Finds YouTube videos you're interested in. PyCon US talks finder example.

youtube_find_relevant.py

import re
import requests


API_KEY = 'xxx'
'''Google API (YouTube Data API v3) key from https://console.developers.google.com/apis/.'''

# Put titles you're interested into RELEVANT string:
# one title per line.
RELEVANT = '''
Trio: Async concurrency for mere mortals
Solve Your Problem With Sloppy Python
Dataclasses: The code generator to end all code generators
Python 3: ten years later
'''

CHANNEL = 'UCsX05-2sVSH7Nx3zuk3NYuQ'  # PyCon US 2018 Channel
'''YouTube channel ID here.'''

TITLE_POSTFIX = 'PyCon 2018'
'''Postfix to strip from titles.'''

#############################################

_SPACES = re.compile('\s+')
_BASE_URL = 'https://www.googleapis.com/youtube/v3/search?order=date&part=snippet&channelId=%(channel)s&maxResults=50&key=%(key)s%(page)s'


def traverse(page=0):

    params = {   
        'channel': CHANNEL,
        'key': API_KEY,
        'page': '',
    }
    
    if page:
        params['page'] = '&pageToken=%s' % page
    
    url = _BASE_URL % params
    
    response = requests.get(url)
    json = response.json()
    
    
    next_page = json.get('nextPageToken')
    
    for item in json['items']:
        
        if item['id']['kind'] != 'youtube#video':
            continue
        
        video_id = item['id']['videoId']
        title = item['snippet']['title']
        
        title = title.replace(TITLE_POSTFIX, '').strip(' -')
        split = title.split(' - ', 1)
        
        prefix = split[0]
        prefix = prefix.replace('/', ',')
        
        if len(prefix.split(' ')) in {2, 3} or (',' in prefix):  
            # strip person name
            try:
                title = split[1]
            except IndexError:
                pass
        
        title = _SPACES.sub(' ', title)
        title = title.strip(' -')
        
        yield video_id, title
        
    if next_page:
        yield from traverse(page=next_page)

        
def find_relevant():
    
    relevant_lines = []
    
    for line in RELEVANT.splitlines():
        line = line.strip()
        if line:
            line = _SPACES.sub(' ', line)
            relevant_lines.append(line)
    
    total_relevant = len(relevant_lines)
    total_traversed = 0
    
    traversed = [item for item in traverse()][::-1]  # eldest first
    
    for idx, (video_id, title) in enumerate(traversed, 1):
        total_traversed += 1
        
        url = ''
        if title in relevant_lines:
            url = 'https://youtu.be/%s' % video_id
            relevant_lines.remove(title)

        print('%s. %s %s'% (idx, title, url))


    total_missing = len(relevant_lines)
       
    print(
        '\nSummary: among %s found %s of %s, missing %s\n' % (
            total_traversed,
            total_relevant - total_missing,
            total_relevant,
            total_missing
        ))

    for idx, line in enumerate(relevant_lines, 1):
        print('%s. %s'% (idx, line))


find_relevant()