feat: add bank file import as core, move Enable Banking to extension
Replace PSD2 bank integration as the default with file-based bank import (CSV/XML), which better suits Swedish sole traders and small companies. Enable Banking is now an opt-in extension. - Phase 1: Extract generic transaction ingestion service (ingest.ts) with dedup, auto-categorization, and OCR-based invoice matching - Phase 2: Bank file parser library supporting Nordea, SEB, Swedbank, Handelsbanken CSV formats and ISO 20022 camt.053 XML - Phase 3: Database migration adding import_source, reference columns and bank_file_imports tracking table - Phase 4: Import wizard UI (5-step flow) and API routes for parse/execute - Phase 5: Move Enable Banking to extensions/enable-banking/ with commented-out loader entry for opt-in activation - Phase 6: 104 new tests (ingestion + all parser formats), fixing Nordea detection overlap and camt.053 XML tag collision bugs Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
Jakob Wennberg
co-authored by
Claude Opus 4.6
Claude Opus 4.6
parent
838dc6b8b5
commit
885f362a29
@@ -0,0 +1,972 @@
|
||||
/**
|
||||
* Comprehensive tests for the bank file parser library.
|
||||
*
|
||||
* Covers auto-detection, parsing for all Swedish bank formats (Nordea, SEB,
|
||||
* Swedbank, Handelsbanken), ISO 20022 camt.053 XML, external ID generation,
|
||||
* file hashing, stats calculation, date range extraction, and edge cases.
|
||||
*/
|
||||
|
||||
import { detectFileFormat, parseBankFile, generateExternalId, generateFileHash, getFormat, getAllFormats } from '../parser'
|
||||
import type { ParsedBankTransaction, BankFileFormatId } from '../types'
|
||||
|
||||
// ---------------------------------------------------------------------------
|
||||
// Test data — realistic CSV/XML content for each Swedish bank format
|
||||
// ---------------------------------------------------------------------------
|
||||
|
||||
const NORDEA_CSV = [
|
||||
'Datum,Transaktion,Kategori,Belopp,Saldo',
|
||||
'2024-01-15,SPOTIFY AB,,"-99,00","12 345,67"',
|
||||
'2024-01-14,ICA MAXI LINDHAGEN,,"-432,50","12 444,67"',
|
||||
'2024-01-13,LÖNEUTBETALNING,,"25 000,00","12 877,17"',
|
||||
].join('\n')
|
||||
|
||||
const NORDEA_CSV_WITH_RESERVED = [
|
||||
'Datum,Transaktion,Kategori,Belopp,Saldo',
|
||||
'2024-01-15,SPOTIFY AB,,"-99,00","12 345,67"',
|
||||
'2024-01-14,Reserverat köp CLAS OHLSON,,"-199,00","12 444,67"',
|
||||
'2024-01-13,LÖNEUTBETALNING,,"25 000,00","12 643,67"',
|
||||
].join('\n')
|
||||
|
||||
const NORDEA_CSV_SWEDISH_CHARS = [
|
||||
'Datum,Transaktion,Kategori,Belopp,Saldo',
|
||||
'2024-03-01,GÖTEBORGS HAMNCAFÉ,,"-85,00","5 000,00"',
|
||||
'2024-03-02,ÅHLENS CITY,,"-249,00","4 751,00"',
|
||||
'2024-03-03,ÄRLA GÅRD AB,,"1 200,00","5 951,00"',
|
||||
].join('\n')
|
||||
|
||||
const SEB_CSV = [
|
||||
'Bokföringsdag;Valutadag;Verifikationsnummer;Text;Belopp;Saldo',
|
||||
'2024-01-15;2024-01-15;12345;SPOTIFY AB;-99,00;12345,67',
|
||||
'2024-01-14;2024-01-14;12346;HEMKÖP FRIDHEMSPLAN;-432,50;12444,67',
|
||||
'2024-01-13;2024-01-13;12347;LÖNEUTBETALNING;25000,00;12877,17',
|
||||
].join('\n')
|
||||
|
||||
const SWEDBANK_CSV = [
|
||||
'Kontouppgifter',
|
||||
'Clearingnummer,Kontonummer,Datum,Text,Belopp,Saldo',
|
||||
'8123,12345678,2024-01-15,SPOTIFY AB,-99.00,12345.67',
|
||||
'8123,12345678,2024-01-14,ICA MAXI,-432.50,12444.67',
|
||||
'8123,12345678,2024-01-13,LÖNEUTBETALNING,25000.00,12877.17',
|
||||
].join('\n')
|
||||
|
||||
const SWEDBANK_CSV_NO_METADATA = [
|
||||
'Clearingnummer,Kontonummer,Datum,Text,Belopp,Saldo',
|
||||
'8123,12345678,2024-02-01,TELIA SVERIGE,-299.00,10000.00',
|
||||
'8123,12345678,2024-02-02,SKATTEVERKET INBETALNING,5000.00,15000.00',
|
||||
].join('\n')
|
||||
|
||||
const HANDELSBANKEN_CSV = [
|
||||
'Reskontradatum;Transaktionsdatum;Text;Belopp;Saldo',
|
||||
'2024-01-15;2024-01-15;SPOTIFY AB;-99,00;12345,67',
|
||||
'2024-01-14;2024-01-14;HEMKÖP;-432,50;12444,67',
|
||||
'2024-01-13;2024-01-13;LÖNEUTBETALNING;25000,00;12877,17',
|
||||
].join('\n')
|
||||
|
||||
const HANDELSBANKEN_CSV_WITH_PREL = [
|
||||
'Reskontradatum;Transaktionsdatum;Text;Belopp;Saldo',
|
||||
'2024-01-15;2024-01-15;SPOTIFY AB;-99,00;12345,67',
|
||||
'2024-01-14;2024-01-14;Prel kortköp CLAS OHLSON;-199,00;12444,67',
|
||||
'2024-01-13;2024-01-13;LÖNEUTBETALNING;25000,00;12643,67',
|
||||
].join('\n')
|
||||
|
||||
const CAMT053_XML = `<?xml version="1.0" encoding="UTF-8"?>
|
||||
<Document xmlns="urn:iso:std:iso:20022:tech:xsd:camt.053.001.02">
|
||||
<BkToCstmrStmt>
|
||||
<Stmt>
|
||||
<Acct><Ccy>SEK</Ccy></Acct>
|
||||
<Ntry>
|
||||
<BookgDt><Dt>2024-01-15</Dt></BookgDt>
|
||||
<Amt Ccy="SEK">99.00</Amt>
|
||||
<CdtDbtInd>DBIT</CdtDbtInd>
|
||||
<NtryRef>REF001</NtryRef>
|
||||
<NtryDtls><TxDtls>
|
||||
<RmtInf><Ustrd>SPOTIFY AB</Ustrd></RmtInf>
|
||||
</TxDtls></NtryDtls>
|
||||
</Ntry>
|
||||
<Ntry>
|
||||
<BookgDt><Dt>2024-01-14</Dt></BookgDt>
|
||||
<Amt Ccy="SEK">25000.00</Amt>
|
||||
<CdtDbtInd>CRDT</CdtDbtInd>
|
||||
<NtryRef>REF002</NtryRef>
|
||||
<NtryDtls><TxDtls>
|
||||
<RmtInf><Ustrd>LÖNEUTBETALNING</Ustrd></RmtInf>
|
||||
</TxDtls></NtryDtls>
|
||||
</Ntry>
|
||||
</Stmt>
|
||||
</BkToCstmrStmt>
|
||||
</Document>`
|
||||
|
||||
const CAMT053_XML_WITH_STRUCTURED_REF = `<?xml version="1.0" encoding="UTF-8"?>
|
||||
<Document xmlns="urn:iso:std:iso:20022:tech:xsd:camt.053.001.02">
|
||||
<BkToCstmrStmt>
|
||||
<Stmt>
|
||||
<Ntry>
|
||||
<BookgDt><Dt>2024-02-01</Dt></BookgDt>
|
||||
<Amt Ccy="SEK">1500.00</Amt>
|
||||
<CdtDbtInd>CRDT</CdtDbtInd>
|
||||
<NtryRef>REF100</NtryRef>
|
||||
<NtryDtls><TxDtls>
|
||||
<RmtInf>
|
||||
<Strd><CdtrRefInf><Ref>OCR123456789</Ref></CdtrRefInf></Strd>
|
||||
<Ustrd>Betalning faktura 1001</Ustrd>
|
||||
</RmtInf>
|
||||
</TxDtls></NtryDtls>
|
||||
</Ntry>
|
||||
</Stmt>
|
||||
</BkToCstmrStmt>
|
||||
</Document>`
|
||||
|
||||
const UNKNOWN_CSV = [
|
||||
'id,name,value,timestamp',
|
||||
'1,Widget A,100,2024-01-15T10:00:00',
|
||||
'2,Widget B,200,2024-01-16T11:00:00',
|
||||
].join('\n')
|
||||
|
||||
const EMPTY_FILE = ''
|
||||
|
||||
const HEADER_ONLY_NORDEA = 'Datum,Transaktion,Kategori,Belopp,Saldo\n'
|
||||
|
||||
// ---------------------------------------------------------------------------
|
||||
// Tests
|
||||
// ---------------------------------------------------------------------------
|
||||
|
||||
describe('detectFileFormat', () => {
|
||||
it('detects Nordea CSV from header keywords', () => {
|
||||
const format = detectFileFormat(NORDEA_CSV, 'transaktioner.csv')
|
||||
expect(format).not.toBeNull()
|
||||
expect(format!.id).toBe('nordea')
|
||||
})
|
||||
|
||||
it('detects SEB CSV from semicolon-delimited header with bokföringsdag', () => {
|
||||
const format = detectFileFormat(SEB_CSV, 'kontoutdrag.csv')
|
||||
expect(format).not.toBeNull()
|
||||
expect(format!.id).toBe('seb')
|
||||
})
|
||||
|
||||
it('detects Swedbank CSV from clearingnummer header', () => {
|
||||
const format = detectFileFormat(SWEDBANK_CSV, 'export.csv')
|
||||
expect(format).not.toBeNull()
|
||||
expect(format!.id).toBe('swedbank')
|
||||
})
|
||||
|
||||
it('detects Swedbank CSV when header is on the first line (no metadata)', () => {
|
||||
const format = detectFileFormat(SWEDBANK_CSV_NO_METADATA, 'export.csv')
|
||||
expect(format).not.toBeNull()
|
||||
expect(format!.id).toBe('swedbank')
|
||||
})
|
||||
|
||||
it('detects Handelsbanken CSV from reskontradatum/transaktionsdatum header', () => {
|
||||
const format = detectFileFormat(HANDELSBANKEN_CSV, 'handelsbanken.csv')
|
||||
expect(format).not.toBeNull()
|
||||
expect(format!.id).toBe('handelsbanken')
|
||||
})
|
||||
|
||||
it('detects camt.053 XML from namespace and .xml extension', () => {
|
||||
const format = detectFileFormat(CAMT053_XML, 'statement.xml')
|
||||
expect(format).not.toBeNull()
|
||||
expect(format!.id).toBe('camt053')
|
||||
})
|
||||
|
||||
it('detects camt.053 XML when content includes BkToCstmrStmt tag', () => {
|
||||
const xmlContent = '<?xml version="1.0"?><Document><BkToCstmrStmt></BkToCstmrStmt></Document>'
|
||||
const format = detectFileFormat(xmlContent, 'data.xml')
|
||||
expect(format).not.toBeNull()
|
||||
expect(format!.id).toBe('camt053')
|
||||
})
|
||||
|
||||
it('does not detect camt.053 without .xml extension', () => {
|
||||
// camt053 detection requires .xml extension
|
||||
const format = detectFileFormat(CAMT053_XML, 'statement.csv')
|
||||
// It should not match camt053 since extension is .csv
|
||||
// But it could match something else if the content resembles a CSV header
|
||||
// The important check is that it does NOT return camt053
|
||||
if (format) {
|
||||
expect(format.id).not.toBe('camt053')
|
||||
}
|
||||
})
|
||||
|
||||
it('returns null for unrecognized CSV content', () => {
|
||||
const format = detectFileFormat(UNKNOWN_CSV, 'data.csv')
|
||||
expect(format).toBeNull()
|
||||
})
|
||||
|
||||
it('returns null for empty content', () => {
|
||||
const format = detectFileFormat(EMPTY_FILE, 'empty.csv')
|
||||
expect(format).toBeNull()
|
||||
})
|
||||
|
||||
it('generic_csv format never auto-detects', () => {
|
||||
// Even with simple CSV content, generic should not be picked
|
||||
const simpleCSV = 'date,description,amount\n2024-01-15,Test,-100'
|
||||
const format = detectFileFormat(simpleCSV, 'test.csv')
|
||||
if (format) {
|
||||
expect(format.id).not.toBe('generic_csv')
|
||||
}
|
||||
})
|
||||
|
||||
it('is case-insensitive on header detection', () => {
|
||||
const upperNordea = 'DATUM,TRANSAKTION,KATEGORI,BELOPP,SALDO\n2024-01-15,Test,,"-100,00","5000,00"'
|
||||
const format = detectFileFormat(upperNordea, 'test.csv')
|
||||
expect(format).not.toBeNull()
|
||||
expect(format!.id).toBe('nordea')
|
||||
})
|
||||
})
|
||||
|
||||
describe('parseBankFile — Nordea format', () => {
|
||||
it('parses comma-delimited CSV with comma decimal separator', () => {
|
||||
const result = parseBankFile(NORDEA_CSV, 'nordea.csv')
|
||||
|
||||
expect(result.format).toBe('nordea')
|
||||
expect(result.format_name).toBe('Nordea')
|
||||
expect(result.transactions).toHaveLength(3)
