feat(transactions): expose the bank account (cash_account_id + ledger) on listings, filter by it (#1809)

* feat(transactions): expose the bank account (cash_account_id + ledger) on listings, filter by it Customer report: neither the MCP transaction listings nor v1 REST said which bank account a transaction belongs to, so per-account reconciliation could not be driven from outside and a difference on one account was hunted on another. - gnubok_list_uncategorized_transactions: cash_account_id + cash_account_ledger (BAS account of the bank account, one lookup per page) on every row, and an optional cash_account_id filter applied to both count and page. - transactions_without_documents RPC (new migration, same signature): rows carry cash_account_id + cash_account_ledger via LEFT JOIN cash_accounts; gnubok_list_transactions_without_documents declares them. - v1 transactions list/detail: cash_account_id column; list accepts ?cash_account_id=<uuid> (400 on non-UUID). - tools/list budget bumped 59.85K -> 59.9K with the usual log entry; no property descriptions added. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(transactions): import insertCashAccount in the pg test, regenerate banking.md, validate cash_account_id Skeptic/CI findings: the new pg-real test referenced insertCashAccount without importing it; the accounted-api agent skill (banking.md) was stale after cash_account_id joined the v1 projections (apiskill:check). Also reject a ledger number passed as cash_account_id on the MCP tool with a clear message instead of a raw uuid cast error, since the ledger now sits next to the id in every row. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
2026-08-23 03:03:23 +02:00
parent d4c42fd8db
commit 6e5694fd03
9 changed files with 346 additions and 12 deletions
@@ -31,6 +31,7 @@ const TransactionDetail = z.object({
  external_id: z.string().nullable(),
  import_source: z.string().nullable(),
  reconciliation_method: z.string().nullable(),
+  cash_account_id: z.string().uuid().nullable(),
  created_at: z.string(),
  updated_at: z.string(),
 })
@@ -41,7 +42,7 @@ const TRANSACTION_DETAIL_COLUMNS =
  'id, date, description, amount, currency, amount_sek, reference, merchant_name, ' +
  'counterparty_account, journal_entry_id, invoice_id, supplier_invoice_id, ' +
  'potential_invoice_id, is_business, category, receipt_id, document_id, ' +
-  'external_id, import_source, reconciliation_method, created_at, updated_at'
+  'external_id, import_source, reconciliation_method, cash_account_id, created_at, updated_at'

 registerEndpoint({
  operation: 'transactions.get',
@@ -48,13 +48,17 @@ function makeFlexibleSupabase(byTable: Record<string, MockResult | MockResult[]>
            resolve(next)
          }
        }
-        return (..._args: unknown[]) => buildChain(table)
+        return (...args: unknown[]) => {
+          calls.push({ table, method: String(prop), args })
+          return buildChain(table)
+        }
      },
    }
    return new Proxy({}, handler)
  }
  return { from: vi.fn((table: string) => buildChain(table)) }
 }
+const calls: Array<{ table: string; method: string; args: unknown[] }> = []

 const COMPANY_ID = 'aaaaaaaa-aaaa-4aaa-8aaa-aaaaaaaaaaaa'
 const TX_ID = 'bbbbbbbb-bbbb-4bbb-8bbb-bbbbbbbbbbbb'
@@ -69,6 +73,7 @@ function makeRequest(url: string): Request {

 beforeEach(() => {
  vi.clearAllMocks()
+  calls.length = 0
  mockValidate.mockResolvedValue({
    userId: USER_ID,
    companyId: COMPANY_ID,
@@ -104,6 +109,47 @@ describe('GET /api/v1/companies/:companyId/transactions', () => {
    expect(body.meta.next_cursor).toBeUndefined()
  })

