Add web export bundle and cookie header parsing

2026-06-15 16:58:29 +03:00 · 2026-06-15 16:58:29 +03:00 · 9fc382ae28
commit 9fc382ae28
parent f655edf578
10 changed files with 648 additions and 1 deletions
--- a/.dockerignore
+++ b/.dockerignore
@ -0,0 +1,7 @@
+.git
+.idea
+.venv
+__pycache__/
+*.py[cod]
+.DS_Store
+data/
--- a/.gitignore
+++ b/.gitignore
@ -1,2 +1,5 @@
 .idea
-data
+data
+__pycache__/
+*.py[cod]
+.DS_Store
--- a/.run/docker-compose.yml_
+++ b/.run/docker-compose.yml_
@ -0,0 +1,11 @@
+<component name="ProjectRunConfigurationManager">
+  <configuration default="false" name="docker-compose.yml: Compose Deployment" type="docker-deploy" factoryName="docker-compose.yml" server-name="3. 1 GB">
+    <deployment type="docker-compose.yml">
+      <settings>
+        <option name="commandLineOptions" value="--build" />
+        <option name="sourceFilePath" value="docker-compose.yml" />
+      </settings>
+    </deployment>
+    <method v="2" />
+  </configuration>
+</component>
--- a/15
+++ b/15
@ -0,0 +1,15 @@
+FROM python:3.13-slim
+
+ENV PYTHONDONTWRITEBYTECODE=1
+ENV PYTHONUNBUFFERED=1
+
+WORKDIR /app
+
+COPY requirements.txt .
+RUN pip install --no-cache-dir -r requirements.txt
+
+COPY . .
+
+EXPOSE 8000
+
+CMD ["python", "web_app.py", "--host", "0.0.0.0", "--port", "8000"]
--- a/README.md
+++ b/README.md
@ -57,6 +57,70 @@ output_path = exporter.save_month(2026, 5, "sber_may_2026.xlsx")
 print(output_path)
 ```

+## Веб-интерфейс
+
+Запустить локальную форму:
+
+```bash
+.venv/bin/python web_app.py
+```
+
+После запуска откройте:
+
+```text
+http://127.0.0.1:8000
+```
+
+Публичный адрес через Traefik:
+
+```text
+https://sber.gb.7af.ru/
+```
+
+В форме можно вставить cookies в одном из поддержанных форматов, формат определяется автоматически:
+
+- Netscape `cookies.txt`
+- JSON-массив cookies
+- обычный HTTP-заголовок `Cookie`
+
+Для получения `cookies.txt` можно использовать расширение Chrome:
+[Get cookies.txt LOCALLY](https://chromewebstore.google.com/detail/get-cookiestxt-locally/cclelndahbckbenkjhflpdbgdldlbecc?pli=1).
+
+Порядок действий:
+
+1. Установить расширение [Get cookies.txt LOCALLY](https://chromewebstore.google.com/detail/get-cookiestxt-locally/cclelndahbckbenkjhflpdbgdldlbecc?pli=1).
+2. Открыть [СберБанк Онлайн](https://online.sberbank.ru/CSAFront/index.do) и авторизоваться.
+3. Скопировать cookies через расширение и вставить их в поле формы.
+
+После вставки cookies форма сразу проверяет, сможет ли сервер распарсить введенный формат.
+
+После выбора месяца экспорт скачивается ZIP-архивом. Внутри архива:
+
+- `.xlsx` с основными полями операций
+- `.json` со списком всех операций в том виде, в котором они пришли из API
+
+Каждый экспорт сохраняется в `data/exports` с уникальным именем ZIP-архива. После скачивания архив не удаляется; отдельные временные `.xlsx` и `.json` остаются только внутри архива.
+
+### Запуск через Docker Compose
+
+```bash
+docker compose up --build
+```
+
+После запуска интерфейс будет доступен здесь:
+
+```text
+http://127.0.0.1:8000
+```
+
+В Docker файлы экспорта сохраняются в named volume `sberhistorytoexcel_sber-history-data`.
