501 lines
18 KiB
TypeScript
501 lines
18 KiB
TypeScript
/**
|
|
* Invoice extraction test using MiniCPM-V only (visual extraction)
|
|
*
|
|
* Multi-query approach with thinking DISABLED for speed.
|
|
* Single pass, no consensus voting.
|
|
*/
|
|
import { tap, expect } from '@git.zone/tstest/tapbundle';
|
|
import * as fs from 'fs';
|
|
import * as path from 'path';
|
|
import { execSync } from 'child_process';
|
|
import * as os from 'os';
|
|
import { ensureMiniCpm } from './helpers/docker.js';
|
|
|
|
const OLLAMA_URL = 'http://localhost:11434';
|
|
const MODEL = 'minicpm-v:latest';
|
|
|
|
interface IInvoice {
|
|
invoice_number: string;
|
|
invoice_date: string;
|
|
vendor_name: string;
|
|
currency: string;
|
|
net_amount: number;
|
|
vat_amount: number;
|
|
total_amount: number;
|
|
}
|
|
|
|
/**
|
|
* Convert PDF to PNG images using ImageMagick
|
|
*/
|
|
function convertPdfToImages(pdfPath: string): string[] {
|
|
const tempDir = fs.mkdtempSync(path.join(os.tmpdir(), 'pdf-convert-'));
|
|
const outputPattern = path.join(tempDir, 'page-%d.png');
|
|
|
|
try {
|
|
execSync(
|
|
`convert -density 200 -quality 90 "${pdfPath}" -background white -alpha remove "${outputPattern}"`,
|
|
{ stdio: 'pipe' }
|
|
);
|
|
|
|
const files = fs.readdirSync(tempDir).filter((f) => f.endsWith('.png')).sort();
|
|
const images: string[] = [];
|
|
|
|
for (const file of files) {
|
|
const imagePath = path.join(tempDir, file);
|
|
const imageData = fs.readFileSync(imagePath);
|
|
images.push(imageData.toString('base64'));
|
|
}
|
|
|
|
return images;
|
|
} finally {
|
|
fs.rmSync(tempDir, { recursive: true, force: true });
|
|
}
|
|
}
|
|
|
|
/**
|
|
* Query MiniCPM-V for a single field (thinking disabled for speed)
|
|
*/
|
|
async function queryField(images: string[], question: string): Promise<string> {
|
|
const response = await fetch(`${OLLAMA_URL}/api/generate`, {
|
|
method: 'POST',
|
|
headers: { 'Content-Type': 'application/json' },
|
|
body: JSON.stringify({
|
|
model: MODEL,
|
|
prompt: `/no_think\n${question}`,
|
|
images: images,
|
|
stream: false,
|
|
options: {
|
|
num_predict: 500,
|
|
temperature: 0.1,
|
|
},
|
|
}),
|
|
});
|
|
|
|
if (!response.ok) {
|
|
throw new Error(`Ollama API error: ${response.status}`);
|
|
}
|
|
|
|
const data = await response.json();
|
|
const content = (data.response || '').trim();
|
|
|
|
// Return full content (no thinking to filter)
|
|
return content;
|
|
}
|
|
|
|
/**
|
|
* Extract invoice data using multiple queries with validation
|
|
*/
|
|
async function extractInvoiceFromImages(images: string[]): Promise<IInvoice> {
|
|
console.log(` [Vision] Processing ${images.length} page(s) with MiniCPM-V (multi-query + validation)`);
|
|
|
|
// Log each result as it comes in
|
|
const queryAndLog = async (name: string, question: string): Promise<string> => {
|
|
const result = await queryField(images, question);
|
|
console.log(` [Query] ${name}: "${result}"`);
|
|
return result;
|
|
};
|
|
|
|
// STRATEGY 1: List-then-pick for invoice number (avoids confusion with VAT/customer IDs)
|
|
// Also ask for invoice number directly as backup
|
|
const [allNumbers, directInvoiceNum] = await Promise.all([
|
|
queryAndLog('All Numbers ', `List ALL document numbers visible on this invoice.
|
|
For each number, identify what type it is.