|
||||
expect(result.issues).toHaveLength(0)
|
||||
})
|
||||
|
||||
it('correctly parses negative amounts with comma decimal and space thousands', () => {
|
||||
const result = parseBankFile(NORDEA_CSV, 'nordea.csv')
|
||||
|
||||
const spotify = result.transactions[0]
|
||||
expect(spotify.amount).toBe(-99)
|
||||
expect(spotify.description).toBe('SPOTIFY AB')
|
||||
expect(spotify.date).toBe('2024-01-15')
|
||||
expect(spotify.currency).toBe('SEK')
|
||||
|
||||
const ica = result.transactions[1]
|
||||
expect(ica.amount).toBe(-432.5)
|
||||
})
|
||||
|
||||
it('correctly parses positive amounts with space thousands separator', () => {
|
||||
const result = parseBankFile(NORDEA_CSV, 'nordea.csv')
|
||||
|
||||
const salary = result.transactions[2]
|
||||
expect(salary.amount).toBe(25000)
|
||||
expect(salary.description).toBe('LÖNEUTBETALNING')
|
||||
})
|
||||
|
||||
it('parses balance field with space thousands separator', () => {
|
||||
const result = parseBankFile(NORDEA_CSV, 'nordea.csv')
|
||||
|
||||
const spotify = result.transactions[0]
|
||||
expect(spotify.balance).toBe(12345.67)
|
||||
})
|
||||
|
||||
it('filters out "Reserverat" (pending) transactions', () => {
|
||||
const result = parseBankFile(NORDEA_CSV_WITH_RESERVED, 'nordea.csv')
|
||||
|
||||
expect(result.transactions).toHaveLength(2)
|
||||
expect(result.stats.skipped_rows).toBe(1)
|
||||
|
||||
const descriptions = result.transactions.map((t) => t.description)
|
||||
expect(descriptions).not.toContain(expect.stringContaining('Reserverat'))
|
||||
})
|
||||
|
||||
it('handles Swedish characters (a-ring, a-diaeresis, o-diaeresis)', () => {
|
||||
const result = parseBankFile(NORDEA_CSV_SWEDISH_CHARS, 'nordea.csv')
|
||||
|
||||
expect(result.transactions).toHaveLength(3)
|
||||
expect(result.transactions[0].description).toBe('GÖTEBORGS HAMNCAFÉ')
|
||||
expect(result.transactions[1].description).toBe('ÅHLENS CITY')
|
||||
expect(result.transactions[2].description).toBe('ÄRLA GÅRD AB')
|
||||
})
|
||||
|
||||
it('stores raw_line for each transaction', () => {
|
||||
const result = parseBankFile(NORDEA_CSV, 'nordea.csv')
|
||||
|
||||
result.transactions.forEach((tx) => {
|
||||
expect(tx.raw_line).toBeDefined()
|
||||
expect(tx.raw_line!.length).toBeGreaterThan(0)
|
||||
})
|
||||
})
|
||||
|
||||
it('handles header-only file with no data rows', () => {
|
||||
const result = parseBankFile(HEADER_ONLY_NORDEA, 'nordea.csv')
|
||||
|
||||
expect(result.format).toBe('nordea')
|
||||
expect(result.transactions).toHaveLength(0)
|
||||
expect(result.date_from).toBeNull()
|
||||
expect(result.date_to).toBeNull()
|
||||
expect(result.stats.parsed_rows).toBe(0)
|
||||
})
|
||||
})
|
||||
|
||||
describe('parseBankFile — SEB format', () => {
|
||||
it('parses semicolon-delimited CSV with comma decimal separator', () => {
|
||||
const result = parseBankFile(SEB_CSV, 'seb.csv')
|
||||
|
||||
expect(result.format).toBe('seb')
|
||||
expect(result.format_name).toBe('SEB')
|
||||
expect(result.transactions).toHaveLength(3)
|
||||
expect(result.issues).toHaveLength(0)
|
||||
})
|
||||
|
||||
it('correctly extracts columns using dynamic header mapping', () => {
|
||||
const result = parseBankFile(SEB_CSV, 'seb.csv')
|
||||
|
||||
const spotify = result.transactions[0]
|
||||
expect(spotify.date).toBe('2024-01-15')
|
||||
expect(spotify.description).toBe('SPOTIFY AB')
|
||||
expect(spotify.amount).toBe(-99)
|
||||
expect(spotify.balance).toBe(12345.67)
|
||||
})
|
||||
|
||||
it('parses positive income amounts correctly', () => {
|
||||
const result = parseBankFile(SEB_CSV, 'seb.csv')
|
||||
|
||||
const salary = result.transactions[2]
|
||||
expect(salary.amount).toBe(25000)
|
||||
expect(salary.description).toBe('LÖNEUTBETALNING')
|
||||
})
|
||||
|
||||
it('handles alternative SEB header names', () => {
|
||||
const altSEB = [
|
||||
'Bokforingsdatum;Valutadag;Verifikationsnummer;Text;Belopp;Saldo',
|
||||
'2024-01-15;2024-01-15;12345;TEST;-50,00;1000,00',
|
||||
].join('\n')
|
||||
|
||||
const result = parseBankFile(altSEB, 'seb_alt.csv')
|
||||
expect(result.format).toBe('seb')
|
||||
expect(result.transactions).toHaveLength(1)
|
||||
expect(result.transactions[0].amount).toBe(-50)
|
||||
})
|
||||
})
|
||||
|
||||
describe('parseBankFile — Swedbank format', () => {
|
||||
it('parses comma-delimited CSV with PERIOD decimal separator', () => {
|
||||
const result = parseBankFile(SWEDBANK_CSV, 'swedbank.csv')
|
||||
|
||||
expect(result.format).toBe('swedbank')
|
||||
expect(result.format_name).toBe('Swedbank')
|
||||
expect(result.transactions).toHaveLength(3)
|
||||
expect(result.issues).toHaveLength(0)
|
||||
})
|
||||
|
||||
it('correctly handles period decimal separator (the Swedish exception)', () => {
|
||||
const result = parseBankFile(SWEDBANK_CSV, 'swedbank.csv')
|
||||
|
||||
const spotify = result.transactions[0]
|
||||
expect(spotify.amount).toBe(-99)
|
||||
expect(spotify.balance).toBe(12345.67)
|
||||
|
||||
const ica = result.transactions[1]
|
||||
expect(ica.amount).toBe(-432.5)
|
||||
})
|
||||
|
||||
it('skips metadata line when present (first line is account info)', () => {
|
||||
const result = parseBankFile(SWEDBANK_CSV, 'swedbank.csv')
|
||||
|
||||
// With metadata line, there are 3 data rows after headerLineIdx=1
|
||||
expect(result.transactions).toHaveLength(3)
|
||||
// No transaction should have "Kontouppgifter" as description
|
||||
const descriptions = result.transactions.map((t) => t.description)
|
||||
expect(descriptions).not.toContain('Kontouppgifter')
|
||||
})
|
||||
|
||||
it('works when header is on the first line (no metadata)', () => {
|
||||
const result = parseBankFile(SWEDBANK_CSV_NO_METADATA, 'swedbank.csv')
|
||||
|
||||
expect(result.format).toBe('swedbank')
|
||||
expect(result.transactions).toHaveLength(2)
|
||||
expect(result.transactions[0].amount).toBe(-299)
|
||||
expect(result.transactions[1].amount).toBe(5000)
|
||||
})
|
||||
|
||||
it('extracts dates correctly', () => {
|
||||
const result = parseBankFile(SWEDBANK_CSV, 'swedbank.csv')
|
||||
|
||||
expect(result.transactions[0].date).toBe('2024-01-15')
|
||||
expect(result.transactions[2].date).toBe('2024-01-13')
|
||||
})
|
||||
})
|
||||
|
||||
describe('parseBankFile — Handelsbanken format', () => {
|
||||
it('parses semicolon-delimited CSV with comma decimal separator', () => {
|
||||
const result = parseBankFile(HANDELSBANKEN_CSV, 'handelsbanken.csv')
|
||||
|
||||
expect(result.format).toBe('handelsbanken')
|
||||
expect(result.format_name).toBe('Handelsbanken')
|
||||
expect(result.transactions).toHaveLength(3)
|
||||
expect(result.issues).toHaveLength(0)
|
||||
})
|
||||
|
||||
it('correctly parses amounts and balances', () => {
|
||||
const result = parseBankFile(HANDELSBANKEN_CSV, 'handelsbanken.csv')
|
||||
|
||||
expect(result.transactions[0].amount).toBe(-99)
|
||||
expect(result.transactions[0].balance).toBe(12345.67)
|
||||
expect(result.transactions[2].amount).toBe(25000)
|
||||
})
|
||||
|
||||
it('filters out "Prel" (preliminary) transactions', () => {
|
||||
const result = parseBankFile(HANDELSBANKEN_CSV_WITH_PREL, 'handelsbanken.csv')
|
||||
|
||||
expect(result.transactions).toHaveLength(2)
|
||||
expect(result.stats.skipped_rows).toBe(1)
|
||||
|
||||
const descriptions = result.transactions.map((t) => t.description)
|
||||
expect(descriptions).not.toContain(expect.stringContaining('Prel'))
|
||||
})
|
||||
|
||||
it('prefers transaktionsdatum over reskontradatum when both are present', () => {
|
||||
// Handelsbanken has both columns; transaktionsdatum should be used
|
||||
const result = parseBankFile(HANDELSBANKEN_CSV, 'handelsbanken.csv')
|
||||
|
||||
// In our test data both dates are the same, but verify it selects dates properly
|
||||
expect(result.transactions[0].date).toBe('2024-01-15')
|
||||
})
|
||||
|
||||
it('uses transaktionsdatum as the primary date field', () => {
|
||||
// Create data where reskontradatum differs from transaktionsdatum
|
||||
const diffDates = [
|
||||
'Reskontradatum;Transaktionsdatum;Text;Belopp;Saldo',
|
||||
'2024-01-16;2024-01-15;PURCHASE;-100,00;5000,00',
|
||||
].join('\n')
|
||||
|
||||
const result = parseBankFile(diffDates, 'shb.csv')
|
||||
expect(result.transactions[0].date).toBe('2024-01-15')
|
||||
})
|
||||
})
|
||||
|
||||
describe('parseBankFile — camt.053 XML format', () => {
|
||||
it('parses XML with credit and debit entries', () => {
|
||||
const result = parseBankFile(CAMT053_XML, 'statement.xml')
|
||||
|
||||
expect(result.format).toBe('camt053')
|
||||
expect(result.format_name).toBe('ISO 20022 camt.053')
|
||||
expect(result.transactions).toHaveLength(2)
|
||||
})
|
||||
|
||||
it('applies DBIT indicator as negative amount', () => {
|
||||
const result = parseBankFile(CAMT053_XML, 'statement.xml')
|
||||
|
||||
const debit = result.transactions.find((t) => t.description === 'SPOTIFY AB')
|
||||
expect(debit).toBeDefined()
|
||||
expect(debit!.amount).toBe(-99)
|
||||
})
|
||||
|
||||
it('applies CRDT indicator as positive amount', () => {
|
||||
const result = parseBankFile(CAMT053_XML, 'statement.xml')
|
||||
|
||||
const credit = result.transactions.find((t) => t.description === 'LÖNEUTBETALNING')
|
||||
expect(credit).toBeDefined()
|
||||
expect(credit!.amount).toBe(25000)
|
||||
})
|
||||
|
||||
it('extracts entry reference into raw_line for external ID generation', () => {
|
||||
const result = parseBankFile(CAMT053_XML, 'statement.xml')
|
||||
|
||||
const debit = result.transactions[0]
|
||||
expect(debit.raw_line).toBe('REF001')
|
||||
})
|
||||
|
||||
it('extracts OCR reference from structured remittance info', () => {
|
||||
const result = parseBankFile(CAMT053_XML_WITH_STRUCTURED_REF, 'statement.xml')
|
||||
|
||||
expect(result.transactions).toHaveLength(1)
|
||||
expect(result.transactions[0].reference).toBe('OCR123456789')
|
||||
})
|
||||
|
||||
it('uses unstructured remittance info as description', () => {
|
||||
const result = parseBankFile(CAMT053_XML, 'statement.xml')
|
||||
|
||||
expect(result.transactions[0].description).toBe('SPOTIFY AB')
|