+
+Остановить контейнер:
+
+```bash
+docker compose down
+```
+
 Можно отдельно получить операции без сохранения:

 ```python
--- a/docker-compose.yml
+++ b/docker-compose.yml
@ -0,0 +1,26 @@
+services:
+  sber-history-web:
+    build: .
+    labels:
+      - "traefik.enable=true"
+      - "traefik.docker.network=traefik"
+      - "traefik.http.routers.sber-history-web.rule=Host(`sber.gb.7af.ru`)"
+      - "traefik.http.routers.sber-history-web.entrypoints=websecure"
+      - "traefik.http.routers.sber-history-web.tls=true"
+      - "traefik.http.routers.sber-history-web.tls.certresolver=letsencrypt"
+      - "traefik.http.services.sber-history-web.loadbalancer.server.port=8000"
+      - "traefik.http.routers.sber-history-web.service=sber-history-web"
+    networks:
+      - traefik
+    ports:
+      - "8000:8000"
+    volumes:
+      - sber-history-data:/app/data
+    restart: unless-stopped
+
+networks:
+  traefik:
+    external: true
+
+volumes:
+  sber-history-data:
--- a/load_data.py
+++ b/load_data.py
@ -1,6 +1,7 @@
 from __future__ import annotations

 import argparse
+import json
 import os
 import zipfile
 from dataclasses import dataclass, field
@ -80,6 +81,24 @@ class SberMonthHistoryExporter:
        self._write_xlsx(path, rows)
        return path

+    def save_month_bundle(
+        self,
+        year: int,
+        month: int,
+        xlsx_path: str | Path | None = None,
+        json_path: str | Path | None = None,
+    ) -> tuple[Path, Path, int]:
+        """Fetch a month and save Excel rows plus the raw operation list as JSON."""
+        operations = self.fetch_month(year, month)
+        xlsx_output = Path(xlsx_path or f"sber_operations_{year}_{month:02d}.xlsx")
+        json_output = Path(json_path or f"sber_operations_{year}_{month:02d}.json")
+
+        rows = [self._operation_to_row(operation) for operation in operations]
+        self._write_xlsx(xlsx_output, rows)
+        self._write_json(json_output, operations)
+
+        return xlsx_output, json_output, len(operations)
+
    def _fetch_page(self, offset: int, month_start: datetime) -> list[dict[str, Any]]:
        payload = {
            "paginationOffset": offset,
@ -230,6 +249,14 @@ class SberMonthHistoryExporter:
            archive.writestr("xl/styles.xml", _styles_xml())
            archive.writestr("xl/worksheets/sheet1.xml", worksheet_xml)

+    @staticmethod
+    def _write_json(path: Path, operations: Iterable[dict[str, Any]]) -> None:
+        path.parent.mkdir(parents=True, exist_ok=True)
+        path.write_text(
+            json.dumps(list(operations), ensure_ascii=False, indent=2),
+            encoding="utf-8",
+        )
+

 def _build_worksheet_xml(headers: list[str], rows: list[dict[str, Any]]) -> str:
    sheet_rows = [_build_row_xml(1, headers)]
--- a/netscape_cookies/README.md
+++ b/netscape_cookies/README.md
@ -50,6 +50,12 @@
 session_id=abc123; theme=dark;
 ```

+Также можно вставить заголовок целиком с префиксом:
+
+```text
+Cookie: session_id=abc123; theme=dark;
+```
+
 У этого формата нет метаданных вроде `domain`, `expires`, `secure` и `httpOnly`, поэтому скрипт заполняет их пустыми или безопасными значениями по умолчанию.