+  it('passes cash_account_id through as a filter and returns it on each row', async () => {
+    const CASH_ACCOUNT_ID = 'cccccccc-cccc-4ccc-8ccc-cccccccccccc'
+    const supabase = makeFlexibleSupabase({
+      company_members: { data: { company_id: COMPANY_ID, role: 'owner' }, error: null },
+      transactions: {
+        data: [
+          {
+            id: TX_ID, date: '2026-05-12', amount: -100, currency: 'SEK', description: 'ICA',
+            cash_account_id: CASH_ACCOUNT_ID, created_at: '2026-05-12T10:00:00Z',
+          },
+        ],
+        error: null,
+      },
+    })
+    mockServiceClient.mockReturnValue(supabase)
+
+    const res = await listTransactions(
+      makeRequest(`https://x.test/api/v1/companies/${COMPANY_ID}/transactions?cash_account_id=${CASH_ACCOUNT_ID}`),
+      { params: Promise.resolve({ companyId: COMPANY_ID }) },
+    )
+    expect(res.status).toBe(200)
+    const body = await res.json()
+    expect(body.data[0].cash_account_id).toBe(CASH_ACCOUNT_ID)
+    const eqCalls = calls.filter((c) => c.table === 'transactions' && c.method === 'eq' && c.args[0] === 'cash_account_id')
+    expect(eqCalls.map((c) => c.args[1])).toEqual([CASH_ACCOUNT_ID])
+  })
+
+  it('rejects a non-UUID cash_account_id filter with 400', async () => {
+    mockServiceClient.mockReturnValue(
+      makeFlexibleSupabase({
+        company_members: { data: { company_id: COMPANY_ID, role: 'owner' }, error: null },
+        transactions: { data: [], error: null },
+      }),
+    )
+    const res = await listTransactions(
+      makeRequest(`https://x.test/api/v1/companies/${COMPANY_ID}/transactions?cash_account_id=1930`),
+      { params: Promise.resolve({ companyId: COMPANY_ID }) },
+    )
+    expect(res.status).toBe(400)
+  })
+
  it('rejects invalid status filter with 400', async () => {
    mockServiceClient.mockReturnValue(
      makeFlexibleSupabase({
@@ -2,8 +2,9 @@
 * GET /api/v1/companies/{companyId}/transactions
 *
 * Cursor-paginated transaction list. Filters: status (booked/unbooked),
- * date range, currency, search (description ilike). Default sort:
- * (date DESC, id ASC): newest first, deterministic tie-break.
+ * date range, currency, cash_account_id (one bank account), search
+ * (description ilike). Default sort: (date DESC, id ASC): newest first,
+ * deterministic tie-break.
 */
 import { z } from 'zod'
 import { paginated } from '@/lib/api/v1/response'
@@ -30,6 +31,7 @@ const TransactionSummary = z.object({
  is_business: z.boolean().nullable(),
  category: z.string().nullable(),
  import_source: z.string().nullable(),
+  cash_account_id: z.string().uuid().nullable(),
  created_at: z.string(),
 })

@@ -40,7 +42,7 @@ const TransactionListResponse = listEnvelope(TransactionSummary)
 const TRANSACTION_SUMMARY_COLUMNS =
  'id, date, description, amount, currency, reference, merchant_name, ' +
  'journal_entry_id, invoice_id, supplier_invoice_id, is_business, category, ' +
-  'import_source, created_at'
+  'import_source, cash_account_id, created_at'

