Fixing the title and book details from AA (#289)

Should fix #288

.github/workflows/build-and-publish-docker-image.yml

@@ -70,6 +70,7 @@ jobs:
           push: ${{ github.event_name != 'pull_request' }}
           build-args: |
             BUILD_VERSION=${{ steps.date.outputs.date }}-${{ github.sha }}
+            RELEASE_VERSION=${{ github.ref_name }}
           tags: ${{ steps.meta.outputs.tags }}
           labels: ${{ steps.meta.outputs.labels }}
       

Dockerfile

@@ -4,6 +4,8 @@ FROM python:3.10-slim AS base
 # Add build argument for version
 ARG BUILD_VERSION
 ENV BUILD_VERSION=${BUILD_VERSION}
+ARG RELEASE_VERSION
+ENV RELEASE_VERSION=${RELEASE_VERSION}
 
 # Set shell to bash with pipefail option
 SHELL ["/bin/bash", "-o", "pipefail", "-c"]

app.py

@@ -12,8 +12,8 @@ from flask import url_for as flask_url_for
 import typing
 
 from logger import setup_logger
-from config import _SUPPORTED_BOOK_LANGUAGE, BOOK_LANGUAGE
-from env import FLASK_HOST, FLASK_PORT, APP_ENV, CWA_DB_PATH, DEBUG, USING_EXTERNAL_BYPASSER
+from config import _SUPPORTED_BOOK_LANGUAGE, BOOK_LANGUAGE, SUPPORTED_FORMATS
+from env import FLASK_HOST, FLASK_PORT, APP_ENV, CWA_DB_PATH, DEBUG, USING_EXTERNAL_BYPASSER, BUILD_VERSION, RELEASE_VERSION
 import backend
 
 from models import SearchFilters
@@ -103,7 +103,15 @@ def index() -> str:
     """
     Render main page with search and status table.
     """
-    return render_template('index.html', book_languages=_SUPPORTED_BOOK_LANGUAGE, default_language=BOOK_LANGUAGE, debug=DEBUG)
+    return render_template('index.html', 
+                           book_languages=_SUPPORTED_BOOK_LANGUAGE, 
+                           default_language=BOOK_LANGUAGE, 
+                           supported_formats=SUPPORTED_FORMATS,
+                           debug=DEBUG,
+                           build_version=BUILD_VERSION,
+                           release_version=RELEASE_VERSION,
+                           app_env=APP_ENV
+                           )
 
  
 

backend.py

@@ -81,6 +81,12 @@ def queue_status() -> Dict[str, Dict[str, Any]]:
         Dict: Queue status organized by status type
     """
     status = book_queue.get_status()
+    for _, books in status.items():
+        for _, book_info in books.items():
+            if book_info.download_path:
+                if not os.path.exists(book_info.download_path):
+                    book_info.download_path = None
+
     # Convert Enum keys to strings and properly format the response
     return {
         status_type.value: books

book_manager.py

@@ -10,7 +10,7 @@ from bs4 import BeautifulSoup, Tag, NavigableString, ResultSet
 import downloader
 from logger import setup_logger
 from config import SUPPORTED_FORMATS, BOOK_LANGUAGE, AA_BASE_URL
-from env import AA_DONATOR_KEY, USE_CF_BYPASS, PRIORITIZE_WELIB
+from env import AA_DONATOR_KEY, USE_CF_BYPASS, PRIORITIZE_WELIB, ALLOW_USE_WELIB
 from models import BookInfo, SearchFilters
 logger = setup_logger(__name__)
 
@@ -169,8 +169,6 @@ def _parse_book_info_page(soup: BeautifulSoup, book_id: str) -> BookInfo:
 
     data = soup.find_all("div", {"class": "main-inner"})[0].find_next("div")
     divs = list(data.children)
-    format = divs[13].text.split(" · ")[-6].strip().lower()
-    size = divs[13].text.split(" · ")[-5].strip().lower()
 
     every_url = soup.find_all("a")
     slow_urls_no_waitlist = set()
@@ -224,20 +222,49 @@ def _parse_book_info_page(soup: BeautifulSoup, book_id: str) -> BookInfo:
     # Remove empty urls
     urls = [url for url in urls if url != ""]
 