||||
})
|
||||
|
||||
it('extracts currency from Amount element', () => {
|
||||
const result = parseBankFile(CAMT053_XML, 'statement.xml')
|
||||
|
||||
result.transactions.forEach((tx) => {
|
||||
expect(tx.currency).toBe('SEK')
|
||||
})
|
||||
})
|
||||
|
||||
it('handles XML with no Ntry elements', () => {
|
||||
const emptyXml = `<?xml version="1.0"?>
|
||||
<Document xmlns="urn:iso:std:iso:20022:tech:xsd:camt.053.001.02">
|
||||
<BkToCstmrStmt><Stmt></Stmt></BkToCstmrStmt></Document>`
|
||||
|
||||
const result = parseBankFile(emptyXml, 'empty.xml')
|
||||
|
||||
expect(result.format).toBe('camt053')
|
||||
expect(result.transactions).toHaveLength(0)
|
||||
expect(result.issues.length).toBeGreaterThan(0)
|
||||
expect(result.issues[0].message).toContain('No <Ntry> elements')
|
||||
})
|
||||
})
|
||||
|
||||
describe('parseBankFile — explicit format override', () => {
|
||||
it('uses the specified format instead of auto-detection', () => {
|
||||
// Force parsing Nordea content as SEB (will produce issues but should use SEB format)
|
||||
const result = parseBankFile(
|
||||
'Datum,Transaktion,Kategori,Belopp,Saldo\n2024-01-15,Test,,"-100,00","5000,00"',
|
||||
'nordea.csv',
|
||||
'seb'
|
||||
)
|
||||
|
||||
expect(result.format).toBe('seb')
|
||||
})
|
||||
|
||||
it('returns error for unknown formatId', () => {
|
||||
const result = parseBankFile(NORDEA_CSV, 'test.csv', 'unknown_format' as BankFileFormatId)
|
||||
|
||||
expect(result.format).toBe('unknown_format')
|
||||
expect(result.format_name).toBe('Unknown')
|
||||
expect(result.transactions).toHaveLength(0)
|
||||
expect(result.issues).toHaveLength(1)
|
||||
expect(result.issues[0].severity).toBe('error')
|
||||
expect(result.issues[0].message).toContain('Unknown format')
|
||||
})
|
||||
|
||||
it('returns format detection error when no format matches and no override given', () => {
|
||||
const result = parseBankFile(UNKNOWN_CSV, 'unknown.csv')
|
||||
|
||||
expect(result.format).toBe('generic_csv')
|
||||
expect(result.format_name).toBe('Unknown')
|
||||
expect(result.transactions).toHaveLength(0)
|
||||
expect(result.issues).toHaveLength(1)
|
||||
expect(result.issues[0].message).toContain('Could not auto-detect')
|
||||
})
|
||||
|
||||
it('can force generic_csv format by explicit ID', () => {
|
||||
const csvContent = '2024-01-15,Some purchase,-50.00\n2024-01-16,Income,1000.00'
|
||||
const result = parseBankFile(csvContent, 'test.csv', 'generic_csv')
|
||||
|
||||
// generic_csv uses a default mapping (date=0, description=1, amount=2)
|
||||
// But the first line is treated as header (skip_rows=1), so only second row is data
|
||||
expect(result.format).toBe('generic_csv')
|
||||
})
|
||||
})
|
||||
|
||||
describe('generateExternalId', () => {
|
||||
const baseTx: ParsedBankTransaction = {
|
||||
date: '2024-01-15',
|
||||
description: 'SPOTIFY AB',
|
||||
amount: -99,
|
||||
currency: 'SEK',
|
||||
balance: 12345.67,
|
||||
reference: null,
|
||||
counterparty: null,
|
||||
raw_line: '2024-01-15,SPOTIFY AB,,"-99,00","12 345,67"',
|
||||
}
|
||||
|
||||
it('generates SHA-256 composite key for CSV formats', () => {
|
||||
const id = generateExternalId(baseTx, 'nordea', 0)
|
||||
|
||||
expect(id).toMatch(/^nordea_[0-9a-f]{16}$/)
|
||||
})
|
||||
|
||||
it('generates different IDs for different row indices (same transaction data)', () => {
|
||||
const id1 = generateExternalId(baseTx, 'nordea', 0)
|
||||
const id2 = generateExternalId(baseTx, 'nordea', 1)
|
||||
|
||||
expect(id1).not.toBe(id2)
|
||||
})
|
||||
|
||||
it('generates different IDs for different formats (same data, same row index)', () => {
|
||||
const nordeaId = generateExternalId(baseTx, 'nordea', 0)
|
||||
const sebId = generateExternalId(baseTx, 'seb', 0)
|
||||
|
||||
expect(nordeaId).not.toBe(sebId)
|
||||
})
|
||||
|
||||
it('generates deterministic IDs for the same inputs', () => {
|
||||
const id1 = generateExternalId(baseTx, 'nordea', 0)
|
||||
const id2 = generateExternalId(baseTx, 'nordea', 0)
|
||||
|
||||
expect(id1).toBe(id2)
|
||||
})
|
||||
|
||||
it('uses entry reference for camt.053 transactions with NtryRef', () => {
|
||||
const camtTx: ParsedBankTransaction = {
|
||||
date: '2024-01-15',
|
||||
description: 'SPOTIFY AB',
|
||||
amount: -99,
|
||||
currency: 'SEK',
|
||||
raw_line: 'REF001', // NtryRef stored in raw_line
|
||||
}
|
||||
|
||||
const id = generateExternalId(camtTx, 'camt053', 0)
|
||||
|
||||
expect(id).toBe('camt053_REF001')
|
||||
})
|
||||
|
||||
it('falls back to hash for camt.053 when raw_line starts with camt053_entry_', () => {
|
||||
const camtTx: ParsedBankTransaction = {
|
||||
date: '2024-01-15',
|
||||
description: 'SPOTIFY AB',
|
||||
amount: -99,
|
||||
currency: 'SEK',
|
||||
raw_line: 'camt053_entry_0', // Auto-generated fallback reference
|
||||
}
|
||||
|
||||
const id = generateExternalId(camtTx, 'camt053', 0)
|
||||
|
||||
// Should fall through to hash-based ID since raw_line starts with 'camt053_entry_'
|
||||
expect(id).toMatch(/^camt053_[0-9a-f]{16}$/)
|
||||
})
|
||||
|
||||
it('falls back to hash for camt.053 when raw_line is undefined', () => {
|
||||
const camtTx: ParsedBankTransaction = {
|
||||
date: '2024-01-15',
|
||||
description: 'SPOTIFY AB',
|
||||
amount: -99,
|
||||
currency: 'SEK',
|
||||
}
|
||||
|
||||
const id = generateExternalId(camtTx, 'camt053', 0)
|
||||
|
||||
expect(id).toMatch(/^camt053_[0-9a-f]{16}$/)
|
||||
})
|
||||
|
||||
it('includes amount in hash so different amounts produce different IDs', () => {
|
||||
const tx1 = { ...baseTx, amount: -99 }
|
||||
const tx2 = { ...baseTx, amount: -100 }
|
||||
|
||||
const id1 = generateExternalId(tx1, 'nordea', 0)
|
||||
const id2 = generateExternalId(tx2, 'nordea', 0)
|
||||
|
||||
expect(id1).not.toBe(id2)
|
||||
})
|
||||
|
||||
it('includes description in hash so different descriptions produce different IDs', () => {
|
||||
const tx1 = { ...baseTx, description: 'SPOTIFY AB' }
|
||||
const tx2 = { ...baseTx, description: 'NETFLIX' }
|
||||
|
||||
const id1 = generateExternalId(tx1, 'nordea', 0)
|
||||
const id2 = generateExternalId(tx2, 'nordea', 0)
|
||||
|
||||
expect(id1).not.toBe(id2)
|
||||
})
|
||||
})
|
||||
|
||||
describe('generateFileHash', () => {
|
||||
it('returns a SHA-256 hex string', () => {
|
||||
const hash = generateFileHash(NORDEA_CSV)
|
||||
|
||||
expect(hash).toMatch(/^[0-9a-f]{64}$/)
|
||||
})
|
||||
|
||||
it('produces deterministic output for the same input', () => {
|
||||
const hash1 = generateFileHash(NORDEA_CSV)
|
||||
const hash2 = generateFileHash(NORDEA_CSV)
|
||||
|
||||
expect(hash1).toBe(hash2)
|
||||
})
|
||||
|
||||
it('produces different hashes for different content', () => {
|
||||
const hash1 = generateFileHash(NORDEA_CSV)
|
||||
const hash2 = generateFileHash(SEB_CSV)
|
||||
|
||||
expect(hash1).not.toBe(hash2)
|
||||
})
|
||||
|
||||
it('produces different hash even for tiny content differences', () => {
|
||||
const hash1 = generateFileHash('abc')
|
||||
const hash2 = generateFileHash('abd')
|
||||
|
||||
expect(hash1).not.toBe(hash2)
|
||||
})
|
||||
|
||||
it('handles empty string', () => {
|
||||
const hash = generateFileHash('')
|
||||
expect(hash).toMatch(/^[0-9a-f]{64}$/)
|
||||
})
|
||||
})
|
||||
|
||||
describe('stats calculation', () => {
|
||||
it('calculates total_income as sum of positive amounts (Nordea)', () => {
|
||||
const result = parseBankFile(NORDEA_CSV, 'nordea.csv')
|
||||
|
||||
expect(result.stats.total_income).toBe(25000)
|
||||
})
|
||||
|
||||
it('calculates total_expenses as sum of negative amounts (Nordea)', () => {
|
||||
const result = parseBankFile(NORDEA_CSV, 'nordea.csv')
|
||||
|
||||
// -99 + -432.5 = -531.5
|
||||
expect(result.stats.total_expenses).toBe(-531.5)
|
||||
})
|
||||
|
||||
it('calculates parsed_rows correctly', () => {
|
||||
const result = parseBankFile(NORDEA_CSV, 'nordea.csv')
|
||||
|
||||
expect(result.stats.parsed_rows).toBe(3)
|
||||
})
|
||||
|
||||
it('calculates total_rows (excluding header)', () => {
|
||||
const result = parseBankFile(NORDEA_CSV, 'nordea.csv')
|
||||
|
||||
// 4 lines total - 1 header = 3 data rows
|
||||
expect(result.stats.total_rows).toBe(3)
|
||||
})
|
||||
|
||||
it('tracks skipped_rows for reserved/preliminary transactions', () => {
|
||||
const result = parseBankFile(NORDEA_CSV_WITH_RESERVED, 'nordea.csv')
|
||||
|
||||
expect(result.stats.skipped_rows).toBe(1)
|
||||
expect(result.stats.parsed_rows).toBe(2)
|
||||
})
|
||||
|
||||
it('calculates stats correctly for SEB format', () => {
|
||||
const result = parseBankFile(SEB_CSV, 'seb.csv')
|
||||
|
||||
expect(result.stats.total_income).toBe(25000)
|
||||
expect(result.stats.total_expenses).toBe(-531.5)
|
||||
expect(result.stats.parsed_rows).toBe(3)
|
||||
expect(result.stats.skipped_rows).toBe(0)
|
||||
})
|
||||
|
||||
it('calculates stats correctly for Swedbank format', () => {
|
||||
const result = parseBankFile(SWEDBANK_CSV, 'swedbank.csv')
|
||||
|
||||
expect(result.stats.total_income).toBe(25000)
|
||||
expect(result.stats.total_expenses).toBe(-531.5)
|
||||
expect(result.stats.parsed_rows).toBe(3)
|
||||
})
|
||||
|
||||
it('calculates stats correctly for camt.053 XML', () => {
|
||||
const result = parseBankFile(CAMT053_XML, 'statement.xml')
|
||||
|
||||
expect(result.stats.total_income).toBe(25000)
|
||||
expect(result.stats.total_expenses).toBe(-99)
|
||||
expect(result.stats.parsed_rows).toBe(2)
|
||||
expect(result.stats.total_rows).toBe(2)
|
||||
})
|
||||
|
||||
it('uses Math.round(x * 100) / 100 for monetary precision', () => {
|
||||
// Create a file that would produce floating point imprecision
|
||||
const precisionCSV = [
|
||||
'Datum,Transaktion,Kategori,Belopp,Saldo',
|
||||
'2024-01-01,TX1,,"-0,10","100,00"',
|
||||
'2024-01-02,TX2,,"-0,20","99,90"',
|
||||
'2024-01-03,TX3,,"-0,30","99,70"',
|
||||
].join('\n')
|
||||
|
||||
const result = parseBankFile(precisionCSV, 'precision.csv')
|
||||
|
||||
// 0.1 + 0.2 + 0.3 = 0.6000000000000001 without rounding
|
||||
// With Math.round(x * 100) / 100, it should be -0.6
|
||||
expect(result.stats.total_expenses).toBe(-0.6)