 registerEndpoint({
  operation: 'transactions.list',
@@ -103,6 +105,7 @@ export const GET = withApiV1<{ params: Promise<{ companyId: string }> }>(
        .regex(/^\d{4}-\d{2}-\d{2}$/)
        .optional(),
      search: z.string().min(1).max(200).optional(),
+      cash_account_id: z.string().uuid().optional(),
    })
    const filtersResult = FiltersSchema.safeParse({
      status: url.searchParams.get('status') ?? undefined,
@@ -110,6 +113,7 @@ export const GET = withApiV1<{ params: Promise<{ companyId: string }> }>(
      date_from: url.searchParams.get('date_from') ?? undefined,
      date_to: url.searchParams.get('date_to') ?? undefined,
      search: url.searchParams.get('search') ?? undefined,
+      cash_account_id: url.searchParams.get('cash_account_id') ?? undefined,
    })
    if (!filtersResult.success) {
      return v1ErrorResponseFromCode('VALIDATION_ERROR', ctx.log, {
@@ -142,6 +146,7 @@ export const GET = withApiV1<{ params: Promise<{ companyId: string }> }>(
    if (f.status === 'booked') query = query.not('journal_entry_id', 'is', null)
    else if (f.status === 'unbooked') query = query.is('journal_entry_id', null)
    if (f.currency) query = query.eq('currency', f.currency)
+    if (f.cash_account_id) query = query.eq('cash_account_id', f.cash_account_id)
    if (f.date_from) query = query.gte('date', f.date_from)
    if (f.date_to) query = query.lte('date', f.date_to)
    if (f.search) {
@@ -52,4 +52,48 @@ describe('gnubok_list_uncategorized_transactions', () => {
    expect(result.transactions[0].reference).toBeNull()
    expect(result.transactions[0].is_business).toBeNull()
  })
+
+  it('exposes cash_account_id and resolves the bank account ledger per row', async () => {
+    // Customer report (A4): nothing on the API said which bank account a
+    // transaction belongs to, so per-account reconciliation was impossible.
+    const rows = [
+      { id: 't-1', date: '2026-03-09', description: 'A', amount: -10, currency: 'SEK', merchant_name: null, reference: null, is_business: null, category: null, cash_account_id: 'ca-1930' },
+      { id: 't-2', date: '2026-03-08', description: 'B', amount: -20, currency: 'SEK', merchant_name: null, reference: null, is_business: null, category: null, cash_account_id: null },
+    ]
+    const { supabase, enqueue, findCalls } = createQueuedMockSupabase()
+    enqueue({ data: null, error: null, count: 2 })
+    enqueue({ data: rows, error: null })
+    enqueue({ data: [{ id: 'ca-1930', ledger_account: '1930' }], error: null })
+
+    const result = (await tool.execute({ limit: 20 }, 'company-1', 'user-1', supabase as never)) as {
+      transactions: Array<{ transaction_id: string; cash_account_id: string | null; cash_account_ledger: string | null }>
+    }
+
+    expect(result.transactions[0]).toMatchObject({ transaction_id: 't-1', cash_account_id: 'ca-1930', cash_account_ledger: '1930' })
+    expect(result.transactions[1]).toMatchObject({ transaction_id: 't-2', cash_account_id: null, cash_account_ledger: null })
+    // The ledger lookup is scoped to the company and the ids on this page.
+    const inCalls = findCalls('cash_accounts', 'in')
+    expect(inCalls).toEqual([['id', ['ca-1930']]])
+  })
+
+  it('narrows both the count and the page to one bank account when cash_account_id is given', async () => {
+    const { supabase, enqueue, findCalls } = createQueuedMockSupabase()
+    enqueue({ data: null, error: null, count: 0 })
+    enqueue({ data: [], error: null })
+
+    const CA = '550e8400-e29b-41d4-a716-446655441940'
+    await tool.execute({ limit: 20, cash_account_id: CA }, 'company-1', 'user-1', supabase as never)
+
+    const eqCalls = findCalls('transactions', 'eq').filter((args) => args[0] === 'cash_account_id')