|
|
Format: type:number, type:number
|
|
Example: >>>invoice:R0014359508, vat:DE123456789, customer:K001234<<<`),
|
|
queryAndLog('Invoice # Dir ', `What is the INVOICE NUMBER (Rechnungsnummer)?
|
|
NOT the VAT number (starts with DE/IE), NOT customer ID.
|
|
Look for "Invoice No.", "Rechnungsnr.", "Invoice #", or "Facture".
|
|
For Adobe: starts with IEE or R followed by digits.
|
|
Return ONLY the number: >>>IEE2022006460244<<<`),
|
|
]);
|
|
|
|
// STRATEGY 2: Query each field with >>> <<< delimiters
|
|
const [invoiceDate, invoiceDateAlt, vendor, currency, totalAmount, netAmount, vatAmount] = await Promise.all([
|
|
queryAndLog('Invoice Date ', `Find the INVOICE DATE (when issued, NOT due date).
|
|
Look for: "Invoice Date", "Rechnungsdatum", "Date", "Datum"
|
|
Return ONLY the date in YYYY-MM-DD format: >>>2024-01-15<<<`),
|
|
|
|
// STRATEGY 3: Ask same question differently for verification
|
|
queryAndLog('Date Alt ', `What date appears next to the invoice number at the top?
|
|
Return ONLY YYYY-MM-DD format: >>>2024-01-15<<<`),
|
|
|
|
queryAndLog('Vendor ', `What company ISSUED this invoice (seller, not buyer)?
|
|
Look at letterhead/logo at top.
|
|
Return ONLY the company name: >>>Adobe Inc.<<<`),
|
|
|
|
queryAndLog('Currency ', `What currency symbol appears next to amounts? € $ or £?
|
|
Return the 3-letter code: >>>EUR<<<`),
|
|
|
|
queryAndLog('Total Amount ', `What is the FINAL TOTAL amount (including tax) the customer must pay?
|
|
Look for "Total", "Grand Total", "Gesamtbetrag" at the bottom.
|
|
Return ONLY the number (no symbol): >>>24.99<<<`),
|
|
|
|
queryAndLog('Net Amount ', `What is the NET/subtotal amount BEFORE tax?
|
|
Look for "Net", "Netto", "Subtotal".
|
|
Return ONLY the number: >>>20.99<<<`),
|
|
|
|
queryAndLog('VAT Amount ', `What is the VAT/tax amount?
|
|
Look for "VAT", "MwSt", "USt", "Tax".
|
|
Return ONLY the number: >>>4.00<<<`),
|
|
]);
|
|
|
|
// Extract value from >>> <<< delimiters, or return original if not found
|
|
const extractDelimited = (s: string): string => {
|
|
const match = s.match(/>>>([^<]+)<<</);
|
|
return match ? match[1].trim() : s.trim();
|
|
};
|
|
|
|
// Parse amount from string (handles European format and prose)
|
|
const parseAmount = (s: string): number => {
|
|
if (!s) return 0;
|
|
|
|
// First try delimited format
|
|
const delimitedMatch = s.match(/>>>([^<]+)<<</);
|
|
if (delimitedMatch) {
|
|
const numMatch = delimitedMatch[1].match(/([\d.,]+)/);
|
|
if (numMatch) {
|
|
const numStr = numMatch[1];
|
|
const normalized = numStr.includes(',') && numStr.indexOf(',') > numStr.lastIndexOf('.')
|
|
? numStr.replace(/\./g, '').replace(',', '.')
|
|
: numStr.replace(/,/g, '');
|
|
return parseFloat(normalized) || 0;
|
|
}
|
|
}
|
|
|
|
// Try to find amount patterns in prose: "24.99", "24,99", "€24.99", "24.99 EUR"
|
|
const amountPatterns = [
|
|
/(?:€|EUR|USD|GBP)\s*([\d.,]+)/i, // €24.99 or EUR 24.99
|
|
/([\d.,]+)\s*(?:€|EUR|USD|GBP)/i, // 24.99 EUR or 24.99€
|
|
/(?:is|amount|total)[:\s]+([\d.,]+)/i, // "is 24.99" or "amount: 24.99"
|
|
/\b(\d{1,3}(?:[.,]\d{2,3})*(?:[.,]\d{2}))\b/, // General number pattern with decimals
|
|
];
|
|
|
|
for (const pattern of amountPatterns) {
|
|
const match = s.match(pattern);
|
|
if (match) {
|
|
const numStr = match[1];
|
|
// European format: 1.234,56 → 1234.56
|
|
const normalized = numStr.includes(',') && numStr.indexOf(',') > numStr.lastIndexOf('.')