|
||||
})
|
||||
})
|
||||
|
||||
describe('date range extraction', () => {
|
||||
it('sets date_from to the earliest date', () => {
|
||||
const result = parseBankFile(NORDEA_CSV, 'nordea.csv')
|
||||
|
||||
expect(result.date_from).toBe('2024-01-13')
|
||||
})
|
||||
|
||||
it('sets date_to to the latest date', () => {
|
||||
const result = parseBankFile(NORDEA_CSV, 'nordea.csv')
|
||||
|
||||
expect(result.date_to).toBe('2024-01-15')
|
||||
})
|
||||
|
||||
it('returns null dates for empty transaction set', () => {
|
||||
const result = parseBankFile(HEADER_ONLY_NORDEA, 'nordea.csv')
|
||||
|
||||
expect(result.date_from).toBeNull()
|
||||
expect(result.date_to).toBeNull()
|
||||
})
|
||||
|
||||
it('handles single-transaction file (date_from equals date_to)', () => {
|
||||
const singleRow = [
|
||||
'Datum,Transaktion,Kategori,Belopp,Saldo',
|
||||
'2024-06-15,ENSKILD BETALNING,,"-500,00","10 000,00"',
|
||||
].join('\n')
|
||||
|
||||
const result = parseBankFile(singleRow, 'single.csv')
|
||||
|
||||
expect(result.date_from).toBe('2024-06-15')
|
||||
expect(result.date_to).toBe('2024-06-15')
|
||||
})
|
||||
|
||||
it('calculates correct date range for camt.053', () => {
|
||||
const result = parseBankFile(CAMT053_XML, 'statement.xml')
|
||||
|
||||
expect(result.date_from).toBe('2024-01-14')
|
||||
expect(result.date_to).toBe('2024-01-15')
|
||||
})
|
||||
|
||||
it('sorts dates lexicographically (YYYY-MM-DD is naturally sortable)', () => {
|
||||
const multiMonth = [
|
||||
'Datum,Transaktion,Kategori,Belopp,Saldo',
|
||||
'2024-12-31,DEC TX,,"-10,00","1000,00"',
|
||||
'2024-01-01,JAN TX,,"-20,00","990,00"',
|
||||
'2024-06-15,JUN TX,,"-30,00","960,00"',
|
||||
].join('\n')
|
||||
|
||||
const result = parseBankFile(multiMonth, 'multimonth.csv')
|
||||
|
||||
expect(result.date_from).toBe('2024-01-01')
|
||||
expect(result.date_to).toBe('2024-12-31')
|
||||
})
|
||||
})
|
||||
|
||||
describe('empty file handling', () => {
|
||||
it('returns error result for completely empty file (no auto-detect match)', () => {
|
||||
const result = parseBankFile(EMPTY_FILE, 'empty.csv')
|
||||
|
||||
expect(result.transactions).toHaveLength(0)
|
||||
expect(result.issues.length).toBeGreaterThan(0)
|
||||
expect(result.date_from).toBeNull()
|
||||
expect(result.date_to).toBeNull()
|
||||
expect(result.stats.parsed_rows).toBe(0)
|
||||
})
|
||||
|
||||
it('returns zero transactions for file with only whitespace', () => {
|
||||
const whitespace = ' \n \n '
|
||||
const result = parseBankFile(whitespace, 'blank.csv')
|
||||
|
||||
expect(result.transactions).toHaveLength(0)
|
||||
})
|
||||
|
||||
it('returns zero transactions for Nordea header-only file', () => {
|
||||
const result = parseBankFile(HEADER_ONLY_NORDEA, 'nordea.csv')
|
||||
|
||||
expect(result.format).toBe('nordea')
|
||||
expect(result.transactions).toHaveLength(0)
|
||||
expect(result.stats.total_rows).toBe(0)
|
||||
})
|
||||
})
|
||||
|
||||
describe('getFormat and getAllFormats', () => {
|
||||
it('getFormat returns the correct format by ID', () => {
|
||||
const nordea = getFormat('nordea')
|
||||
expect(nordea).toBeDefined()
|
||||
expect(nordea!.id).toBe('nordea')
|
||||
expect(nordea!.name).toBe('Nordea')
|
||||
|
||||
const seb = getFormat('seb')
|
||||
expect(seb).toBeDefined()
|
||||
expect(seb!.id).toBe('seb')
|
||||
})
|
||||
|
||||
it('getFormat returns undefined for unknown ID', () => {
|
||||
const unknown = getFormat('nonexistent' as BankFileFormatId)
|
||||
expect(unknown).toBeUndefined()
|
||||
})
|
||||
|
||||
it('getAllFormats returns all registered formats', () => {
|
||||
const formats = getAllFormats()
|
||||
|
||||
expect(formats.length).toBeGreaterThanOrEqual(6)
|
||||
|
||||
const ids = formats.map((f) => f.id)
|
||||
expect(ids).toContain('nordea')
|
||||
expect(ids).toContain('seb')
|
||||
expect(ids).toContain('swedbank')
|
||||
expect(ids).toContain('handelsbanken')
|
||||
expect(ids).toContain('camt053')
|
||||
expect(ids).toContain('generic_csv')
|
||||
})
|
||||
|
||||
it('camt053 is listed before bank-specific CSV formats (detection priority)', () => {
|
||||
const formats = getAllFormats()
|
||||
const camtIdx = formats.findIndex((f) => f.id === 'camt053')
|
||||
const nordeaIdx = formats.findIndex((f) => f.id === 'nordea')
|
||||
|
||||
expect(camtIdx).toBeLessThan(nordeaIdx)
|
||||
})
|
||||
|
||||
it('generic_csv is listed last (manual fallback only)', () => {
|
||||
const formats = getAllFormats()
|
||||
const genericIdx = formats.findIndex((f) => f.id === 'generic_csv')
|
||||
|
||||
expect(genericIdx).toBe(formats.length - 1)
|
||||
})
|
||||
})
|
||||
|
||||
describe('edge cases and robustness', () => {
|
||||
it('handles Windows-style line endings (CRLF)', () => {
|
||||
const crlfContent = 'Datum,Transaktion,Kategori,Belopp,Saldo\r\n2024-01-15,SPOTIFY AB,,"-99,00","12 345,67"\r\n'
|
||||
|
||||
const result = parseBankFile(crlfContent, 'nordea.csv')
|
||||
|
||||
expect(result.format).toBe('nordea')
|
||||
expect(result.transactions).toHaveLength(1)
|
||||
expect(result.transactions[0].amount).toBe(-99)
|
||||
})
|
||||
|
||||
it('handles BOM (Byte Order Mark) prefix', () => {
|
||||
const bomContent = '\uFEFF' + NORDEA_CSV
|
||||
|
||||
const result = parseBankFile(bomContent, 'nordea.csv')
|
||||
|
||||
expect(result.format).toBe('nordea')
|
||||
expect(result.transactions).toHaveLength(3)
|
||||
})
|
||||
|
||||
it('handles rows with invalid dates gracefully', () => {
|
||||
const invalidDate = [
|
||||
'Bokföringsdag;Valutadag;Verifikationsnummer;Text;Belopp;Saldo',
|
||||
'not-a-date;2024-01-15;12345;SPOTIFY AB;-99,00;12345,67',
|
||||
'2024-01-14;2024-01-14;12346;VALID TX;-50,00;12395,67',
|
||||
].join('\n')
|
||||
|
||||
const result = parseBankFile(invalidDate, 'seb.csv')
|
||||
|
||||
expect(result.transactions).toHaveLength(1)
|
||||
expect(result.transactions[0].description).toBe('VALID TX')
|
||||
expect(result.issues.length).toBeGreaterThan(0)
|
||||
expect(result.stats.skipped_rows).toBe(1)
|
||||
})
|
||||
|
||||
it('handles rows with invalid amounts gracefully', () => {
|
||||
const invalidAmount = [
|
||||
'Datum,Transaktion,Kategori,Belopp,Saldo',
|
||||
'2024-01-15,SPOTIFY AB,,"abc","12 345,67"',
|
||||
'2024-01-14,VALID TX,,"-50,00","12 395,67"',
|
||||
].join('\n')
|
||||
|
||||
const result = parseBankFile(invalidAmount, 'nordea.csv')
|
||||
|
||||
expect(result.transactions).toHaveLength(1)
|
||||
expect(result.transactions[0].description).toBe('VALID TX')
|
||||
expect(result.stats.skipped_rows).toBe(1)
|
||||
})
|
||||
|
||||
it('handles trailing blank lines', () => {
|
||||
const trailing = NORDEA_CSV + '\n\n\n'
|
||||
|
||||
const result = parseBankFile(trailing, 'nordea.csv')
|
||||
|
||||
expect(result.transactions).toHaveLength(3)
|
||||
})
|
||||
|
||||
it('handles Handelsbanken CSV with only reskontradatum (no transaktionsdatum)', () => {
|
||||
const onlyReskontra = [
|
||||
'Reskontradatum;Text;Belopp;Saldo',
|
||||
'2024-01-15;SPOTIFY AB;-99,00;12345,67',
|
||||
].join('\n')
|
||||
|
||||
const format = detectFileFormat(onlyReskontra, 'shb.csv')
|
||||
expect(format).not.toBeNull()
|
||||
expect(format!.id).toBe('handelsbanken')
|
||||
})
|
||||
|
||||
it('handles large amounts without overflow', () => {
|
||||
const largeAmounts = [
|
||||
'Datum,Transaktion,Kategori,Belopp,Saldo',
|
||||
'2024-01-15,BIG TRANSFER,,"1 500 000,00","2 000 000,00"',
|
||||
].join('\n')
|
||||
|
||||
const result = parseBankFile(largeAmounts, 'nordea.csv')
|
||||
|
||||
expect(result.transactions).toHaveLength(1)
|
||||
expect(result.transactions[0].amount).toBe(1500000)
|
||||
expect(result.transactions[0].balance).toBe(2000000)
|
||||
})
|
||||
|
||||
it('handles zero amounts', () => {
|
||||
const zeroAmount = [
|
||||
'Datum,Transaktion,Kategori,Belopp,Saldo',
|
||||
'2024-01-15,FEE REVERSAL,,"0,00","5 000,00"',
|
||||
].join('\n')
|
||||
|
||||
const result = parseBankFile(zeroAmount, 'nordea.csv')
|
||||
|
||||
expect(result.transactions).toHaveLength(1)
|
||||
expect(result.transactions[0].amount).toBe(0)
|
||||
})
|
||||
})
|
||||
@@ -0,0 +1,87 @@
|
||||
/**
|
||||
* Encoding detection and conversion for Swedish bank files.
|
||||
*
|
||||
* Swedish bank exports use either UTF-8 or Windows-1252 (ISO-8859-1).
|
||||
* We detect encoding by checking for valid Swedish characters.
|
||||
*/
|
||||
|
||||
/**
|
||||
* Decode file content, handling both UTF-8 and Windows-1252 encodings.
|
||||
*
|
||||
* Strategy: Try UTF-8 first. If the result contains replacement characters
|
||||
* (U+FFFD) or garbled Swedish chars, fall back to Windows-1252.
|
||||
*/
|
||||
export function decodeFileContent(buffer: ArrayBuffer): string {
|
||||
// Try UTF-8 first
|
||||
const utf8Decoder = new TextDecoder('utf-8', { fatal: false })
|
||||
const utf8Result = utf8Decoder.decode(buffer)
|
||||
|
||||
// Check if UTF-8 decode produced valid Swedish text
|
||||
if (!hasEncodingIssues(utf8Result)) {
|
||||
return utf8Result
|
||||
}
|
||||
|
||||
// Fall back to Windows-1252 (superset of ISO-8859-1)
|
||||
const latin1Decoder = new TextDecoder('windows-1252', { fatal: false })
|
||||
return latin1Decoder.decode(buffer)
|
||||
}
|
||||
|
||||
/**
|
||||
* Decode a string that may have been incorrectly decoded as UTF-8
|
||||
* when the source was actually Windows-1252.
|
||||
*/
|
||||
export function decodeStringContent(content: string): string {
|
||||
// If the string already contains valid Swedish chars, return as-is
|
||||
if (!hasEncodingIssues(content)) {
|
||||
return content
|
||||
}
|
||||
|
||||
// Try re-encoding as Latin-1 and decoding as Windows-1252
|
||||
try {
|
||||
const bytes = new Uint8Array(content.length)
|
||||
for (let i = 0; i < content.length; i++) {
|
||||
bytes[i] = content.charCodeAt(i) & 0xff
|
||||
}
|
||||
const decoder = new TextDecoder('windows-1252', { fatal: false })
|
||||
return decoder.decode(bytes)
|
||||
} catch {
|
||||
return content
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
* Check if a string has encoding issues (garbled Swedish characters).