+    expect(eqCalls).toEqual([['cash_account_id', CA], ['cash_account_id', CA]])
+    // No ledger lookup when the page is empty.
+    expect(findCalls('cash_accounts', 'in')).toHaveLength(0)
+  })
+
+  it('rejects a ledger number passed as cash_account_id with a clear message', async () => {
+    const { supabase } = createQueuedMockSupabase()
+    await expect(
+      tool.execute({ limit: 20, cash_account_id: '1930' }, 'company-1', 'user-1', supabase as never),
+    ).rejects.toThrow(/cash_account_id must be a cash account UUID/)
+  })
 })
@@ -183,9 +183,17 @@ describe('tools/list payload size guard', () => {
    //     the contract; its description and the org_number/payment_terms
    //     descriptions were trimmed to one short sentence first; headroom
    //     before the change was ~11 tokens, so even the trimmed form crossed.
+    //   * 59.85K to 59.9K with the bank account on transaction listings
+    //     (customer A4): cash_account_id + cash_account_ledger on
+    //     gnubok_list_uncategorized_transactions and
+    //     gnubok_list_transactions_without_documents, plus a cash_account_id
+    //     filter on the former, so per-account reconciliation can be driven
+    //     from outside. No property descriptions (names are the contract);
+    //     the tool description gained six words; headroom before the change
+    //     was ~50 tokens, so even the bare contract crossed by ~10.
    // Long-term answer to growth is leaning harder on gnubok_search_tools: if this
    // fires again, prefer trimming descriptions or making a tool opt-in via search
    // before bumping further.
-    expect(approxTokens).toBeLessThan(59_850)
+    expect(approxTokens).toBeLessThan(59_900)
  })
 })
@@ -4163,13 +4163,14 @@ export const tools: McpTool[] = [
  {
    name: 'gnubok_list_uncategorized_transactions',
    title: 'List Uncategorized Transactions',
-    description: 'List bank transactions with no journal entry yet, newest first. Paginated.',
+    description: 'List bank transactions with no journal entry yet, newest first. Paginated. cash_account_id narrows to one bank account.',
    inputSchema: {
      type: 'object',
      additionalProperties: false,
      properties: {
        limit: { type: 'number', description: 'Max results to return, 1-100 (default 20)' },
        offset: { type: 'number', description: 'Number of results to skip for pagination (default 0)' },
+        cash_account_id: { type: 'string' },
      },
    },
    outputSchema: paginatedSchema('transactions', {
@@ -4186,6 +4187,8 @@ export const tools: McpTool[] = [
        reference: { type: ['string', 'null'] },
        is_business: { type: ['boolean', 'null'] },
        category: { type: ['string', 'null'] },
+        cash_account_id: { type: ['string', 'null'] },
+        cash_account_ledger: { type: ['string', 'null'] },
      },
    }),
    annotations: {
@@ -4197,29 +4200,67 @@ export const tools: McpTool[] = [
    async execute(args, companyId, userId, supabase) {
      const limit = Math.min(Math.max(1, Number(args.limit) || 20), 100)
      const offset = Math.max(0, Number(args.offset) || 0)
+      const cashAccountId =
+        typeof args.cash_account_id === 'string' && args.cash_account_id.trim()
+          ? args.cash_account_id.trim()
+          : null
+      // The ledger number now sits next to the id in every row, so an agent
+      // may well pass "1930" here: fail with a clear message instead of a
+      // raw Postgres uuid cast error.
+      if (cashAccountId && !/^[0-9a-f]{8}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{12}$/i.test(cashAccountId)) {
+        throw new Error('cash_account_id must be a cash account UUID (cash_accounts.id), not a ledger account number')
+      }

