Files
accounted/lib/invariants/org-number.ts
T
MattssonandClaude Fable 5.1 2303f75a7b fix(suppliers): one 10-digit org number key for matching and storage (#2405)
* fix(suppliers): one 10-digit org number key for matching and storage

Why the problem occurred: the supplier register was written in three
spellings (the form asks for XXXXXX-XXXX, the v1 API and the MCP tool stored
whatever the caller sent, the AI extractor emits bare digits) while
matchSupplierByIdentity compared raw strings with .eq(). The canonical rule
existed three times (normalizeOrgNumber, the MCP fuzzy pass's orgNumberKey,
the extractor's toOrg10) and nowhere on the path that decides a match, so
every AI-extracted invoice from a hyphen-registered supplier missed the
strongest key and fell to exact-name matching. Prod holds 1738 hyphenated
rows against 493 bare ones.

What was removed or simplified: orgNumberKey (digits only, 10 kept, last 10
of 12, no Luhn) moves into lib/invariants/org-number.ts and replaces the two
other copies. The matcher scans the company's suppliers with an org_number
and compares keys, the same shape as its vat_number branch, so rows written
before the backfill (and self-hosted instances that never run it) match too.
CreateSupplierSchema, UpdateSupplierSchema and the staged create_supplier
schema store the key; the form renders it through formatOrgNumberDisplay.
A backfill migration strips the formatting from existing rows, skipping
migration-reset source companies.

Why this and not the proposed one: the issue's third layer (CHECK plus a
unique index) would fail to create on prod, which holds 94 duplicate
(company_id, key) groups across 18 companies, one of them 124 rows under a
single placeholder-looking number; that needs a merge decision first and is
filed as #2404. Rejecting anything that is not 10 or 12 digits on write was
also dropped: 68 prod rows carry foreign registration numbers (DK, DE, NL,
FI, GB, IE, US, CZ, IT) in org_number, so Swedish-shaped input is
canonicalised and anything else is stored as typed. Luhn stays lenient on
suppliers because two rows with the same mistyped number are one supplier
and parties is Luhn-strict at promotion already.

Fixes #2391

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013yCehdxm8yUubGAmoDFZag

* fix(suppliers): key only Swedish-shaped org numbers, search and dedup through the key

Skeptic pass on the previous commit. Three refutations, all confirmed:

1. orgNumberKey took the last 10 of any 12 digits and stripped letters. A
   VAT number typed into the org field (SE556012579001, orgnr + 01) keyed to
   6012579001, another company's identity, on every write path and in the
   backfill; 26 prod rows hold exactly that shape (prefixes 55/52/87). A
   Belgian BE0123456789 lost its country letters the same way. The key now
   strips only hyphens and spaces and unprefixes 12 digits only behind
   16/18/19/20; everything else is null, stored and compared as typed. The
   migration carries the same rule.
2. The supplier list search, the v1 ?search= filter and the list column all
   used the raw stored value, so a user searching 556677-88 after the
   backfill found nothing. Both searches now compare without separators and
   the column renders XXXXXX-XXXX.
3. Storage was not canonical on every path: the CSV import and the provider
   migration orchestrator wrote as typed and keyed their re-sync dedup by
   the raw value, so a Fortnox re-sync sending 556677-8899 would have
   duplicated the now-bare row. Both write and key through orgNumberKey.

Also: the matcher scans live suppliers only, so a register holding an
archived hyphenated row next to its live replacement resolves to the live
one instead of whichever id sorts first.

Refs #2391

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013yCehdxm8yUubGAmoDFZag

* fix(suppliers): review pass: foreign numbers survive display and dedup, stub key canonical

CodeRabbit findings on PR #2405, all verified against the code:

- The supplier list rendered through formatOrgNumber, which strips letters
  and would show BE0123456789 as 012345-6789; it now uses
  formatOrgNumberDisplay, which leaves anything not Swedish-shaped alone.
- The CSV import dedup fell back to digits-only, so BE0123456789 and
  FR0123456789 collided; the fallback is now the value as typed, in both
  the parse preview and the execute route.
- The provider migration's supplier-invoice stub map was keyed by the raw
  provider value while the stored row was canonical, so 556677-8899 and
  5566778899 on two invoices produced two stubs; the key goes through
  orgMapKey like the other maps.
- v1 response examples show the stored 10-digit form; the request example
  keeps the hyphenated input.