|
||||
*/
|
||||
function hasEncodingIssues(text: string): boolean {
|
||||
// U+FFFD = replacement character (means invalid UTF-8 byte sequences)
|
||||
if (text.includes('\uFFFD')) return true
|
||||
|
||||
// Common garbled patterns when Windows-1252 is read as UTF-8:
|
||||
// Ã¥ = å, ä = ä, ö = ö, Ã… = Å, Ä = Ä, Ö = Ö
|
||||
const garbledPatterns = ['Ã¥', 'ä', 'ö', 'Ã\u0085', 'Ã\u0084', 'Ã\u0096']
|
||||
return garbledPatterns.some((pattern) => text.includes(pattern))
|
||||
}
|
||||
|
||||
/**
|
||||
* Normalize line endings to \n
|
||||
*/
|
||||
export function normalizeLineEndings(content: string): string {
|
||||
return content.replace(/\r\n/g, '\n').replace(/\r/g, '\n')
|
||||
}
|
||||
|
||||
/**
|
||||
* Strip BOM (Byte Order Mark) from start of content
|
||||
*/
|
||||
export function stripBOM(content: string): string {
|
||||
if (content.charCodeAt(0) === 0xfeff) {
|
||||
return content.slice(1)
|
||||
}
|
||||
return content
|
||||
}
|
||||
|
||||
/**
|
||||
* Prepare file content for parsing: strip BOM, normalize line endings, handle encoding
|
||||
*/
|
||||
export function prepareContent(content: string): string {
|
||||
return normalizeLineEndings(stripBOM(decodeStringContent(content)))
|
||||
}
|
||||
@@ -0,0 +1,208 @@
|
||||
/**
|
||||
* ISO 20022 camt.053 (BankToCustomerStatement) XML parser
|
||||
*
|
||||
* This is the EU standard for bank statements, increasingly used by Swedish banks.
|
||||
* Namespace: urn:iso:std:iso:20022:tech:xsd:camt.053.001.XX (various versions)
|
||||
*
|
||||
* Key elements:
|
||||
* - <BkToCstmrStmt> — root container
|
||||
* - <Stmt> — one statement per account
|
||||
* - <Ntry> — individual transaction entries
|
||||
* - <NtryRef> / <AcctSvcrRef> — unique entry reference (external_id)
|
||||
* - <CdtDbtInd> — CRDT/DBIT indicator
|
||||
* - <RmtInf><Strd><CdtrRefInf> — structured remittance (OCR/Bankgiro reference)
|
||||
*/
|
||||
|
||||
import type { BankFileFormat, BankFileParseResult, ParsedBankTransaction, BankFileParseIssue } from '../types'
|
||||
import { prepareContent } from '../encoding'
|
||||
|
||||
export const camt053Format: BankFileFormat = {
|
||||
id: 'camt053',
|
||||
name: 'ISO 20022 camt.053',
|
||||
description: 'ISO 20022 BankToCustomerStatement (XML)',
|
||||
fileExtensions: ['.xml'],
|
||||
|
||||
detect(content: string, filename: string): boolean {
|
||||
// Check for XML with camt.053 namespace
|
||||
if (filename.toLowerCase().endsWith('.xml')) {
|
||||
const lower = content.toLowerCase()
|
||||
return (
|
||||
lower.includes('camt.053') ||
|
||||
lower.includes('bktocstmrstmt') ||
|
||||
lower.includes('banktoCustomerstatement'.toLowerCase())
|
||||
)
|
||||
}
|
||||
return false
|
||||
},
|
||||
|
||||
parse(content: string): BankFileParseResult {
|
||||
const prepared = prepareContent(content)
|
||||
|
||||
const transactions: ParsedBankTransaction[] = []
|
||||
const issues: BankFileParseIssue[] = []
|
||||
|
||||
// Simple XML parsing without external dependencies
|
||||
// Extract <Ntry> elements
|
||||
const entries = extractElements(prepared, 'Ntry')
|
||||
|
||||
if (entries.length === 0) {
|
||||
issues.push({
|
||||
row: 0,
|
||||
message: 'No <Ntry> elements found in camt.053 file',
|
||||
severity: 'error',
|
||||
})
|
||||
}
|
||||
|
||||
// Try to extract currency from statement level
|
||||
const stmtCcy = extractTextContent(prepared, 'Ccy') || 'SEK'
|
||||
|
||||
for (let i = 0; i < entries.length; i++) {
|
||||
const entry = entries[i]
|
||||
|
||||
try {
|
||||
// Date: <BookgDt><Dt> or <ValDt><Dt>
|
||||
const bookingDate = extractTextContent(entry, 'BookgDt>.*?<Dt') ||
|
||||
extractNestedText(entry, 'BookgDt', 'Dt')
|
||||
const valueDate = extractTextContent(entry, 'ValDt>.*?<Dt') ||
|
||||
extractNestedText(entry, 'ValDt', 'Dt')
|
||||
const date = bookingDate || valueDate
|
||||
|
||||
if (!date || !/^\d{4}-\d{2}-\d{2}$/.test(date)) {
|
||||
issues.push({ row: i + 1, message: `Invalid or missing date in entry ${i}`, severity: 'warning' })
|
||||
continue
|
||||
}
|
||||
|
||||
// Amount: <Amt Ccy="SEK">1234.56</Amt>
|
||||
const amountMatch = entry.match(/<Amt[^>]*>([^<]+)<\/Amt>/i)
|
||||
const amountStr = amountMatch?.[1]
|
||||
const currencyMatch = entry.match(/<Amt[^>]*Ccy="([^"]+)"[^>]*>/i)
|
||||
const currency = currencyMatch?.[1] || stmtCcy
|
||||
|
||||
if (!amountStr) {
|
||||
issues.push({ row: i + 1, message: `Missing amount in entry ${i}`, severity: 'warning' })
|
||||
continue
|
||||
}
|
||||
|
||||
let amount = parseFloat(amountStr)
|
||||
if (isNaN(amount)) {
|
||||
issues.push({ row: i + 1, message: `Invalid amount: ${amountStr}`, severity: 'warning' })
|
||||
continue
|
||||
}
|
||||
|
||||
// Credit/Debit indicator: <CdtDbtInd>CRDT</CdtDbtInd> or DBIT
|
||||
const cdtDbtInd = extractTextContent(entry, 'CdtDbtInd')
|
||||
if (cdtDbtInd === 'DBIT') {
|
||||
amount = -Math.abs(amount)
|
||||
} else {
|
||||
amount = Math.abs(amount)
|
||||
}
|
||||
|
||||
// Description: <AddtlNtryInf> or <RmtInf><Ustrd>
|
||||
const additionalInfo = extractTextContent(entry, 'AddtlNtryInf')
|
||||
const unstructuredRemittance = extractTextContent(entry, 'Ustrd')
|
||||
const description = additionalInfo || unstructuredRemittance || 'Unknown'
|
||||
|
||||
// Reference: Try structured remittance info first, then entry reference
|
||||
const structuredRef = extractTextContent(entry, 'Ref') // Inside <CdtrRefInf><Ref>
|
||||
const entryRef = extractTextContent(entry, 'NtryRef')
|
||||
const acctSvcrRef = extractTextContent(entry, 'AcctSvcrRef')
|
||||
const reference = structuredRef || null
|
||||
|
||||
// Counterparty
|
||||
const creditorName = extractTextContent(entry, 'CdtrNm') ||
|
||||
extractNestedText(entry, 'Cdtr', 'Nm')
|
||||
const debtorName = extractTextContent(entry, 'DbtrNm') ||
|
||||
extractNestedText(entry, 'Dbtr', 'Nm')
|
||||
const counterparty = cdtDbtInd === 'DBIT' ? creditorName : debtorName
|
||||
|
||||
// Balance after entry
|
||||
const balanceStr = extractTextContent(entry, 'ClsgAvlblAmt') ||
|
||||
extractTextContent(entry, 'ClsgBookdAmt')
|
||||
const balance = balanceStr ? parseFloat(balanceStr) : null
|
||||
|
||||
transactions.push({
|
||||
date,
|
||||
description: description.trim(),
|
||||
amount: Math.round(amount * 100) / 100,
|
||||
currency,
|
||||
balance: balance !== null && !isNaN(balance) ? balance : null,
|
||||
reference,
|
||||
counterparty: counterparty?.trim() || null,
|
||||
raw_line: entryRef || acctSvcrRef || `camt053_entry_${i}`,
|
||||
})
|
||||
} catch (err) {
|
||||
issues.push({
|
||||
row: i + 1,
|
||||
message: `Error parsing entry ${i}: ${err instanceof Error ? err.message : 'Unknown'}`,
|
||||
severity: 'warning',
|
||||
})
|
||||
}
|
||||
}
|
||||
|
||||
const dates = transactions.map((t) => t.date).sort()
|
||||
|
||||
return {
|
||||
format: 'camt053',
|
||||
format_name: 'ISO 20022 camt.053',
|
||||
transactions,
|
||||
date_from: dates[0] || null,
|
||||
date_to: dates[dates.length - 1] || null,
|
||||
issues,
|
||||
stats: {
|
||||
total_rows: entries.length,
|
||||
parsed_rows: transactions.length,
|
||||
skipped_rows: entries.length - transactions.length,
|
||||
total_income: Math.round(transactions.filter((t) => t.amount > 0).reduce((s, t) => s + t.amount, 0) * 100) / 100,
|
||||
total_expenses: Math.round(transactions.filter((t) => t.amount < 0).reduce((s, t) => s + t.amount, 0) * 100) / 100,
|
||||
},
|
||||
}
|
||||
},
|
||||
}
|
||||
|
||||
/**
|
||||
* Extract all occurrences of a named XML element (simple parser, no dependencies)
|
||||
*/
|
||||
function extractElements(xml: string, tagName: string): string[] {
|
||||
const elements: string[] = []
|
||||
// Match exact tag name (followed by > or whitespace, not a longer tag name)
|
||||
const regex = new RegExp(`<${tagName}(?=[\\s>/])`, 'gi')
|
||||
let match
|
||||
|
||||
while ((match = regex.exec(xml)) !== null) {
|
||||
const startIdx = match.index
|
||||
// Find matching closing tag
|
||||
const closeTag = `</${tagName}>`
|
||||
const closeIdx = xml.indexOf(closeTag, startIdx + match[0].length)
|
||||
|
||||
if (closeIdx === -1) continue
|
||||
|
||||
elements.push(xml.substring(startIdx, closeIdx + closeTag.length))
|
||||
|
||||
// Advance regex past this element to avoid re-matching inside it
|
||||
regex.lastIndex = closeIdx + closeTag.length
|
||||
}
|
||||
|
||||
return elements
|
||||
}
|
||||
|
||||
/**
|
||||
* Extract text content of the first occurrence of a tag
|
||||
*/
|
||||
function extractTextContent(xml: string, tagName: string): string | null {
|
||||
const regex = new RegExp(`<${tagName}[^>]*>([^<]+)<`, 'i')
|
||||
const match = xml.match(regex)
|
||||
return match?.[1]?.trim() || null
|
||||
}
|
||||
|
||||
/**
|
||||
* Extract text content of a nested tag within a parent tag
|
||||
*/
|
||||
function extractNestedText(xml: string, parentTag: string, childTag: string): string | null {
|
||||
const parentRegex = new RegExp(`<${parentTag}[^>]*>([\\s\\S]*?)<\\/${parentTag}>`, 'i')
|
||||
const parentMatch = xml.match(parentRegex)
|
||||
if (!parentMatch) return null
|
||||
|
||||
const childRegex = new RegExp(`<${childTag}[^>]*>([^<]+)<`, 'i')
|
||||
const childMatch = parentMatch[1].match(childRegex)
|
||||
return childMatch?.[1]?.trim() || null
|
||||
}
|
||||
@@ -0,0 +1,161 @@
|
||||
/**
|
||||
* Generic CSV format parser
|
||||
*
|
||||
* Fallback parser that requires the user to map columns manually.
|
||||
* Supports configurable delimiter, decimal separator, and column mapping.
|
||||
*/
|
||||
|
||||
import type { BankFileFormat, BankFileParseResult, ParsedBankTransaction, BankFileParseIssue, GenericCSVColumnMapping } from '../types'
|
||||
import { prepareContent } from '../encoding'
|
||||
import { parseCSVLine } from './nordea'
|
||||
|
||||
/**
|
||||
* Parse a generic CSV with user-provided column mapping
|
||||
*/
|
||||
export function parseGenericCSV(
|
||||
content: string,
|
||||
mapping: GenericCSVColumnMapping
|
||||
): BankFileParseResult {
|
||||
const prepared = prepareContent(content)
|
||||
const lines = prepared.split('\n').filter((line) => line.trim() !== '')
|
||||
|
||||
const transactions: ParsedBankTransaction[] = []
|
||||
const issues: BankFileParseIssue[] = []
|
||||
let skippedRows = 0
|
||||
|
||||
// Skip configured number of header/metadata rows
|
||||
const startRow = mapping.skip_rows
|
||||
|
||||
for (let i = startRow; i < lines.length; i++) {
|
||||
const line = lines[i].trim()
|
||||
if (!line) continue
|
||||
|
||||
const fields = parseCSVLine(line, mapping.delimiter).map((f) =>
|
||||
f.trim().replace(/^"|"$/g, '')
|
||||
)
|
||||
|
||||
const dateStr = fields[mapping.date]
|
||||
const description = fields[mapping.description] || 'Unknown'
|
||||
const amountStr = fields[mapping.amount]
|
||||
const referenceStr = mapping.reference !== undefined ? fields[mapping.reference] : undefined
|
||||
const counterpartyStr = mapping.counterparty !== undefined ? fields[mapping.counterparty] : undefined
|
||||
const balanceStr = mapping.balance !== undefined ? fields[mapping.balance] : undefined
|
||||
|
||||
if (!dateStr || !amountStr) {
|
||||
skippedRows++
|
||||
continue
|
||||
}
|
||||
|
||||
// Parse amount based on configured decimal separator
|
||||
let amount: number
|
||||
if (mapping.decimal_separator === ',') {
|
||||
amount = parseFloat(amountStr.replace(/\s/g, '').replace(',', '.'))
