoice, so the denominator was measuring the document kind rather than the reading of it. It now says how many fields are filled, and says nothing when none are. The failure banner told people their mailbox had not answered even when the failure was ours, sending them to check a healthy Gmail. It now reads searchFailures and only blames the mailbox when a mailbox actually refused. Opacity-suffixed borders on the sources panel, which design.md forbids on surfaces: the border token is calibrated for full opacity. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat(inbox): translate the new strings, and name the mailbox that fetched a receipt Both of these were deferred with reasons, and one of the reasons was wrong. 57 keys in inbox_workspace, in both locales, covering every string this branch added. The component already had 27 t() calls, so hardcoding beside them was an inconsistency rather than a convention. The message-keys guard caught an invented journal_form.no_document on the way, which is what it is for. The provider mark claimed nothing recorded which mailbox fetched a document. It does: lib/receipt-hunt/ingest.ts writes mail_provider and mail_mailbox into channel_context on every ingest, and GET /items already selects that column. A hunted receipt now carries the mark of the mailbox it came from; forwarded mail has no connection behind it and keeps the envelope, which is the honest distinction rather than a guess. InboxChannelContext was WhatsApp-shaped and is now a union over the two intakes that write it. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(agent-context): keep the clarification channel narrow Widening InboxChannelContext.channel to cover the mail hunt broke this: only WhatsApp asks a human anything, so only WhatsApp produces clarifications. The mail hunt writes the same column with its own shape and never carries answers, so the provenance field stays 'whatsapp' rather than following the union. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(inbox): book the transaction we preserved, and date the verifikat by the event Three from PR review, two of them real. Preserving matched_transaction_id without booking it was the worse half of the bug it fixed. The transaction update was still guarded on the caller having sent transaction_id, so an omitted field left the item looking resolved while its bank line stayed open forever. Both the update and the item now use the same resolved id: the one the caller named, or the one the item was already matched to. Reverting the guard fails a test. The verifikat date fell back to today when there was no proposal, which is exactly the unknown-supplier case the dialog exists for. BFL 5 kap 6-7 § asks for datum för affärshändelsen; the day somebody opened a dialog is nobody's business event. It now falls back to the document's own date first, and only then to today. An en dash had crept in as a placeholder glyph, which the repo bans. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Mattsson
|
9dbaebcc50 |
fix(invoices): ROT/RUT credit notes; verifikat amount sort, HTML underlag, source chip (#1523)
* feat(invoice-inbox): store HTML mails as underlag, expandable field editor Body-only mails and .html attachments (including forwarded .eml bodies) no longer dead-end as "Fel vid bearbetning": the mail body is wrapped into a self-contained text/html document, stored through the normal upload/extract pipeline, and extracted via a new HTML-to-text Bedrock path, so the mail itself can serve as bookable underlag. Empty mails keep the error row, unsupported types are still rejected, and webhook retries dedupe on resend_email_id. Mail HTML is attacker-controlled, so rendering is fully sandboxed: iframe sandbox in the workspace preview and a CSP sandbox header on /api/documents/:id/inline for text/html. The type is accepted only from the email pipeline (EMAIL_ALLOWED_MIME_TYPES), never from manual upload. The "Extraherade falt" rail gains an expand button opening a centered dialog with the same autosaving field editor at a readable size (two columns), which also gives every failed or skipped extraction a manual fallback. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * feat(bookkeeping): sortable verifikat list headers with amount sort - clickable sort toggles on the verifikat list headers (asc -> desc -> default) - total_amount computed column + sort_by total/description on the list route - failed list loads render an error card with retry, never the empty-ledger state Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(import): decode bank CSV as Windows-1252 fallback in column mapping The client read the uploaded file with file.text(), which is UTF-8-only, so Windows-1252 exports (e.g. Handelsbanken) rendered and re-parsed with U+FFFD in place of Swedish characters. Decode from bytes with the shared decodeFileContent() helper, matching what the server parse route does. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * feat(bookkeeping): stackable sort keys on verifikat list headers - shift-click adds a column as secondary/tertiary sort key (max 3), plain click keeps the single-key tri-state cycle - sort_by accepts a comma-separated priority list; single tokens stay valid - voucher tiebreak follows the last key's direction (#972 parity) - priority numbers on stacked headers; hint text in the filter dialog Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(invoices): keep ROT/RUT deduction fields positive on credit notes Crediting an invoice with a ROT/RUT deduction failed 100% of the time: the credit-note path negated deduction_total (and per-item deduction_amount) like the other amounts, but both columns carry CHECK (>= 0), so Postgres rejected the insert and the user only saw 'Kunde inte skapa kreditfaktura'. Store the deduction fields as positive magnitudes, matching the convention everywhere else. The stored sign is inert on credit notes: the reversing verifikat recomputes the ROT/RUT split from the items, and the PDF and amount-to-pay logic skip deductions on credit notes. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * refactor(transactions): share the source chip across inbox and history modes Move SourceFilter to transaction-types.ts (widened with 'bank:other' and 'acct:<id>'), render the one toolbar ContextPicker in both view modes, and drop the narrower duplicate chip inside TransactionHistoryList. The history list now applies the acct:/bank:other narrowing itself and hides skattekonto rows under any bank-side selection. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * chore(deps): bump js-yaml to 4.3.1 Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * test(schema): recognize PostgREST computed columns in the migration parser The verifikat amount sort orders by total_amount, a PostgREST computed column (a function on the journal_entries row type, migration 20260811100000). The schema guard only modeled real columns, so no-phantom-columns flagged the order as a phantom. Teach the parser that a function whose only argument is a table's row type joins that table's column set, with DROP FUNCTION retraction when the signature names the row type. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix: resolve PR #1523 review findings - journal-entries route: append the globally unique id tiebreak to every direct-query sort; voucher series+number repeat across fiscal years, so the all-years scope could duplicate or drop rows at page boundaries. Existing order assertions updated, new all-years tiebreak test. - documents inline route: CSP source policy on HTML previews; sandbox alone still loads remote resources, letting a tracking pixel notify the sender on open. New route test asserts the full header. - JournalEntryList: catch rejected list requests so loading cannot stick forever, and gate every post-await state write behind a request generation so a slow earlier request cannot overwrite the current sort. - TransactionHistoryList: pagination follows the selected source scope (reachable with zero matches on the current page, hidden for the skattekonto scope it cannot affect). - transactions page: bank:other picker availability derives from history rows too, not only the pending inbox dataset. - DECISIONS.md: mark the superseded single-sort decision; record the credit-note deduction positive-magnitude invariant and its verified reader inventory (Swedish review flag). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix: guard metadata refetches behind the list request generation fetchAttachmentCounts and fetchRattelseFlags write state after their own awaits; a stale list request's late completion could overwrite attachment counts and rattelse flags for rows a newer request just rendered, showing false missing-underlag warnings. Both helpers now take the caller's generation guard and discard stale completions, including the attachment-counts loaded flag. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