 ## Использование
--- a/netscape_cookies/parse_netscape_cookies.py
+++ b/netscape_cookies/parse_netscape_cookies.py
@ -157,6 +157,8 @@ def parse_json_cookies(text: str, *, decode_value: bool = False) -> list[Netscap

 def parse_cookie_header(text: str, *, decode_value: bool = False) -> list[NetscapeCookie]:
    header = text.strip().rstrip(";")
+    if header.lower().startswith("cookie:"):
+        header = header.split(":", 1)[1].strip().rstrip(";")
    if not header:
        return []

--- a/web_app.py
+++ b/web_app.py
@ -0,0 +1,486 @@
+#!/usr/bin/env python3
+"""Small web UI for exporting Sber operations to XLSX and JSON."""
+
+from __future__ import annotations
+
+import argparse
+import html
+import json
+import re
+import zipfile
+from datetime import datetime
+from http import HTTPStatus
+from http.server import BaseHTTPRequestHandler, ThreadingHTTPServer
+from pathlib import Path
+from tempfile import TemporaryDirectory
+from urllib.parse import parse_qs
+from uuid import uuid4
+
+from load_data import SberMonthHistoryExporter
+from netscape_cookies.parse_netscape_cookies import build_cookie_header, parse_cookies
+
+
+MAX_FORM_BYTES = 2 * 1024 * 1024
+EXPORT_DIR = Path("data") / "exports"
+PERIOD_RE = re.compile(r"^(\d{4})-(\d{2})(?:-\d{2})?$")
+
+
+PAGE_HTML = """<!doctype html>
+<html lang="ru">
+<head>
+  <meta charset="utf-8">
+  <meta name="viewport" content="width=device-width, initial-scale=1">
+  <title>Sber History Export</title>
+  <style>
+    :root {
+      color-scheme: light;
+      --bg: #f5f7fa;
+      --panel: #ffffff;
+      --text: #17202a;
+      --muted: #5f6b7a;
+      --line: #d9e0e8;
+      --accent: #12805c;
+      --accent-dark: #0b6248;
+      --danger: #a33131;
+      --danger-bg: #fff1f1;
+    }
+
+    * {
+      box-sizing: border-box;
+    }
+
+    body {
+      margin: 0;
+      min-height: 100vh;
+      background: var(--bg);
+      color: var(--text);
+      font: 16px/1.5 -apple-system, BlinkMacSystemFont, "Segoe UI", sans-serif;
+    }
+
+    main {
+      width: min(920px, calc(100% - 32px));
+      margin: 0 auto;
+      padding: 40px 0;
+    }
+
+    .shell {
+      background: var(--panel);
+      border: 1px solid var(--line);
+      border-radius: 8px;
+      box-shadow: 0 12px 28px rgba(31, 45, 61, 0.08);
+      overflow: hidden;
+    }
+
+    header {
+      padding: 28px 32px 22px;
+      border-bottom: 1px solid var(--line);
+    }
+
+    h1 {
+      margin: 0 0 8px;
+      font-size: 28px;
+      line-height: 1.15;
+      letter-spacing: 0;
+    }
+
+    p {
+      margin: 0;
+      color: var(--muted);
+    }
+
+    form {
+      padding: 28px 32px 32px;
+      display: grid;
+      gap: 22px;
+    }
+
+    label {
+      display: grid;
+      gap: 8px;
+      font-weight: 650;
+    }
+
+    .steps {
+      margin: 0;
+      padding-left: 22px;
+      color: var(--text);
+    }
+
+    .steps li {
+      margin: 6px 0;
+    }
+
+    .steps a {
+      color: var(--accent-dark);
+      font-weight: 650;
+    }
+
+    textarea,
+    input,
+    select {
+      width: 100%;
+      border: 1px solid var(--line);
+      border-radius: 6px;
+      color: var(--text);
+      background: #fff;
+      font: inherit;
+    }
+
+    textarea {
+      min-height: 220px;
+      resize: vertical;
+      padding: 14px;
+      font-family: ui-monospace, SFMono-Regular, Menlo, Consolas, monospace;