|
||||
} else {
|
||||
amount = parseFloat(amountStr.replace(/\s/g, ''))
|
||||
}
|
||||
|
||||
if (isNaN(amount)) {
|
||||
issues.push({ row: i + 1, message: `Invalid amount: ${amountStr}`, severity: 'warning' })
|
||||
skippedRows++
|
||||
continue
|
||||
}
|
||||
|
||||
// Parse date - expect YYYY-MM-DD
|
||||
const date = dateStr.trim()
|
||||
if (!/^\d{4}-\d{2}-\d{2}$/.test(date)) {
|
||||
issues.push({ row: i + 1, message: `Invalid date format: ${date} (expected YYYY-MM-DD)`, severity: 'warning' })
|
||||
skippedRows++
|
||||
continue
|
||||
}
|
||||
|
||||
let balance: number | null = null
|
||||
if (balanceStr) {
|
||||
if (mapping.decimal_separator === ',') {
|
||||
balance = parseFloat(balanceStr.replace(/\s/g, '').replace(',', '.'))
|
||||
} else {
|
||||
balance = parseFloat(balanceStr.replace(/\s/g, ''))
|
||||
}
|
||||
if (isNaN(balance)) balance = null
|
||||
}
|
||||
|
||||
transactions.push({
|
||||
date,
|
||||
description: description.trim(),
|
||||
amount,
|
||||
currency: 'SEK',
|
||||
balance,
|
||||
reference: referenceStr?.trim() || null,
|
||||
counterparty: counterpartyStr?.trim() || null,
|
||||
raw_line: line,
|
||||
})
|
||||
}
|
||||
|
||||
const dates = transactions.map((t) => t.date).sort()
|
||||
|
||||
return {
|
||||
format: 'generic_csv',
|
||||
format_name: 'CSV (manuell mappning)',
|
||||
transactions,
|
||||
date_from: dates[0] || null,
|
||||
date_to: dates[dates.length - 1] || null,
|
||||
issues,
|
||||
stats: {
|
||||
total_rows: lines.length - startRow,
|
||||
parsed_rows: transactions.length,
|
||||
skipped_rows: skippedRows,
|
||||
total_income: Math.round(transactions.filter((t) => t.amount > 0).reduce((s, t) => s + t.amount, 0) * 100) / 100,
|
||||
total_expenses: Math.round(transactions.filter((t) => t.amount < 0).reduce((s, t) => s + t.amount, 0) * 100) / 100,
|
||||
},
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
* Get column headers from a CSV file for the mapping UI
|
||||
*/
|
||||
export function getCSVHeaders(content: string, delimiter: string = ','): string[] {
|
||||
const prepared = prepareContent(content)
|
||||
const firstLine = prepared.split('\n')[0] || ''
|
||||
return parseCSVLine(firstLine, delimiter).map((h) => h.trim().replace(/^"|"$/g, ''))
|
||||
}
|
||||
|
||||
/**
|
||||
* Get a preview of the first few rows of a CSV file
|
||||
*/
|
||||
export function getCSVPreview(content: string, delimiter: string = ',', rows: number = 5): string[][] {
|
||||
const prepared = prepareContent(content)
|
||||
const lines = prepared.split('\n').filter((line) => line.trim() !== '')
|
||||
|
||||
return lines.slice(0, rows).map((line) =>
|
||||
parseCSVLine(line, delimiter).map((f) => f.trim().replace(/^"|"$/g, ''))
|
||||
)
|
||||
}
|
||||
|
||||
/**
|
||||
* Generic CSV format definition (used for format detection)
|
||||
* Always returns false for detect() since it's a fallback requiring user mapping
|
||||
*/
|
||||
export const genericCSVFormat: BankFileFormat = {
|
||||
id: 'generic_csv',
|
||||
name: 'CSV (manuell mappning)',
|
||||
description: 'Generisk CSV-fil med manuell kolumnmappning',
|
||||
fileExtensions: ['.csv', '.txt'],
|
||||
|
||||
detect(_content: string, _filename: string): boolean {
|
||||
// Generic CSV never auto-detects — it's the manual fallback
|
||||
return false
|
||||
},
|
||||
|
||||
parse(content: string): BankFileParseResult {
|
||||
// Default mapping for a basic CSV: date, description, amount
|
||||
const defaultMapping: GenericCSVColumnMapping = {
|
||||
date: 0,
|
||||
description: 1,
|
||||
amount: 2,
|
||||
delimiter: ',',
|
||||
decimal_separator: ',',
|
||||
skip_rows: 1,
|
||||
date_format: 'YYYY-MM-DD',
|
||||
}
|
||||
return parseGenericCSV(content, defaultMapping)
|
||||
},
|
||||
}
|
||||
@@ -0,0 +1,143 @@
|
||||
/**
|
||||
* Handelsbanken CSV format parser
|
||||
*
|
||||
* Format: Semicolon-delimited, comma decimal separator
|
||||
* Columns: Reskontradatum, Transaktionsdatum, Text, Belopp, Saldo
|
||||
* Date format: YYYY-MM-DD
|
||||
* Encoding: UTF-8 or Windows-1252
|
||||
*
|
||||
* Notes:
|
||||
* - Filter rows with "Prel" prefix (preliminary/pending transactions)
|
||||
*/
|
||||
|
||||
import type { BankFileFormat, BankFileParseResult, ParsedBankTransaction, BankFileParseIssue } from '../types'
|
||||
import { prepareContent } from '../encoding'
|
||||
|
||||
function parseCommaDecimal(value: string): number {
|
||||
const cleaned = value.replace(/\s/g, '').replace(',', '.')
|
||||
return parseFloat(cleaned)
|
||||
}
|
||||
|
||||
export const handelsbankenFormat: BankFileFormat = {
|
||||
id: 'handelsbanken',
|
||||
name: 'Handelsbanken',
|
||||
description: 'Handelsbanken CSV (semicolon-delimited)',
|
||||
fileExtensions: ['.csv', '.txt'],
|
||||
|
||||
detect(content: string, _filename: string): boolean {
|
||||
const prepared = prepareContent(content)
|
||||
const firstLine = prepared.split('\n')[0]?.toLowerCase() || ''
|
||||
return (
|
||||
firstLine.includes(';') &&
|
||||
(firstLine.includes('reskontradatum') || firstLine.includes('transaktionsdatum'))
|
||||
)
|
||||
},
|
||||
|
||||
parse(content: string): BankFileParseResult {
|
||||
const prepared = prepareContent(content)
|
||||
const lines = prepared.split('\n').filter((line) => line.trim() !== '')
|
||||
|
||||
const transactions: ParsedBankTransaction[] = []
|
||||
const issues: BankFileParseIssue[] = []
|
||||
let skippedRows = 0
|
||||
|
||||
// Parse header
|
||||
const headerLine = lines[0] || ''
|
||||
const headers = headerLine.split(';').map((h) => h.trim().toLowerCase().replace(/"/g, ''))
|
||||
|
||||
const dateIdx = headers.findIndex(
|
||||
(h) => h.includes('reskontradatum') || h.includes('transaktionsdatum')
|
||||
)
|
||||
const txDateIdx = headers.findIndex((h) => h.includes('transaktionsdatum'))
|
||||
const descIdx = headers.findIndex((h) => h === 'text' || h.includes('beskrivning'))
|
||||
const amountIdx = headers.findIndex((h) => h.includes('belopp'))
|
||||
const balanceIdx = headers.findIndex((h) => h.includes('saldo'))
|
||||
|
||||
if (dateIdx === -1 || amountIdx === -1) {
|
||||
issues.push({
|
||||
row: 1,
|
||||
message: 'Could not identify required columns',
|
||||
severity: 'error',
|
||||
})
|
||||
return {
|
||||
format: 'handelsbanken',
|
||||
format_name: 'Handelsbanken',
|
||||
transactions: [],
|
||||
date_from: null,
|
||||
date_to: null,
|
||||
issues,
|
||||
stats: { total_rows: 0, parsed_rows: 0, skipped_rows: 0, total_income: 0, total_expenses: 0 },
|
||||
}
|
||||
}
|
||||
|
||||
// Prefer transaktionsdatum over reskontradatum if available
|
||||
const primaryDateIdx = txDateIdx >= 0 ? txDateIdx : dateIdx
|
||||
|
||||
for (let i = 1; i < lines.length; i++) {
|
||||
const line = lines[i].trim()
|
||||
if (!line) continue
|
||||
|
||||
const fields = line.split(';').map((f) => f.trim().replace(/^"|"$/g, ''))
|
||||
|
||||
const date = fields[primaryDateIdx]
|
||||
const description = descIdx >= 0 ? fields[descIdx] : 'Unknown'
|
||||
const amountStr = fields[amountIdx]
|
||||
const balanceStr = balanceIdx >= 0 ? fields[balanceIdx] : undefined
|
||||
|
||||
// Skip preliminary transactions
|
||||
if (description?.toLowerCase().startsWith('prel')) {
|
||||
skippedRows++
|
||||
continue
|
||||
}
|
||||
|
||||
if (!date || !amountStr) {
|
||||
skippedRows++
|
||||
continue
|
||||
}
|
||||
|
||||
const amount = parseCommaDecimal(amountStr)
|
||||
if (isNaN(amount)) {
|
||||
issues.push({ row: i + 1, message: `Invalid amount: ${amountStr}`, severity: 'warning' })
|
||||
skippedRows++
|
||||
continue
|
||||
}
|
||||
|
||||
if (!/^\d{4}-\d{2}-\d{2}$/.test(date)) {
|
||||
issues.push({ row: i + 1, message: `Invalid date: ${date}`, severity: 'warning' })
|
||||
skippedRows++
|
||||
continue
|
||||
}
|
||||
|
||||
const balance = balanceStr ? parseCommaDecimal(balanceStr) : null
|
||||
|
||||
transactions.push({
|
||||
date,
|
||||
description: (description || 'Unknown').trim(),
|
||||
amount,
|
||||
currency: 'SEK',
|
||||
balance: isNaN(balance as number) ? null : balance,
|
||||
reference: null,
|
||||
counterparty: null,
|
||||
raw_line: line,
|
||||
})
|
||||
}
|
||||
|
||||
const dates = transactions.map((t) => t.date).sort()
|
||||
|
||||
return {
|
||||
format: 'handelsbanken',
|
||||
format_name: 'Handelsbanken',
|
||||
transactions,
|
||||
date_from: dates[0] || null,
|
||||
date_to: dates[dates.length - 1] || null,
|
||||
issues,
|
||||
stats: {
|
||||
total_rows: lines.length - 1,
|
||||
parsed_rows: transactions.length,
|
||||
skipped_rows: skippedRows,
|
||||
total_income: Math.round(transactions.filter((t) => t.amount > 0).reduce((s, t) => s + t.amount, 0) * 100) / 100,
|
||||
total_expenses: Math.round(transactions.filter((t) => t.amount < 0).reduce((s, t) => s + t.amount, 0) * 100) / 100,
|
||||
},
|
||||
}
|
||||
},
|
||||
}
|
||||
@@ -0,0 +1,151 @@
|
||||
/**
|
||||
* Nordea CSV format parser
|
||||
*
|
||||
* Format: Comma-delimited, comma decimal separator
|
||||
* Columns: Datum, Transaktion, Kategori, Belopp, Saldo
|
||||
* Date format: YYYY-MM-DD
|
||||
* Encoding: UTF-8 or Windows-1252
|
||||
*
|
||||
* Notes:
|
||||
* - Skip rows with "Reserverat" in Transaktion (pending transactions)
|
||||
* - Skip trailing blank lines
|
||||
*/
|
||||
|
||||
import type { BankFileFormat, BankFileParseResult, ParsedBankTransaction, BankFileParseIssue } from '../types'
|
||||
import { prepareContent } from '../encoding'
|
||||
|
||||
function parseCommaDecimal(value: string): number {
|
||||
// Swedish format: "1 234,56" or "-1 234,56"
|
||||
const cleaned = value.replace(/\s/g, '').replace(',', '.')