+    # Filter out divs that are not text
+    original_divs = divs
+    divs = [div.text.strip() for div in divs if div.text.strip() != ""]
+
+    separator_index = 6
+    for i, div in enumerate(divs):
+        if "·" in div.strip():
+            separator_index = i
+            break
+            
+    _details = divs[separator_index].lower().split(" · ")
+    format = ""
+    size = ""
+    for f in _details:
+        if format == "" and f.strip().lower() in SUPPORTED_FORMATS:
+            format = f.strip().lower()
+        if size == "" and any(u in f.strip().lower() for u in ["mb", "kb", "gb"]):
+            size = f.strip().lower()
+
+    if format == "" or size == "":
+        for f in _details:
+            if f == "" and not " " in f.strip().lower():
+                format = f.strip().lower()
+            if size == "" and "." in f.strip().lower():
+                size = f.strip().lower()
+
+    
+    book_title = divs[separator_index-3].strip("🔍")
+
     # Extract basic information
     book_info = BookInfo(
         id=book_id,
         preview=preview,
-        title=divs[7].next.strip(),
-        publisher=divs[11].text.strip(),
-        author=divs[9].text.strip(),
+        title=book_title,
+        publisher=divs[separator_index-1],
+        author=divs[separator_index-2],
         format=format,
         size=size,
         download_urls=urls,
     )
 
     # Extract additional metadata
-    info = _extract_book_metadata(divs[-6])
+    info = _extract_book_metadata(original_divs[-6])
     book_info.info = info
 
     # Set language and year from metadata if available
@@ -249,6 +276,8 @@ def _parse_book_info_page(soup: BeautifulSoup, book_id: str) -> BookInfo:
     return book_info
 
 def _get_download_urls_from_welib(book_id: str) -> set[str]:
+    if ALLOW_USE_WELIB == False:
+        return set()
     """Get download urls from welib.org."""
     url = f"https://welib.org/md5/{book_id}"
     logger.info(f"Getting download urls from welib.org for {book_id}. While this uses the bypasser, it will not start downloading them yet.")

entrypoint.sh

@@ -20,6 +20,7 @@ set -e
 
 # Print build version
 echo "Build version: $BUILD_VERSION"
+echo "Release version: $RELEASE_VERSION"
 
 # Configure timezone
 if [ "$TZ" ]; then

env.py

@@ -26,7 +26,13 @@ _CUSTOM_SCRIPT = os.getenv("CUSTOM_SCRIPT", "").strip()
 FLASK_HOST = os.getenv("FLASK_HOST", "0.0.0.0")
 FLASK_PORT = int(os.getenv("FLASK_PORT", "8084"))
 DEBUG = string_to_bool(os.getenv("DEBUG", "false"))
+APP_ENV = os.getenv("APP_ENV", "N/A").lower()
 PRIORITIZE_WELIB = string_to_bool(os.getenv("PRIORITIZE_WELIB", "false"))
+ALLOW_USE_WELIB = string_to_bool(os.getenv("ALLOW_USE_WELIB", "true"))
+
+# Version information from Docker build
+BUILD_VERSION = os.getenv("BUILD_VERSION", "N/A")
+RELEASE_VERSION = os.getenv("RELEASE_VERSION", "N/A")
 