+      font-size: 14px;
+      line-height: 1.45;
+    }
+
+    input,
+    select {
+      min-height: 44px;
+      padding: 0 12px;
+    }
+
+    .grid {
+      display: grid;
+      grid-template-columns: minmax(0, 1fr) minmax(180px, 240px);
+      gap: 18px;
+      align-items: end;
+    }
+
+    .grid.single {
+      grid-template-columns: minmax(180px, 320px);
+    }
+
+    .hint {
+      color: var(--muted);
+      font-size: 14px;
+      font-weight: 400;
+    }
+
+    .hint a {
+      color: var(--accent-dark);
+      font-weight: 650;
+    }
+
+    .cookie-status {
+      min-height: 22px;
+      font-size: 14px;
+      font-weight: 500;
+    }
+
+    .cookie-status.ok {
+      color: var(--accent-dark);
+    }
+
+    .cookie-status.error {
+      color: var(--danger);
+    }
+
+    .actions {
+      display: flex;
+      gap: 12px;
+      align-items: center;
+      flex-wrap: wrap;
+    }
+
+    button {
+      min-height: 46px;
+      border: 0;
+      border-radius: 6px;
+      padding: 0 18px;
+      background: var(--accent);
+      color: #fff;
+      font: inherit;
+      font-weight: 700;
+      cursor: pointer;
+    }
+
+    button:hover {
+      background: var(--accent-dark);
+    }
+
+    .notice {
+      border: 1px solid var(--line);
+      border-radius: 6px;
+      padding: 12px 14px;
+      background: #f9fbfc;
+      color: var(--muted);
+      font-size: 14px;
+    }
+
+    .error {
+      border-color: #e7b5b5;
+      background: var(--danger-bg);
+      color: var(--danger);
+    }
+
+    @media (max-width: 680px) {
+      main {
+        width: min(100% - 20px, 920px);
+        padding: 18px 0;
+      }
+
+      header,
+      form {
+        padding-left: 18px;
+        padding-right: 18px;
+      }
+
+      .grid {
+        grid-template-columns: 1fr;
+      }
+
+      h1 {
+        font-size: 24px;
+      }
+    }
+  </style>
+</head>
+<body>
+  <main>
+    <section class="shell">
+      <header>
+        <h1>Экспорт операций Сбера</h1>
+        <p>Вставьте cookies в формате Netscape, JSON-массива или обычного Cookie header.</p>
+      </header>
+      <form method="post" action="/export">
+        __MESSAGE__
+        <div>
+          <ol class="steps">
+            <li>Установите расширение <a href="https://chromewebstore.google.com/detail/get-cookiestxt-locally/cclelndahbckbenkjhflpdbgdldlbecc?pli=1" target="_blank" rel="noopener noreferrer">Get cookies.txt LOCALLY</a>.</li>
+            <li>Откройте <a href="https://online.sberbank.ru/CSAFront/index.do" target="_blank" rel="noopener noreferrer">СберБанк Онлайн</a> и авторизуйтесь.</li>
+            <li>Скопируйте cookies через расширение и вставьте их в поле ниже.</li>
+          </ol>
+        </div>
+        <label>
+          Cookies
+          <textarea name="cookies" spellcheck="false" autocomplete="off" required placeholder="session_id=...; other_cookie=..."></textarea>
+          <span class="cookie-status" id="cookie-status" role="status" aria-live="polite"></span>
+          <span class="hint">
+            Cookies используются только для запроса к Сберу и не сохраняются.
+            Получить `cookies.txt` можно через
+            <a href="https://chromewebstore.google.com/detail/get-cookiestxt-locally/cclelndahbckbenkjhflpdbgdldlbecc?pli=1" target="_blank" rel="noopener noreferrer">Get cookies.txt LOCALLY</a>.