|
|
? numStr.replace(/\./g, '').replace(',', '.')
|
|
: numStr.replace(/,/g, '');
|
|
const value = parseFloat(normalized);
|
|
if (value > 0) return value;
|
|
}
|
|
}
|
|
|
|
return 0;
|
|
};
|
|
|
|
// STRATEGY 1: Parse "all numbers" to find invoice number
|
|
const extractInvoiceFromList = (allNums: string): string | null => {
|
|
const delimited = extractDelimited(allNums);
|
|
|
|
// Find ALL "invoice:XXX" matches
|
|
const invoiceMatches = delimited.matchAll(/invoice[:\s]*([A-Z0-9-]+)/gi);
|
|
const candidates: string[] = [];
|
|
for (const match of invoiceMatches) {
|
|
const value = match[1];
|
|
// Filter out labels like "USt-IdNr", "INVOICE", short strings
|
|
if (value.length > 5 && /\d{4,}/.test(value) && !/^(ust|vat|tax|nr|id|no)/i.test(value)) {
|
|
candidates.push(value);
|
|
}
|
|
}
|
|
if (candidates.length > 0) return candidates[0];
|
|
|
|
// Look for "rechnungsnr:XXX" pattern
|
|
const rechnungMatch = delimited.match(/rechnung[snr]*[:\s]*([A-Z0-9-]{6,})/i);
|
|
if (rechnungMatch && /\d{4,}/.test(rechnungMatch[1])) return rechnungMatch[1];
|
|
|
|
// Look for patterns like IEE2022..., R001... (Adobe invoice number patterns)
|
|
const adobeMatch = delimited.match(/\b(IEE\d{10,})\b/i);
|
|
if (adobeMatch) return adobeMatch[1];
|
|
const rInvoiceMatch = delimited.match(/\b(R\d{8,})\b/i);
|
|
if (rInvoiceMatch) return rInvoiceMatch[1];
|
|
|
|
return null;
|
|
};
|
|
|
|
// Fallback invoice number extraction
|
|
const extractInvoiceNumber = (s: string): string => {
|
|
const delimited = extractDelimited(s);
|
|
if (delimited !== s.trim()) return delimited;
|
|
|
|
let clean = s.replace(/\*\*/g, '').replace(/`/g, '');
|
|
const patterns = [
|
|
/\b([A-Z]{2,3}\d{10,})\b/i,
|
|
/\b([A-Z]\d{8,})\b/i,
|
|
/\b(INV[-\s]?\d{4}[-\s]?\d+)\b/i,
|
|
/\b(\d{7,})\b/,
|
|
];
|
|
for (const pattern of patterns) {
|
|
const match = clean.match(pattern);
|
|
if (match) return match[1];
|
|
}
|
|
return clean.replace(/[^A-Z0-9-]/gi, '').trim() || clean.trim();
|
|
};
|
|
|
|
// Extract date with fallback
|
|
const extractDate = (s: string): string => {
|
|
const delimited = extractDelimited(s);
|
|
if (/^\d{4}-\d{2}-\d{2}$/.test(delimited)) return delimited;
|
|
|
|
let clean = s.replace(/\*\*/g, '').replace(/`/g, '');
|
|
const isoMatch = clean.match(/(\d{4}-\d{2}-\d{2})/);
|
|
if (isoMatch) return isoMatch[1];
|
|
const dmmyMatch = clean.match(/(\d{1,2})[-\/]([A-Z]{3})[-\/](\d{4})/i);
|
|
if (dmmyMatch) {
|
|
const monthMap: Record<string, string> = {
|
|
JAN: '01', FEB: '02', MAR: '03', APR: '04', MAY: '05', JUN: '06',
|
|
JUL: '07', AUG: '08', SEP: '09', OCT: '10', NOV: '11', DEC: '12',
|
|
};
|
|
return `${dmmyMatch[3]}-${monthMap[dmmyMatch[2].toUpperCase()] || '01'}-${dmmyMatch[1].padStart(2, '0')}`;
|
|
}
|
|
const dmyMatch = clean.match(/(\d{1,2})[\/.](\d{1,2})[\/.](\d{4})/);
|
|
if (dmyMatch) {
|
|
return `${dmyMatch[3]}-${dmyMatch[2].padStart(2, '0')}-${dmyMatch[1].padStart(2, '0')}`;
|
|
}
|
|
return '';
|
|
};
|
|
|
|
// Extract currency
|
|
const extractCurrency = (s: string): string => {
|
