List all chapters on mangadex, use correct download server

2 files changed, 50 insertions, 20 deletions

diff --git a/automedia.py b/automedia.py
index 4158d44..5670988 100755
--- a/automedia.py
+++ b/automedia.py
@@ -582,6 +582,10 @@ def sync(rss_config_dir, html_config_dir, download_dir, sync_rate_sec):
             sync_rss(rss, tc)
             # Add last synced timestamp. This together with "updated" file is used to remove series
             # that haven't updated in a long time (either finished series or axes series)
+
+            # TODO: Fix. This will fail if the path is removed while its running,
+            # same with other places.
+            # To fix, add a command to remove tracked items safely
             with open(os.path.join(rss_tracked_dir, rss.title, "synced"), "w") as file:
                 file.write(str(time.time()))
             #else:
diff --git a/plugins/mangadex.py b/plugins/mangadex.py
index 238a264..4890417 100755
--- a/plugins/mangadex.py
+++ b/plugins/mangadex.py
@@ -9,6 +9,10 @@ import re
 
 from lxml import etree
 
+headers = {
+    'User-Agent': "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/80.0.3987.163 Safari/537.36"
+}
+
 def usage():
     print("mangadex.py command")
     print("commands:")
@@ -32,7 +36,7 @@ if len(sys.argv) < 2:
     usage()
 
 def download_file(url, save_path):
-    with requests.get(url, stream=True) as response:
+    with requests.get(url, headers=headers, stream=True) as response:
         if not response.ok:
             return False
         with open(save_path, "wb") as file:
@@ -46,6 +50,9 @@ def title_url_extract_manga_id(url):
     if result and len(result.groups()) > 0:
         return result.groups()[0]
 
+def chapter_sort_func(chapter_data):
+    return chapter_data[1].get("timestamp", 0)
+
 # TODO: Support pagination. Currently only the n latest chapters are listed,
 # but going through all pages might be too slow for large manga like naruto
 def list_chapters(url, chapter_list_input):
@@ -54,7 +61,8 @@ def list_chapters(url, chapter_list_input):
         print("Failed to extract manga id from url: %s. Note: url is expected to be in this format: mangadex.org/title/<number>/..." % url)
         exit(2)
 
-    response = requests.get(url)
+    url = "https://mangadex.org/api/?id=%s&type=manga" % manga_id;
+    response = requests.get(url, headers=headers)
     if response.status_code != 200:
         print("Failed to list chapters, server responded with status code %d" % response.status_code)
         exit(2)
@@ -71,29 +79,46 @@ def list_chapters(url, chapter_list_input):
         if url and len(url) > 0:
             seen_urls.add(url)
 
-    lang = "1" # english
+    lang = "gb" # english
+
+    json_response = response.json()
+    status = json_response["status"]
+    if status != "OK":
+        print("Expected server response OK, got %s" % status)
+        exit(2)
 
-    tree = etree.HTML(response.text)
+    chapter_json = json_response["chapter"]
     chapters = []
-    prev_chapter_number = None
-    for element in tree.xpath("//div[@data-manga-id='%s']" % manga_id):
-        chapter_lang = element.attrib.get("data-lang")
-        if chapter_lang != lang:
+    for key, value in chapter_json.items():
+        chapters.append((key, value))
+
+    chapters = sorted(chapters, key=chapter_sort_func, reverse=True)
+    prev_chapter_number = ""
+    output_chapters = []
+    for chapter_data in chapters:
+        chapter_id = chapter_data[0]
+        chapter = chapter_data[1]
+
+        lang_code = chapter.get("lang_code", "")
+        if lang_code != lang:
             continue
-        chapter_id = element.attrib.get("data-id")
-        chapter_url = "https://mangadex.org/chapter/%s" % chapter_id
-        chapter_title = element.attrib.get("data-title") # optional
-        chapter_number = element.attrib.get("data-chapter")
-        if prev_chapter_number and chapter_number == prev_chapter_number:
+
+        chapter_number_str = chapter.get("chapter", "0")
+        if chapter_number_str == prev_chapter_number:
             continue
-        prev_chapter_number = chapter_number
-        chapter_name = "Ch. %s" % chapter_number
+        prev_chapter_number = chapter_number_str
+
+        chapter_title = chapter.get("title")
+        chapter_url = "https://mangadex.org/chapter/" + chapter_id
+        chapter_name = "Ch. " + chapter_number_str
         if chapter_title:
-            chapter_name += " - %s" % chapter_title
+            chapter_name += " - " + chapter_title
+
         if chapter_title.lower().replace(" ", "") in seen_titles or chapter_url in seen_urls:
             break
-        chapters.append({ "name": chapter_name, "url": chapter_url })
-    print(json.dumps(chapters))
+
+        output_chapters.append({ "name": chapter_name, "url": chapter_url })
+    print(json.dumps(output_chapters))
 
 def chapter_url_extract_manga_id(url):
     result = re.search("mangadex.org/chapter/([0-9]+)", url)
@@ -109,7 +134,7 @@ def download_chapter(url, download_dir):
 
     url = "https://mangadex.org/api/?id=%s&server=null&type=chapter" % manga_id
 
-    response = requests.get(url)
+    response = requests.get(url, headers=headers)
     if response.status_code != 200:
         print("Failed to list chapter images, server responded with status code %d" % response.status_code)
         exit(2)
@@ -126,8 +151,9 @@ def download_chapter(url, download_dir):
         exit(2)
 
     chapter_hash = json_response["hash"]
+    server = json_response.get("server", "https://mangadex.org/data/")
     for image_name in json_response["page_array"]:
-        image_url = "https://mangadex.org/data/%s/%s" % (chapter_hash, image_name)
+        image_url = "%s%s/%s" % (server, chapter_hash, image_name)
         ext = image_url[image_url.rfind("."):]
         image_name = str(img_number) + ext
         image_path = os.path.join(download_dir, image_name)