Remove ad-hoc response saving from code, create a debug setting for fetch_url

8 files changed, 33 insertions, 46 deletions

diff --git a/settings.py b/settings.py
index 2a3885f..5b9520d 100644
--- a/settings.py
+++ b/settings.py
@@ -24,9 +24,12 @@ default_comment_sorting = 0
 
 # developer use to debug 403s
 gather_googlevideo_domains = False
+
+# save all responses from youtube for debugging
+debugging_save_responses = False
 '''
 exec(default_settings)
-allowed_targets = set(("route_tor", "port_number", "allow_foreign_addresses", "subtitles_mode", "subtitles_language", "enable_related_videos", "enable_comments", "enable_comment_avatars", "default_comment_sorting", "gather_googlevideo_domains"))
+allowed_targets = set(("route_tor", "port_number", "allow_foreign_addresses", "subtitles_mode", "subtitles_language", "enable_related_videos", "enable_comments", "enable_comment_avatars", "default_comment_sorting", "gather_googlevideo_domains", "debugging_save_responses"))
 
 def log_ignored_line(line_number, message):
     print("settings.txt: Ignoring line " + str(node.lineno) + " (" + message + ")")
diff --git a/youtube/accounts.py b/youtube/accounts.py
index c35b6cc..d2e8a41 100644
--- a/youtube/accounts.py
+++ b/youtube/accounts.py
@@ -162,10 +162,8 @@ def _login(username, password, cookiejar, use_tor):
     Taken from youtube-dl
     """
 
-    login_page = util.fetch_url(_LOGIN_URL, yt_dl_headers, report_text='Downloaded login page', cookiejar_receive=cookiejar, use_tor=use_tor).decode('utf-8')
-    '''with open('debug/login_page', 'w', encoding='utf-8') as f:
-        f.write(login_page)'''
-    #print(cookiejar.as_lwp_str())
+    login_page = util.fetch_url(_LOGIN_URL, yt_dl_headers, report_text='Downloaded login page', cookiejar_receive=cookiejar, use_tor=use_tor, debug_name='login_page').decode('utf-8')
+
     if login_page is False:
         return
 
@@ -189,10 +187,7 @@ def _login(username, password, cookiejar, use_tor):
             'Google-Accounts-XSRF': 1,
         }
         headers.update(yt_dl_headers)
-        result = util.fetch_url(url, headers, report_text=note, data=data, cookiejar_send=cookiejar, cookiejar_receive=cookiejar, use_tor=use_tor).decode('utf-8')
-        #print(cookiejar.as_lwp_str())
-        '''with open('debug/' + note, 'w', encoding='utf-8') as f:
-            f.write(result)'''
+        result = util.fetch_url(url, headers, report_text=note, data=data, cookiejar_send=cookiejar, cookiejar_receive=cookiejar, use_tor=use_tor, debug_name=note).decode('utf-8')
         result = re.sub(r'^[^\[]*', '', result)
         return json.loads(result)
 
@@ -321,12 +316,10 @@ def _login(username, password, cookiejar, use_tor):
         return False
 
     try:
-        check_cookie_results = util.fetch_url(check_cookie_url, headers=yt_dl_headers, report_text="Checked cookie", cookiejar_send=cookiejar, cookiejar_receive=cookiejar, use_tor=use_tor).decode('utf-8')
+        check_cookie_results = util.fetch_url(check_cookie_url, headers=yt_dl_headers, report_text="Checked cookie", cookiejar_send=cookiejar, cookiejar_receive=cookiejar, use_tor=use_tor, debug_name='check_cookie_results').decode('utf-8')
     except (urllib.error.URLError, compat_http_client.HTTPException, socket.error) as err:
         return False
 