Refs #2391

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013yCehdxm8yUubGAmoDFZag

* docs(api-skill): regenerate suppliers reference for the canonical org_number example

Refs #2391

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013yCehdxm8yUubGAmoDFZag

---------

Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
2026-09-08 10:59:20 +02:00

179 lines
7.3 KiB
TypeScript

import { luhnValidate } from '@/lib/bankgiro/luhn'
/**
* Swedish organisationsnummer / personnummer: the one place that decides what
* "a valid org number" means.
*
* ## The rule
*
* - **Canonical storage form is 10 digits, no separators** (`5560125790`).
* - Input may arrive as 10 or 12 digits, with spaces or hyphens, because that
* is what users type and what provider APIs return. Both forms normalize to
* the same 10 digits; the century prefix is dropped.
* - The last digit is a Luhn (mod-10) check digit, the structural rule
* Bolagsverket and personnummer share.
*
* ## Why this module exists
*
* Before it, seven call sites each had their own idea of the rule, and four of
* them fed Skatteverket-bound output that must agree:
*
* | Site | Old rule | Failure |
* |---|---|---|
* | `lib/skatteverket/format.ts` | strip `-` only | threw on any input containing a space |
* | `lib/salary/ku/ku10-generator.ts` | `replace('-', '')` | first hyphen only, no space handling |
* | `lib/salary/agi/xml-generator.ts` | strip non-digits | no check-digit validation |
* | `lib/bokslut/ixbrl/validate/rules.ts` | `/^\d{6}-?\d{4}$/` | rejected the 12-digit form outright |
*
* A company stored with a space or in 12-digit form could file AGI all year and
* then fail on the årsredovisning, with no way for the user to tell why. The
* rules only stay in agreement if there is exactly one of them.
*
* ## Deliberate asymmetry: normalize everywhere, Luhn only at the boundary
*
* `normalizeOrgNumber` (Luhn-checked) guards data coming *in*. The export-time
* converter `toRedovisare12` is structural only: it must not start rejecting
* numbers that are already stored and have been filing successfully, because a
* failed export at a deadline is worse than a number Skatteverket will reject
* with its own message. Tighten the intake, not the outflow.
*/
/**
* Strip the separators Swedish users and provider APIs put in org numbers.
* Does not validate: use {@link isOrgNumberShaped} or {@link normalizeOrgNumber}.
*/
export function stripOrgNumberFormatting(raw: string): string {
return raw.replace(/[\s-]/g, '')
}
/**
* True when the input is structurally an org number (10 or 12 digits after
* separators are stripped), regardless of check digit.
*/
export function isOrgNumberShaped(raw: string | null | undefined): boolean {
if (!raw) return false
const cleaned = stripOrgNumberFormatting(raw)
return /^\d{10}$/.test(cleaned) || /^\d{12}$/.test(cleaned)
}
/**
* Lenient identity key for a Swedish org number: the 10 significant digits,
* or null when the input is not org-number shaped.
*
* Strips separators (hyphens, spaces), keeps 10 digits as they are and takes
* the last 10 of a 12-digit century-prefixed form: "16" for organisations,
* "18"/"19"/"20" for the personnummer an enskild firma uses. Only those
* prefixes: a 12-digit value that starts with anything else is a Swedish VAT
* number typed into the wrong field (556012579001 = orgnr + "01"), and its
* last 10 digits are somebody else's identity. Letters are not stripped for
* the same reason: BE0123456789 is a Belgian enterprise number, not the
* Swedish 0123456789. Anything not shaped like a Swedish org number keys to
* null and is stored and compared exactly as typed.
*
* No Luhn check on purpose: this key answers "do these two strings denote
* the same counterparty", and two rows holding the same mistyped number are
* still one supplier. Use {@link normalizeOrgNumber} where a number is
