Merge pull request 'fix-pdf-parser-crash' (#2) from fix-pdf-parser-crash into main

Reviewed-on: #2
2026-03-14 10:43:21 +00:00
parent 0638885812 feb756cfe2
commit 5fa6b921d8
10 changed files with 81 additions and 13 deletions
--- a/Dockerfile.backend
+++ b/Dockerfile.backend
@@ -5,6 +5,11 @@ COPY package.json package-lock.json* ./
 COPY shared ./shared
 COPY backend ./backend

+# Увеличенные таймауты для нестабильной сети
+RUN npm config set fetch-retry-mintimeout 20000 && \
+    npm config set fetch-retry-maxtimeout 120000 && \
+    npm config set fetch-timeout 300000
+
 RUN npm install

 FROM node:20-alpine AS build
@@ -22,6 +27,10 @@ FROM node:20-alpine AS runner
 WORKDIR /app
 ENV NODE_ENV=production

+# @napi-rs/canvas (dep of pdf-parse) ships a musl pre-built binary that
+# needs these compatibility / font libraries at runtime.
+RUN apk add --no-cache libc6-compat fontconfig freetype
+
 COPY --from=build /app/backend/dist ./dist
 COPY --from=build /app/backend/package.json ./package.json
 COPY --from=build /app/node_modules ./node_modules
--- a/Dockerfile.frontend
+++ b/Dockerfile.frontend
@@ -9,6 +9,11 @@ COPY package.json package-lock.json* tsconfig.json* ./
 COPY shared ./shared
 COPY frontend ./frontend

+# Увеличиваем таймауты и повторы для нестабильной сети
+RUN npm config set fetch-retry-mintimeout 20000 && \
+    npm config set fetch-retry-maxtimeout 120000 && \
+    npm config set fetch-timeout 300000
+
 # Устанавливаем зависимости из корня
 RUN npm install

--- a/backend/.env.example
+++ b/backend/.env.example
@@ -18,3 +18,6 @@ LLM_API_KEY=
 # Базовый URL API LLM (опционально). По умолчанию https://api.openai.com
 # Примеры: Ollama — http://localhost:11434/v1, Azure — https://YOUR_RESOURCE.openai.azure.com/openai/deployments/YOUR_DEPLOYMENT
 LLM_API_BASE_URL=
+
+# Имя модели LLM (опционально). Для OpenAI: gpt-4o-mini. Для Ollama: qwen2.5:7b, qwen3:7b
+LLM_MODEL=
--- a/backend/package.json
+++ b/backend/package.json
@@ -8,7 +8,7 @@
    "start": "node dist/app.js",
    "migrate": "tsx src/db/migrate.ts",
    "migrate:prod": "node dist/db/migrate.js",
-    "migrate:prod": "node dist/db/migrate.js"
+    "test:llm": "tsx src/scripts/testLlm.ts"
  },
  "dependencies": {
    "@family-budget/shared": "*",
--- a/backend/src/config.ts
+++ b/backend/src/config.ts
@@ -24,4 +24,7 @@ export const config = {

