From 5f25f348f9eb5db842b1ec6799f95bebb7ba35a7 Mon Sep 17 00:00:00 2001
From: Simon Sawicki <contact@grub4k.xyz>
Date: Tue, 23 Jan 2024 23:20:13 +0100
Subject: [PATCH] [ie/pr0gramm] Enable POL filter and provide tags without
 login (#9051)

Authored by: Grub4K
---
 yt_dlp/extractor/pr0gramm.py | 41 ++++++++++++++++++++++++------------
 1 file changed, 27 insertions(+), 14 deletions(-)

diff --git a/yt_dlp/extractor/pr0gramm.py b/yt_dlp/extractor/pr0gramm.py
index 2a6794208..36e415f4a 100644
--- a/yt_dlp/extractor/pr0gramm.py
+++ b/yt_dlp/extractor/pr0gramm.py
@@ -18,7 +18,6 @@
 class Pr0grammIE(InfoExtractor):
     _VALID_URL = r'https?://pr0gramm\.com\/(?:[^/?#]+/)+(?P<id>[\d]+)(?:[/?#:]|$)'
     _TESTS = [{
-        # Tags require account
         'url': 'https://pr0gramm.com/new/video/5466437',
         'info_dict': {
             'id': '5466437',
@@ -36,7 +35,6 @@ class Pr0grammIE(InfoExtractor):
             '_old_archive_ids': ['pr0grammstatic 5466437'],
         },
     }, {
-        # Tags require account
         'url': 'https://pr0gramm.com/new/3052805:comment28391322',
         'info_dict': {
             'id': '3052805',
@@ -71,6 +69,23 @@ class Pr0grammIE(InfoExtractor):
             'thumbnail': r're:^https://thumb\.pr0gramm\.com/.*\.jpg',
             '_old_archive_ids': ['pr0grammstatic 5848332'],
         },
+    }, {
+        'url': 'https://pr0gramm.com/top/5895149',
+        'info_dict': {
+            'id': '5895149',
+            'ext': 'mp4',
+            'title': 'pr0gramm-5895149 by algoholigSeeManThrower',
+            'tags': 'count:19',
+            'uploader': 'algoholigSeeManThrower',
+            'uploader_id': 457556,
+            'upload_timestamp': 1697580902,
+            'upload_date': '20231018',
+            'like_count': int,
+            'dislike_count': int,
+            'age_limit': 0,
+            'thumbnail': 'https://thumb.pr0gramm.com/2023/10/18/db47bb3db5e1a1b3.jpg',
+            '_old_archive_ids': ['pr0grammstatic 5895149'],
+        },
     }, {
         'url': 'https://pr0gramm.com/static/5466437',
         'only_matching': True,
@@ -92,15 +107,15 @@ def _is_logged_in(self):
     def _maximum_flags(self):
         # We need to guess the flags for the content otherwise the api will raise an error
         # We can guess the maximum allowed flags for the account from the cookies
-        # Bitflags are (msbf): nsfp, nsfl, nsfw, sfw
-        flags = 0b0001
+        # Bitflags are (msbf): pol, nsfp, nsfl, nsfw, sfw
+        flags = 0b10001
         if self._is_logged_in:
-            flags |= 0b1000
+            flags |= 0b01000
             cookies = self._get_cookies(self.BASE_URL)
             if 'me' not in cookies:
                 self._download_webpage(self.BASE_URL, None, 'Refreshing verification information')
             if traverse_obj(cookies, ('me', {lambda x: x.value}, {unquote}, {json.loads}, 'verified')):
-                flags |= 0b0110
+                flags |= 0b00110
 
         return flags
 
@@ -134,14 +149,12 @@ def _real_extract(self, url):
         if not source or not source.endswith('mp4'):
             self.raise_no_formats('Could not extract a video', expected=bool(source), video_id=video_id)
 
-        tags = None
-        if self._is_logged_in:
-            metadata = self._call_api('info', video_id, {'itemId': video_id}, note='Downloading tags')
-            tags = traverse_obj(metadata, ('tags', ..., 'tag', {str}))
-            # Sorted by "confidence", higher confidence = earlier in list
-            confidences = traverse_obj(metadata, ('tags', ..., 'confidence', ({int}, {float})))
-            if confidences:
-                tags = [tag for _, tag in sorted(zip(confidences, tags), reverse=True)]
+        metadata = self._call_api('info', video_id, {'itemId': video_id}, note='Downloading tags')
+        tags = traverse_obj(metadata, ('tags', ..., 'tag', {str}))
+        # Sorted by "confidence", higher confidence = earlier in list
+        confidences = traverse_obj(metadata, ('tags', ..., 'confidence', ({int}, {float})))
+        if confidences:
+            tags = [tag for _, tag in sorted(zip(confidences, tags), reverse=True)]
 
         formats = traverse_obj(video_info, ('variants', ..., {
             'format_id': ('name', {str}),