 # If debug is true, we want to log everything
 if DEBUG:
@@ -41,7 +47,7 @@ DOCKERMODE = string_to_bool(os.getenv("DOCKERMODE", "false"))
 _CUSTOM_DNS = os.getenv("CUSTOM_DNS", "").strip()
 USE_DOH = string_to_bool(os.getenv("USE_DOH", "false"))
 BYPASS_RELEASE_INACTIVE_MIN = int(os.getenv("BYPASS_RELEASE_INACTIVE_MIN", "5"))
-APP_ENV = os.getenv("APP_ENV", "prod").lower()
+
 # Logging settings
 LOG_FILE = LOG_DIR / "cwa-book-downloader.log"
 

genDebug.sh

@@ -3,7 +3,7 @@
 # Set up log paths
 LOG_ROOT=${LOG_ROOT:-"/var/log"}
 LOG_DIR="$LOG_ROOT/cwa-book-downloader"
-OUTPUT_FILE_NAME="cwa-book-downloader-debug_BUILD-${BUILD_VERSION:-local}_$(date +%Y%m%d-%H%M%S)"
+OUTPUT_FILE_NAME="cwa-book-downloader-debug_BUILD-${BUILD_VERSION:-local}_RELEASE-${RELEASE_VERSION:-NA}_$(date +%Y%m%d-%H%M%S)"
 OUTPUT_FILE="/tmp/$OUTPUT_FILE_NAME.zip"
 
 # Create LOG_DIR if it doesn't exist
@@ -125,16 +125,16 @@ fi
 env | grep -v -E "(AA_DONATOR_KEY)" | sort > "$LOG_DIR/environment.txt"
 
 echo "--- HTTPBin ---" > $LOG_DIR/network_info.txt
-pyrequests https://httpbin.org/get >> $LOG_DIR/network_info.txt
+curl -s https://httpbin.org/get >> $LOG_DIR/network_info.txt
 ehco ""
 echo "--- HowsMySSL ---" >> $LOG_DIR/network_info.txt
-pyrequests https://www.howsmyssl.com/a/check >> $LOG_DIR/network_info.txt
+curl -s https://www.howsmyssl.com/a/check >> $LOG_DIR/network_info.txt
 ehco ""
 echo "--- IPInfo ---" >> $LOG_DIR/network_info.txt
-pyrequests https://ipinfo.io >> $LOG_DIR/network_info.txt
+curl -s https://ipinfo.io >> $LOG_DIR/network_info.txt
 ehco ""
 echo "--- Cloudflare Trace ---" >> $LOG_DIR/network_info.txt
-pyrequests https://1.1.1.1/cdn-cgi/trace >> $LOG_DIR/network_info.txt
+curl -s https://1.1.1.1/cdn-cgi/trace >> $LOG_DIR/network_info.txt
 
 # Create the zip file directly from LOG_DIR
 ln -s "$LOG_DIR" /tmp/$OUTPUT_FILE_NAME

readme.md

@@ -83,6 +83,7 @@ Note that if using TOR, the TZ will be calculated automatically based on IP.
 | `AA_DONATOR_KEY`       | Optional Donator key for Anna's Archive fast download API | ``                                |
 | `USE_BOOK_TITLE`       | Use book title as filename instead of ID                  | `false`                           |
 | `PRIORITIZE_WELIB`     | When downloading, download from WELIB first instead of AA | `false`                           |
+| `ALLOW_USE_WELIB`       | Allow usage of welib for downloading books if found there | `true`                            |
 
 If you change `BOOK_LANGUAGE`, you can add multiple comma separated languages, such as `en,fr,ru` etc.  
 

static/js/main.js

@@ -94,8 +94,8 @@
 
   // ---- Cards ----
   function renderCard(book) {
-    const cover = book.preview ? `<img src="${utils.e(book.preview)}" alt="Cover" class="w-full h-44 object-cover rounded">` :
-      `<div class="w-full h-44 rounded flex items-center justify-center opacity-70" style="background: var(--bg-soft)">No Cover</div>`;
+    const cover = book.preview ? `<img src="${utils.e(book.preview)}" alt="Cover" class="w-full h-88 object-cover rounded">` :
+      `<div class="w-full h-88 rounded flex items-center justify-center opacity-70" style="background: var(--bg-soft)">No Cover</div>`;
 