+          </span>
+        </label>
+        <div class="grid single">
+          <label>
+            Период
+            <input type="month" name="period" value="__PERIOD__" required>
+          </label>
+        </div>
+        <div class="actions">
+          <button type="submit">Экспортировать ZIP</button>
+          <span class="hint">В архиве будут Excel и JSON со списком всех операций.</span>
+        </div>
+      </form>
+    </section>
+  </main>
+  <script>
+    const cookiesInput = document.querySelector('textarea[name="cookies"]');
+    const statusNode = document.querySelector('#cookie-status');
+    let validateTimer = null;
+
+    function setCookieStatus(text, mode) {
+      statusNode.textContent = text;
+      statusNode.className = mode ? `cookie-status ${mode}` : 'cookie-status';
+    }
+
+    async function validateCookies() {
+      const cookies = cookiesInput.value.trim();
+      if (!cookies) {
+        setCookieStatus('', '');
+        return;
+      }
+
+      try {
+        const response = await fetch('/validate-cookies', {
+          method: 'POST',
+          headers: { 'Content-Type': 'application/x-www-form-urlencoded;charset=UTF-8' },
+          body: new URLSearchParams({ cookies }),
+        });
+        const result = await response.json();
+        if (result.ok) {
+          setCookieStatus(`Формат распознан, cookies: ${result.count}`, 'ok');
+        } else {
+          setCookieStatus(result.error || 'Cookies не удалось распарсить.', 'error');
+        }
+      } catch {
+        setCookieStatus('Не удалось проверить cookies.', 'error');
+      }
+    }
+
+    cookiesInput.addEventListener('input', () => {
+      clearTimeout(validateTimer);
+      validateTimer = setTimeout(validateCookies, 250);
+    });
+  </script>
+</body>
+</html>
+"""
+
+
+class ExportHandler(BaseHTTPRequestHandler):
+    server_version = "SberExportHTTP/1.0"
+
+    def do_GET(self) -> None:
+        if self.path not in ("/", "/index.html"):
+            self.send_error(HTTPStatus.NOT_FOUND, "Not found")
+            return
+        self._send_page()
+
+    def do_POST(self) -> None:
+        if self.path == "/validate-cookies":
+            self._handle_validate_cookies()
+            return
+
+        if self.path != "/export":
+            self.send_error(HTTPStatus.NOT_FOUND, "Not found")
+            return
+
+        try:
+            fields = self._read_form()
+            cookies_text = self._field(fields, "cookies").strip()
+            year, month = parse_period(self._field(fields, "period"))
+
+            cookie_header, _ = parse_cookie_header_from_text(cookies_text)
+            archive_name, archive_bytes, count = export_archive(cookie_header, year, month)
+        except Exception as exc:
+            self._send_page(str(exc), status=HTTPStatus.BAD_REQUEST)
+            return
+
+        self.send_response(HTTPStatus.OK)
+        self.send_header("Content-Type", "application/zip")
+        self.send_header("Content-Disposition", f'attachment; filename="{archive_name}"')
+        self.send_header("Content-Length", str(len(archive_bytes)))
+        self.send_header("X-Operations-Count", str(count))
+        self.end_headers()
+        self.wfile.write(archive_bytes)
+
+    def _handle_validate_cookies(self) -> None:
+        try:
+            fields = self._read_form()
+            _, cookie_count = parse_cookie_header_from_text(self._field(fields, "cookies").strip())
+            payload = {"ok": True, "count": cookie_count}
+            status = HTTPStatus.OK
+        except Exception as exc:
+            payload = {"ok": False, "error": str(exc)}
+            status = HTTPStatus.BAD_REQUEST
+
+        self._send_json(payload, status=status)
+
+    def log_message(self, format: str, *args: object) -> None:
+        print(f"{self.address_string()} - {format % args}")
+
+    def _read_form(self) -> dict[str, list[str]]:
+        content_length = int(self.headers.get("Content-Length", "0"))
+        if content_length > MAX_FORM_BYTES:
+            raise ValueError("Слишком большой ввод cookies.")