f2d9e98af3 |
fix(mail): stop Gmail refusing the search, and stop calling that "hittade inget" (#1521)
Pressing Leta produced mails=25, documents=0 on a real two-mailbox run. Nothing was found because nothing was searched: every request came back 429 "Too many concurrent requests for user". Two bugs, and the second is the one that matters. The search fanned out with Promise.all over every message id at once, one Gmail request per message, per connection. Gmail enforces a per-user concurrency ceiling as well as a daily quota, and this sailed past it long before any volume worth worrying about. It now runs through a pool of five per connection, which is comfortably under and still finishes a page of results in a couple of round trips. The catch turned each refusal into an empty array, with a comment saying one mailbox's failure must not become the company's. Right instinct, wrong consequence: an empty array is also what an empty mailbox returns, and the manual hunt loop stops on fetched === 0 because that is its signal for "the mailboxes hold nothing more for what is open". So a rate-limited search told the user their receipts do not exist, and stopped looking. searchFailureCount() now separates "could not look" from "nothing there". The run route reports it, and the loop treats a pass with failures as failed rather than finished, so pressing again is the obvious next move instead of a pointless one. This is the failure this feature exists to catch, happening inside the feature: silence that reads as an answer. Restoring the unbounded fan-out fails one test; removing the failure counter fails three. Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
709c0c817a |
fix(auth): serve /docs and llms.txt to anonymous agents (#1520)
The middleware allowlist never included the agent-discovery surfaces, so every anonymous request to /llms.txt, /llms-full.txt, or /docs/* was 307-bounced to /login on hosted. The llms.txt convention exists for logged-out crawlers and IDE agents, and /docs is the public API documentation that the OpenAPI spec and the installable accounted-api skill link to. openapi.json and /.well-known/* only escaped because the proxy matcher skips .json and .well-known paths. Signed-in users fall through to the same content with no redirect. Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
11b82cbb91 |
feat(api): installable accounted-api agent skill + openapi-to-skill generator (#1516)
* feat(api): installable accounted-api agent skill + openapi-to-skill generator Three layers, per the July/August 2026 agent-skills ecosystem (skills.sh / npx skills add, as used by Stripe/Cloudflare/Supabase for their APIs): - skills/openapi-to-skill/: generic, installable skill that turns any OpenAPI spec into a consumer-side integration skill, with a portable stdlib-only inventory/condenser tool and an output template + quality checklist encoding the distill-not-restate methodology. - skills/accounted-api/: the installable skill for our own API, rendered deterministically by scripts/api-skill/generate.ts from the v1 endpoint registry + hand-authored overlays (auth, conventions, domain gotchas). CI gate: npm run apiskill:check (core-build.yml). - lib/api/v1/registry.ts: generateOpenApiSpec now emits requestBody (incl. multipart binary parts) and path parameters, and the Zod converter learned .default()/z.record()/.pipe()/.transform(), so the public spec carries request contracts instead of prose-only. Docs: /docs/api landing + /llms.txt now point agents at the skill install; corrected the stale test-key description in the landing (test keys read real data and force dry-run writes; they are not sandbox-company bound). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(skills): escape backslashes in markdown table cells (CodeQL js/incomplete-sanitization) Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
a004990041 |
feat(inbox): the purchases that are still missing their receipt (#1517)
The page lists documents, so a purchase with no document at all could not appear on it. That is precisely the gap the receipt hunt exists to close, and the half a user can do something about: fetch the invoice from the supplier's portal, or ask whoever made the purchase. GET /purchases supplies it, read-only, with the portal link attached when the directory knows where that supplier keeps its invoices. Salary and tax get no link: they have no invoice to fetch, and a link there implies somewhere to go. The predicate moves into lib/transactions/purchases-without-underlag.ts, and it is not the hunt's filter copied across. `journal_entry_id IS NULL` is not the same as "not booked": bulk-booking many transactions onto one verifikat records it in transaction_voucher_links, and a payment split across invoices records it in the payment tables, and both leave that column null. The hunt tolerates the false candidate because the worst case is a proposal nobody accepts. A list shown to a person does not: those rows would sit under "saknar underlag" forever, already booked, with nothing the user could do to clear them. So the column filter stays as the cheap indexed first pass and isTransactionBooked settles it afterwards. That predicate is canonical and nothing here re-implements it. The hunt is deliberately not rewired in this change: its query is identical apart from that check, but moving a nightly cron onto new code belongs in its own PR. The thresholds are kept equal to the hunt's so the two cannot drift meanwhile. Removing the booked filter fails five of the seven predicate tests. Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
ca6b962662 |
feat(receipt-hunt): a directory of where suppliers keep their invoices (#1506)
Some vendors never attach anything. They mail "your invoice is ready", or nothing at all, and the invoice waits behind a login. Seeing "OPENAI CHATGPT SUBSCR" on a statement is enough to say which page it is on, and that turns a dead end into a one-click errand. Deliberately not what Presto does. They log in for you across a thousand portals, with a desktop app so credentials never leave the machine. Holding a customer's supplier passwords is the expensive and legally heavy half of that product, and skipping it costs little of the value: knowing *where* is most of it. The lookup runs at read time and nothing is stored against a transaction, so adding an entry fixes every ledger at once, retroactively, with no migration and no backfill. A supplier added today starts answering for purchases made last year. Eighteen entries, ordered by how many companies actually pay them, measured across production ledgers and cross-checked against a customer poll. The two agree closely: Google and OpenAI lead both by a wide margin. Measured coverage: 769 of 10 182 undocumented purchases, which is 7.6% of rows but 120 of 224 companies, so more than half of all companies have at least one purchase this can now answer. Two things it refuses to do. It offers nothing for payments that have no invoice, reusing the rule that already keeps the hunt from searching mail for salary and tax: a link there implies somewhere to go. And longer aliases win, so a Google Cloud bill is not sent to the Workspace console. Vendors whose billing page could not be pinned down were left out rather than guessed at, including some with real volume: Fortnox, Apple, Klarna and Mynt. A wrong URL is worse than a missing one, because it spends the trust the feature runs on. Every entry here still deserves a human clicking it before anyone relies on it. No interface yet: this is the data and the lookup, so the workspace can use it the day it lands. Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Mattsson
|
614b7e60b9 |
fix(salary): apply percent brackets for monthly incomes above 80 000 kr (#1510)
* fix(salary): apply percent brackets for monthly incomes above 80 000 kr
Skatteverket's monthly tax tables switch from fixed krona amounts to
percent-of-income rows above 80 000 kr/month. The lookup only loaded the
krona ("30B") rows and clamped higher incomes to the last bracket,
under-withholding every salary above 80 000 kr (e.g. 100 000 kr, tabell
31 kolumn 1: 25 294 kr instead of 35 000 kr).