|
||||
return parseFloat(cleaned)
|
||||
}
|
||||
|
||||
export const nordeaFormat: BankFileFormat = {
|
||||
id: 'nordea',
|
||||
name: 'Nordea',
|
||||
description: 'Nordea CSV (Datum, Transaktion, Kategori, Belopp, Saldo)',
|
||||
fileExtensions: ['.csv', '.txt'],
|
||||
|
||||
detect(content: string, _filename: string): boolean {
|
||||
const prepared = prepareContent(content)
|
||||
const firstLine = prepared.split('\n')[0]?.toLowerCase() || ''
|
||||
// Nordea header: comma-delimited with "datum", "transaktion", "belopp"
|
||||
// Must NOT contain semicolons (that would be SEB or Handelsbanken)
|
||||
return (
|
||||
!firstLine.includes(';') &&
|
||||
firstLine.includes('datum') &&
|
||||
firstLine.includes('transaktion') &&
|
||||
firstLine.includes('belopp')
|
||||
)
|
||||
},
|
||||
|
||||
parse(content: string): BankFileParseResult {
|
||||
const prepared = prepareContent(content)
|
||||
const lines = prepared.split('\n').filter((line) => line.trim() !== '')
|
||||
|
||||
const transactions: ParsedBankTransaction[] = []
|
||||
const issues: BankFileParseIssue[] = []
|
||||
let skippedRows = 0
|
||||
|
||||
// Skip header row
|
||||
for (let i = 1; i < lines.length; i++) {
|
||||
const line = lines[i].trim()
|
||||
if (!line) continue
|
||||
|
||||
// Parse CSV with comma delimiter
|
||||
// Handle quoted fields that may contain commas
|
||||
const fields = parseCSVLine(line, ',')
|
||||
|
||||
if (fields.length < 4) {
|
||||
issues.push({ row: i + 1, message: 'Too few columns', severity: 'warning' })
|
||||
skippedRows++
|
||||
continue
|
||||
}
|
||||
|
||||
const [date, description, _category, amountStr, balanceStr] = fields
|
||||
|
||||
// Skip reserved/pending transactions
|
||||
if (description?.toLowerCase().includes('reserverat')) {
|
||||
skippedRows++
|
||||
continue
|
||||
}
|
||||
|
||||
const amount = parseCommaDecimal(amountStr)
|
||||
if (isNaN(amount)) {
|
||||
issues.push({ row: i + 1, message: `Invalid amount: ${amountStr}`, severity: 'warning' })
|
||||
skippedRows++
|
||||
continue
|
||||
}
|
||||
|
||||
// Validate date format
|
||||
if (!/^\d{4}-\d{2}-\d{2}$/.test(date?.trim())) {
|
||||
issues.push({ row: i + 1, message: `Invalid date: ${date}`, severity: 'warning' })
|
||||
skippedRows++
|
||||
continue
|
||||
}
|
||||
|
||||
const balance = balanceStr ? parseCommaDecimal(balanceStr) : null
|
||||
|
||||
transactions.push({
|
||||
date: date.trim(),
|
||||
description: description?.trim() || 'Unknown',
|
||||
amount,
|
||||
currency: 'SEK',
|
||||
balance: isNaN(balance as number) ? null : balance,
|
||||
reference: null,
|
||||
counterparty: null,
|
||||
raw_line: line,
|
||||
})
|
||||
}
|
||||
|
||||
const dates = transactions.map((t) => t.date).sort()
|
||||
|
||||
return {
|
||||
format: 'nordea',
|
||||
format_name: 'Nordea',
|
||||
transactions,
|
||||
date_from: dates[0] || null,
|
||||
date_to: dates[dates.length - 1] || null,
|
||||
issues,
|
||||
stats: {
|
||||
total_rows: lines.length - 1,
|
||||
parsed_rows: transactions.length,
|
||||
skipped_rows: skippedRows,
|
||||
total_income: Math.round(transactions.filter((t) => t.amount > 0).reduce((s, t) => s + t.amount, 0) * 100) / 100,
|
||||
total_expenses: Math.round(transactions.filter((t) => t.amount < 0).reduce((s, t) => s + t.amount, 0) * 100) / 100,
|
||||
},
|
||||
}
|
||||
},
|
||||
}
|
||||
|
||||
/**
|
||||
* Parse a CSV line respecting quoted fields
|
||||
*/
|
||||
function parseCSVLine(line: string, delimiter: string): string[] {
|
||||
const fields: string[] = []
|
||||
let current = ''
|
||||
let inQuotes = false
|
||||
|
||||
for (let i = 0; i < line.length; i++) {
|
||||
const char = line[i]
|
||||
|
||||
if (char === '"') {
|
||||
if (inQuotes && line[i + 1] === '"') {
|
||||
current += '"'
|
||||
i++ // Skip escaped quote
|
||||
} else {
|
||||
inQuotes = !inQuotes
|
||||
}
|
||||
} else if (char === delimiter && !inQuotes) {
|
||||
fields.push(current)
|
||||
current = ''
|
||||
} else {
|
||||
current += char
|
||||
}
|
||||
}
|
||||
|
||||
fields.push(current)
|
||||
return fields
|
||||
}
|
||||
|
||||
export { parseCSVLine }
|
||||
@@ -0,0 +1,138 @@
|
||||
/**
|
||||
* SEB CSV format parser
|
||||
*
|
||||
* Format: Semicolon-delimited, comma decimal separator
|
||||
* Columns vary but typically: Bokföringsdag, Valutadag, Verifikationsnummer,
|
||||
* Text/mottagare, Belopp, Saldo
|
||||
* Date format: YYYY-MM-DD
|
||||
* Encoding: UTF-8 or Windows-1252
|
||||
*/
|
||||
|
||||
import type { BankFileFormat, BankFileParseResult, ParsedBankTransaction, BankFileParseIssue } from '../types'
|
||||
import { prepareContent } from '../encoding'
|
||||
|
||||
function parseCommaDecimal(value: string): number {
|
||||
const cleaned = value.replace(/\s/g, '').replace(',', '.')
|
||||
return parseFloat(cleaned)
|
||||
}
|
||||
|
||||
export const sebFormat: BankFileFormat = {
|
||||
id: 'seb',
|
||||
name: 'SEB',
|
||||
description: 'SEB CSV (semicolon-delimited)',
|
||||
fileExtensions: ['.csv', '.txt'],
|
||||
|
||||
detect(content: string, _filename: string): boolean {
|
||||
const prepared = prepareContent(content)
|
||||
const firstLine = prepared.split('\n')[0]?.toLowerCase() || ''
|
||||
// SEB headers contain "bokföringsdag" or "bokforingsdatum" and use semicolons
|
||||
return (
|
||||
firstLine.includes(';') &&
|
||||
(firstLine.includes('bokföringsdag') ||
|
||||
firstLine.includes('bokforingsdatum') ||
|
||||
firstLine.includes('bokföringsdag'))
|
||||
)
|
||||
},
|
||||
|
||||
parse(content: string): BankFileParseResult {
|
||||
const prepared = prepareContent(content)
|
||||
const lines = prepared.split('\n').filter((line) => line.trim() !== '')
|
||||
|
||||
const transactions: ParsedBankTransaction[] = []
|
||||
const issues: BankFileParseIssue[] = []
|
||||
let skippedRows = 0
|
||||
|
||||
// Parse header to find column indices
|
||||
const headerLine = lines[0] || ''
|
||||
const headers = headerLine.split(';').map((h) => h.trim().toLowerCase().replace(/"/g, ''))
|
||||
|
||||
// Find column indices dynamically
|
||||
const dateIdx = headers.findIndex(
|
||||
(h) => h.includes('bokföringsdag') || h.includes('bokforingsdatum') || h.includes('bokföringsdag')
|
||||
)
|
||||
const descIdx = headers.findIndex(
|
||||
(h) => h.includes('text') || h.includes('mottagare') || h.includes('beskrivning')
|
||||
)
|
||||
const amountIdx = headers.findIndex((h) => h.includes('belopp'))
|
||||
const balanceIdx = headers.findIndex((h) => h.includes('saldo'))
|
||||
|
||||
if (dateIdx === -1 || amountIdx === -1) {
|
||||
issues.push({
|
||||
row: 1,
|
||||
message: 'Could not identify required columns (date, amount)',
|
||||
severity: 'error',
|
||||
})
|
||||
return {
|
||||
format: 'seb',
|
||||
format_name: 'SEB',
|
||||
transactions: [],
|
||||
date_from: null,
|
||||
date_to: null,
|
||||
issues,
|
||||
stats: { total_rows: 0, parsed_rows: 0, skipped_rows: 0, total_income: 0, total_expenses: 0 },
|
||||
}
|
||||
}
|
||||
|
||||
for (let i = 1; i < lines.length; i++) {
|
||||
const line = lines[i].trim()
|
||||
if (!line) continue
|
||||
|
||||
const fields = line.split(';').map((f) => f.trim().replace(/^"|"$/g, ''))
|
||||
|
||||
const date = fields[dateIdx]
|
||||
const description = fields[descIdx >= 0 ? descIdx : dateIdx + 1] || 'Unknown'
|
||||
const amountStr = fields[amountIdx]
|
||||
const balanceStr = balanceIdx >= 0 ? fields[balanceIdx] : undefined
|
||||
|
||||
if (!date || !amountStr) {
|
||||
issues.push({ row: i + 1, message: 'Missing required fields', severity: 'warning' })
|
||||
skippedRows++
|
||||
continue
|
||||
}
|
||||
|
||||
const amount = parseCommaDecimal(amountStr)
|
||||
if (isNaN(amount)) {
|
||||
issues.push({ row: i + 1, message: `Invalid amount: ${amountStr}`, severity: 'warning' })
|
||||
skippedRows++
|
||||
continue
|
||||
}
|
||||
|
||||
if (!/^\d{4}-\d{2}-\d{2}$/.test(date)) {
|
||||
issues.push({ row: i + 1, message: `Invalid date: ${date}`, severity: 'warning' })
|
||||
skippedRows++
|
||||
continue
|
||||
}
|
||||
|
||||
const balance = balanceStr ? parseCommaDecimal(balanceStr) : null
|
||||
|
||||
transactions.push({
|
||||
date,
|
||||
description: description.trim(),
|
||||
amount,
|
||||
currency: 'SEK',
|
||||
balance: isNaN(balance as number) ? null : balance,
|
||||
reference: null,
|
||||
counterparty: null,
|
||||
raw_line: line,
|
||||
})
|
||||
}
|
||||
|
||||
const dates = transactions.map((t) => t.date).sort()
|
||||
|
||||
return {
|
||||
format: 'seb',
|
||||
format_name: 'SEB',
|
||||
transactions,
|
||||
date_from: dates[0] || null,
|
||||
date_to: dates[dates.length - 1] || null,
|
||||
issues,
|
||||
stats: {
|
||||
total_rows: lines.length - 1,
|
||||
parsed_rows: transactions.length,
|
||||
skipped_rows: skippedRows,
|
||||
total_income: Math.round(transactions.filter((t) => t.amount > 0).reduce((s, t) => s + t.amount, 0) * 100) / 100,
|
||||
total_expenses: Math.round(transactions.filter((t) => t.amount < 0).reduce((s, t) => s + t.amount, 0) * 100) / 100,
|
||||
},
|
||||
}
|
||||
},
|
||||
}
|
||||
@@ -0,0 +1,149 @@
|
||||
/**
|
||||
* Swedbank CSV format parser
|
||||
*
|
||||
* Format: Comma-delimited, PERIOD decimal separator (exception among Swedish banks!)
