Difference between revisions of "YouTube parser"

From Scriptwiki
Jump to: navigation, search
(Bugfix: script didn't produce output on unrated videos)
m (Added note about this script no longer working because of API deprecation)
 
(2 intermediate revisions by the same user not shown)
Line 1: Line 1:
 +
<div class="boilerplate metadata" id="stub" style="padding: 7px; background: #ffeeee; border: 1px solid #ff0000; text-align: center; font-size:95%;">'''''On May 6th, 2015, YouTube shut down their API v2 which is being used in this script to retrieve video metadata. One now has to use their API v3, which unfortunately seems to require an API key (as opposed to being publicly accessible by anyone).<br>As a consequence, this script is now useless, unless you view it as a code example. I'm working on a version that works with the new API.''''' </div>
 +
 +
 +
 
As the description says, this script retrieves information about youtube videos from the YouTube API and posts them to the channel whenever a youtube link is posted. You can install it by opening the script editor (Alt+R), creating a new script file (File -> New) and then copy paste the script into it.
 
As the description says, this script retrieves information about youtube videos from the YouTube API and posts them to the channel whenever a youtube link is posted. You can install it by opening the script editor (Alt+R), creating a new script file (File -> New) and then copy paste the script into it.
  
Line 18: Line 22:
 
   
 
   
 
  ; on text event: regex check for messages containing youtube.com or youtu.be, match the URL
 
  ; on text event: regex check for messages containing youtube.com or youtu.be, match the URL
  on $*:TEXT:/(youtu\.be\/.*?(\s|$)|youtube\.com\/.*?(\s|$))/:#pda:{
+
  on $*:TEXT:/(youtu\.be\/.*?(\s|$)|youtube\.com\/.*?(\s|$))/:#foo,#bar:{
 
   
 
   
 
   ; use another regex to extract the video ID from the matched URL
 
   ; use another regex to extract the video ID from the matched URL
Line 59: Line 63:
 
   
 
   
 
     ; extract things of interest out of the source code into a hash table
 
     ; extract things of interest out of the source code into a hash table
     [[If-Then-Else|if]] (*<title>*</title>* iswm %data) { ylog found title! | [[hadd]] -m %t title [[$regsubex]](%data,/(.*<title>|</title>.*)/g,) }
+
     [[If-Then-Else|if]] (*<title>*</title>* iswm %data) { [[hadd]] -m %t title [[$regsubex]](%data,/(.*<title>|</title>.*)/g,) }
     [[If-Then-Else|if]] (*<yt:duration seconds='*'/>* iswm %data) { ylog found duration! | [[hadd]] -m %t dur [[$duration]]([[$regsubex]](%data,/(.*seconds='|'/>.*)/g,),3) }
+
     [[If-Then-Else|if]] (*<yt:duration seconds='*'/>* iswm %data) { [[hadd]] -m %t dur [[$duration]]([[$regsubex]](%data,/(.*seconds='|'/>.*)/g,),3) }
     [[If-Then-Else|if]] (*<yt:statistics favoriteCount='*' viewCount='*'/>* iswm %data) { ylog found views! | [[hadd]] -m %t views [[$regsubex]](%data,/(.*viewCount='|'/>)/g,) }
+
     [[If-Then-Else|if]] (*<yt:statistics favoriteCount='*' viewCount='*'/>* iswm %data) { [[hadd]] -m %t views [[$regsubex]](%data,/(.*viewCount='|'/>)/g,) }
     [[If-Then-Else|if]] (*<yt:rating numDislikes='*' numLikes='*'/>* iswm %data) { ylog found rating! | [[hadd]] -m %t likes [[$regsubex]](%data,/(.*numLikes='|'/>.*)/g,) | [[hadd]] -m %t dislikes [[$regsubex]](%data,/(.*numDislikes='|'.*)/g,) }
+
     [[If-Then-Else|if]] (*<yt:rating numDislikes='*' numLikes='*'/>* iswm %data) { [[hadd]] -m %t likes [[$regsubex]](%data,/(.*numLikes='|'/>.*)/g,) | [[hadd]] -m %t dislikes [[$regsubex]](%data,/(.*numDislikes='|'.*)/g,) }
     [[If-Then-Else|if]] (*<name>*</name>* iswm %data) { ylog found uploader name! | [[hadd]] -m %t uploader [[$regsubex]](%data,/(.*<name>|</name>.*)/g,) }
+
     [[If-Then-Else|if]] (*<name>*</name>* iswm %data) { [[hadd]] -m %t uploader [[$regsubex]](%data,/(.*<name>|</name>.*)/g,) }
     [[If-Then-Else|if]] (*<yt:uploaded>*</yt:uploaded>* iswm %data) { ylog found upload date! | [[hadd]] -m %t uploaded [[$left]]([[$regsubex]](%data,/(.*<yt:uploaded>|</yt:uploaded>.*)/g,),10) }
+
     [[If-Then-Else|if]] (*<yt:uploaded>*</yt:uploaded>* iswm %data) { [[hadd]] -m %t uploaded [[$left]]([[$regsubex]](%data,/(.*<yt:uploaded>|</yt:uploaded>.*)/g,),10) }
     [[If-Then-Else|if]] (*accessControl action='rate' permission='denied'* iswm %data)  { ylog found rating disabled | [[hadd]] -m %t rating disabled or no votes }
+
     [[If-Then-Else|if]] (*accessControl action='rate' permission='denied'* iswm %data)  { [[hadd]] -m %t rating disabled or no votes }
 
   
 
   
 
     ; the </entry> tag marks the end of the data.
 
     ; the </entry> tag marks the end of the data.

Latest revision as of 23:12, 6 May 2015


As the description says, this script retrieves information about youtube videos from the YouTube API and posts them to the channel whenever a youtube link is posted. You can install it by opening the script editor (Alt+R), creating a new script file (File -> New) and then copy paste the script into it.

Note: Make sure to replace the channels (#foo,#bar) in the on TEXT event line with the channels you want the script to be active in.

/*
********************************************************************************************************
*
*  youtube.mrc by Jay2k1 @ QuakeNet, 2013
*
*  Whenever a youtube link is posted in the channels #foo or #bar, this script retrieves the video's details
*  from the youtube API and post them to the channel. It looks like this:
*
*  <Jay2k1> http://www.youtube.com/watch?v=oHg5SJYRHA0
*  <Bot> 'RickRoll' by cotter548, 00:03:34, 68213084 views, rating: 87% (259236/38751)
*
*********************************************************************************************************
*/

; on text event: regex check for messages containing youtube.com or youtu.be, match the URL
on $*:TEXT:/(youtu\.be\/.*?(\s|$)|youtube\.com\/.*?(\s|$))/:#foo,#bar:{

  ; use another regex to extract the video ID from the matched URL
  noop $regex($regml(1),/(\?v=|&v=|\/\d\/|\/embed\/|\/v\/|\.be\/)([a-zA-Z0-9\-\_]{11})/)

  ; call the youtube script with the channel and the video ID, if there is one
  youtube $$regml(2) $chan
}

; alias youtube: opens a socket connection to youtube
; expects: youtube video ID as $1, #channel as $2
; returns: -
alias -l youtube {
  ; create a unique socket name to allow for multiple connections at a time
  var %sock = youtube. $+ $ticks
  sockopen %sock gdata.youtube.com 80
  ; store video ID and channel in socket mark
  sockmark %sock $1-
}

on *:SOCKOPEN:youtube.*:{
  ; obligatory error check
  if $sockerr { return }

  ; send our request
  sockwrite -n $sockname GET /feeds/mobile/videos/ $+ $gettok($sock($sockname).mark,1,32) $+ ?v=2&prettyprint=true HTTP/1.1
  sockwrite -n $sockname Host: $sock($sockname).addr
  sockwrite -n $sockname $crlf
}

on *:SOCKREAD:youtube.*:{
  if $sockerr { return }

  ; receive answer to the request into a variable
  var %t = $sockname, %data
  sockread %data

  ; while there's bytes in the sockread buffer...
  while $sockbr {

    ; extract things of interest out of the source code into a hash table
    if (*<title>*</title>* iswm %data) { hadd -m %t title $regsubex(%data,/(.*<title>|</title>.*)/g,) }
    if (*<yt:duration seconds='*'/>* iswm %data) { hadd -m %t dur $duration($regsubex(%data,/(.*seconds='|'/>.*)/g,),3) }
    if (*<yt:statistics favoriteCount='*' viewCount='*'/>* iswm %data) { hadd -m %t views $regsubex(%data,/(.*viewCount='|'/>)/g,) }
    if (*<yt:rating numDislikes='*' numLikes='*'/>* iswm %data) { hadd -m %t likes $regsubex(%data,/(.*numLikes='|'/>.*)/g,) | hadd -m %t dislikes $regsubex(%data,/(.*numDislikes='|'.*)/g,) }
    if (*<name>*</name>* iswm %data) { hadd -m %t uploader $regsubex(%data,/(.*<name>|</name>.*)/g,) }
    if (*<yt:uploaded>*</yt:uploaded>* iswm %data) { hadd -m %t uploaded $left($regsubex(%data,/(.*<yt:uploaded>|</yt:uploaded>.*)/g,),10) }
    if (*accessControl action='rate' permission='denied'* iswm %data)  { hadd -m %t rating disabled or no votes }

    ; the </entry> tag marks the end of the data.
    if (*</entry>* iswm %data) || ($hget(%t,0).item == 7)  || (($hget(%t,0).item == 7) && ($hget(%t,rating) == disabled)) {
      var %ytmsg = ' $+ $hget(%t,title) $+ ' by $hget(%t,uploader) $+ $chr(44) $hget(%t,dur) $+ $chr(44) $hget(%t,views) views
      var %ytmsg = %ytmsg $+ , $iif($hget(%t,likes),rating: $round($calc($hget(%t,likes) / ($hget(%t,likes) + $hget(%t,dislikes)) * 100),1) $+ % ( $+ $hget(%t,likes) $+ / $+ $hget(%t,dislikes) $+ ),rating disabled or not rated yet)
      var %ytmsg = $replace(%ytmsg, ,$chr(160),",",',',<,<,>,>,&,&, £,$chr(163), €, $chr(8364), ©, $chr(169), &trade, $chr(8482), ®, $chr(174))                  
      msg $gettok($sock(%t).mark,2,32) %ytmsg

      sockclose %t
      hfree %t
      return
    }
    sockread %data
  }
}
Contributed by Jay2k1