-    '''with open('debug/check_cookie_results', 'w', encoding='utf-8') as f:
-        f.write(check_cookie_results)'''
 
     if 'https://myaccount.google.com/' not in check_cookie_results:
         warn('Unable to log in')
diff --git a/youtube/channel.py b/youtube/channel.py
index 04f698b..daf4791 100644
--- a/youtube/channel.py
+++ b/youtube/channel.py
@@ -88,11 +88,9 @@ def get_channel_tab(channel_id, page="1", sort=3, tab='videos', view=1):
     url = "https://www.youtube.com/browse_ajax?ctoken=" + ctoken
 
     print("Sending channel tab ajax request")
-    content = util.fetch_url(url, util.desktop_ua + headers_1)
+    content = util.fetch_url(url, util.desktop_ua + headers_1, debug_name='channel_tab')
     print("Finished recieving channel tab response")
 
-    '''with open('debug/channel_debug', 'wb') as f:
-        f.write(content)'''
     return content
 
 def get_number_of_videos(channel_id):
@@ -103,15 +101,13 @@ def get_number_of_videos(channel_id):
 
     # Sometimes retrieving playlist info fails with 403 for no discernable reason
     try:
-        response = util.fetch_url(url, util.mobile_ua + headers_pbj)
+        response = util.fetch_url(url, util.mobile_ua + headers_pbj, debug_name='number_of_videos')
     except urllib.error.HTTPError as e:
         if e.code != 403:
             raise
         print("Couldn't retrieve number of videos")
         return 1000
 
-    '''with open('debug/playlist_debug_metadata', 'wb') as f:
-        f.write(response)'''
     response = response.decode('utf-8')
     print("Got response for number of videos")
 
@@ -135,9 +131,7 @@ def get_channel_search_json(channel_id, query, page):
     ctoken = proto.string(2, channel_id) + proto.string(3, params) + proto.string(11, query)
     ctoken = base64.urlsafe_b64encode(proto.nested(80226972, ctoken)).decode('ascii')
 
-    polymer_json = util.fetch_url("https://www.youtube.com/browse_ajax?ctoken=" + ctoken, util.desktop_ua + headers_1)
-    '''with open('debug/channel_search_debug', 'wb') as f:
-        f.write(polymer_json)'''
+    polymer_json = util.fetch_url("https://www.youtube.com/browse_ajax?ctoken=" + ctoken, util.desktop_ua + headers_1, debug_name='channel_search')
 
     return polymer_json
 
@@ -293,9 +287,9 @@ def get_channel_page(channel_id, tab='videos'):
         number_of_videos, polymer_json = tasks[0].value, tasks[1].value
 
     elif tab == 'about':
-        polymer_json = util.fetch_url('https://www.youtube.com/channel/' + channel_id + '/about?pbj=1', util.desktop_ua + headers_1)
+        polymer_json = util.fetch_url('https://www.youtube.com/channel/' + channel_id + '/about?pbj=1', util.desktop_ua + headers_1, debug_name='channel_about')
     elif tab == 'playlists':
-        polymer_json = util.fetch_url('https://www.youtube.com/channel/' + channel_id + '/playlists?pbj=1&view=1&sort=' + playlist_sort_codes[sort], util.desktop_ua + headers_1)
+        polymer_json = util.fetch_url('https://www.youtube.com/channel/' + channel_id + '/playlists?pbj=1&view=1&sort=' + playlist_sort_codes[sort], util.desktop_ua + headers_1, debug_name='channel_playlists')
     elif tab == 'search':
         tasks = (
             gevent.spawn(get_number_of_videos, channel_id ), 
@@ -336,13 +330,11 @@ def get_channel_page_general_url(base_url, tab, request):
     query = request.args.get('query', '')
 
     if tab == 'videos':
-        polymer_json = util.fetch_url(base_url + '/videos?pbj=1&view=0', util.desktop_ua + headers_1)
-        with open('debug/channel_debug', 'wb') as f:
-            f.write(polymer_json)
+        polymer_json = util.fetch_url(base_url + '/videos?pbj=1&view=0', util.desktop_ua + headers_1, debug_name='gen_channel_videos')
     elif tab == 'about':
-        polymer_json = util.fetch_url(base_url + '/about?pbj=1', util.desktop_ua + headers_1)
+        polymer_json = util.fetch_url(base_url + '/about?pbj=1', util.desktop_ua + headers_1, debug_name='gen_channel_about')
     elif tab == 'playlists':
-        polymer_json = util.fetch_url(base_url+ '/playlists?pbj=1&view=1', util.desktop_ua + headers_1)
+        polymer_json = util.fetch_url(base_url+ '/playlists?pbj=1&view=1', util.desktop_ua + headers_1, debug_name='gen_channel_playlists')
     elif tab == 'search':
         raise NotImplementedError()
     else:
diff --git a/youtube/comments.py b/youtube/comments.py
index ba82154..4485ad2 100644
--- a/youtube/comments.py
+++ b/youtube/comments.py
@@ -83,7 +83,7 @@ def request_comments(ctoken, replies=False):
     url = base_url + ctoken.replace("=", "%3D") + "&pbj=1"
 
     for i in range(0,8):    # don't retry more than 8 times
-        content = util.fetch_url(url, headers=mobile_headers, report_text="Retrieved comments")
+        content = util.fetch_url(url, headers=mobile_headers, report_text="Retrieved comments", debug_name='request_comments')
         if content[0:4] == b")]}'":             # random closing characters included at beginning of response for some reason
             content = content[4:]
         elif content[0:10] == b'\n<!DOCTYPE':   # occasionally returns html instead of json for no reason
@@ -91,8 +91,6 @@ def request_comments(ctoken, replies=False):
             print("got <!DOCTYPE>, retrying")
             continue
         break
-    '''with open('debug/comments_debug', 'wb') as f:
-        f.write(content)'''
     return content
 
 
diff --git a/youtube/playlist.py b/youtube/playlist.py
index 18ddf49..5df7074 100644
--- a/youtube/playlist.py
+++ b/youtube/playlist.py
@@ -47,9 +47,7 @@ headers_1 = (
 
 def playlist_first_page(playlist_id, report_text = "Retrieved playlist"):
     url = 'https://m.youtube.com/playlist?list=' + playlist_id + '&pbj=1'
-    content = util.fetch_url(url, util.mobile_ua + headers_1, report_text=report_text)
-    '''with open('debug/playlist_debug', 'wb') as f:
-        f.write(content)'''
+    content = util.fetch_url(url, util.mobile_ua + headers_1, report_text=report_text, debug_name='playlist_first_page')
     content = json.loads(util.uppercase_escape(content.decode('utf-8')))
 
     return content
@@ -67,9 +65,7 @@ def get_videos(playlist_id, page):
         'X-YouTube-Client-Version': '2.20180508',
     }
 
-    content = util.fetch_url(url, headers, report_text="Retrieved playlist")
-    '''with open('debug/playlist_debug', 'wb') as f:
-        f.write(content)'''
+    content = util.fetch_url(url, headers, report_text="Retrieved playlist", debug_name='playlist_videos')
 
     info = json.loads(util.uppercase_escape(content.decode('utf-8')))
     return info
diff --git a/youtube/post_comment.py b/youtube/post_comment.py
index c4ffb9d..25d0e3a 100644
--- a/youtube/post_comment.py
+++ b/youtube/post_comment.py
@@ -35,13 +35,11 @@ def _post_comment(text, video_id, session_token, cookiejar):
     data = urllib.parse.urlencode(data_dict).encode()
 
 
-    content = util.fetch_url("https://m.youtube.com/service_ajax?name=createCommentEndpoint", headers=headers, data=data, cookiejar_send=cookiejar)
+    content = util.fetch_url("https://m.youtube.com/service_ajax?name=createCommentEndpoint", headers=headers, data=data, cookiejar_send=cookiejar, debug_name='post_comment')
 
     code = json.loads(content)['code']
     print("Comment posting code: " + code)
     return code
-    '''with open('debug/post_comment_response', 'wb') as f:
-        f.write(content)'''
 
 
 def _post_comment_reply(text, video_id, parent_comment_id, session_token, cookiejar):
@@ -66,13 +64,11 @@ def _post_comment_reply(text, video_id, parent_comment_id, session_token, cookie
     }
     data = urllib.parse.urlencode(data_dict).encode()
 
-    content = util.fetch_url("https://m.youtube.com/service_ajax?name=createCommentReplyEndpoint", headers=headers, data=data, cookiejar_send=cookiejar)
+    content = util.fetch_url("https://m.youtube.com/service_ajax?name=createCommentReplyEndpoint", headers=headers, data=data, cookiejar_send=cookiejar, debug_name='post_reply')
 
     code = json.loads(content)['code']
     print("Comment posting code: " + code)
     return code
-    '''with open('debug/post_comment_response', 'wb') as f:
-        f.write(content)'''
 
 def _delete_comment(video_id, comment_id, author_id, session_token, cookiejar):
     headers = {
diff --git a/youtube/search.py b/youtube/search.py
index 39a80bf..e35d0cb 100644
--- a/youtube/search.py
+++ b/youtube/search.py
@@ -53,7 +53,7 @@ def get_search_json(query, page, autocorrect, sort, filters):
         'X-YouTube-Client-Version': '2.20180418',
     }
     url += "&pbj=1&sp=" + page_number_to_sp_parameter(page, autocorrect, sort, filters).replace("=", "%3D")
-    content = util.fetch_url(url, headers=headers, report_text="Got search results")
+    content = util.fetch_url(url, headers=headers, report_text="Got search results", debug_name='search_results')
     info = json.loads(content)
     return info
 
diff --git a/youtube/util.py b/youtube/util.py
index 9950815..2f80f11 100644
--- a/youtube/util.py
+++ b/youtube/util.py
@@ -5,6 +5,7 @@ import brotli
 import urllib.parse
 import re
 import time
+import os
 
 # The trouble with the requests library: It ships its own certificate bundle via certifi
 #  instead of using the system certificate store, meaning self-signed certificates
@@ -103,7 +104,7 @@ def decode_content(content, encoding_header):
             content = gzip.decompress(content)
     return content
 
-def fetch_url(url, headers=(), timeout=15, report_text=None, data=None, cookiejar_send=None, cookiejar_receive=None, use_tor=True, return_response=False):
+def fetch_url(url, headers=(), timeout=15, report_text=None, data=None, cookiejar_send=None, cookiejar_receive=None, use_tor=True, return_response=False, debug_name=None):
     '''
     When cookiejar_send is set to a CookieJar object,
      those cookies will be sent in the request (but cookies in response will not be merged into it)
@@ -160,6 +161,14 @@ def fetch_url(url, headers=(), timeout=15, report_text=None, data=None, cookieja
         print(report_text, '    Latency:', round(response_time - start_time,3), '    Read time:', round(read_finish - response_time,3))
     content = decode_content(content, response.getheader('Content-Encoding', default='identity'))
 
+    if settings.debugging_save_responses and debug_name is not None:
+        save_dir = os.path.join(settings.data_dir, 'debug')
+        if not os.path.exists(save_dir):
+            os.makedirs(save_dir)
+
+        with open(os.path.join(save_dir, debug_name), 'wb') as f:
+            f.write(content)
+
     if return_response:
         return content, response
     return content
@@ -226,4 +235,4 @@ def update_query_string(query_string, items):
 def uppercase_escape(s):
      return re.sub(
          r'\\U([0-9a-fA-F]{8})',
-         lambda m: chr(int(m.group(1), base=16)), s)
\ No newline at end of file
+         lambda m: chr(int(m.group(1), base=16)), s)