make book language configurable (#1)

book_manager.py

@@ -7,7 +7,7 @@ from bs4 import BeautifulSoup
 from io import BytesIO
 
 from logger import setup_logger
-from config import SUPPORTED_FORMATS
+from config import SUPPORTED_FORMATS, BOOK_LANGUAGE
 from models import BookInfo
 import network
 
@@ -28,8 +28,8 @@ def search_books(query: str) -> List[BookInfo]:
     query_html = quote(query)
     url = (
         f"https://annas-archive.org/search?index=&page=1&display=table"
-        f"&acc=aa_download&acc=external_download&lang=en&sort="
-        f"&ext={'&ext='.join(SUPPORTED_FORMATS)}&lang=en&q={query_html}"
+        f"&acc=aa_download&acc=external_download&lang={BOOK_LANGUAGE}&sort="
+        f"&ext={'&ext='.join(SUPPORTED_FORMATS)}&lang={BOOK_LANGUAGE}&q={query_html}"
     )
     
     html = network.html_get_page(url)

config.py

@@ -29,6 +29,8 @@ CLOUDFLARE_PROXY = os.getenv("CLOUDFLARE_PROXY_URL", "http://localhost:8000")
 SUPPORTED_FORMATS = os.getenv("SUPPORTED_FORMATS", "epub,mobi,azw3,fb2,djvu,cbz,cbr")
 SUPPORTED_FORMATS = SUPPORTED_FORMATS.split(",")
 
+BOOK_LANGUAGE = os.getenv("BOOK_LANGUAGE", "en")
+
 # API settings
 FLASK_HOST = os.getenv("FLASK_HOST",  "0.0.0.0")
 FLASK_PORT = int(os.getenv("FLASK_PORT", 5003))

docker-compose.yml

@@ -6,6 +6,7 @@ services:
       FLASK_DEBUG: false
       CLOUDFLARE_PROXY_URL: http://cloudflarebypassforscraping:8000
       INGEST_DIR: /cwa-book-ingest
+      BOOK_LANGUAGE: en
     ports:
       - "${FLASK_PORT:-8084}:${FLASK_PORT:-8084}"
     healthcheck:

readme.md

@@ -60,6 +60,7 @@ An intuitive web interface for searching and requesting book downloads, designed
 | `DEFAULT_SLEEP` | Retry delay (seconds) | `5` |
 | `MAIN_LOOP_SLEEP_TIME` | Processing loop delay (seconds) | `5` |
 | `SUPPORTED_FORMATS` | Supported book formats | `epub,mobi,azw3,fb2,djvu,cbz,cbr` |
+| `BOOK_LANGUAGE` | Preferred language for books | `en` |
 
 Note that PDF are NOT supported at the moment (they do not get ingested by CWA, but if you want to just dowload them loclaly, you can add `pdf` to the `SUPPORTED_FORMATS` env