- fetch both 30B and 30% sections from the Skatteverket API; treat a
missing section as API failure so the bundled fallback wins over
incomplete data
- TaxTableRate is a discriminated union; percent brackets withhold
percent of the whole monthly income, ore dropped per SFF 2011:1261
22 kap. 1 (oretal bortfaller)
- fallback generator parses %-rows too; regenerated with 1 232 percent
rows and a guard that every table carries both sections
- keep the old clamp only as a warn-logging last resort
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(salary): fail loudly on incomplete tax table data (review findings)
Address CodeRabbit and Swedish accounting review findings on #1510:
- lookupTaxAmount throws TaxTableUnavailableError when loaded brackets
contain a gap instead of silently withholding 0
- a failed or empty pagination page fails the whole API fetch so the
bundled fallback serves complete data
- kolumn values are parsed strictly (decimal-aware, comma accepted);
malformed values fail the fetch instead of becoming 0 kr / 0 %
- importer rejects malformed column values instead of emitting 0
(regenerated fallback is byte-identical)
- close the bracket gap in the calculation-engine test fixture
- clarify the ore-truncation citation and use an absolute date in
DECISIONS.md
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(salary): validate income boundaries in tax table parsers
Round-2 CodeRabbit finding on #1510: income boundaries were still parsed
with parseInt, which accepts "100abc" and turns garbage into 0 or an
open-ended bracket. Both the importer and the API loader now require
digits-only boundaries; an empty upper bound is legal only on percent
rows (the open-ended top row). Malformed API data fails the fetch so the
bundled fallback runs; malformed TXT data fails the import. Regenerated
fallback is byte-identical.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
|
||
|
Jakob Wennberg
|
e45218bcc6 |
fix(payments): creditor address on every payment (Validex round 3) (#1509)
Rule 237 fires on the BGNR-to-BGNR path too, so the round-2 reading of rule 020 was wrong: it only relaxes the Ctry element, not the address. Cdtr now always carries PstlAdr (TwnNm from the payee_city snapshot when known, Ctry SE), and the preview warns when the supplier register lacks a city, since rules 237 + 222 together make a town effectively mandatory at Swedbank. Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
4655f3da48 |
fix(payments): creditor address per Swedbank TwnNm rule (Validex round 2) (#1508)
A present PstlAdr must carry TwnNm from November 2026 (PFH_222), so BGNR-to-BGNR payments now carry no creditor address at all (rule 020 requires none there), IBAN-debited payments carry the supplier's town (snapshotted as payee_city) plus Ctry SE, and the debtor address comes from company settings, clearing the info-level rule 236 as well. Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
fea5dfd1f9 |
fix(payments): correct pain.001 dialect per Swedbank Validex run (#1507)
* fix(payments): correct pain.001 dialect per Swedbank Validex run Real MIG validation (eken.validex.net) rejected the first generated file on four rules: character set (e-acute in names), missing InitgPty OrgId, BGNR creditors demanding a BGNR debtor, and Strd lacking RfrdDocAmt. Names and messages now transliterate to the MIG set, the org number is required at batch creation (settings first, companies fallback), bankgiro payees debit the company bankgiro in their own PmtInf group when one exists (IBAN otherwise, with Cdtr PstlAdr/Ctry SE always present), and structured OCR remittance repeats the amount as RfrdDocAmt. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(payments): review quick wins on the MIG pass NFC-normalize before transliteration (decomposed marks from PDF-pasted names fold to the precomposed forms the map knows), a dedicated settings link label for the missing-org state, and coverage for an invalid company bankgiro being dropped from the debtor snapshot. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
f0bedc14af |
feat(payments): betalfil UI (betalfil 3/3) (#1505)
* feat(payments): betalfil UI (betalfil 3/3) Bulk-select + Skapa betalfil bulkbar on the supplier-invoices list, preview dialog with per-line editable amount/date and exclusion reasons, payment-files history page with re-download, cancel and a sequential bulk mark-paid (duplicate guard respected, never forced), I betalfil chip on rows in active batches, clearing/kontonummer fields on the supplier form, and the supplier_payment_files namespace in sv+en. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(sandbox): make the demo AP data betalfil-ready Demo supplier bankgiro numbers were not Luhn-valid, the unpaid demo invoice had remaining_amount 0 (no trigger derives it, so the list said 0 kr kvar att betala), and the company had no IBAN/BIC, all of which excluded the seeded data from the betalfil flow. Numbers swapped for Luhn-valid ones (991-2346 is Bankgirot's test number), a valid OCR added, and both bulk-insert rows set remaining_amount explicitly per the PostgREST normalization rule already documented inline. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
576a34750a |
feat(payments): payment-batch API surface (betalfil 2/3) (#1504)
Preview/create/list/get/file/cancel routes under /api/supplier-invoices/payment-batches, SI_BATCH_* structured errors, and a fail-closed batch-membership pre-check in the supplier invoice DELETE route (the FK RESTRICT is the backstop). File downloads stamp file_generated_at + download_count but regenerate byte-identically. Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
f776e375c5 |
feat(payments): supplier payment batch schema + pain.001 domain lib (betalfil 1/3) (#1500)
* feat(payments): supplier payment batch schema + pain.001 domain lib Betalfil for leverantorsfakturor, part 1 of 3. New tables supplier_payment_batches + supplier_payment_batch_items (RLS, immutable item snapshots, FK RESTRICT on invoices), payee/reference resolution, eligibility rules shared by preview and create, and a supplier-dialect pain.001.001.03 generator (SESBA 9900 BGNR / 9960 BBAN / clearing BBAN, SCOR for Luhn-valid OCR, Ustrd fallback, no SvcLvl/CtgyPurp). Deterministic regeneration: msg_id derives from the batch id, CreDtTm from created_at, so re-downloads are byte-identical. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * refactor(payments): use lib/money helpers instead of raw ore rounding The naive-ore-round ratchet flags new Math.round(x*100)/100 sites; roundOre/sumOre/ORE_TOLERANCE are the sanctioned forms. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(payments): classify batch tables in full-archive contract + fixture The no-phantom-columns contract requires every company-scoped table to be triaged in full-archive-export; the batch rows are underlag for the payments they initiated, so they dump with the archive. makeSupplier gains the clearing/account columns the Supplier type now carries. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(payments): harden batch integrity per review Composite (id, company_id) FKs so items can never cross-link a batch and an invoice from different companies; BEFORE UPDATE trigger keeps batches immutable outside lifecycle + download metadata and one-way on created -> cancelled; active-batch lookup now fails closed (an error no longer reads as no active batches, which would have silently disabled the duplicate-batch guard); today derives from Europe/Stockholm, not UTC; pain.001 control sums add the amounts as rendered so CtrlSum always equals sum(InstdAmt); event-bus reset in test hooks; Danske LB date claim in DECISIONS verified against the primary page (the bot's 12 May date is the alias-initiation date, not LB retirement). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(payments): bind cancellation metadata to the cancel transition cancelled_at/cancelled_by may only be written by created -> cancelled; cancelled_by may still become NULL so the FK's ON DELETE SET NULL keeps working when the cancelling user's account is deleted (proven in pg). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
88140e5774 |
feat(receipt-hunt): let a model settle the pairs the arithmetic cannot (#1498)
The weights in the matcher are chosen by hand, and two of them were
adjusted last week because they produced a wrong answer on real data. A
pair agreeing to within 1%, from a merchant the matcher recognised, still
scored 0.62 because a date had drifted. That says more about the
constants than about the receipt.
So the formula keeps what it is good at and hands over what it is not.
Above 0.80 it decides alone: an exact amount from a recognised merchant
needs no second opinion, and buying one for every pair would be latency
and cost for a verdict nobody doubts. Below 0.60 nothing is asked either,
because inviting a model to rescue a pair the evidence does not support
is how plausible wrong answers get made. Only the band between goes for
adjudication, which on a real ledger was five pairs against one the
formula had already settled.
The question is a yes or no with a reason, never a score. An earlier
design in this feature asked a model to rate its own certainty and it
anchored on round numbers, which is what the calibration literature
predicts. Judging concrete evidence and explaining the judgement is a
different task.
A verdict is checked, not trusted: a key nobody asked about is dropped, a
pair answered twice keeps its first answer, and a failed call accepts
nothing, leaving the run exactly where the arithmetic left it. The
formula's own score is stored unflattered next to the verdict, because
dressing it up would hide the uncertainty that sent the pair for a second
opinion in the first place, and agent_metadata records which instrument
decided.
Adjudication runs on a dry run too. A provkörning is meant to show what a
real run would propose, and skipping it would show a smaller, different
answer than the one that lands.
Measured on a real ledger: five uncertain pairs asked, two accepted with
reasons a human can check ("Samma leverantör och belopp stämmer inom
rimlig valutamarginal"), three rejected. Proposals went from one to three.
Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
||
|
Jakob Wennberg
|
4c479f84f4 |
feat(receipt-hunt): convert a foreign receipt instead of refusing to compare it (#1497)
* feat(receipt-hunt): compare a foreign receipt by converting it, not by refusing A Swedish bank posts a converted figure for a purchase abroad while the receipt states the original: Anthropic bills 180,00 EUR and the statement reads -2 014,32 kr. Neither number appears in the other document, so the matcher refused the pair rather than guess. On a SaaS-heavy ledger that is not an edge case: of 25 receipts fetched from a real company's mailboxes, 14 were in USD or EUR and none could ever pair. The receipt's total is now resolved into kronor with Riksbanken's rate for its own date, and handed to the same matcher, which still wants the merchant and the date to agree. The seam already existed: the scorer passed null where a SEK value would go, with a comment explaining that cross-currency pairs were deliberately incomparable. Surfaces that do not resolve a rate still pass nothing and behave exactly as before. Rates are fetched once per currency and day. Riksbanken answers 429 to a caller that asks per document, and a run holds a dozen receipts from one vendor in one month. A rate that cannot be resolved leaves the receipt exactly as incomparable as it was. Two calibration faults surfaced once the amounts became comparable: A converted total is judged at 9% rather than 5%. Riksbanken publishes a mid rate and a card issuer charges its own, so the two carry a known spread on top of any disagreement about the sum: measured against real statements, 1.2% to 3%. Holding both to one bar treats a rate spread as if it were a discrepancy. The date tolerance moves from 3 days to 10. A card settles days after the purchase, an international one routinely a week later, and a forwarded receipt carries the purchase date while the statement carries the posting. At three days the signal scored zero for ordinary correct pairs and took a quarter of the weight with it: a receipt agreeing to within 1%, from a merchant the matcher recognised, still capped at 0.62. The 27 human- confirmed pairs and the 7 near-misses that must not match all still hold. Measured on that ledger: purchases with any candidate at all go 9 -> 13, and the Anthropic pair proposes at 0.82 where it was previously unscoreable. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(receipt-hunt): an undated receipt is not converted at today's rate Raised in review. A foreign receipt with no invoiceDate fell back to the current date, which put a receipt of unknown age into amount matching on the strength of a guess: a rate two years out is how something incomparable acquires confidence it has not earned. No date, no conversion, and the receipt stays exactly as it was. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
1d635b0d25 |
feat(receipt-hunt): look for receipts on request, from the mailbox settings page (#1496)
* feat(receipt-hunt): a button that looks for receipts on request The nightly cron exists but still does not search mailboxes, and for a good reason: a sweep of one real 172-message mailbox took over 600s, against a scheduled function's 300. Pressing a button is the honest shape for work that big. A bounded pass reports what it found and how much is left, and the person decides whether to press again; a nightly run could only truncate silently. POST /api/receipt-hunt/run searches the mailboxes for eight purchases and fetches at most ten receipts per press. Gated on the AI tier, because reading the amount out of a PDF is what makes a fetched attachment matchable at all: without it the hunt would file documents that can never pair, which is worse than not running. Writes no journal entries; every pairing is still a proposal waiting for approval. The button lives on the mailbox settings page, which already ships, and says what happened in words rather than a spinner that stops: "3 underlag hämtade. 12 köp kvar att söka igenom." huntCompany gains maxReceipts so a manual pass can carry a different budget from a nightly one. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(receipt-hunt): stop a manual press spending its budget on the wrong purchases The first real press searched eight purchases, read forty mails and found nothing, which looked like a broken model. It was the ordering. Purchases are searched largest first, which is right for a nightly sweep that eventually covers everything and wrong for a button pressed a few times. On a real ledger the largest rows are the least likely to have a findable receipt: rent already invoiced, bare payment references, direct debits. Those filled the forty-mail cap, so the productive purchases further down the list, the ones whose receipts are actually sitting in the mailbox, were never read at all. The cap was the binding constraint, not the time: eight purchases and forty mails took 43s of the 300 available. A press now searches 25 purchases and reads 100 mails, measured at 85s and finding 7 underlag on the same ledger that returned 0 before. huntCompany gains maxMails alongside maxReceipts, so a manual pass can carry a different budget from a nightly one rather than sharing an environment default with it. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(receipt-hunt): one underlag per purchase, and a press that fits its budget Your second press exposed three things, none of which a dry run reaches. It took 5.8 minutes. The 85s I measured was a dry run, which never fetches, uploads or extracts; each fetched receipt costs about another 37s because it is downloaded, stored, and then read by a model that opens the PDF. Seven of them ran past the 300s a serverless function gets, so in production that press would have been killed. Four receipts per press keeps a full pass inside the budget. It fetched seven receipts and proposed nothing. A single mail carries the invoice AND the receipt for one purchase under different names ("Invoice-E19DBF63-0021.pdf" beside "Receipt-2066-0204-8388.pdf"), and the same receipt reaches a second mailbox on a different message. Each was fetched separately, so the pool filled with identical candidates and the matcher refused to propose any of them rather than flip a coin. The per-run key is now the vendor and the total, which is what identifies a purchase; the filename only decides when no amount was read. Nine duplicates already in the pool were removed. And with the duplicates gone it still proposed nothing, for a separate reason: "Utlägg Norwegian" scored 0.18 against "Norwegian Air Shuttle AOC AS". Utlägg is Swedish for an expense reimbursement, bank vocabulary rather than a company, and leaving it in broke the token-subset match, so an exact 1 998 kr pair leaned entirely on a date eight days out and fell under the floor. Stripped, along with överföring, via internet, bg-bet and autogiro, in the comparison path only. normalizeMerchantName is untouched: it is the persisted konteringskarta key with a SQL mirror, and its 22 string pins and the 27-pair golden set still pass. Measured after: the Norwegian pair proposes at 0.72. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * perf(mail): read each message once per press, not once per query The third press proposed a pairing, which the second had not, but still ran 5.2 minutes against a function's 300s. Cutting receipts from seven to four had only saved 36s, which said the receipts were never the cost. Every search fetched a full message for every hit, and a press searches many purchases across every connected mailbox. One receipt mail answers several of those queries, so 25 purchases against 2 mailboxes could ask Gmail for well over a thousand messages to end up with a hundred distinct ones. Deduplication happened in the caller, too late to save the work. A mail's content never changes, so it is now read once per mailbox and kept, bounded at a thousand entries and evicting oldest first. Measured on the same ledger: 55 purchases and 100 mails now take 102s, where 25 purchases alone previously cost around 264s before a single receipt was fetched. clearMessageCache exists because tests reuse message ids and production does not. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(receipt-hunt): do not refetch a receipt the company already holds The fourth press fetched four documents the company already had: Bolagsverket, Supabase twice, Uber. They came back because I had deleted them as duplicates, and the cross-run check is the message and attachment id, which lives in the rows I removed. That was my mistake, but it exposed a real gap. The vendor-and-total key only deduplicates inside a single pass. Across passes the same purchase still arrives as an invoice in one mail and a receipt in another, with different file keys, and both were fetched: the pool fills with identical candidates and the matcher then refuses to choose between them, which is how a press can fetch four documents and propose nothing. The pass now starts from what the company already holds, so its budget goes on documents that are actually missing. Receipts per press drops to three. Measured on this ledger, a fetched receipt costs about 50s from download to a stored amount, and that is the model reading the PDF rather than the network: seven took 5.8 minutes and four took 5.1, both past the 300s a function gets. Three fits, but it is a stopgap. Doing the fetch inside the request is the wrong shape for work this slow, and the fix is to move it off the request rather than keep shaving this number. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(mail): a mail body must not outlive the run that read it Two findings from the review, both correct. The message cache held whole MailCandidate values, and one of those fields is the mail body. The contract says a body is read once to extract fields and discarded, and a process-global cache quietly broke that: bodies of one company's mail sat in memory across requests until eviction or a restart. The MailSearchService contract now has releaseCache, the Gmail adapter clears its messages, and the hunt calls it in a finally so a failed run releases them too. The duplicate key accepted an empty vendor, so two unrelated documents that happened to cost the same collapsed into one candidate. Those now fall back to the file they came from: without a vendor there is nothing to anchor an amount to. The same finding caught something worse that I had introduced one commit earlier. The persistent check derived its key from the stored extraction while the fetch derived one from the reading model, so a document filed as "Norwegian Air Shuttle AOC AS" did not recognise an incoming "Norwegian" and was fetched again. Rather than guess at aliases, which would fold "Google Cloud" into "Google Workspace", the identity is now written onto the row when the receipt is filed and read back verbatim. Rows filed before that fall back to the extraction. receiptIdentity is one exported helper with its own tests, used by both sides, instead of the same expression written twice. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(receipt-hunt): a monthly subscription is not a duplicate of last month The review caught that my duplicate key was worse than the problem it solved. Anthropic bills the same amount every month, and keying on vendor and total alone made July look like a duplicate of June: every later receipt from any recurring supplier would have been suppressed forever, silently. Duplicates block one proposal; that would have lost a receipt per month per subscription. The identity now carries the document date. Two documents for one purchase share a date; June and July do not. Two smaller faults in the same key. The amount was serialised as a raw float, so 0.1 + 0.2 read as a different total from 0.3; it is rounded to öre like every other money comparison in this codebase. And a document with no vendor was identified by its filename alone, which collapses two unrelated papers whenever a billing system attaches "invoice.pdf": those now carry the message they came from. The key is versioned so a future change to its shape cannot be mistaken for a match against rows written under the old one. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
0d0415e19c |
fix(vat): danstillställningar 25% → 6% from 2026-07-01 in guidance and suggestions (#1491)
* fix(vat): danstillställningar 25% to 6% from 2026-07-01 in guidance and suggestions (#1483) From 2026-07-01 tillträde till danstillställningar is 6% VAT, aligned with other cultural events. 6% was already a supported rate; every guidance surface was silent on the change, so a dance-event customer plausibly got 25% suggested. - swedish-vat skill: rate table row + a July 2026 change note under rate misclassification (cutover date, mixed-venue split vs 25% alcohol), and the 2631 account table mentions dance admission - revenue_reduced_6 descriptor: dans/danstillstallning/dansband/entre keywords and updated description, which flows to both the UI suggestions and gnubok_suggest_categories via findMatchingTemplates - atom seed regenerated; the generator now emits a version-downgrade guard on the ON CONFLICT so two branches each carrying a full 108-atom seed can no longer clobber each other's atom bodies depending on merge order Closes #1483 Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * review(vat): drop overbroad entré keyword, cite SkU25 and the prepayment rule Two findings from the Swedish compliance review bot: - bare 'entré' matched generic admission that is not always reduced-rate; the dance-specific keywords stay - the rate claim now cites its primary sources (riksdagen 2025/26:SkU25, Skatteverket halvårsskiftet 2026) and records that tickets sold and paid before 2026-07-01 keep 25% Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * review(vat): admission-specific keywords only, explicit cutoff wording, seed rebuilt post-merge - bare 'dans' also matched dance courses and artist fees; keep danstillställning/dansband and add danskväll - rate table states the boundary explicitly: 25% through 30 June 2026 - atom seed regenerated from the tree that now includes #1489, emitted as 20260810121001 Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
c77367ac1a |
feat(audit): log trusted-role committed_at overrides to behandlingshistorik (#1490)
* feat(audit): log trusted-role committed_at overrides to behandlingshistorik (#1444) Migrations 20260806150000/160000 let backend writers (service_role, no-claims direct SQL) preserve a preset committed_at on the draft-to-posted transition, but nothing recorded that the transition timestamp was overridden: BFNAR 2013:2 kap 8 wants a log of who did what, when. If a trusted-role connection is ever used against a live company, there is now an audit row separating real from preset commit time. - audit_log accepts action COMMITTED_AT_OVERRIDE (types + Zod filter too) - log_committed_at_override() SECURITY DEFINER writer: audit_log has RLS with no INSERT policy and service_role is not guaranteed BYPASSRLS in every harness; EXECUTE revoked from anon/authenticated so PostgREST cannot expose it as an RPC - set_committed_at() calls the writer in the preserve branch; the stamping branch is unchanged from 20260806160000 - pg tests: override row content (preset value, wall clock, jwt role) for postgres and service_role writers, absence on the stamp paths, and the writer's privilege lockdown Closes #1444 Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * review(audit): stamp wall_clock with clock_timestamp(), not now() Seeding flows post many entries inside one transaction; now() would pin every override row's wall_clock to the BEGIN instead of the actual transition moment. Captured once so new_state and description agree. Test posts inside an explicit transaction after pg_sleep and asserts wall_clock moved past the transaction start. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
fe1ff8649b |
fix(bookkeeping): drop non-standard account 2012, book EF F-skatt on 2013 (#1409) (#1489)
Primary-source check against bas.se (BAS 2026 v2): the official kontoplan has no account 2012; the enskild firma equity block is 2010, 2011, 2013, 2017, 2018, 2019. 2012 'Avräkning för skatter och avgifter' is a program convention (Visma, Bokio, Björn Lundén), not standard BAS, and a non-standard account in BAS_REFERENCE leaks via the backfill into charts, SIE export and SRU filing. - remove 2012 from class-2-equity-liabilities.ts, with a tombstone comment - migration retargets 'Preliminär F-skatt (EF)' lines 2012 -> 2013 (system row plus any clones still carrying the seeded shape) - pin 2012's absence in bas-ef-equity-accounts.test.ts (2113 precedent) - correct the swedish-year-end-closing references that motivated #1388, regenerate atom seed migration Companies whose charts already got 2012 backfilled keep it: existing history stays valid; only future template use books 2013. Closes #1409 Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Mattsson
|
3ff12faa77 |
fix(bookkeeping): reverse-charge VAT in booking templates is 25% of the base, not 20% (#1494)
* fix(bookkeeping): reverse-charge VAT in booking templates is 25% of the base, not 20%
applyTemplate() extracted VAT out of the total (rate/(1+rate)) for every
vat line, including the fiktiv-moms pair of reverse-charge templates.
Under omvand skattskyldighet the supplier charges no VAT, so the total IS
the beskattningsunderlag: on 807.99 kr the seeded EU-purchase template
booked 161.60 kr (20%) on 2614/2645 instead of 202.00 kr (25%),
understating Ruta 30-32 and Ruta 48 on the momsdeklaration.
Fiktiv-moms lines (2614/2624/2634 output, 2615/2625/2635 import,
2645/2647 input) now compute amount x rate on top of the base.
deriveTemplateLinesFromBooking ("Spara som mall") gets the mirror fix:
RC legs no longer inflate the derived total (they net to zero), and RC
rates snap against the base, so a correct RC booking round-trips.
Counterparty/SIE learned patterns already strip RC legs and regenerate
them via generateReverseChargeLines with the gross base; those paths
were correct and are unchanged.
User-reported: "Er automatiska utrakning ar pa 20%, inte 25%".