     const html = `
       <article class="rounded border p-3 flex flex-col gap-3" style="border-color: var(--border-muted); background: var(--bg-soft)">
@@ -172,7 +172,7 @@
       }
     },
     tpl(book) {
-      const cover = book.preview ? `<img src="${utils.e(book.preview)}" alt="Cover" class="w-full h-56 object-cover rounded">` : '';
+      const cover = book.preview ? `<img src="${utils.e(book.preview)}" alt="Cover" class="w-full h-88 object-cover rounded">` : '';
       const infoList = book.info ? Object.entries(book.info).map(([k, v]) => `<li><strong>${utils.e(k)}:</strong> ${utils.e((v||[]).join 
         ? v.join(', ') : v)}</li>`).join('') : '';
       return `
@@ -229,6 +229,10 @@
       for (const [name, items] of Object.entries(data || {})) {
         if (!items || Object.keys(items).length === 0) continue;
         const rows = Object.values(items).map((b) => {
+          const titleText = utils.e(b.title) || '-';
+          const maybeLinkedTitle = b.download_path
+            ? `<a href="/request/api/localdownload?id=${encodeURIComponent(b.id)}" class="text-blue-600 hover:underline">${titleText}</a>`
+            : titleText;
           const actions = (name === 'queued' || name === 'downloading')
             ? `<button class="px-2 py-1 rounded border text-xs" data-cancel="${utils.e(b.id)}" style="border-color: var(--border-muted);">Cancel</button>`
             : '';
@@ -236,7 +240,7 @@
             ? `<div class="h-2 bg-black/10 rounded overflow-hidden"><div class="h-2 bg-blue-600" style="width:${Math.round(b.progress)}%"></div></div>`
             : '';
           return `<li class="p-3 rounded border flex flex-col gap-2" style="border-color: var(--border-muted); background: var(--bg-soft)">
-            <div class="text-sm"><span class="opacity-70">${utils.e(name)}</span> • <strong>${utils.e(b.title || '-') }</strong></div>
+            <div class="text-sm"><span class="opacity-70">${utils.e(name)}</span> • <strong>${maybeLinkedTitle}</strong></div>
             ${progress}
             <div class="flex items-center gap-2">${actions}</div>
           </li>`;