      // Get total count
-      const { count: totalCount, error: countError } = await supabase
+      let countQuery = supabase
        .from('transactions')
        .select('id', { count: 'exact', head: true })
        .eq('company_id', companyId)
        .is('journal_entry_id', null)
+      if (cashAccountId) countQuery = countQuery.eq('cash_account_id', cashAccountId)
+      const { count: totalCount, error: countError } = await countQuery

      if (countError) throw new Error(`Database error: ${countError.message}`)

-      const { data, error } = await supabase
+      let listQuery = supabase
        .from('transactions')
        .select(
-          'id, date, description, amount, currency, merchant_name, reference, is_business, category'
+          'id, date, description, amount, currency, merchant_name, reference, is_business, category, cash_account_id'
        )
        .eq('company_id', companyId)
        .is('journal_entry_id', null)
+      if (cashAccountId) listQuery = listQuery.eq('cash_account_id', cashAccountId)
+      const { data, error } = await listQuery
        .order('date', { ascending: false })
        .range(offset, offset + limit - 1)

      if (error) throw new Error(`Database error: ${error.message}`)

-      const rows = (data ?? []).map((t: { id: string }) => ({ ...t, transaction_id: t.id }))
+      // Resolve the bank account's BAS ledger for the rows on this page so a
+      // per-account reconciliation can be driven from outside (customer
+      // report: the API never said which bank account a transaction belongs
+      // to). One lookup per page, only when any row carries a cash_account_id.
+      const pageRows = (data ?? []) as Array<{ id: string; cash_account_id?: string | null }>
+      const cashAccountIds = [...new Set(pageRows.map((t) => t.cash_account_id).filter((v): v is string => !!v))]
+      const ledgerByCashAccount = new Map<string, string>()
+      if (cashAccountIds.length > 0) {
+        const { data: cashRows, error: cashError } = await supabase
+          .from('cash_accounts')
+          .select('id, ledger_account')
+          .eq('company_id', companyId)
+          .in('id', cashAccountIds)
+        if (cashError) throw new Error(`Database error: ${cashError.message}`)
+        for (const c of (cashRows ?? []) as Array<{ id: string; ledger_account: string }>) {
+          ledgerByCashAccount.set(c.id, c.ledger_account)
+        }
+      }
+
+      const rows = pageRows.map((t) => ({
+        ...t,
+        transaction_id: t.id,
+        cash_account_id: t.cash_account_id ?? null,
+        cash_account_ledger: t.cash_account_id ? ledgerByCashAccount.get(t.cash_account_id) ?? null : null,
+      }))
      const total = totalCount ?? 0
      const hasMore = total > offset + rows.length

