YouTube module for willie - using v3 API

youtube.py

# coding=utf8

"""
youtube.py - Willie YouTube v3 Module
Copyright (c) 2015 Calvin Montgomery, All rights reserved.

Redistribution and use in source and binary forms, with or without modification,
are permitted provided that the following conditions are met:

Redistributions of source code must retain the above copyright notice, this list
of conditions and the following disclaimer. Redistributions in binary form must
reproduce the above copyright notice, this list of conditions and the following
disclaimer in the documentation and/or other materials provided with the
distribution. THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND
CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR
CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY,
OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING
IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY
OF SUCH DAMAGE.

Behavior based on the original YouTube module by the following authors:
Dimitri Molenaars, Tyrope.nl.
Elad Alfassa, <[email protected]>
Edward Powell, embolalia.net

Usage:

    1. Go to https://console.developers.google.com/project
    2. Create a new API project
    3. On the left sidebar, click "Credentials" under "APIs & auth"
    4. Click "Create new Key" under "Public API access"
    5. Click "Server key"
    6. Under "APIs & auth" click "YouTube Data API" and then click "Enable API"

Once the key has been generated, add the following to your willie config
(replace dummy_key with the key obtained from the API console):

[youtube]
api_key = dummy_key

"""

import datetime
import json
import re
import sys
from willie import web, tools
from willie.module import rule, commands, example

URL_REGEX = re.compile(r'(youtube.com/watch\S*v=|youtu.be/)([\w-]+)')
INFO_URL = ('https://www.googleapis.com/youtube/v3/videos'
            '?key={}&part=contentDetails,status,snippet,statistics&id={}')
SEARCH_URL = (u'https://www.googleapis.com/youtube/v3/search'
              '?key={}&part=id&maxResults=1&q={}&type=video')

class YouTubeError(Exception):
    pass

def setup(bot):
    if not bot.memory.contains('url_callbacks'):
        bot.memory['url_callbacks'] = tools.WillieMemory()
    bot.memory['url_callbacks'][URL_REGEX] = youtube_info

def shutdown(bot):
    del bot.memory['url_callbacks'][URL_REGEX]

def get_api_key(bot):
    if not bot.config.has_option('youtube', 'api_key'):
        raise KeyError('Missing YouTube API key')
    return bot.config.youtube.api_key

def configure(config):
    if config.option('Configure YouTube v3 API', False):
        config.interactive_add('youtube', 'api_key', 'Google Developers '
                'Console API key (Server key)')

def convert_date(date):
    """Parses an ISO 8601 datestamp and reformats it to be a bit nicer"""
    date = datetime.datetime.strptime(date, '%Y-%m-%dT%H:%M:%S.000Z')
    return date.strftime('%Y-%m-%d %H:%M:%S UTC')

def convert_duration(duration):
    """Converts an ISO 8601 duration to a human-readable duration"""
    units = {
        'hour': 0,
        'minute': 0,
        'second': 0
    }
    for symbol, unit in zip(('H', 'M', 'S'), ('hour', 'minute', 'second')):
        match = re.search(r'(\d+)' + symbol, duration)
        if match:
            units[unit] = int(match.group(1))

    time = datetime.time(**units)
    output = str(time)
    match = re.search('(\d+)D', duration)
    if match:
        output = match.group(1) + ' days, ' + output
    return output

def fetch_video_info(bot, id):
    """Retrieves video metadata from YouTube"""
    url = INFO_URL.format(get_api_key(bot), id)
    raw, headers = web.get(url, return_headers=True)

    if headers['_http_status'] == 403:
        bot.say(u'[YouTube Search] Access denied.  Check that your API key is '
                u'configured to allow access to your IP address.')
        return

    try:
        result = json.loads(raw)
    except ValueError as e:
        raise YouTubeError(u'Failed to decode: ' + raw)

    if 'error' in result:
        raise YouTubeError(result['error']['message'])

    if len(result['items']) == 0:
        raise YouTubeError('YouTube API returned empty result')

    video = result['items'][0]
    info = {
        'title': video['snippet']['title'],
        'uploader': video['snippet']['channelTitle'],
        'uploaded': convert_date(video['snippet']['publishedAt']),
        'duration': convert_duration(video['contentDetails']['duration']),
        'views': video['statistics']['viewCount'],
        'comments': video['statistics']['commentCount'],
        'likes': video['statistics']['likeCount'],
        'dislikes': video['statistics']['dislikeCount'],
        'link': 'https://youtu.be/' + video['id']
    }

    return info

def fix_count(count):
    """Adds commas to a number representing a count"""
    return '{:,}'.format(int(count))

def format_info(tag, info, include_link=False):
    """Formats video information for sending to IRC.