+
+        raw_body = self.rfile.read(content_length).decode("utf-8")
+        return parse_qs(raw_body, keep_blank_values=True)
+
+    def _field(self, fields: dict[str, list[str]], name: str, default: str = "") -> str:
+        values = fields.get(name)
+        return values[0] if values else default
+
+    def _send_page(self, message: str = "", status: HTTPStatus = HTTPStatus.OK) -> None:
+        now = datetime.now()
+        notice = ""
+        if message:
+            notice = f'<div class="notice error">{html.escape(message)}</div>'
+        body = (
+            PAGE_HTML
+            .replace("__MESSAGE__", notice)
+            .replace("__PERIOD__", f"{now.year}-{now.month:02d}")
+        )
+
+        encoded = body.encode("utf-8")
+        self.send_response(status)
+        self.send_header("Content-Type", "text/html; charset=utf-8")
+        self.send_header("Content-Length", str(len(encoded)))
+        self.end_headers()
+        self.wfile.write(encoded)
+
+    def _send_json(self, payload: dict[str, object], status: HTTPStatus = HTTPStatus.OK) -> None:
+        encoded = json.dumps(payload, ensure_ascii=False).encode("utf-8")
+        self.send_response(status)
+        self.send_header("Content-Type", "application/json; charset=utf-8")
+        self.send_header("Cache-Control", "no-store")
+        self.send_header("Content-Length", str(len(encoded)))
+        self.end_headers()
+        self.wfile.write(encoded)
+
+
+def parse_cookie_header_from_text(cookies_text: str) -> tuple[str, int]:
+    if not cookies_text:
+        raise ValueError("Вставьте cookies.")
+
+    cookies = parse_cookies(cookies_text, input_format="auto")
+    cookie_header = build_cookie_header(cookies)
+    if not cookie_header:
+        raise ValueError("Не удалось получить ни одной cookie из введенного текста.")
+
+    return cookie_header, len(cookies)
+
+
+def parse_period(value: str) -> tuple[int, int]:
+    match = PERIOD_RE.match(value.strip())
+    if not match:
+        raise ValueError("Выберите период в формате ГГГГ-ММ.")
+
+    year = int(match.group(1))
+    month = int(match.group(2))
+    if month < 1 or month > 12:
+        raise ValueError("Месяц должен быть от 1 до 12.")
+    return year, month
+
+
+def export_archive(cookie_header: str, year: int, month: int) -> tuple[str, bytes, int]:
+    timestamp = datetime.now().strftime("%Y%m%d_%H%M%S_%f")
+    base_name = f"sber_operations_{year}_{month:02d}_{timestamp}_{uuid4().hex[:8]}"
+    EXPORT_DIR.mkdir(parents=True, exist_ok=True)
+
+    archive_path = EXPORT_DIR / f"{base_name}.zip"
+
+    with TemporaryDirectory() as tmp_dir:
+        tmp_path = Path(tmp_dir)
+        xlsx_path = tmp_path / f"{base_name}.xlsx"
+        json_path = tmp_path / f"{base_name}.json"
+
+        exporter = SberMonthHistoryExporter(cookie=cookie_header)
+        _, _, count = exporter.save_month_bundle(year, month, xlsx_path, json_path)
+
+        with zipfile.ZipFile(archive_path, "w", compression=zipfile.ZIP_DEFLATED) as archive:
+            archive.write(xlsx_path, arcname=xlsx_path.name)
+            archive.write(json_path, arcname=json_path.name)
+
+    return archive_path.name, archive_path.read_bytes(), count
+
+
+def create_server(host: str, port: int) -> ThreadingHTTPServer:
+    return ThreadingHTTPServer((host, port), ExportHandler)
+
+
+def parse_args() -> argparse.Namespace:
+    parser = argparse.ArgumentParser(description="Run web UI for Sber operations export.")
+    parser.add_argument("--host", default="127.0.0.1", help="Host to bind. Default: 127.0.0.1")
+    parser.add_argument("--port", type=int, default=8000, help="Port to bind. Default: 8000")
+    return parser.parse_args()
+
+
+def main() -> int:
+    args = parse_args()
+    server = create_server(args.host, args.port)
+    print(f"Open http://{args.host}:{args.port}")
+    try:
+        server.serve_forever()
+    except KeyboardInterrupt:
+        print("\nStopping server")
+    finally:
+        server.server_close()
+    return 0
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())