@@ -4261,6 +4302,8 @@ export const tools: McpTool[] = [
        is_business: { type: ['boolean', 'null'] },
        category: { type: ['string', 'null'] },
        journal_entry_id: { type: 'string' },
+        cash_account_id: { type: ['string', 'null'] },
+        cash_account_ledger: { type: ['string', 'null'] },
      },
    }),
    annotations: {
@@ -205,7 +205,7 @@ Cursor-paginated transaction list ordered by created_at DESC, id ASC (newest-imp
 Response `200`:
 ```ts
 {
-  data: { id: string, date: string, description: string, amount: number, currency: string, reference: string, merchant_name: string, journal_entry_id: string, invoice_id: string, supplier_invoice_id: string, is_business: boolean, category: string, import_source: string, created_at: string }[],
+  data: { id: string, date: string, description: string, amount: number, currency: string, reference: string, merchant_name: string, journal_entry_id: string, invoice_id: string, supplier_invoice_id: string, is_business: boolean, category: string, import_source: string, cash_account_id: string, created_at: string }[],
  meta: {
    request_id: string,
    api_version: string,
@@ -261,6 +261,7 @@ Response `200`:
    external_id: string,
    import_source: string,
    reconciliation_method: string,
+    cash_account_id: string,
    created_at: string,
    updated_at: string
  },
@@ -0,0 +1,141 @@
+-- Migration: transactions_without_documents exposes the bank account
+--
+-- Customer report (A4): neither MCP transaction listing nor v1 REST says
+-- which bank account a transaction belongs to, so per-account reconciliation
+-- cannot be done from outside. The predicate, ordering and paging of this
+-- RPC are unchanged; every row now also carries cash_account_id and the
+-- cash account's BAS ledger (cash_account_ledger), LEFT JOINed so rows
+-- without a backfilled cash_account_id still appear with nulls.
+--
+-- Same signature as 20260724090000 (uuid, date, integer, integer): CREATE OR
+-- REPLACE keeps the existing grants; they are restated for clarity.
+
+CREATE OR REPLACE FUNCTION public.transactions_without_documents(
+  p_company_id uuid,
+  p_since date DEFAULT NULL,
+  p_limit integer DEFAULT 20,
+  p_offset integer DEFAULT 0
+)
+RETURNS jsonb
+LANGUAGE plpgsql
+STABLE
+SECURITY DEFINER
+SET search_path TO 'public'
+AS $$
+DECLARE
+  v_jwt_role text := coalesce(nullif(current_setting('request.jwt.claims', true), '')::jsonb ->> 'role', '');
+  v_limit integer := least(greatest(coalesce(p_limit, 20), 1), 100);
+  v_offset integer := greatest(coalesce(p_offset, 0), 0);
+  v_result jsonb;
+BEGIN
+  IF v_jwt_role IN ('anon', 'authenticated') THEN
+    IF p_company_id IS NULL OR NOT EXISTS (
+      SELECT 1 FROM public.user_company_ids() AS c(id) WHERE c.id = p_company_id
+    ) THEN
+      RETURN jsonb_build_object('ok', false, 'code', 'TRANSACTIONS_WITHOUT_DOCUMENTS_FORBIDDEN');
+    END IF;
+  END IF;
+
+  WITH candidates AS (
+    SELECT
+      t.id,
+      t.date,
+      t.description,
+      t.amount,
+      t.currency,
+      t.merchant_name,
+      t.reference,
+      t.is_business,
+      t.category,
+      t.journal_entry_id,
+      t.cash_account_id,
+      ca.ledger_account AS cash_account_ledger
+    FROM transactions t
+    JOIN journal_entries je ON je.id = t.journal_entry_id
+    LEFT JOIN cash_accounts ca
+      ON ca.id = t.cash_account_id
+     AND ca.company_id = t.company_id
+    WHERE t.company_id = p_company_id
+      AND je.status = 'posted'
+      -- Same predicate as verifikat_without_documents: this surface is the
+      -- bank-driven subset, keyed on the SAME document truth
+      -- (document_attachments), never transactions.document_id.
+      AND je.source_type IN (
+        'manual',
+        'bank_transaction',
+        'supplier_invoice_registered',
+        'supplier_invoice_paid',
+        'supplier_invoice_cash_payment',
+        'import'
+      )
+      AND NOT EXISTS (
+        SELECT 1 FROM document_attachments d
+        WHERE d.journal_entry_id = je.id AND d.is_current_version = true
+      )
+      AND NOT EXISTS (
+        SELECT 1 FROM journal_entry_no_doc_required x
+        WHERE x.journal_entry_id = je.id
+      )
+      -- BFL 5 kap 7 § hänvisning till underlag (anchored docs only); see
+      -- verifikat_without_documents.
+      AND NOT EXISTS (