templates/index.html

@@ -132,14 +132,38 @@
                     <div class="md:col-span-2 lg:col-span-3">
                         <label class="block text-sm mb-1 opacity-80">Formats</label>
                         <div class="flex flex-wrap gap-3 text-sm">
-                            <label class="inline-flex items-center gap-2"><input type="checkbox" id="format-pdf" value="pdf"> PDF</label>
-                            <label class="inline-flex items-center gap-2"><input type="checkbox" id="format-epub" value="epub" checked> EPUB</label>
-                            <label class="inline-flex items-center gap-2"><input type="checkbox" id="format-mobi" value="mobi" checked> MOBI</label>
-                            <label class="inline-flex items-center gap-2"><input type="checkbox" id="format-azw3" value="azw3" checked> AZW3</label>
-                            <label class="inline-flex items-center gap-2"><input type="checkbox" id="format-fb2" value="fb2" checked> FB2</label>
-                            <label class="inline-flex items-center gap-2"><input type="checkbox" id="format-djvu" value="djvu" checked> DJVU</label>
-                            <label class="inline-flex items-center gap-2"><input type="checkbox" id="format-cbz" value="cbz" checked> CBZ</label>
-                            <label class="inline-flex items-center gap-2"><input type="checkbox" id="format-cbr" value="cbr" checked> CBR</label>
+                            <label class="inline-flex items-center gap-2 {{ 'opacity-50 cursor-not-allowed' if 'pdf' not in supported_formats else '' }}">
+                                <input type="checkbox" id="format-pdf" value="pdf" {% if 'pdf' not in supported_formats %}disabled{% endif %}>
+                                PDF
+                            </label>
+                            <label class="inline-flex items-center gap-2 {{ 'opacity-50 cursor-not-allowed' if 'epub' not in supported_formats else '' }}">
+                                <input type="checkbox" id="format-epub" value="epub" {% if 'epub' in supported_formats %}checked{% endif %} {% if 'epub' not in supported_formats %}disabled{% endif %}>
+                                EPUB
+                            </label>
+                            <label class="inline-flex items-center gap-2 {{ 'opacity-50 cursor-not-allowed' if 'mobi' not in supported_formats else '' }}">
+                                <input type="checkbox" id="format-mobi" value="mobi" {% if 'mobi' in supported_formats %}checked{% endif %} {% if 'mobi' not in supported_formats %}disabled{% endif %}>
+                                MOBI
+                            </label>
+                            <label class="inline-flex items-center gap-2 {{ 'opacity-50 cursor-not-allowed' if 'azw3' not in supported_formats else '' }}">
+                                <input type="checkbox" id="format-azw3" value="azw3" {% if 'azw3' in supported_formats %}checked{% endif %} {% if 'azw3' not in supported_formats %}disabled{% endif %}>
+                                AZW3
+                            </label>
+                            <label class="inline-flex items-center gap-2 {{ 'opacity-50 cursor-not-allowed' if 'fb2' not in supported_formats else '' }}">
+                                <input type="checkbox" id="format-fb2" value="fb2" {% if 'fb2' in supported_formats %}checked{% endif %} {% if 'fb2' not in supported_formats %}disabled{% endif %}>
+                                FB2
+                            </label>
+                            <label class="inline-flex items-center gap-2 {{ 'opacity-50 cursor-not-allowed' if 'djvu' not in supported_formats else '' }}">
+                                <input type="checkbox" id="format-djvu" value="djvu" {% if 'djvu' in supported_formats %}checked{% endif %} {% if 'djvu' not in supported_formats %}disabled{% endif %}>
+                                DJVU
+                            </label>
+                            <label class="inline-flex items-center gap-2 {{ 'opacity-50 cursor-not-allowed' if 'cbz' not in supported_formats else '' }}">
+                                <input type="checkbox" id="format-cbz" value="cbz" {% if 'cbz' in supported_formats %}checked{% endif %} {% if 'cbz' not in supported_formats %}disabled{% endif %}>
+                                CBZ
+                            </label>
+                            <label class="inline-flex items-center gap-2 {{ 'opacity-50 cursor-not-allowed' if 'cbr' not in supported_formats else '' }}">
+                                <input type="checkbox" id="format-cbr" value="cbr" {% if 'cbr' in supported_formats %}checked{% endif %} {% if 'cbr' not in supported_formats %}disabled{% endif %}>
+                                CBR
+                            </label>
                         </div>
                     </div>
                     <div class="md:col-span-2 lg:col-span-3 flex justify-end">
@@ -193,7 +217,12 @@
 
     <footer class="mt-10 border-t pt-6 pb-10" style="border-color: var(--border-muted); background: var(--footer-bg);">
         <div class="max-w-7xl mx-auto px-4 sm:px-6 lg:px-8 flex items-center justify-between">
-            <p class="text-sm opacity-80">Calibre Web Book Downloader</p>
+            <div>
+                <p class="text-sm opacity-80">Calibre Web Book Downloader</p>
+                <p class="text-xs opacity-60 mt-1">
+                    Build: {{ build_version }}  • Release: {{ release_version }}  • Env: {{ app_env }}
+                </p>
+            </div>
             <a href="https://github.com/calibrain/calibre-web-automated-book-downloader" class="opacity-80 hover:opacity-100" aria-label="GitHub">
                 <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 16 16" fill="currentColor" class="w-6 h-6">
                     <path d="M8 0C3.58 0 0 3.58 0 8a8 8 0 005.47 7.59c.4.07.55-.17.55-.38

tor.sh

@@ -14,6 +14,7 @@ set -e
 echo "[*] Running tor script..."
 
 echo "Build version: $BUILD_VERSION"
+echo "Release version: $RELEASE_VERSION"
 
 echo "[*] Installing Tor and dependencies..."
 echo "[*] Writing Tor transparent proxy config..."