|
||||
* Columns: Clearingnummer, Kontonummer, Datum, Text, Belopp, Saldo, and more (12 columns)
|
||||
* Date format: YYYY-MM-DD
|
||||
* Encoding: UTF-8 or Windows-1252
|
||||
*
|
||||
* Notes:
|
||||
* - First line is metadata (account info), SKIP it
|
||||
* - Second line is the actual header
|
||||
* - Uses period as decimal separator (unlike Nordea/SEB/Handelsbanken)
|
||||
*/
|
||||
|
||||
import type { BankFileFormat, BankFileParseResult, ParsedBankTransaction, BankFileParseIssue } from '../types'
|
||||
import { prepareContent } from '../encoding'
|
||||
import { parseCSVLine } from './nordea'
|
||||
|
||||
export const swedbankFormat: BankFileFormat = {
|
||||
id: 'swedbank',
|
||||
name: 'Swedbank',
|
||||
description: 'Swedbank CSV (comma-delimited, period decimal)',
|
||||
fileExtensions: ['.csv', '.txt'],
|
||||
|
||||
detect(content: string, _filename: string): boolean {
|
||||
const prepared = prepareContent(content)
|
||||
const lines = prepared.split('\n')
|
||||
// Check first two lines — Swedbank has metadata line, then header
|
||||
const line1 = lines[0]?.toLowerCase() || ''
|
||||
const line2 = lines[1]?.toLowerCase() || ''
|
||||
|
||||
return (
|
||||
(line1.includes('clearingnummer') || line2.includes('clearingnummer') ||
|
||||
line1.includes('radnummer') || line2.includes('radnummer'))
|
||||
)
|
||||
},
|
||||
|
||||
parse(content: string): BankFileParseResult {
|
||||
const prepared = prepareContent(content)
|
||||
const lines = prepared.split('\n').filter((line) => line.trim() !== '')
|
||||
|
||||
const transactions: ParsedBankTransaction[] = []
|
||||
const issues: BankFileParseIssue[] = []
|
||||
let skippedRows = 0
|
||||
|
||||
// Determine where the header is
|
||||
// Line 0 might be metadata, line 1 might be header
|
||||
let headerLineIdx = 0
|
||||
const line0Lower = lines[0]?.toLowerCase() || ''
|
||||
const line1Lower = lines[1]?.toLowerCase() || ''
|
||||
|
||||
if (line1Lower.includes('clearingnummer') || line1Lower.includes('radnummer')) {
|
||||
headerLineIdx = 1
|
||||
} else if (line0Lower.includes('clearingnummer') || line0Lower.includes('radnummer')) {
|
||||
headerLineIdx = 0
|
||||
}
|
||||
|
||||
const headerLine = lines[headerLineIdx] || ''
|
||||
const headers = parseCSVLine(headerLine, ',').map((h) =>
|
||||
h.trim().toLowerCase().replace(/"/g, '')
|
||||
)
|
||||
|
||||
// Find column indices
|
||||
const dateIdx = headers.findIndex((h) => h === 'datum' || h.includes('bokföringsdatum'))
|
||||
const descIdx = headers.findIndex((h) => h === 'text' || h.includes('beskrivning'))
|
||||
const amountIdx = headers.findIndex((h) => h === 'belopp')
|
||||
const balanceIdx = headers.findIndex((h) => h === 'saldo')
|
||||
|
||||
if (dateIdx === -1 || amountIdx === -1) {
|
||||
issues.push({
|
||||
row: 1,
|
||||
message: 'Could not identify required columns (datum, belopp)',
|
||||
severity: 'error',
|
||||
})
|
||||
return {
|
||||
format: 'swedbank',
|
||||
format_name: 'Swedbank',
|
||||
transactions: [],
|
||||
date_from: null,
|
||||
date_to: null,
|
||||
issues,
|
||||
stats: { total_rows: 0, parsed_rows: 0, skipped_rows: 0, total_income: 0, total_expenses: 0 },
|
||||
}
|
||||
}
|
||||
|
||||
// Data starts after header
|
||||
for (let i = headerLineIdx + 1; i < lines.length; i++) {
|
||||
const line = lines[i].trim()
|
||||
if (!line) continue
|
||||
|
||||
const fields = parseCSVLine(line, ',').map((f) => f.trim().replace(/^"|"$/g, ''))
|
||||
|
||||
const date = fields[dateIdx]
|
||||
const description = descIdx >= 0 ? fields[descIdx] : 'Unknown'
|
||||
const amountStr = fields[amountIdx]
|
||||
const balanceStr = balanceIdx >= 0 ? fields[balanceIdx] : undefined
|
||||
|
||||
if (!date || !amountStr) {
|
||||
skippedRows++
|
||||
continue
|
||||
}
|
||||
|
||||
// Swedbank uses PERIOD decimal separator
|
||||
const amount = parseFloat(amountStr.replace(/\s/g, ''))
|
||||
if (isNaN(amount)) {
|
||||
issues.push({ row: i + 1, message: `Invalid amount: ${amountStr}`, severity: 'warning' })
|
||||
skippedRows++
|
||||
continue
|
||||
}
|
||||
|
||||
if (!/^\d{4}-\d{2}-\d{2}$/.test(date)) {
|
||||
issues.push({ row: i + 1, message: `Invalid date: ${date}`, severity: 'warning' })
|
||||
skippedRows++
|
||||
continue
|
||||
}
|
||||
|
||||
const balance = balanceStr ? parseFloat(balanceStr.replace(/\s/g, '')) : null
|
||||
|
||||
transactions.push({
|
||||
date,
|
||||
description: (description || 'Unknown').trim(),
|
||||
amount,
|
||||
currency: 'SEK',
|
||||
balance: isNaN(balance as number) ? null : balance,
|
||||
reference: null,
|
||||
counterparty: null,
|
||||
raw_line: line,
|
||||
})
|
||||
}
|
||||
|
||||
const dates = transactions.map((t) => t.date).sort()
|
||||
|
||||
return {
|
||||
format: 'swedbank',
|
||||
format_name: 'Swedbank',
|
||||
transactions,
|
||||
date_from: dates[0] || null,
|
||||
date_to: dates[dates.length - 1] || null,
|
||||
issues,
|
||||
stats: {
|
||||
total_rows: lines.length - headerLineIdx - 1,
|
||||
parsed_rows: transactions.length,
|
||||
skipped_rows: skippedRows,
|
||||
total_income: Math.round(transactions.filter((t) => t.amount > 0).reduce((s, t) => s + t.amount, 0) * 100) / 100,
|
||||
total_expenses: Math.round(transactions.filter((t) => t.amount < 0).reduce((s, t) => s + t.amount, 0) * 100) / 100,
|
||||
},
|
||||
}
|
||||
},
|
||||
}
|
||||
@@ -0,0 +1,138 @@
|
||||
/**
|
||||
* Bank file parser — main entry point
|
||||
*
|
||||
* Auto-detects Swedish bank file formats and parses to normalized transactions.
|
||||
* Supports Nordea, SEB, Swedbank, Handelsbanken CSV and ISO 20022 camt.053 XML.
|
||||
*/
|
||||
|
||||
import * as crypto from 'crypto'
|
||||
import type { BankFileFormat, BankFileFormatId, BankFileParseResult, ParsedBankTransaction } from './types'
|
||||
import { nordeaFormat } from './formats/nordea'
|
||||
import { sebFormat } from './formats/seb'
|
||||
import { swedbankFormat } from './formats/swedbank'
|
||||
import { handelsbankenFormat } from './formats/handelsbanken'
|
||||
import { camt053Format } from './formats/camt053'
|
||||
import { genericCSVFormat } from './formats/generic-csv'
|
||||
|
||||
/**
|
||||
* Ordered list of format detectors.
|
||||
* camt.053 first (XML detection is unambiguous), then bank-specific CSV formats.
|
||||
* Generic CSV is last — it never auto-detects (manual fallback only).
|
||||
*/
|
||||
const FORMATS: BankFileFormat[] = [
|
||||
camt053Format,
|
||||
nordeaFormat,
|
||||
sebFormat,
|
||||
swedbankFormat,
|
||||
handelsbankenFormat,
|
||||
genericCSVFormat,
|
||||
]
|
||||
|
||||
/**
|
||||
* Get a format by its ID
|
||||
*/
|
||||
export function getFormat(id: BankFileFormatId): BankFileFormat | undefined {
|
||||
return FORMATS.find((f) => f.id === id)
|
||||
}
|
||||
|
||||
/**
|
||||
* Get all available formats
|
||||
*/
|
||||
export function getAllFormats(): BankFileFormat[] {
|
||||
return FORMATS
|
||||
}
|
||||
|
||||
/**
|
||||
* Auto-detect the bank file format from content and filename
|
||||
*
|
||||
* Returns the first matching format, or null if no format matches.
|
||||
* Uses filename extension as a hint (e.g. .xml for camt.053).
|
||||
*/
|
||||
export function detectFileFormat(content: string, filename: string): BankFileFormat | null {
|
||||
for (const format of FORMATS) {
|
||||
if (format.detect(content, filename)) {
|
||||
return format
|
||||
}
|
||||
}
|
||||
return null
|
||||
}
|
||||
|
||||
/**
|
||||
* Parse a bank file with auto-detection or explicit format
|
||||
*
|
||||
* @param content - File content as string (already decoded)
|
||||
* @param filename - Original filename (used for format detection hints)
|
||||
* @param formatId - Optional explicit format to use (skips auto-detection)
|
||||
*/
|
||||
export function parseBankFile(
|
||||
content: string,
|
||||
filename: string,
|
||||
formatId?: BankFileFormatId
|
||||
): BankFileParseResult {
|
||||
let format: BankFileFormat | undefined
|
||||
|
||||
if (formatId) {
|
||||
format = getFormat(formatId)
|
||||
if (!format) {
|
||||
return {
|
||||
format: formatId,
|
||||
format_name: 'Unknown',
|
||||
transactions: [],
|
||||
date_from: null,
|
||||
date_to: null,
|
||||
issues: [{ row: 0, message: `Unknown format: ${formatId}`, severity: 'error' }],
|
||||
stats: { total_rows: 0, parsed_rows: 0, skipped_rows: 0, total_income: 0, total_expenses: 0 },
|
||||
}
|
||||
}
|
||||
} else {
|
||||
format = detectFileFormat(content, filename) || undefined
|
||||
if (!format) {
|
||||
return {
|
||||
format: 'generic_csv',
|
||||
format_name: 'Unknown',
|
||||
transactions: [],
|
||||
date_from: null,
|
||||
date_to: null,
|
||||
issues: [{
|
||||
row: 0,
|
||||
message: 'Could not auto-detect file format. Please select your bank manually.',
|
||||
severity: 'error',
|
||||
}],
|
||||
stats: { total_rows: 0, parsed_rows: 0, skipped_rows: 0, total_income: 0, total_expenses: 0 },
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
return format.parse(content)
|
||||
}
|
||||
|
||||
/**
|
||||
* Generate a stable external_id for a parsed bank transaction.
|
||||
*
|
||||
* For CSV files: SHA-256 of (format + date + description + amount + row_index)
|
||||
* For camt.053: Uses the entry reference from the XML if available
|
||||
*
|
||||
* Two identical transactions on the same day will get different IDs due to row_index.
|
||||
*/
|
||||
export function generateExternalId(
|
||||
tx: ParsedBankTransaction,
|
||||
formatId: BankFileFormatId,
|
||||
rowIndex: number
|
||||
): string {
|
||||
// For camt.053, prefer the raw_line which contains the entry reference
|
||||
if (formatId === 'camt053' && tx.raw_line && !tx.raw_line.startsWith('camt053_entry_')) {
|
||||
return `camt053_${tx.raw_line}`
|
||||
}
|
||||
|
||||
// For CSV formats, create a composite hash
|
||||
const composite = `${formatId}|${tx.date}|${tx.description}|${tx.amount}|${rowIndex}`
|
||||
const hash = crypto.createHash('sha256').update(composite).digest('hex').substring(0, 16)
|
||||
return `${formatId}_${hash}`
|
||||
}
|
||||
|
||||
/**
|
||||
* Generate a file hash for dedup of the same file being uploaded twice
|
||||
*/
|
||||
export function generateFileHash(content: string): string {
|
||||
return crypto.createHash('sha256').update(content).digest('hex')
|
||||
}
|
||||
@@ -0,0 +1,94 @@
|
||||
/**
|
||||
* Bank file import types
|
||||
*
|
||||
* Supports Swedish bank CSV formats (Nordea, SEB, Swedbank, Handelsbanken)
|
||||
* and ISO 20022 camt.053 XML.
|
||||
*/
|
||||
|
||||
/** Parsed and normalized bank transaction from any file format */
|
||||
export interface ParsedBankTransaction {
|
||||
date: string // YYYY-MM-DD
|
||||
description: string
|
||||
amount: number // Positive = income, negative = expense
|
||||
currency: string
|
||||
balance?: number | null
|
||||
reference?: string | null // OCR number, Bankgiro reference
|
||||
counterparty?: string | null
|
||||
raw_line?: string // Original CSV line for debugging
|
||||
}
|
||||
|
||||
/** Result from parsing a bank file */
|
||||
export interface BankFileParseResult {
|
||||
format: BankFileFormatId
|
||||
format_name: string
|
||||
transactions: ParsedBankTransaction[]
|
||||
date_from: string | null
|
||||
date_to: string | null
|
||||
issues: BankFileParseIssue[]
|
||||
stats: {
|
||||
total_rows: number
|
||||
parsed_rows: number
|
||||
skipped_rows: number
|
||||
total_income: number
|
||||
total_expenses: number
|
||||
}
|
||||
}
|
||||
|
||||
/** Issue encountered during parsing */
|
||||
export interface BankFileParseIssue {
|
||||
row: number
|
||||
message: string
|
||||
severity: 'warning' | 'error'
|
||||
}
|
||||
|
||||
/** Supported bank file format identifiers */
|
||||
export type BankFileFormatId =
|
||||
| 'nordea'
|
||||
| 'seb'
|
||||
| 'swedbank'
|
||||
| 'handelsbanken'
|
||||
| 'generic_csv'
|
||||
| 'camt053'
|
||||
|
||||
/** Format definition with detection and parsing capability */
|
||||
export interface BankFileFormat {
|
||||
id: BankFileFormatId
|
||||
name: string
|
||||
description: string
|
||||
fileExtensions: string[]
|
||||
detect: (content: string, filename: string) => boolean
|
||||
parse: (content: string) => BankFileParseResult
|
||||
}
|
||||
|
||||
/** Import tracking record stored in DB */
|
||||
export interface BankFileImport {
|
||||
id: string
|
||||
user_id: string
|
||||
filename: string
|
||||
file_hash: string
|
||||
file_format: string
|
||||
transaction_count: number
|
||||
imported_count: number
|
||||
duplicate_count: number
|
||||
matched_count: number
|
||||
date_from: string | null
|
||||
date_to: string | null
|
||||
status: 'pending' | 'processing' | 'completed' | 'failed'
|
||||
error_message: string | null
|
||||
created_at: string
|
||||
updated_at: string
|
||||
}
|
||||
|
||||
/** Column mapping for generic CSV format */
|
||||
export interface GenericCSVColumnMapping {
|
||||
date: number
|
||||
description: number
|
||||
amount: number
|
||||
reference?: number
|
||||
counterparty?: number
|
||||
balance?: number
|
||||
delimiter: string
|
||||
decimal_separator: ',' | '.'
|
||||
skip_rows: number
|
||||
date_format: string // e.g. 'YYYY-MM-DD'
|
||||
}
|
||||
Reference in New Issue
Block a user