+        SELECT 1
+        FROM supplier_invoices si
+        JOIN document_attachments sd ON sd.id = si.document_id
+        WHERE si.company_id = p_company_id
+          AND sd.journal_entry_id IS NOT NULL
+          AND (si.registration_journal_entry_id = je.id
+            OR si.payment_journal_entry_id = je.id)
+      )
+      AND NOT EXISTS (
+        SELECT 1
+        FROM supplier_invoice_payments sip
+        JOIN supplier_invoices sip_si ON sip_si.id = sip.supplier_invoice_id
+        JOIN document_attachments sipd ON sipd.id = sip_si.document_id
+        WHERE sip.journal_entry_id = je.id
+          AND sip_si.company_id = p_company_id
+          AND sipd.journal_entry_id IS NOT NULL
+      )
+      AND (p_since IS NULL OR t.date >= p_since)
+  ),
+  total AS (
+    SELECT count(*) AS n FROM candidates
+  ),
+  page AS (
+    SELECT * FROM candidates
+    ORDER BY date DESC, id DESC
+    LIMIT v_limit OFFSET v_offset
+  )
+  SELECT jsonb_build_object(
+    'ok', true,
+    'total_count', (SELECT n FROM total),
+    'transactions', coalesce(
+      (SELECT jsonb_agg(
+         jsonb_build_object(
+           'id', p.id,
+           'transaction_id', p.id,
+           'date', p.date,
+           'description', p.description,
+           'amount', p.amount,
+           'currency', p.currency,
+           'merchant_name', p.merchant_name,
+           'reference', p.reference,
+           'is_business', p.is_business,
+           'category', p.category,
+           'journal_entry_id', p.journal_entry_id,
+           'cash_account_id', p.cash_account_id,
+           'cash_account_ledger', p.cash_account_ledger
+         )
+         ORDER BY p.date DESC, p.id DESC
+       ) FROM page p),
+      '[]'::jsonb
+    )
+  )
+  INTO v_result;
+
+  RETURN v_result;
+END;
+$$;
+
+REVOKE ALL ON FUNCTION public.transactions_without_documents(uuid, date, integer, integer) FROM PUBLIC, anon;
+GRANT EXECUTE ON FUNCTION public.transactions_without_documents(uuid, date, integer, integer) TO authenticated, service_role;
@@ -4,6 +4,7 @@ import { NEEDS_DOC_SOURCE_TYPES } from '@/lib/worklist/categories'
 import { getPool } from './setup'
 import {
  seedCompany,
+  insertCashAccount,
  insertDraftJournalEntry,
  insertPostedJournalEntry,
  insertBalancedLines,
@@ -697,3 +698,47 @@ describe('floating supplier-invoice document backfill (migration 20260727180000)
    expect(await anchorOf(doc)).toBeNull()
  })
 })
+
+describe('transactions_without_documents: bank account on each row (A4)', () => {
+  it('returns cash_account_id and the cash account ledger, null when unbackfilled', async () => {
+    const s = await seedCompany()
+    const cashAccountId = await insertCashAccount({ companyId: s.companyId, ledgerAccount: '1940' })
+    const mkJe = (n: number) =>
+      insertPostedJournalEntry({
+        userId: s.userId,
+        companyId: s.companyId,
+        fiscalPeriodId: s.fiscalPeriodId,
+        voucherNumber: n,
+        entryDate: `2026-06-${String(n).padStart(2, '0')}`,
+        description: `bank ${n}`,
+        sourceType: 'bank_transaction',
+        lines: [
+          { accountNumber: '1940', debitAmount: 100, creditAmount: 0 },
+          { accountNumber: '3001', debitAmount: 0, creditAmount: 100 },
+        ],
+      })
+    const jeWithAccount = await mkJe(1)
+    const jeWithoutAccount = await mkJe(2)
+    const txWith = await insertTransaction({
+      companyId: s.companyId,
+      userId: s.userId,
+      journalEntryId: jeWithAccount,
+      cashAccountId,
+      date: '2026-06-01',
+    })
+    const txWithout = await insertTransaction({
+      companyId: s.companyId,
+      userId: s.userId,
+      journalEntryId: jeWithoutAccount,
+      cashAccountId: null,
+      date: '2026-06-02',
+    })
+
+    const { rows } = await getPool().query<{
+      r: { ok: boolean; transactions: Array<{ id: string; cash_account_id: string | null; cash_account_ledger: string | null }> }
+    }>(`SELECT public.transactions_without_documents($1, NULL, 100, 0) AS r`, [s.companyId])
+    const byId = new Map(rows[0].r.transactions.map((t) => [t.id, t]))
+    expect(byId.get(txWith)).toMatchObject({ cash_account_id: cashAccountId, cash_account_ledger: '1940' })
+    expect(byId.get(txWithout)).toMatchObject({ cash_account_id: null, cash_account_ledger: null })
+  })
+})