|
const delimited = extractDelimited(s);
|
|
if (['EUR', 'USD', 'GBP'].includes(delimited.toUpperCase())) return delimited.toUpperCase();
|
|
const upper = s.toUpperCase();
|
|
if (upper.includes('EUR') || upper.includes('€')) return 'EUR';
|
|
if (upper.includes('USD') || upper.includes('$')) return 'USD';
|
|
if (upper.includes('GBP') || upper.includes('£')) return 'GBP';
|
|
return 'EUR';
|
|
};
|
|
|
|
// Extract vendor
|
|
const extractVendor = (s: string): string => {
|
|
const delimited = extractDelimited(s);
|
|
if (delimited !== s.trim()) return delimited;
|
|
let clean = s.replace(/\*\*/g, '').replace(/`/g, '').trim();
|
|
if (clean.length < 50) return clean.replace(/[."]+$/, '').trim();
|
|
const companyMatch = clean.match(/([A-Z][A-Za-z0-9\s&]+(?:Ltd|Limited|GmbH|Inc|BV|AG|SE|LLC|Co|Corp)[.]?)/i);
|
|
if (companyMatch) return companyMatch[1].trim();
|
|
return clean;
|
|
};
|
|
|
|
// STRATEGY 1: Get invoice number - try multiple approaches
|
|
// 1. From list with type labels
|
|
// 2. From direct query
|
|
// 3. From pattern matching
|
|
const fromList = extractInvoiceFromList(allNumbers);
|
|
const fromDirect = extractInvoiceNumber(directInvoiceNum);
|
|
const fromFallback = extractInvoiceNumber(allNumbers);
|
|
|
|
// Prefer direct query if it has digits, otherwise use list
|
|
const invoiceNumber = (fromDirect && /\d{6,}/.test(fromDirect)) ? fromDirect :
|
|
(fromList && /\d{4,}/.test(fromList)) ? fromList :
|
|
fromDirect || fromList || fromFallback;
|
|
console.log(` [Parsed] Invoice Number: "${invoiceNumber}" (list: ${fromList}, direct: ${fromDirect})`);
|
|
|
|
// STRATEGY 3: Compare two date responses, pick the valid one
|
|
const date1 = extractDate(invoiceDate);
|
|
const date2 = extractDate(invoiceDateAlt);
|
|
const finalDate = date1 || date2;
|
|
if (date1 && date2 && date1 !== date2) {
|
|
console.log(` [Validate] Date mismatch: "${date1}" vs "${date2}" - using first`);
|
|
}
|
|
|
|
// Parse amounts
|
|
let total = parseAmount(totalAmount);
|
|
let net = parseAmount(netAmount);
|
|
let vat = parseAmount(vatAmount);
|
|
|
|
// STRATEGY 4: Cross-field validation for amounts
|
|
// If amounts seem wrong (e.g., 1690 instead of 1.69), try to fix
|
|
if (total > 10000 && net < 100) {
|
|
console.log(` [Validate] Total ${total} seems too high vs net ${net}, dividing by 100`);
|
|
total = total / 100;
|
|
}
|
|
if (net > 10000 && total < 100) {
|
|
console.log(` [Validate] Net ${net} seems too high vs total ${total}, dividing by 100`);
|
|
net = net / 100;
|
|
}
|
|
|
|
// Check if Net + VAT ≈ Total
|
|
if (net > 0 && vat >= 0 && total > 0) {
|
|
const calculated = net + vat;
|
|
if (Math.abs(calculated - total) > 1) {
|
|
console.log(` [Validate] Math check: ${net} + ${vat} = ${calculated} ≠ ${total}`);
|
|
}
|
|
}
|
|
|
|
return {
|
|
invoice_number: invoiceNumber,
|
|
invoice_date: finalDate,
|
|
vendor_name: extractVendor(vendor),
|
|
currency: extractCurrency(currency),
|
|
net_amount: net,
|
|
vat_amount: vat,
|
|
total_amount: total,
|
|
};
|
|
}
|
|
|
|
/**
|
|
