Merge pull request #58 from Euklios/master

Use rss requests for nyaa frontpage

NyaaPy/nyaa.py

@@ -37,20 +37,29 @@ class Nyaa:
             user_uri = ""
 
         if page > 0:
-            r = requests.get("{}/{}?f={}&c={}_{}&q={}&p={}".format(
-                url, user_uri, filters, category, subcategory, keyword,
-                page))
+            uri = f"{url}/{user_uri}?f={filters}&c={category}_{subcategory}&q={keyword}&p={page}"
         else:
-            r = requests.get("{}/{}?f={}&c={}_{}&q={}".format(
-                url, user_uri, filters, category, subcategory, keyword))
+            uri = f"{url}/{user_uri}?f={filters}&c={category}_{subcategory}&q={keyword}"
 
-        r.raise_for_status()
+        if not user:
+            uri += "&page=rss"
 
-        json_data = utils.parse_nyaa(
-            request_text=r.text,
-            limit=None,
-            site=self.SITE
-        )
+        http_response = requests.get(uri)
+
+        http_response.raise_for_status()
+
+        if user:
+            json_data = utils.parse_nyaa(
+                request_text=http_response.text,
+                limit=None,
+                site=self.SITE
+            )
+        else:
+            json_data = utils.parse_nyaa_rss(
+                request_text=http_response.text,
+                limit=None,
+                site=self.SITE
+            )
 
         return torrent.json_to_class(json_data)
 

NyaaPy/utils.py

@@ -1,4 +1,7 @@
+import urllib
 from enum import Enum
+from urllib.parse import urlencode
+
 from lxml import etree
 
 
@@ -80,6 +83,48 @@ def nyaa_categories(b):
     return category_name
 
 
+def parse_nyaa_rss(request_text, limit, site):
+    """
+    Extracts torrent information from a given rss response.
+    """
+    root = etree.fromstring(request_text)
+    torrents = []
+
+    for item in root.xpath("channel/item")[:limit]:
+        # Decide category.
+        if site in [TorrentSite.NYAASI, TorrentSite.NYAANET]:
+            category = item.findtext("nyaa:categoryId", namespaces=item.nsmap)
+        elif site in [TorrentSite.SUKEBEINYAASI, TorrentSite.SUKEBEINYAANET]:
+            category = item.findtext("nyaa:categoryId", namespaces=item.nsmap)
+        else:
+            raise ValueError("Unknown TorrentSite received!")
+
+        try:
+            is_remake = item.findtext("nyaa:remake", namespaces=item.nsmap) == "Yes"
+            is_trusted = item.findtext("nyaa:trusted", namespaces=item.nsmap) == "Yes"
+            item_type = "remake" if is_remake else "trusted" if is_trusted else "default"
+
+            torrent = {
+                'id': item.findtext("guid").split("/")[-1],
+                'category': category,
+                'url': item.findtext("guid"),
+                'name': item.findtext("title"),
+                'download_url': item.findtext("link"),
+                'magnet': magnet_builder(item.findtext("nyaa:infoHash", namespaces=item.nsmap), item.findtext("title")),
+                'size': item.findtext("nyaa:size", namespaces=item.nsmap),
+                'date': item.findtext("pubDate"),
+                'seeders': item.findtext("nyaa:seeders", namespaces=item.nsmap),
+                'leechers': item.findtext("nyaa:leechers", namespaces=item.nsmap),
+                'completed_downloads': None,
+                'type': item_type
+            }
+            torrents.append(torrent)
+        except IndexError:
+            pass
+
+    return torrents
+
+
 def parse_nyaa(request_text, limit, site):
     parser = etree.HTMLParser()
     tree = etree.fromstring(request_text, parser)
@@ -230,6 +275,25 @@ def sukebei_categories(b):
     return category_name
 
 
+def magnet_builder(info_hash, title):
+    """
+    Generates a magnet link using the info_hash and title of a given file.
+    """
+    known_trackers = [
+        "http://nyaa.tracker.wf:7777/announce",
+        "udp://open.stealth.si:80/announce",
+        "udp://tracker.opentrackr.org:1337/announce",
+        "udp://exodus.desync.com:6969/announce",
+        "udp://tracker.torrent.eu.org:451/announce"
+    ]
+
+    magnet_link = f"magnet:?xt=urn:btih:{info_hash}&" + urlencode({"dn": title}, quote_via=urllib.parse.quote)
+    for tracker in known_trackers:
+        magnet_link += f"&{urlencode({'tr': tracker})}"
+
+    return magnet_link
+
+
 # Pantsu Utils
 def query_builder(q, params):
     available_params = ["category", "page", "limit", "userID", "fromID",