* accepted into the system as valid; use this where existing values are
* compared or canonicalised.
*
* `suppliers.org_number` is stored in this form: the supplier matcher, the
* write schemas (web, v1, MCP, CSV import, provider migration) and the
* extractor's self-invoice guard all go through it, so a hyphenated register
* entry and a bare extracted number meet (#2391).
*/
export function orgNumberKey(raw: string | null | undefined): string | null {
if (!raw) return null
const cleaned = stripOrgNumberFormatting(raw)
if (/^\d{10}$/.test(cleaned)) return cleaned
if (/^(16|18|19|20)\d{10}$/.test(cleaned)) return cleaned.slice(2)
return null
}
/**
* Normalize an org number to Accounted's canonical 10-digit storage form.
*
* Accepts hyphen/space-formatted input in either of the two shapes Swedish
* users commonly type:
* - 10 digits (5560125790 or 8001011231): stored as-is
* - 12 digits (198001011231): century prefix stripped
*
* Returns null for any other length, non-digit content, or invalid Luhn check
* digit. Storing a structurally invalid org number would later be caught by
* Skatteverket SRU and any receiving SIE4 system: refusing at the boundary
* keeps Accounted's bookkeeping from accumulating under an unusable identifier.
*/
export function normalizeOrgNumber(raw: string | null | undefined): string | null {
if (!raw) return null
const cleaned = stripOrgNumberFormatting(raw)
let canonical: string
if (/^\d{10}$/.test(cleaned)) {
canonical = cleaned
} else if (/^\d{12}$/.test(cleaned)) {
canonical = cleaned.substring(2)
} else {
return null
}
return luhnValidate(canonical) ? canonical : null
}
/** True when {@link normalizeOrgNumber} accepts the input. */
export function isValidOrgNumber(raw: string | null | undefined): boolean {
return normalizeOrgNumber(raw) !== null
}
/**
* True when the input is shaped like an org number but its check digit is
* wrong. Lets a validator tell the user *which* problem they have instead of
* one undifferentiated "ogiltigt organisationsnummer".
*/
export function hasInvalidOrgNumberCheckDigit(raw: string | null | undefined): boolean {
return isOrgNumberShaped(raw) && !isValidOrgNumber(raw)
}
/**
* Format a canonical org number for display: `NNNNNN-NNNN`.
* Returns the input unchanged when it is not org-number shaped.
*/
export function formatOrgNumberDisplay(raw: string | null | undefined): string {
if (!raw) return ''
const cleaned = stripOrgNumberFormatting(raw)
const ten = /^\d{12}$/.test(cleaned) ? cleaned.substring(2) : cleaned
if (!/^\d{10}$/.test(ten)) return raw
return `${ten.substring(0, 6)}-${ten.substring(6)}`
}
/**
* Convert an org number to Skatteverket's 12-digit "redovisare" format.
*
* - Organisationsnummer (aktiebolag): prefix `16` (5020000013 -> 165020000013)
* - Personnummer (enskild firma): prefix `19` or `20` by century
* - Input already in 12-digit form passes through untouched
*
* Structural only, no check-digit validation: see the module docblock for why
* the export path stays permissive.
*
* @throws when the input is not 10 or 12 digits after separators are stripped.
*/
export function toRedovisare12(
orgNumber: string,
entityType: 'enskild_firma' | 'aktiebolag',
): string {
const clean = stripOrgNumberFormatting(orgNumber)
if (/^\d{12}$/.test(clean)) return clean
if (!/^\d{10}$/.test(clean)) {
throw new Error(`Ogiltigt organisationsnummer: ${orgNumber} (förväntar 10 eller 12 siffror)`)
}
if (entityType === 'aktiebolag') return `16${clean}`
// Enskild firma: personnummer. A two-digit year above the current one must
// belong to the previous century (someone born in 98 is 1998, not 2098).
const yearDigits = parseInt(clean.substring(0, 2), 10)
const currentTwoDigitYear = new Date().getFullYear() % 100
const prefix = yearDigits > currentTwoDigitYear ? '19' : '20'
return `${prefix}${clean}`
}