* Normalize date to YYYY-MM-DD
|
|
*/
|
|
function normalizeDate(dateStr: string | null): string {
|
|
if (!dateStr) return '';
|
|
if (/^\d{4}-\d{2}-\d{2}$/.test(dateStr)) return dateStr;
|
|
|
|
const monthMap: Record<string, string> = {
|
|
JAN: '01', FEB: '02', MAR: '03', APR: '04', MAY: '05', JUN: '06',
|
|
JUL: '07', AUG: '08', SEP: '09', OCT: '10', NOV: '11', DEC: '12',
|
|
};
|
|
|
|
let match = dateStr.match(/^(\d{1,2})-([A-Z]{3})-(\d{4})$/i);
|
|
if (match) {
|
|
return `${match[3]}-${monthMap[match[2].toUpperCase()] || '01'}-${match[1].padStart(2, '0')}`;
|
|
}
|
|
|
|
match = dateStr.match(/^(\d{1,2})[\/.](\d{1,2})[\/.](\d{4})$/);
|
|
if (match) {
|
|
return `${match[3]}-${match[2].padStart(2, '0')}-${match[1].padStart(2, '0')}`;
|
|
}
|
|
|
|
return dateStr;
|
|
}
|
|
|
|
/**
|
|
* Compare extracted invoice against expected
|
|
*/
|
|
function compareInvoice(
|
|
extracted: IInvoice,
|
|
expected: IInvoice
|
|
): { match: boolean; errors: string[] } {
|
|
const errors: string[] = [];
|
|
|
|
// Compare invoice number (normalize by removing spaces and case)
|
|
const extNum = extracted.invoice_number?.replace(/\s/g, '').toLowerCase() || '';
|
|
const expNum = expected.invoice_number?.replace(/\s/g, '').toLowerCase() || '';
|
|
if (extNum !== expNum) {
|
|
errors.push(`invoice_number: expected "${expected.invoice_number}", got "${extracted.invoice_number}"`);
|
|
}
|
|
|
|
// Compare date
|
|
if (normalizeDate(extracted.invoice_date) !== normalizeDate(expected.invoice_date)) {
|
|
errors.push(`invoice_date: expected "${expected.invoice_date}", got "${extracted.invoice_date}"`);
|
|
}
|
|
|
|
// Compare total amount (with tolerance)
|
|
if (Math.abs(extracted.total_amount - expected.total_amount) > 0.02) {
|
|
errors.push(`total_amount: expected ${expected.total_amount}, got ${extracted.total_amount}`);
|
|
}
|
|
|
|
// Compare currency
|
|
if (extracted.currency?.toUpperCase() !== expected.currency?.toUpperCase()) {
|
|
errors.push(`currency: expected "${expected.currency}", got "${extracted.currency}"`);
|
|
}
|
|
|
|
return { match: errors.length === 0, errors };
|
|
}
|
|
|
|
/**
|
|
* Find all test cases (PDF + JSON pairs) in .nogit/invoices/
|
|
*/
|
|
function findTestCases(): Array<{ name: string; pdfPath: string; jsonPath: string }> {
|
|
const testDir = path.join(process.cwd(), '.nogit/invoices');
|
|
if (!fs.existsSync(testDir)) {
|
|
return [];
|
|
}
|
|
|
|
const files = fs.readdirSync(testDir);
|
|
const pdfFiles = files.filter((f) => f.endsWith('.pdf'));
|
|
const testCases: Array<{ name: string; pdfPath: string; jsonPath: string }> = [];
|
|
|
|
for (const pdf of pdfFiles) {
|
|
const baseName = pdf.replace('.pdf', '');
|
|
const jsonFile = `${baseName}.json`;
|
|
if (files.includes(jsonFile)) {
|
|
testCases.push({
|
|
name: baseName,
|
|
pdfPath: path.join(testDir, pdf),
|
|
jsonPath: path.join(testDir, jsonFile),
|
|
});
|
|
}
|
|
}
|
|
|
|
testCases.sort((a, b) => a.name.localeCompare(b.name));
|
|
return testCases;
|
|
}
|
|
|
|
// Tests
|
|
|
|
tap.test('setup: ensure Docker containers are running', async () => {
|
|
console.log('\n[Setup] Checking Docker containers...\n');
|
|
const minicpmOk = await ensureMiniCpm();
|
|
expect(minicpmOk).toBeTrue();
|
|
console.log('\n[Setup] All containers ready!\n');
|
|
});
|
|
|
|
tap.test('should have MiniCPM-V model loaded', async () => {
|
|
const response = await fetch(`${OLLAMA_URL}/api/tags`);
|
|
const data = await response.json();
|
|
const modelNames = data.models.map((m: { name: string }) => m.name);
|
|
expect(modelNames.some((name: string) => name.includes('minicpm'))).toBeTrue();
|
|
});
|
|
|
|
const testCases = findTestCases();
|
|
console.log(`\nFound ${testCases.length} invoice test cases (MiniCPM-V)\n`);
|
|
|
|
let passedCount = 0;
|
|
let failedCount = 0;
|
|
const processingTimes: number[] = [];
|
|
|
|
for (const testCase of testCases) {
|
|
tap.test(`should extract invoice: ${testCase.name}`, async () => {
|
|
const expected: IInvoice = JSON.parse(fs.readFileSync(testCase.jsonPath, 'utf-8'));
|
|
console.log(`\n=== ${testCase.name} ===`);
|
|
console.log(`Expected: ${expected.invoice_number} | ${expected.invoice_date} | ${expected.total_amount} ${expected.currency}`);
|
|
|
|
const startTime = Date.now();
|
|
const images = convertPdfToImages(testCase.pdfPath);
|
|
console.log(` Pages: ${images.length}`);
|
|
|
|
const extracted = await extractInvoiceFromImages(images);
|
|
console.log(` Extracted: ${extracted.invoice_number} | ${extracted.invoice_date} | ${extracted.total_amount} ${extracted.currency}`);
|
|
|
|
const elapsedMs = Date.now() - startTime;
|
|
processingTimes.push(elapsedMs);
|
|
|
|
const result = compareInvoice(extracted, expected);
|
|
|
|
if (result.match) {
|
|
passedCount++;
|
|
console.log(` Result: MATCH (${(elapsedMs / 1000).toFixed(1)}s)`);
|
|
} else {
|
|
failedCount++;
|
|
console.log(` Result: MISMATCH (${(elapsedMs / 1000).toFixed(1)}s)`);
|
|
result.errors.forEach((e) => console.log(` - ${e}`));
|
|
}
|
|
|
|
expect(result.match).toBeTrue();
|
|
});
|
|
}
|
|
|
|
tap.test('summary', async () => {
|
|
const totalInvoices = testCases.length;
|
|
const accuracy = totalInvoices > 0 ? (passedCount / totalInvoices) * 100 : 0;
|
|
const totalTimeMs = processingTimes.reduce((a, b) => a + b, 0);
|
|
const avgTimeSec = processingTimes.length > 0 ? totalTimeMs / processingTimes.length / 1000 : 0;
|
|
|
|
console.log(`\n========================================`);
|
|
console.log(` Invoice Extraction Summary (MiniCPM)`);
|
|
console.log(`========================================`);
|
|
console.log(` Method: Multi-query (no_think)`);
|
|
console.log(` Passed: ${passedCount}/${totalInvoices}`);
|
|
console.log(` Failed: ${failedCount}/${totalInvoices}`);
|
|
console.log(` Accuracy: ${accuracy.toFixed(1)}%`);
|
|
console.log(`----------------------------------------`);
|
|
console.log(` Total time: ${(totalTimeMs / 1000).toFixed(1)}s`);
|
|
console.log(` Avg per inv: ${avgTimeSec.toFixed(1)}s`);
|
|
console.log(`========================================\n`);
|
|
});
|
|
|
|
export default tap.start();
|