    If include_link is True, then the video link will be included in the
    output (this is useful for search results), otherwise it is not (no
    reason to include a link if we are simply printing information about
    a video that was already linked in chat).
    """
    output = [
        u'[{}] Title: {}'.format(tag, info['title']),
        u'Uploader: ' + info['uploader'],
        u'Uploaded: ' + info['uploaded'],
        u'Duration: ' + info['duration'],
        u'Views: ' + fix_count(info['views']),
        u'Comments: ' + fix_count(info['comments']),
        u'Likes: ' + fix_count(info['likes']),
        u'Dislikes: ' + fix_count(info['dislikes'])
    ]

    if include_link:
        output.append(u'Link: ' + info['link'])

    return u' | '.join(output)

@rule('.*(youtube.com/watch\S*v=|youtu.be/)([\w-]+).*')
def youtube_info(bot, trigger, found_match=None):
    """Catches youtube links said in chat and fetches video information"""
    match = found_match or trigger
    try:
        info = fetch_video_info(bot, match.group(2))
    except YouTubeError as e:
        bot.say(u'[YouTube] Lookup failed: {}'.format(e))
        return
    bot.say(format_info('YouTube', info))

@commands('yt', 'youtube')
@example('.yt Mystery Skulls - Ghost')
def ytsearch(bot, trigger):
    """Allows users to search for YouTube videos with .yt <search query>"""
    if not trigger.group(2):
        return

    # Note that web.get() quotes the query parameters, so the
    # trigger is purposely left unquoted (double-quoting breaks things)
    url = SEARCH_URL.format(get_api_key(bot), trigger.group(2))
    raw, headers = web.get(url, return_headers=True)
    if headers['_http_status'] == 403:
        bot.say(u'[YouTube Search] Access denied.  Check that your API key is '
                u'configured to allow access to your IP address.')
        return

    try:
        result = json.loads(raw)
    except ValueError as e:
        bot.say(u'[YouTube Search] Failed to decode: ' + raw)
        return

    if 'error' in result:
        bot.say(u'[YouTube Search] ' + result['error']['message'])
        return

    if len(result['items']) == 0:
        bot.say(u'[YouTube Search] No results for ' + trigger.group(2))
        return

    # YouTube v3 API does not include useful video metadata in search results.
    # Searching gives us the video ID, now we have to do a regular lookup to
    # get the information we want.
    try:
        info = fetch_video_info(bot, result['items'][0]['id']['videoId'])
    except YouTubeError as e:
        bot.say(u'[YouTube] Lookup failed: {}'.format(e))
        return

    bot.say(format_info('YouTube Search', info, include_link=True))

Added that step to the instructions and added error handling.

Great work, thank you! Seems like searching with more than one search term returns an error, though.

Here's the traceback.

Traceback (most recent call last):
  File "/usr/local/lib/python2.7/dist-packages/willie/bot.py", line 581, in call
    exit_code = func(willie, trigger)
  File "/usr/local/lib/python2.7/dist-packages/willie/modules/youtube.py", line 180, in ytsearch
    result = json.loads(web.get(url))
  File "/usr/lib/python2.7/json/__init__.py", line 326, in loads
    return _default_decoder.decode(s)
  File "/usr/lib/python2.7/json/decoder.py", line 366, in decode
    obj, end = self.raw_decode(s, idx=_w(s, 0).end())
  File "/usr/lib/python2.7/json/decoder.py", line 384, in raw_decode
    raise ValueError("No JSON object could be decoded")
ValueError: No JSON object could be decoded

In IRC it spits out just this:

ValueError: No JSON object could be decoded (file "/usr/lib/python2.7/json/decoder.py", line 384, in raw_decode)

Did I miss something?

I'm also getting that JSON error for a string that has spaces in it. If you search with URL entities it works fine. For example

.yt test%20search

would work, but

.yt test search

will throw an error

I apologize for not replying promptly; I didn't realize that GitHub doesn't email me for comments on gists.

@rtil5 Are you sure you're using the latest version of willie? URL escaping is handled by web.get(), hence test search should automatically be encoded as test%20search by that function (and in fact, from my testing, this appears to be the case)

02:28:12     <@calzoneman> .yt test search
02:28:13       <willieDev> [YouTube Search] Title: Google AdWords Exam Practice Test Questions and Answers (Search Advertising Advance) | Uploader: Gokuldas K | Uploaded: 2013-07-25 12:41:30 UTC | Duration: 00:21:45 | Views: 21,247 | Comments: 8 | Likes: 25 | Dislikes: 18 | Link: https://youtu.be/kW1_YIXqmUA

@zwerxyplous Are you sure that you have the API key set up correctly? I may need to add some more debug information to figure out what is going on in your case. I'm guessing you're running into the same issue as @rtil5.

I just updated willie to be sure, and it still doesn't work.

Just had someone else report this issue; it turned out he didn't have his IP address right for the API key.

Updated gist with explicit error handling for HTTP 403.

Here's a new error.

KeyError: '_http_status' (file "/usr/lib64/python2.7/rfc822.py", line 388, in __getitem__)

a KeyError is thrown on line 134 if comments are disabled. there must be an easy fix for this, but i'm not sure what.

here's this crude workaround for now:

try:
    info = {
        'title': video['snippet']['title'],
        'uploader': video['snippet']['channelTitle'],
        'uploaded': convert_date(video['snippet']['publishedAt']),
        'duration': convert_duration(video['contentDetails']['duration']),
        'views': video['statistics']['viewCount'],
        'comments': video['statistics']['commentCount'],
        'likes': video['statistics']['likeCount'],
        'dislikes': video['statistics']['dislikeCount'],
        'link': 'https://youtu.be/' + video['id']
    }
except KeyError:
    info = {
        'title': video['snippet']['title'],
        'uploader': video['snippet']['channelTitle'],
        'uploaded': convert_date(video['snippet']['publishedAt']),
        'duration': convert_duration(video['contentDetails']['duration']),
        'views': video['statistics']['viewCount'],
        'comments': 0,
        'likes': video['statistics']['likeCount'],
        'dislikes': video['statistics']['dislikeCount'],
        'link': 'https://youtu.be/' + video['id']
    }