Simplify scripts/findTreatedChannelWithMostSubscribers.py

Otherwise `Linux, is in millions of computers` doesn't match the not automatically generated caption of [`o8NPllzkFhE`](https://www.youtube.com/watch?v=o8NPllzkFhE). Note to be confused with the search across captions that already used to work with for instance `is in millions of computers, it`.

README.md

@@ -1,3 +1,5 @@
+A video introducing this project is available [here](https://crawler.yt.lemnoslife.com/presentation).
+
 # The algorithm:
 
 To retrieve the most YouTube video ids in order to retrieve the most video captions, we need to retrieve the most YouTube channels.
@@ -9,6 +11,24 @@ A ready to be used by the end-user website instance of this project is hosted at
 
 See more details on [the Wiki](https://gitea.lemnoslife.com/Benjamin_Loison/YouTube_captions_search_engine/wiki).
 
+# The project structure:
+
+- `main.cpp` contains the C++ multi-threaded algorithm proceeding to the YouTube channels discovery. It is notably made of the following functions:
+  - `main` which takes into account the command line arguments, load variables from files (`channels.txt`, `keys.txt`, `channels/` content) and start the threads as executing `treatChannels` function
+  - `treatChannels` gets a YouTube channel to treat, treat it in `treatChannelOrVideo` function and compress the retrieved data
+  - `treatChannelOrVideo` which provided a YouTube channel id or a video id, treats this resource. In both cases it treats comments left on this resource. In the case of a channel it also treats its `CHANNELS`, `COMMUNITY`, `PLAYLISTS` and `LIVE` tabs and downloads the captions of the channel videos.
+  - `markChannelAsRequiringTreatmentIfNeeded` which provided a YouTube channel id marks it as requiring treatment if it wasn't already treated
+  - `execute` which provided an `yt-dlp` command executes it in a shell
+  - `getJson` which provided an API request returns a JSON structure with its result. In the case that the API requested is YouTube Data API v3 and a set of keys is provided (see below `keys.txt`), it rotates the keys as required
+- `channels.txt` contains a starting set of channels which contains mostly the 100 most subscribed French channels
+- `keys.txt` contains a set of YouTube Data API v3 keys (not provided) to have the ability to request this API (see an alternative to filling it in the section below with `--no-keys` command line argument)
+- `scripts/` contains Python scripts to:
+  - generate the `channels.txt` as described above (`retrieveTop100SubscribersFrance.py`)
+  - remove channels being treated before a restart of the algorithm as described in [the `main` function documentation](https://gitea.lemnoslife.com/Benjamin_Loison/YouTube_captions_search_engine/src/commit/8dd89e6e881da0a905b6fa4b23775c4344dd0d9d/main.cpp#L126-L128) (`removeChannelsBeingTreated.py`)
+- `website/` is a PHP website using WebSocket to allow the end-user to proceed to requests on the retrieved dataset. When fetching the website, the end-user receives the interpreted `index.php` which upon making a request interacts with `websocket.php` which in the back-end dispatches the requests from various end-users to `search.py` (which treats the actual end-user request on the compressed dataset) by using `users/` to make the inter-process communication.
+
+Note that this project heavily relies on [YouTube operational API](https://github.com/Benjamin-Loison/YouTube-operational-API) [which was modified for this project](https://gitea.lemnoslife.com/Benjamin_Loison/YouTube_captions_search_engine/wiki/YouTube-operational-API-commits).
+
 # Running the YouTube graph discovery algorithm:
 
 Because of [the current compression mechanism](https://gitea.lemnoslife.com/Benjamin_Loison/YouTube_captions_search_engine/issues/30), Linux is the only known OS able to run this algorithm.

scripts/findAlreadyTreatedCommentsCount.py

@@ -1,6 +1,6 @@
 #!/usr/bin/python3
 
-PREFIX = 'Comments per second: '
+PREFIX = 'Channels per second: '
 alreadyTreatedCommentsCount = 0
 
 with open('nohup.out') as f:
@@ -8,5 +8,7 @@ with open('nohup.out') as f:
     for line in lines:
         if PREFIX in line:
             alreadyTreatedCommentsCount += int(line.split(PREFIX)[-1])
+        #if 'UCsT0YIqwnpJCM-mx7-gSA4Q' in line:
+        #    break
 
 print(alreadyTreatedCommentsCount)

scripts/findLatestTreatedCommentsForChannelsBeingTreated.py

@@ -1,5 +1,7 @@
 #!/usr/bin/python3
 
+# This algorithm should also take in account other features that we use to retrieve channels.
+
 import os, requests, json, time, datetime
 
 path = 'channels/'
@@ -7,23 +9,24 @@ path = 'channels/'
 os.chdir(path)
 
 def getTimestampFromDateString(dateString):
-    return int(time.mktime(datetime.datetime.strptime(dateString, "%Y-%m-%dT%H:%M:%SZ").timetuple()))
+    return int(time.mktime(datetime.datetime.strptime(dateString, '%Y-%m-%dT%H:%M:%SZ').timetuple()))
 
 for channelId in list(os.walk('.'))[1]:
     channelId = channelId[2:]
     #print(channelId)
-    numberOfRequests = len(list(os.walk(channelId))[0][2])
+    numberOfRequests = len(list(os.walk(f'{channelId}/requests'))[0][2]) - 1
     # Assume that the folder isn't empty (may not be the case, but it is most of the time).
-    with open(f'{channelId}/{str(numberOfRequests - 1)}.json') as f:
+    filePath = f'{channelId}/requests/{str(numberOfRequests - 1)}.json'
-        content = "\n".join(f.read().splitlines()[1:])
+    with open(filePath) as f:
-        data = json.loads(content)
+        print(filePath)
+        #content = "\n".join(f.read().splitlines()[1:])
+        data = json.load(f)#json.loads(content)
         snippet = data['items'][-1]['snippet']
         if 'topLevelComment' in snippet:
             snippet = snippet['topLevelComment']['snippet']
         latestTreatedCommentDate = snippet['publishedAt']
     url = f'https://yt.lemnoslife.com/noKey/channels?part=snippet&id={channelId}'
-    content = requests.get(url).text
+    data = requests.get(url).json()
-    data = json.loads(content)
     channelCreationDate = data['items'][0]['snippet']['publishedAt']
     #print(channelCreationDate)
     # Timing percentage not taking into account the not uniform in time distribution of comments. Note that in the case of the last request is to list replies to a comment, the percentage might goes a bit backward, as replies are posted after the initial comment.

scripts/findTreatedChannelWithMostSubscribers.py

@@ -1,8 +1,8 @@
 #!/usr/bin/python3
 
-import os, requests, json
+import os, requests
 
-channelIds = next(os.walk('channels/'))[1]
+channelIds = [channelId.replace('.zip', '') for channelId in next(os.walk('channels/'))[2]]
 maxResults = 50
 
 channelIdsChunks = [channelIds[i : i + maxResults] for i in range(0, len(channelIds), maxResults)]
@@ -11,8 +11,7 @@ mostSubscriberChannel = None
 
 for channelIds in channelIdsChunks:
     url = 'https://yt.lemnoslife.com/noKey/channels?part=statistics&id=' + ','.join(channelIds)
-    content = requests.get(url).text
+    data = requests.get(url).json()
-    data = json.loads(content)
     items = data['items']
     for item in items:
         subscriberCount = int(item['statistics']['subscriberCount'])

scripts/retrieveTop100SubscribersFrance.py

@@ -1,4 +1,7 @@
+#!/usr/bin/python3
+
 # We can't proceed automatically by using `requests` Python module because https://socialblade.com/youtube/top/country/fr/mostsubscribed is protected by CloudFlare.
+# Note that `undetected-chromedriver` might be a workaround this limitation.
 
 with open('mostsubscribed.html') as f:
     lines = f.read().splitlines()

website/search.py

@@ -10,6 +10,11 @@ pathSearchMessageParts = sys.argv[2].split(' ')
 pathSearch = pathSearchMessageParts[1]
 message = ' '.join(pathSearchMessageParts[2:])
 
+pathSearchRegex = re.compile(pathSearch)
+messageRegex = re.compile(message)
+
+isPathSearchAChannelId = re.match(r'[a-zA-Z0-9-_]{24}', pathSearch)
+
 searchOnlyCaptions = pathSearchMessageParts[0] == 'search-only-captions'
 
 clientFilePath = f'users/{clientId}.txt'
@@ -33,23 +38,31 @@ def cleanCaption(caption):
     return caption.replace('\n', ' ')
 
 # As `zipgrep` doesn't support arguments to stop on first match for each file, we proceed manually to keep a good theoretical complexity.
+if isPathSearchAChannelId:
+    file = pathSearch + '.zip'
+    if os.path.isfile(path + file):
+        files = [file]
+    else:
+        write(f'progress:0 / 0')
+else:
     files = [file for file in os.listdir(path) if file.endswith('.zip')]
+
 for fileIndex, file in enumerate(files):
-    write(f'progress:{fileIndex + 1} / {len(files)}')
+    write(f'progress:{fileIndex} / {len(files)}')
     zip = zipfile.ZipFile(path + file)
     for fileInZip in zip.namelist():
         endsWithVtt = fileInZip.endswith('.vtt')
         if searchOnlyCaptions and not endsWithVtt:
             continue
         toWrite = f'{file}/{fileInZip}'
-        if not bool(re.search(pathSearch, toWrite)):
+        if not bool(pathSearchRegex.search(toWrite)):
             continue
         with zip.open(fileInZip) as f:
             if endsWithVtt:
                 content = f.read().decode('utf-8')
                 stringIOf = StringIO(content)
                 wholeCaption = ' '.join([cleanCaption(caption.text) for caption in webvtt.read_buffer(stringIOf)])
-                messagePositions = [m.start() for m in re.finditer(message, wholeCaption)]
+                messagePositions = [m.start() for m in messageRegex.finditer(wholeCaption)]
                 if messagePositions != []:
                     timestamps = []
                     for messagePosition in messagePositions:
@@ -67,6 +80,7 @@ for fileIndex, file in enumerate(files):
                     if message in str(line):
                         write(toWrite)
                         break
+    write(f'progress:{fileIndex + 1} / {len(files)}')
 
 with open(clientFilePath) as f:
     while True: