Use BeautifulSoup to extract information from YT searches

Google is blocking IP's the instances of invidio, because of these reasons
livie will use BeautifulSoup to extract information from YT searches

3 files changed, 32 insertions, 28 deletions

diff --git a/README.md b/README.md
index 53b212b..2ab5f07 100644
--- a/README.md
+++ b/README.md
@@ -14,9 +14,10 @@ Livie allows the user to search youtube.com and play the video from `mpv`.
 
 - `python >= 3.5`
 - `python-requests`
+- `python-beautifulsoup4`
 - `mpv`
 
-  `sudo pacman -S python mpv python-requests`
+  `sudo pacman -S python mpv python-requests python-beautifulsoup4`
 
 ## Installation
 
diff --git a/livie.el b/livie.el
index 11e990c..226d3f7 100644
--- a/livie.el
+++ b/livie.el
@@ -47,8 +47,7 @@
   :group 'livie
   :type 'string)
 
-(defvar livie-youtube-regexp
-  "https://invidio.us/latest_version\\?id=[A-Za-z0-9_\\-]\\{11\\}&itag=\\<\\([0-9]*\\.[0-9]+\\|[0-9]+\\)[df]?\\>&local=true")
+(defvar livie-youtube-regexp "https://www.youtube.com/watch\\?v=[A-Za-z0-9_\\-]\\{11\\}")
 
 (define-derived-mode livie-mode
   special-mode "livie"
diff --git a/livie.py b/livie.py
index b03d1c4..5b5028a 100644
--- a/livie.py
+++ b/livie.py
@@ -1,34 +1,38 @@
 """This module does render video"""
 
 import sys
-import datetime
-import json
 import requests
+from bs4 import BeautifulSoup
 
-URL = 'https://invidio.us'
+URL = 'https://www.youtube.com'
+FILTER = '&sp=EgIQAQ%253D%253D'
 INPUT = sys.argv[1]
-SEARCH = '%s/api/v1/search?q=%s' % (URL, INPUT)
+SEARCH = '%s/results?search_query=%s%s' % (URL, INPUT, FILTER)
 REQUEST = requests.get(SEARCH)
-SD = '&itag=18&local=true'
-HD = '&itag=22&local=true'
+SOUP = BeautifulSoup(REQUEST.content, 'lxml', from_encoding=REQUEST.encoding)
+FIRST = True
 
-FIRST = True  # skip line loop
 
-VIDEOS = json.loads(REQUEST.content.decode('utf-8'))
+def replace(string):
+    """Remove unnecessary characters"""
+    string = string.replace(' - Duration: ', '')
+    string = string.replace('.', '')
+    string = string.replace(' views', '')
+    return string
 
-for video in VIDEOS:
+for vid in SOUP.find_all(class_='yt-lockup-content'):
     try:
-        title = video.get('title', '')
-        videoid = video.get('videoId', '')
-        author = video.get('author', '')
+        link = URL + vid.h3.a['href']
+        title = vid.h3.a.text
+        description = vid.h3.span.text
+        author = vid.find(class_='yt-lockup-byline').a.text
+        meta = vid.find(class_='yt-lockup-meta').ul.contents
+        time_srt = vid.find(class_='yt-lockup-title').span.text
+        time = replace(time_srt)
+        uploaded = meta[0].text
+        views_str = vid.find(class_='yt-lockup-meta').ul.li.find_next()
+        views = replace(views_str.text)
 
-        # Make URL
-        sd = '%s/latest_version?id=%s%s' % (URL, videoid, SD)
-        hd = '%s/latest_version?id=%s%s' % (URL, videoid, HD)
-
-        timer = video.get('lengthSeconds', '')
-        time = str(datetime.timedelta(seconds=timer))
-        publish = video.get('publishedText', '')
     except TypeError:
         continue
 
@@ -38,9 +42,9 @@ for video in VIDEOS:
         print()  # print skip line
 
     # prints
-    print('    title: %s' % (title))
-    print('       SD: %s' % (sd))
-    print('       HD: %s' % (hd))
-    print('           HD ^ Only some videos available caused by DRM')
-    print('  channel: %s' % (author))
-    print('     time: %s' % (time))
+    print('    title: %s' % title)
+    print('      url: %s' % link)
+    print('  channel: %s' % author)
+    print(' uploaded: %s' % uploaded)
+    print('     time: %s' % time)
+    print('    views: %s' % views)