  /** Базовый URL API LLM. По умолчанию https://api.openai.com. Для Ollama: http://localhost:11434/v1 */
  llmApiBaseUrl: process.env.LLM_API_BASE_URL || undefined,
+
+  /** Имя модели LLM. Для OpenAI: gpt-4o-mini. Для Ollama: qwen2.5:7b, qwen3:7b и т.п. */
+  llmModel: process.env.LLM_MODEL || 'gpt-4o-mini',
 };
--- a/backend/src/db/pool.ts
+++ b/backend/src/db/pool.ts
@@ -7,4 +7,5 @@ export const pool = new Pool({
  database: config.db.database,
  user: config.db.user,
  password: config.db.password,
+  connectionTimeoutMillis: 5000,
 });
--- a/backend/src/scripts/testLlm.ts
+++ b/backend/src/scripts/testLlm.ts
@@ -0,0 +1,42 @@
+/**
+ * Тестовый запрос к LLM серверу.
+ * Запуск: npm run test:llm
+ */
+import OpenAI from 'openai';
+import { config } from '../config';
+
+async function main() {
+  if (!config.llmApiKey || config.llmApiKey.trim() === '') {
+    console.error('Ошибка: LLM_API_KEY не задан в .env');
+    process.exit(1);
+  }
+
+  const openai = new OpenAI({
+    apiKey: config.llmApiKey,
+    ...(config.llmApiBaseUrl && { baseURL: config.llmApiBaseUrl }),
+  });
+
+  const url = config.llmApiBaseUrl || 'https://api.openai.com/v1';
+  console.log('LLM сервер:', url);
+  console.log('Модель:', config.llmModel);
+  console.log('---');
+
+  try {
+    const completion = await openai.chat.completions.create({
+      model: config.llmModel,
+      messages: [{ role: 'user', content: 'Ответь одним словом: какой цвет у неба?' }],
+      temperature: 0,
+      max_tokens: 50,
+    });
+
+    const content = completion.choices[0]?.message?.content;
+    console.log('Ответ:', content || '(пусто)');
+    console.log('---');
+    console.log('OK');
+  } catch (err) {
+    console.error('Ошибка:', err instanceof Error ? err.message : err);
+    process.exit(1);
+  }
+}
+
+main();
--- a/backend/src/services/pdfToStatement.ts
+++ b/backend/src/services/pdfToStatement.ts
@@ -1,4 +1,3 @@
-import { PDFParse } from 'pdf-parse';
 import OpenAI from 'openai';
 import { config } from '../config';
 import type { StatementFile } from '@family-budget/shared';
@@ -64,6 +63,18 @@ export function isPdfConversionError(r: unknown): r is PdfConversionError {
  );
 }

+// Lazy-loaded to avoid crashing the app at startup — pdf-parse pulls in
+// @napi-rs/canvas (native Skia binary) which may fail on Alpine.
+let _PDFParse: typeof import('pdf-parse')['PDFParse'] | undefined;
+function loadPDFParse() {
+  if (!_PDFParse) {
+    // eslint-disable-next-line @typescript-eslint/no-require-imports
+    const mod = require('pdf-parse') as typeof import('pdf-parse');
+    _PDFParse = mod.PDFParse;
+  }
+  return _PDFParse;
+}
+
 export async function convertPdfToStatement(
  buffer: Buffer,
 ): Promise<StatementFile | PdfConversionError> {
@@ -77,6 +88,7 @@ export async function convertPdfToStatement(

  let text: string;
  try {
+    const PDFParse = loadPDFParse();
    const parser = new PDFParse({ data: buffer });
    const result = await parser.getText();
    text = result.text || '';
@@ -105,12 +117,14 @@ export async function convertPdfToStatement(

  try {
    const completion = await openai.chat.completions.create({
-      model: 'gpt-4o-mini',
+      model: config.llmModel,
      messages: [
        { role: 'system', content: PDF2JSON_PROMPT },
        { role: 'user', content: `Текст выписки:\n\n${text}` },
      ],
      temperature: 0,
+      max_tokens: 32768,
+      response_format: { type: 'json_object' },
    });

    const content = completion.choices[0]?.message?.content?.trim();
--- a/frontend/src/components/ImportModal.tsx
+++ b/frontend/src/components/ImportModal.tsx
@@ -89,7 +89,7 @@ export function ImportModal({ onClose, onDone }: Props) {

          {result && (
            <div className="import-result">
-              <div className="import-result-icon">&check;</div>
+              <div className="import-result-icon" aria-hidden="true">✓</div>
              <h3>Импорт завершён</h3>
              <table className="import-stats">
                <tbody>
--- a/package-lock.json
+++ b/package-lock.json
@@ -89,7 +89,6 @@
      "integrity": "sha512-CGOfOJqWjg2qW/Mb6zNsDm+u5vFQ8DxXfbM09z69p5Z6+mE1ikP2jUXw+j42Pf1XTYED2Rni5f95npYeuwMDQA==",
      "dev": true,
      "license": "MIT",
-      "peer": true,
      "dependencies": {
        "@babel/code-frame": "^7.29.0",
        "@babel/generator": "^7.29.0",
@@ -1557,7 +1556,6 @@
      "integrity": "sha512-sKYVuV7Sv9fbPIt/442koC7+IIwK5olP1KWeD88e/idgoJqDm3JV/YUiPwkoKK92ylff2MGxSz1CSjsXelx0YA==",
      "dev": true,
      "license": "MIT",
-      "peer": true,
      "dependencies": {
        "@types/body-parser": "*",
        "@types/express-serve-static-core": "^5.0.0",
@@ -1636,7 +1634,6 @@
      "integrity": "sha512-ilcTH/UniCkMdtexkoCN0bI7pMcJDvmQFPvuPvmEaYA/NSfFTAgdUSLAoVjaRJm7+6PvcM+q1zYOwS4wTYMF9w==",
      "dev": true,
      "license": "MIT",
-      "peer": true,
      "dependencies": {
        "csstype": "^3.2.2"
      }
@@ -1776,7 +1773,6 @@
        }
      ],
      "license": "MIT",
-      "peer": true,
      "dependencies": {
        "baseline-browser-mapping": "^2.9.0",
        "caniuse-lite": "^1.0.30001759",
@@ -2926,7 +2922,6 @@
      "resolved": "https://registry.npmjs.org/pg/-/pg-8.19.0.tgz",
      "integrity": "sha512-QIcLGi508BAHkQ3pJNptsFz5WQMlpGbuBGBaIaXsWK8mel2kQ/rThYI+DbgjUvZrIr7MiuEuc9LcChJoEZK1xQ==",
      "license": "MIT",
-      "peer": true,
      "dependencies": {
        "pg-connection-string": "^2.11.0",
        "pg-pool": "^3.12.0",
@@ -3024,7 +3019,6 @@
      "integrity": "sha512-5gTmgEY/sqK6gFXLIsQNH19lWb4ebPDLA4SdLP7dsWkIXHWlG66oPuVvXSGFPppYZz8ZDZq0dYYrbHfBCVUb1Q==",
      "dev": true,
      "license": "MIT",
-      "peer": true,
      "engines": {
        "node": ">=12"
      },
@@ -3174,7 +3168,6 @@
      "resolved": "https://registry.npmjs.org/react/-/react-19.2.4.tgz",
      "integrity": "sha512-9nfp2hYpCwOjAN+8TZFGhtWEwgvWHXqESH8qT89AT/lWklpLON22Lc8pEtnpsZz7VmawabSU0gCjnj8aC0euHQ==",
      "license": "MIT",
-      "peer": true,
      "engines": {
        "node": ">=0.10.0"
      }
@@ -3184,7 +3177,6 @@
      "resolved": "https://registry.npmjs.org/react-dom/-/react-dom-19.2.4.tgz",
      "integrity": "sha512-AXJdLo8kgMbimY95O2aKQqsz2iWi9jMgKJhRBAxECE4IFxfcazB2LmzloIoibJI3C12IlY20+KFaLv+71bUJeQ==",
      "license": "MIT",
-      "peer": true,
      "dependencies": {
        "scheduler": "^0.27.0"
      },
@@ -4296,7 +4288,6 @@
      "integrity": "sha512-+Oxm7q9hDoLMyJOYfUYBuHQo+dkAloi33apOPP56pzj+vsdJDzr+j1NISE5pyaAuKL4A3UD34qd0lx5+kfKp2g==",
      "dev": true,
      "license": "MIT",
-      "peer": true,
      "dependencies": {
        "esbuild": "^0.25.0",
        "fdir": "^6.4.4",