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(bookkeeping): use roundOre for template VAT rounding (guard ratchet)
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* docs(bookkeeping): clarify fiktiv-moms comment: total is the base, booked amount is the VAT
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
|
||
|
Jakob Wennberg
|
38f5d9812e |
feat(receipt-hunt): find receipts in connected mailboxes and pair them on the amount (#1492)
* feat(receipt-hunt): nightly matcher pairing unbooked purchases with held receipts Stages an attach_document_to_transaction proposal for every unbooked card purchase whose receipt the company already holds, so the underlag is attached before the transaction is booked and the gap never forms. When the user later books it, categorize-core.ts propagates the document onto the new verifikat through the matched_transaction_id link the executor writes. Deliberately scoped to UNBOOKED transactions. The posted-verifikat backlog is 96% imported history whose originals live in the previous system, so it stays a pull (the verifikat_missing_document worklist) rather than a nightly push. Ranking reuses scoreUnderlagCandidates; the pool is loaded once per company instead of per transaction, which removes both the N+1 and the newest-50 truncation a per-transaction lookup imposes on a deep backlog. Five guards, each mutation-tested: a confidence floor above the shared candidate floor, an ambiguity margin so two equally-good receipts are left to the picker rather than coin-flipped, one-receipt-one-purchase, one live proposal per purchase, and permanent suppression of pairs a human rejected. Suppression is derived from pending_operations history rather than a new table: terminal rows are immutable and a rejection is already the durable "no". Runs 05:30 UTC, after the 05:00 bank sync. Gated on RECEIPT_HUNT_COMPANY_IDS, which hunts nobody when unset so enabling it stays a deliberate act. No migration, no journal writes, no UI: proposals land in the existing Granskning queue. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat(receipt-hunt): dry-run mode for provkörning against a real ledger Returns the pairings a run would stage without writing any of them, so a company can see tonight's proposals before they reach the granskningskö and so the matcher can be validated against production data without staging an operation. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(matching): fold Swedish bank descriptors so receipts reach their purchases calculateMerchantSimilarity compared raw bank descriptors, so a receipt from "Alviks kött och fisk" scored 0.125 against the bank's own row for it, "Alviks koett och fisk K3667 Kortköp/uttag" — an öre-exact pair no threshold could reach. Adds normalizeForMatch, used for similarity only, which folds what the card rails add and never changes identity: the K#### token, Kortköp/uttag verbs, a leading "Kortköp YYMMDD", trailing /YY-MM-DD dates, reference numbers glued to the name, domain wrappers, legal forms, and the three ways banks mangle Swedish letters (ö, transliterated "oe", and ?? mojibake). Processor markers become spaces because the merchant sits before the star in GOOGLE*PLAY and after it in K*IKEA GALLE. Token-subset containment is scored level with substring containment so a receipt's legal name matches the bank's trading name. normalizeMerchantName is left byte-identical and now documents why: it is a transitive input to categorization_templates.counterparty_name, a persisted UNIQUE key with a hand-written SQL mirror the ledger-context RPC recomputes at query time. Changing it would make stored keys stop equalling computed ones, so the konteringskarta join misses and insertOrUpdateTemplate inserts a second row per merchant instead of migrating the occurrence counts. Aggressive folding is safe because it is applied to both sides of every comparison, so an over-eager fold still matches; the risk is collision between different merchants, which the new tests guard. Measured on 27 receipt/transaction pairs humans actually confirmed in production: recall 27/27, and 0/7 false positives on deliberately similar but distinct merchants. Full unit suite unchanged (13,004 passing), including the 22 string pins on the frozen key path. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat(mail): read-only Gmail connector so receipts are found without forwarding Forwarding was the only way a receipt reached Accounted, and it is both unpopular (97% of companies with the problem have never used their inbox address) and fragile: Arcim's own forward has been off for weeks and nobody noticed. This lets the hunt look in the mailbox instead. Scope is gmail.readonly and nothing else. It can search and download attachment bytes, and it structurally cannot send, modify or delete: the promise the consent screen makes is enforced by the grant, not by our code being careful. The consequence is deliberate: the agent can prepare a forward for a portal-link receipt but can never send one itself. Query-then-classify, never sync. For each unexplained purchase we run a provider-side search in a -3/+10 day window, pull metadata for a handful of hits, and keep nothing. No mailbox is mirrored and no message body is stored, which is what keeps this inside Google's Limited Use terms and GDPR data minimisation. Mail is searched only for purchases Underlag could not already explain, so a receipt we already hold never costs a mailbox read. The query ORs merchant against amount rather than requiring both: demanding both misses every rebrand and reseller (Anthropic bills as Claude), while the amount alone is a strong filter inside two weeks. mail_connections is service-role only with RLS enabled and zero policies, because the row holds a live refresh token and RLS cannot hide a column. Uniqueness is (company, provider, address) so a second mailbox is additive and a reconnect updates in place. Tokens are AES-256-GCM under their own key by preference, since a mail grant reads correspondence rather than backups. Core reaches the extension through a registered service, mirroring lib/email/service.ts, so lib/receipt-hunt never imports from @/extensions and a zero-extension build still compiles. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat(mail): connect UI and ingest, making the hunt reach into the mailbox Two halves that together make the connector usable. Ingest (lib/receipt-hunt/ingest.ts, core): fetches the attachment, files it as a document and an inbox item with source 'mail_hunt', then stages the pairing. It lives in core because it writes documents and inbox items, and an extension may never import another extension; the mail extension only ever hands over bytes. No re-matching for a hunted receipt: it was fetched WHILE SEARCHING for a specific purchase, so the pairing is known by construction. The search is a deliberately broad OR query, which is exactly why the proposal still goes to a human with the mailbox, sender and subject written on it rather than being linked automatically. Provenance goes in channel_context, never extracted_data, because retrying extraction overwrites extracted_data wholesale and the record of which mailbox a receipt came from has to survive that. A partial unique index on (company_id, channel_context->>'mail_message_id') makes re-runs and the same receipt arriving in two mailboxes idempotent, and a 23505 is treated as success rather than an error. Guards, both mutation-tested: a duplicate message costs no provider call, and an oversized attachment is skipped rather than stored. One unreadable attachment falls through to the next and never aborts a night's hunt. UI: /settings/mail lists connected mailboxes with their health, connects a new one through a user-gesture tab (opened before the await, so popup blockers do not eat it), and disconnects behind a ConfirmDialog that states the outcome up front, including that already-approved receipts stay because they belong to the bookkeeping now. Strings in sv and en; the read-only promise is spelled out on the page rather than buried in a consent screen. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(mail): renumber migrations to clear a version collision on main 20260806150000 was already taken by preserve_preset_committed_at, and woocommerce_connections plus enforce_balance_on_posted_insert landed after this branch was cut. Two files sharing a version breaks every fresh database, which only shows up on a clean setup rather than on an already-migrated one. Applied to prod under the new versions (20260807090000 / 20260807090100), so schema_migrations matches these filenames exactly. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(receipt-hunt): make the mailbox search actually able to find an underlag A provkörning against a real ledger returned the same seven unrelated messages for every purchase, all reporting no attachments. Three separate causes, each fixed and pinned: 1. `getMessageSummary` asked Gmail for `format=metadata`, which returns headers and omits `payload.parts` entirely. Every message therefore looked attachment-free, `bodyIsReceipt` was always true, and the `found.find(c => c.attachmentIds.length > 0)` guard in the hunt could never select anything: the feature could not file a single receipt. Gmail has no format that returns MIME structure without the body, so the body now comes down the wire; it is read for nothing and stored nowhere. 2. The bank's description is not a merchant name. "Lön Juli Jakob Överföring via internet" searched for "Juli" and matched most of the mailbox. Month names and payment-rail boilerplate are now stopwords. 3. Salary and tax runs are a company's largest outgoing rows, so they consumed the whole search budget hunting receipts that cannot exist. `canHaveEmailReceipt` skips them for the mail leg only. Deliberately narrow: a supplier invoice paid over bankgiro does arrive by mail, and an "Utlägg" reimbursement has a real receipt behind it. Measured on the same ledger: 22 hits, 0 with attachments, 0 ingestable -> 4 hits, all with attachments, 3 of 4 correct (Elgiganten, Sting, Anthropic). The fourth matched a Stockholm billing address, which is why every proposal still waits for a human. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat(receipt-hunt): let a model resolve merchants and pick the receipt The keyword hunt was failing for reasons regex tuning cannot reach, all measured against a real mailbox rather than assumed: - `from:anthropic.com` returns 0. Receipts arrive here by being forwarded, so the sender is the user, not the vendor. - The exact charged amount returns 0. The bank posts a converted SEK figure that appears nowhere in a USD receipt. - A date window around the purchase returns 0, while the same merchant search without one returns 10+. A forward is stamped when it was forwarded, sometimes months later. So the query now searches merchant names across the whole mailbox, and precision is restored by judgement rather than by syntax. Two model calls per run, both through forced tool use so the reply is a shape and not prose to be parsed: 1. `planMerchantGroups` resolves bank descriptors to merchants and merges repeats. Six Anthropic subscriptions become one search and one decision instead of six of each. 2. `assignReceipts` decides which mail, and which attachment on it, is the receipt for which charge, and says why in a sentence the reviewer reads. The attachment, not the message, is the unit of an underlag: a single forward routinely carries receipts for several purchases ("Fwd: Kvitton februari" has five). Migration 20260807103000 moves the dedupe key from message to message+attachment, with a backfill, because the old index would have silently blocked every receipt after the first in a forward. The model may not produce any number that reaches the ledger. It returns ids, a confidence and a reason; amounts, dates and the write stay in deterministic code. Its answer is validated, not trusted: an unknown message id, an invented filename or a low confidence drops the pairing, and any failed call proposes nothing at all. Every result still waits for a human. Measured on the same ledger: 0 receipts that could ever be filed -> 3 correct pairings (Elgiganten, Sting office invoice, Anthropic), each with a stated reason. The five remaining Anthropic charges are dated after 2026-06-15, when forwarding to the connected mailbox stopped; the model declined them correctly. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * refactor(receipt-hunt): amount first, and drop the confidence scoring Three findings from how others build this, applied. Production email search (Superhuman, Haystack 2026) reports that recall comes from loosening retrieval and letting the model filter downstream, not from tightening the query. Retrieval depth per merchant 12 -> 25, and purchases the planner cannot name a merchant for are now searched by amount alone instead of skipped: a line like "1260525758758 Europabetalning" identifies no merchant but is a real supplier payment whose invoice may carry exactly that total. Reconciliation engines weight amount far above date (Midday: 35% vs 5%) because banks post late while amounts do not drift. The Gmail query now leads with the amount and ORs the merchant, rather than dropping the amount whenever a merchant alias exists. Still an OR: a receipt billed in USD never contains the SEK figure the bank charged. The confidence score is gone entirely. Research on verbalised confidence finds it badly calibrated, clustered on round-number anchors and barely better than chance at separating a model's own right answers from its wrong ones. That matched what this ran into: the model anchored on 0.6 / 0.7 / 0.75 / 0.9, and the 0.7 threshold discarded two correct pairings. It is replaced by an observation rather than a self-assessment, whether the charged amount is actually visible in the mail, which is what a reviewer checks first and what sorts the queue. Also fixes a real defect the run exposed: the one-file-one-purchase guard only held within a merchant group, so when the planner split one landlord into "Sting" and "Kontorsplatser" both 15 000 kr charges were assigned the same invoice. A file is now claimed once per run, which is the duplicate underlag BFL forbids. Measured on the same ledger: 3 -> 5 pairings, no duplicate. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * refactor(receipt-hunt): harvest receipts, then pair them on the amount Splits the mailbox leg in two along the line of what each side can actually know. The model was being asked which purchase a mail belonged to. Deciding that needs the amount; the amount lives inside the PDF; a Gmail preview essentially never shows it. Measured over a real mailbox, every single pairing came back "belopp ej synligt": it was answering without the deciding evidence, which is why it declined five of six repeat subscriptions and why two correct pairings sat just under a threshold. Now it answers only what a subject, a sender and a preview line support: is this mail an underlag, and which attachment is it. Then the receipt is fetched, the extraction that already runs on document.uploaded reads its amount, date and vendor, and the pairing is the same deterministic amount-and-merchant match every other underlag goes through. Amount becomes decisive for real rather than as an instruction the model could not act on. The load-bearing fix is small: ingest now copies the extraction result onto the inbox item. The pool is read from invoice_inbox_items, so a hunted receipt with no extracted_data could never have matched anything, and the whole mail leg was quietly incapable of producing a pairing on amount. Consequences, all deliberate: - Harvesting runs BEFORE the pool is read, so a receipt found tonight is paired tonight rather than a night later. - One staging path instead of two. Mail-sourced proposals carry the same preview and confidence as every other, plus where they came from. - Deduped on the attachment filename, not on the message: the same invoice arrives as an original, a reminder and two forwards, and the old key filed "Invoice_13041840.pdf" four times over. - Capped at 8 receipts per merchant per run. Measured on the same ledger: 5 pairings attempted from thin evidence -> 16 real documents identified, each waiting on an amount it can be checked against. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * refactor(receipt-hunt): the model reads mail, arithmetic does the matching Collapses the mailbox leg to one model call that extracts fields, and hands every judgement back to deterministic code. Gone: resolving bank descriptors to merchant names, deciding which mail belongs to which charge, and the confidence score gating the result. Three prompts and two model calls become one, and mail-intelligence.ts drops from 450 lines to 250. What made this possible was measuring what a mail actually contains. The body was being downloaded and thrown away in favour of a 200-character snippet, and the body is where a forwarded receipt quotes its original sender and its original date. That is the purchase date, the thing whose absence forced the date window off entirely and made the old design miss five of six repeat subscriptions. It was there all along. So the model now answers only what text can support: is this an underlag, from whom, when, and for how much if the mail says so. Fields, not judgements. Everything after is arithmetic: - Retrieval is deterministic. No model decides what to search for. - Fetching is gated by worthFetching(): a stated amount is enough on its own, a vendor needs a plausible date, and a mail found by a purchase's own search is evidence in itself. That last rule is what handles a supplier the bank and the invoice name differently ("Kontorsplatser j BG" against "Stockholm Innovation & Growth AB"), which is what the deleted merchant-resolution call used to buy. - The pairing is the existing scorer, reached the same way as every other underlag: fetch, let the extraction that already runs on upload read the PDF, match on the amount. Amount is decisive in fact rather than as an instruction the model could not act on. Also adds the Swedish thousands-space amount formats to the query. Measured: the Sting invoice is findable as "15 000,00" and "15 000" and by no ungrouped form at all, so every amount search was missing them. Measured on the same ledger: 5 thin pairings -> 8 real documents, each with a vendor and a true purchase date, waiting on the amount in its own PDF. Currency is never converted to make a number agree. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(receipt-hunt): trust the bytes, not the mail, when filing an attachment Found by the first live run, which fetched nothing and reported success. Three defects, each invisible to a dry run because a dry run never downloads anything. 1. Gmail declares a forwarded PDF as application/octet-stream, and uploadDocument validates content against the declared type, so the upload was rejected: "Filinnehållet matchar inte den angivna filtypen". Every forwarded receipt with a generic MIME type would have failed this way, silently, since ingest swallows one bad attachment to protect the rest of the run. The type is now sniffed from the magic bytes, then the filename, and only then from what the mail claimed. 2. The filename was re-derived by a second full message fetch inside fetchAttachment, which came back empty and fell back to a generic "underlag.pdf", discarding the real "2332687551.pdf" the search had already reported. The known name now wins. 3. The provkörning script imported lib/init instead of calling ensureInitialized(), so document.uploaded reached no handler and nothing was ever extracted. It also used static imports, which are hoisted and ran before .env.local was read, leaving the extraction extension unable to build a Supabase client. Both are script defects, not product defects: the cron route calls ensureInitialized() at module level as the architecture requires. The script now loads the environment first and imports dynamically. Also makes the per-run fetch cap tunable (RECEIPT_HUNT_MAX_RECEIPTS) so a pilot can be held to a couple of documents, and adds --live to the script, which is the only way it writes anything. Verified end to end against a real ledger, every link exercised for the first time: two attachments fetched from Gmail, stored with their real names and types, extraction run on both, the amount copied onto the inbox item, and the deterministic matcher pairing Elgiganten 21 639,00 kr from the PDF against the -21 639 kr card purchase at 0.85, staged into Granskning as attach_document_to_transaction. The second document, a Bolagsverket filing receipt, carries no total and correctly paired with nothing. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat(receipt-hunt): sweep a whole mailbox, and stop lending one receipt twice A backfill on a real ledger, 22 documents fetched from 172 messages. Batches the extraction (25 mails per call) so a first run on an existing company can read the whole mailbox instead of the 40 mails one call can carry, and makes the per-run caps tunable (RECEIPT_HUNT_MAX_MAILS, RECEIPT_HUNT_MAX_RECEIPTS) so a pilot can be bounded. The nightly caps stay where they are: they pace the review queue, and a backlog is a different job from a nightly tick. Two defects the backfill exposed, neither reachable from a dry run: The one-receipt-one-purchase rule only held inside a single run. `spentDocumentIds` is per-invocation, so an H&M receipt was proposed against a -358 kr purchase on one pass and a -354 kr purchase on the next, and approving both would have put the same underlag on two verifikat. A live proposal now claims its document across runs, the same way it already claimed its transaction. A document reported with no filename, on a message carrying five attachments, was not an answer but a shrug: the caller fetched attachment number one and hoped. Those are dropped now. A body-only receipt, where there is nothing to choose between, still passes. Measured after the sweep: 21 of 22 documents read correctly, and the binding constraint on this ledger is no longer retrieval but currency. Ten receipts are in SEK and five of those pair on the amount; twelve are in USD or EUR, where the bank charged a converted figure that appears nowhere in the receipt, so no comparison is possible and none is attempted. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat(mail): show the provider's own mark on the mailbox settings page Someone connecting a mailbox is picking an account at a provider, and the provider's mark is how they recognise which one. A generic envelope glyph said "mail" when the question is "whose". The Google "G" already existed, drawn inline inside GoogleAuthButton for the sign-in flow. It moves to components/ui/provider-marks so there is one definition rather than two, and a Microsoft square joins it for the Graph connector. Both stay inline: no external host is contacted for an icon before anyone has agreed to anything. These are the only coloured glyphs in an achromatic interface, which is deliberate rather than an oversight. A brand mark is identity, not chrome, and Google's terms require its mark unaltered rather than tinted to match a palette. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(archive): drop the duplicate mail_connections exclusion left by the rebase Main added the table to ARCHIVE_EXCLUDED_TABLES while this branch was open, so rebasing produced the key twice and the zero-extension build failed to type check. Main's entry stays, in its alphabetical place, and keeps the sentence that answers the retention question: the grants are not räkenskapsinformation, but the receipts they find are archived as documents. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(mail): record who disconnected a mailbox, without keeping the token Raised by the compliance review: disconnect() hard-deleted the row with no trace, and which mailboxes feed underlag into the books is a control over how räkenskapsinformation is produced (BFNAR 2013:2 kap 8), so switching one off should be reconstructable years later. Written by hand rather than by the write_audit_log trigger the accounting tables use. That trigger copies the whole row into audit_log, which here would mean copying an encrypted refresh token into a second table and keeping it after the entire point of the delete was to destroy it. The sibling credential table shopify_connections omits the trigger for the same reason. Only the address and provider are recorded, pinned by a test that fails if a credential ever reaches the audit entry. The review's two other flags were checked rather than assumed: nothing purges mail_hunt documents, and categorize-core.ts:403 does carry the attached document onto the verifikat when the transaction is booked. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(mail): bound every outbound call, and stop the token widening itself Four findings from the review, each checked against the code first. Neither the Gmail API nor Google's token endpoint had a deadline. Both are awaited inside Promise.all across mailboxes, so one stalled request held the whole company's hunt open until the platform killed the run. Both now carry a 15s AbortSignal, which turns a stall into one mailbox missing from tonight's sweep. `include_granted_scopes: 'true'` let Google fold scopes this app was granted elsewhere into the token issued for a mailbox, so a grant could carry more authority than the consent screen showed. Removed, and pinned by a test asserting the parameter is absent. disconnect() ignored both statement results: a failed delete still wrote an audit entry claiming the mailbox was disconnected while the credential was live, and a failed audit insert passed silently. The delete now throws, so the entry is never written for a delete that did not happen. The audit failure is logged rather than rolled back: the two can now only diverge one way, credential gone and note missing, and recreating a credential to keep them in step would be worse than a missing note. The fifth finding is real and stays open by choice, recorded in DECISIONS.md: the cron still passes searchMail=false. A sweep of one 172-message mailbox took over 600s against a maxDuration of 300, so enabling the mailbox leg nightly would time out mid-run. That flag and RECEIPT_HUNT_COMPANY_IDS get flipped together once the per-company budget is measured. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(receipt-hunt): file each attachment under its own identity Four more findings from the review. The first is a real defect. ingestMailCandidate loops over candidate.attachmentIds, but the dedupe key, the mail_attachment_id provenance and the filename were all read from index 0. Storing the second attachment therefore recorded the first one's key and name, which mislabels the row and, because the key is unique, permanently blocks the first attachment from ever landing. Masked today only because the hunt narrows to a single attachment before calling in, so nothing in the current path exercises it. All three now come from the attachment actually being stored, and the duplicate pre-check moved inside the loop so trying a second attachment is not suppressed by the first already being filed. Mutation-tested. The per-run fetch key was the bare filename, which is not an identity: "invoice.pdf" is what half the world's billing systems attach, so a second supplier's invoice would be dropped as a duplicate of the first. Scoped by vendor as well, keeping the behaviour it was written for, one fetch for an invoice that arrives as an original, a reminder and two forwards. Adds tests/pg/mail-hunt-file-dedupe.pg.test.ts for the new unique index: five attachments from one forward all land, the same attachment is refused twice, two companies hold the same file independently, other inbox sources are untouched by the partial predicate, and the message-scoped predecessor is gone. Written against CI's Postgres; there is no local DATABASE_URL here, so CI is what exercises it. --live now refuses unless RECEIPT_HUNT_CONFIRM names the same company. The script writes to whatever .env.local points at, which for this repo is production, and a recalled command should not be able to fire it. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(test): cast the jsonb parameter so Postgres can type it pg-real could not determine the type of $3 inside jsonb_build_object. An explicit ::text is what the other pg tests do. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
d4ef4f8bc4 |
feat(import): the import theater during SIE execute (#1471)
* feat(onboarding): branch question on the journey done screen Second slice of the approved activation concept: the moment the company exists, the done screen asks "Var fanns bokföringen innan?" with provider chips (real logos), SIE file, and new-business options, plus a quiet look-around escape. Choices persist initial_setup_path (fire-and-forget) and deep-link into the existing flows: providers jump straight to the migration wizard's connect step (sieViaApi providers only; Visma/Bokio land on the provider list where the SIE-first gate lives), the SIE chip opens the upload step, new business lands on Hem with step one checked off. mode='add' keeps the plain "Öppna Accounted" button: the concept's own guard, and it avoids writing the path onto the previous company if setActiveCompany silently failed. Routing lives in a pure helper with tests; anonymous onboarding_branch_chosen funnel event follows the guarded capture pattern. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(onboarding): review triage: single-choice latch, preselect reset Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix: restore package-lock.json to main (worktree npm install mutated it) Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * feat(import): the import theater: a knowledge graph draws itself during SIE execute Third slice of the activation concept. While the SIE import commits server-side (one opaque call, up to ~5 min), the client parses the same file locally (the parser is browser-clean) and a canvas constellation builds itself: company hub, fiscal years as tree rings, account-class anchors, top accounts and recognized counterparties, with paced narration lines alongside. The final line holds with the elapsed counter until the server answers, so the theater never outruns the truth. - lib/import/theater-model.ts: pure aggregation of ParsedSIEFile into a capped display model (14 accounts, 12 counterparties, >=2 sightings, internal accounting texts skipped, counterparty attached to its counter account rather than the bank leg). Tested with fixture-string SIE per the sie-parser test pattern. - components/import/ImportTheater.tsx: ink-on-paper canvas + narration, tokens read per frame (theme/palette reactive, JourneyOrb idiom), reduced motion renders the settled graph and all lines instantly. - Wizard: client parse kicks off at execute start via dynamic import, capped at 8 MB; any failure silently leaves the existing spinner takeover, which also remains for oversized files. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |