View source for URL-logger

 Screenshots:<br />
 [[:Image:ss_url_color.jpg|Screenshot 1: Colored links]]<br />
 [[:Image:ss_url_search.jpg|Screenshot 2: Links search]]
  
 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; 
 ; 
 ; URL-logger with colored and underlined links by Cail at Quakenet -> #help.script
 ;
 ; $urlc -alias will color and underline links from the input, and log them into a file for later searching
 ; Tested and works on mIRC 6.2 and above (i recommend using >6.21 for $regsubex to work correctly)
 ; 
 ; This script uses spopup.dll (OPTIONAL! you only need this if you want a new popup with info and some new features.),
 ; to replace mirc's builtin url-popup
 ; spopup.dll is made by Saturn, and you can get it from http://www.xise.nl/mirc/
 ; NOTE! Depending on how your mIRC is set up, your working path might be mirc's root dir, or Users/AppData folder
 ; For example in Windows 7 the AppData folder would be: C:\Users\HappyUser\AppData\Roaming\mIRC
 ; Note this when placing the .dll somewhere and not giving full path in the script.
 ;
 ; Formats recognized links like:
 ;   - <nowiki>"http://google.com"</nowiki> -> <nowiki>http://google.com</nowiki>
 ;   - url('<nowiki>www.google.com</nowiki>'); -> <nowiki>www.google.com</nowiki>
 ;   - (<nowiki>www.google.com/ahihi(kiaa)</nowiki>) -> <nowiki>www.google.com/ahihi(kiaa)</nowiki>
 ;   
 ;   And some more cleanups, and adds '<nowiki>http://</nowiki>' in front of '<nowiki>www.</nowiki>' for storing.
 ;   so that <nowiki>www.google.com</nowiki> and <nowiki>http://www.google.com</nowiki> are treated as a same link, and stored only once
 ;
 ; Uses 3 different colors for coloring links:
 ;   1) new link - not found from the file
 ;   2) opened link - is found from the file, and you have opened it via mirc
 ;   3) old link - is found from the file, but you haven't opened it
 ;
 ; Fetches title of youtube video's:
 ;   - Gets title of the video and stores it into the links.db (for searching)
 ;   - Echo's the title under the link in chat
 ;   - Hold down ctrl key when pressing enter if you post a link, to /msg the title into the channel
 ;   - Fill %sendYTtitleChans with channels separated by spaces (eg. %sendYTtitleChans #chan1 #chan2) to automatically
 ;     send the title into the channel when ever a youtube link is posted.
 ;
 ; Thanks to:
 ;   - Saturn for the spopup.dll
 ;   - Msmo for the cleanup regex
 ;
 ; 
 ; UPDATE 06/07/2014: Performance update!
 ;                     - Added hashtables to be used on regular checks, speeding up the normal link flow, no more slowdowns with hundreds of thousands of links
 ;                     - Changed youtube title fetcher to use youtube API for faster download & faster parsing
 ;                     - Some bug fixes and general performance improvements
 ; UPDATE 14/06/2013: Added support for fetching and storing title from youtube links
 ;                     - fetches youtube video's title, stores it into the logfile, echoes the title on the channel
 ;                     - hold down ctrl key when pressing enter, when you post a link, to send the title as /msg to the channel (overrides %sendYTtitleChans variable)
 ;                     - set %sendYTtitleChans variable for channels you want to /msg the title automatically when a link is posted
 ;                       (for example: %sendYTtitleChans #channel1 #channel2 #channel3) would send the title into the channel on those 3 channels
 ; UPDATE 17/06/2007: Updated regex (fixed '<nowiki>www.someurl.com/something's_here</nowiki>' to show correctly), added a timer with 200ms delay
 ;                    into the on keyup event (to prevent filtering while writing fast).
 ; UPDATE 23/03/2007: Changed on keydown -> on keyup, much more efficient now, won't need so much checks, and allows you to
 ;                    insert text infront/into the middle of the line.
 ; UPDATE 29/01/2007: added a regex into on keydown event to halt the script if something else than \S was pressed
 ;                    (now holding down shift/alt/ctrl doesn't affect it), and optimized the on keydown event a bit.
 ; UPDATE 18/01/2007: added on input event for the search (optional), rearranged the dll-stuff into same place, 
 ;                    fixed a bug that didn't color some opened links without http:// (only www.)
 ;                    fixed if (!$exists($spopupdll)) { return } from linkpopup to 'halt' (now shows the original popup)
 ;                    added a $db alias to return the filename, users can choose their own path/filename for the file.
 ;
 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;  
 
 ; $db returns the name of the database file, change this to whatever you want
 alias -l db { return $qt($deltok($mircexe,-1,92) $+ \ $+ links.db) } 
 
 ; on start event for creating and loading the hash table
 on *:start:{
   linkhash_load
 }
 
 alias linkhash_load {
   ; free the old hashtable if it exists, and hmake a new one
   if ($hget(links.hash)) { hfree links.hash }
   hmake links.hash 1000
 
   ; filter links from DB file to the hashtable
   filter -fk $db LinkHash_copy *
 }
 
 alias linkhash_copy {
   tokenize 32 $1
   hadd links.hash $strip($3) $1-
 }
 
 ; alias to open the links window, and filter all of the urls into it
 alias urllog { 
   ; if the window isn't open, create it, -ea to create an editbox and activate the window
   ; filter will copy the lines from the file by a matchtext (* = everything), -fw = from File to Window, and -p = wrap the text
   if (!$window(@links)) { 
     window -ea @links 
     filter -fwp $db @links * 
   } 
 
   ; else the window is already open, so activate it
   else { window -a @links } 
 } 
 
 ; alias to return colorcodes for the script, change these values to whatever you like.
 ; new, old and opened are for echo text, log_link and log_info are for the @links window
 alias -l url_color {
   if ($1 == new) { return 12 }
   elseif ($1 == old) { return 6 }
   elseif ($1 == opened) { return 10 }
   elseif ($1 == log_link) { return 10 }
   elseif ($1 == log_info) { return 14 }
 }
 
 ; $clr alias to return $chr(3) and formatted color code, ie. $clr(3) == $+($chr(3),03) == ctrl+k 03
 alias -l clr { if ($isid) { return $+($chr(3),$base($1,10,10,2)) $+ $iif($2,$+($chr(44),$base($2,10,10,2)),) } }
 
 ; this alias handles checking the input for links, looping through tokens (words in this case), and replacing links with colored ones
 alias urlc {
   if (!$isid) { return }
   var %string = $1-
 
   ; if the line even has links, useless to loop through it if there are no links
   if ($regex(%string,/(?:\b(?:www.?\.|https?:\/\/|s?ftp:\/\/).)/i)) { 
 
     ; so the line has links, loop through tokens then
     var %x = 1 
     while (%x <= $numtok(%string,32)) { 
 
       ; we'll assign the current token into a variable %tok, for easier use later
       ; and we'll run it through $url_parse, and assign the result into a variable %url 
       ; (it'll be checked if it's a 'valid' url, and useless stuff be removed around it, if it's not an url the alias will return $null
       var %tok = $gettok(%string,%x,32)
       var %url = $url_parse(%tok) 
 
       ; check if there is an url in variable %url
       if (%url) { 
 
         ; $url_log will take care of storing the url into a database file, and return appropriate color (in $1) to use in coloring the link
         ; also returns the linenumber of newly added link (in $2) if we want to store the title for it
         ; and if we already have the title in the log, return that too in %titleLine
         var %indb = $url_log(%url)
         var %linkColor = $gettok(%indb,1,32)
         var %titleLine = $gettok(%indb,2-,32)
 
         ; now we color and underline the url and assign it into %newurl, (%indb will be holding the color code number)
         ; then we replace the url part of the token, so that from "url('www.something.com');" only the "www.something.com" part will be colored and underlined
         var %newurl = $+($clr(%linkColor),$chr(31),%url,$chr(31),$clr)
         var %newurl = $replace(%tok,%url,%newurl) 
 
         ; and finally, we put the modified token back to the original string
         %string = $puttok(%string,%newurl,%x,32)
 
         ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; OPTIONAL: GET YOUTUBE TITLE 
         ; this part is for checking if it's a youtube link, and initiates the fetching of the video's title
         ; this can be commented out or removed if you don't want this functionality
         if (*youtube* iswm %url) || (*youtu.be* iswm %url) {
           getTitle youtube %url $iif($chan,$chan,$iif($target != $me,$target,$nick)) %titleLine
         }
         ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; END OF OPTIONAL 
       }
       inc %x 
     }
   }
 
   ; modified or not, return the string
   return %string
 }
 
 ; alias to trim the given url, returns $null if input isn't an url
 ; big thanks to Msmo on Quakenet for this regex
 alias -l url_parse {
   var %p = /(?:^|[][()<>{}'"\s])((?:(?:irc|s?ftp|https?:)\/\/|www\d*\.)[^\s./]+\.(?:\[\S*?]|\(\S*?\)|\{\S*?}|[^][()<>{}'"\s]|['"](?!\s|$|[()\[\]<>]))++)/iS
   noop $regex($1,%p)
   return $regml(1)
 }
 
 ; alias to store the given url into database file, and return color according to whether it's new, opened, or unopened in database
 alias -l url_log {
   ; this first $iif is to append "http://" in front of "www.", so "www.google.com" and "<nowiki>http://www.google.com</nowiki>" will be a same link, and not to be written twice
   ; we'll assign the target channel/query into variable %t
   ; the $read will search the database file for the given url  
   var %url = $iif(www* iswm $1,$+(<nowiki>http://,$1</nowiki>),$1)
   var %t = $iif($left($target,1) != $chr(35),query,$target)
   var %isOld = $hget(links.hash,%url)
 
   ; if %isOld is empty (the link wasn't found from the file), write it in there and return color of "new link" & the line number the link was added to
   ; also add the info to the hashtable
   if (!%isOld) { 
     var %insertLine = $asctime(yyyy.mm.dd HH:nn:ss) $+($clr(03),$chr(31),%url,$chr(31),$clr()) $+($clr(07),$iif($nick,$nick,$me),$clr(08),@,$clr(07),%t,$clr()) 
     write $db %insertLine
     hadd links.hash %url %insertLine
 
     return $url_color(new) $lines($db)
   }
 
   ; otherwise, if the link was found, check if it has been opened via mirc (4th word will be !), return color of "opened link"
   elseif ($gettok(%isOld,4,32) == !) { 
     return $url_color(opened) __old %isOld
   }
 
   ; and if none of above, it's in the database, but you haven't opened it, return color of "old link"
   else { 
     return $url_color(old) __old %isOld
   }
 }
 
 ; this on keydown event triggers every time you press a key in @links window, 
 ; and filters the database file into the window with the text in editbox
 ; so, performs a new search everytime you write something in that window, i like this kind of search, but you can make it with on input
 ; if you wish to use search with on input (searches when you press enter), disable #onKEYsearch -group (change 'on' -> 'off')
 
 #onKEYsearch on
 on *:KEYUP:@links:*:{
   ; filter with the content of editbox
   ; timer to give a little delay for the filter (not filtering if writing fast), change the 200 (in milliseconds) to
   ; a bit higher value, if you're a slower writer.
   .timerURL -m 1 200 filter -cpfw $db @links $+(*,$editbox(@links),*)
 }
 #onKEYsearch end
 
 ; on input search for the @links window, if you want to use this instead of the on keydown, enable #onINPUTsearch -group (change 'off' -> 'on')
 #onINPUTsearch off
 on *:INPUT:@links: {
   filter -cpfw $db @links $+(*,$1-,*)
 }
 #onINPUTsearch end
 
 ; on hotlink to trigger on links, this first one is for doubleclick
 on *$:HOTLINK:/((www.?\.|https?:\/\/|ftp:\/\/).)/i:*:{ 
   ; clean up the triggered url with $url_parse and append "http://" in front of "www." 
   ; read the url info from database file, and if it doesn't have '!' as fourth token, add the '!' in there and write it back
   ; then open the url with url -n %url (change this, if you have your own alias for opening urls)
   var %url = $url_parse($1)
   var %url = $iif(www* iswm %url,$+(<nowiki>http://,%url</nowiki>),%url)
 
   url -n %url 
 
   var %oldurl = $read($db,w,$+(*,$chr(31),%url,$chr(31),*))
 
   if ($gettok(%oldurl,4,32) != !) {
     write $+(-l,$readn) $db $instok(%oldurl,!,4,32) 
     hadd links.hash %url $instok(%oldurl,!,4,32)
   } 
 }
 
 ; this second on hotlink is for the right click, shows the popup (leave it even if you're not going to use the spopup.dll)
 on ^*$:HOTLINK:/((www.?\.|https?:\/\/|ftp:\/\/).)/i:*:{ 
   ; check if the spopup.dll file is in the right place, if not, halt so it'll show the default popup
   if (!$exists($spopupdll)) { halt } 
 
   ; check if mouse button 2 have been pressed
   if ($mouse.key & 16) { popup_url $url_parse($1) }
 }
 
 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
 ;
 ; here starts the functionality with the spopup.dll
 ; you won't need this part if you don't want to use the dll
 ;
 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
 
 ; the path to spopup.dll (this default snipped uses the spopup.dll in the same folder where mirc.exe is located)
 alias -l spopupdll { return $qt($deltok($mircexe,-1,92) $+ \ $+ spopup.dll) }
 
 ; the new popup to replace the original one with a dll
 alias -l popup_url {
 
   ; strip colors etc. from given url ($1), and find the url info from the database file
   var %url = $strip($1)
   var %url2 = $strip($hget(links.hash,%url))
 
   ; assign the info variables
   var %datetime = $strip($gettok(%url2,1-2,32))
   var %nickchan = $strip($gettok(%url2,4-,32))
   var %nickchan = $iif($left(%nickchan,1) == !,$right(%nickchan,-2),%nickchan) 
 
   ; show the popup
   dll $spopupdll clear
   dll $spopupdll add 1 0 0 open
   dll $spopupdll add 2 0 0 copy url
   dll $spopupdll add 3 0 0 copy info
   dll $spopupdll add 4 0 0 %datetime %nickchan
 
   ; show additional "remove item" if you are in the linklist
   if ($active == @links) {
     dll $spopupdll add 5 0 0 -
     dll $spopupdll add 6 0 16 remove url from db
     dll $spopupdll add 7 1 0 ok
   }
 
   ; assign the popup selection into a variable
   var %c =  $dll($spopupdll,popup,$mouse.cx $mouse.cy 1)
 
   ; check what the selection was
   ; these are the ID's from the popup, 2) copy url 3) copy info 4) the info (echo's the info into active) 7) remove link (only in @links window) 
   if (%c == 1) { url -n %url }
   elseif (%c == 2) { clipboard %url }
   elseif (%c == 3) { clipboard %datetime %nickchan %url }
   elseif (%c == 4) { echo -agi30 url info: %datetime %nickchan %url }
   elseif (%c == 7) { write $+(-dl,$readn) $db | filter -fwp $db @links * | LinkHash_Load }
 }
 
 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
 ;
 ; this section is for the title fetching (youtube only at this time)
 ; you won't need this part if you don't want to use the title fetching funcitonality
 ;
 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
 
 ; this is the output handler alias that gets called from different parts of the script, depending on if we're dealing with an old title or new
 alias titleOutput {
   ; format the title, we'll need timers here to output text _after_ the link has been processed and is echoed
   var %title = $replace($gettok($remtok($iif($4 == __new,$1-,$5-),!,1,32),5-,32),|,$!chr(124)) 
 
   ; %inputSendLink will be true if you pressed enter holding down ctrl key, and it will message the title to the channel
   if (%inputSendLink) { 
     .timerYTTitle $+ _msg_ $+ $2 -m 1 1 msg $3 Youtube: %title 
   }
 
   ; checking if the channel where the link was posted, is in %sendYTtitleChans variable, if so, sends the title to the channel
   else if ($istok(%sendYTtitleChans,$3,32)) { 
     .timerYTTitle $+ _msg_ $+ $2 -m 1 1 msg $3 Youtube: %title 
   }
 
   ; echo the title of the link in the channel for you to see it
   .timerYTTitle $+ _echo_ $+ $2 -m 1 1 echo -t $3 Youtube: %title 
 }
 
 ; a helper alias to clean HTML entities from the titles
 alias stripHTMLEntities { 
   return $replace($1,&nbsp;,$chr(160),&quot;,",&lt;,<,&gt;,>,&amp;,&, $&
     &pound;, $chr(163), &euro;, $chr(8364), &copy;, $chr(169), &trade, $chr(8482), &reg;, $chr(174))
 }
 
 ; alias to handle different possible sources for the title (youtube only for now)
 alias gettitle {
   ; first we check if we already have the title in the log, if so, use a quick timer to echo it out after the link
   if ($4 == __old) { 
     titleOutput $1- 
     return 
   }
 
   ; get the title for youtube links, also put some info on the sockmark, $2 = youtube URL, $3- = rest of the info
   else if ($1 == youtube) { 
     var %sock = yttitle. $+ $ticks
     sockopen %sock gdata.youtube.com 80
     sockmark %sock $2 $3-
   }
 }
 
 ; socket events for youtube - sockopen
 on *:sockopen:yttitle.*:{ 
   if $sockerr { return }
 
   noop $regex($gettok($sock($sockname).mark,1,32),/(\?v=|&v=|\/\d\/|\/embed\/|\/v\/|\.be\/)([a-zA-Z0-9\-\_]{11})/)
 
   sockwrite -n $sockname GET /feeds/mobile/videos/ $+ $$regml(2) $+ ?v=2&prettyprint=true HTTP/1.1
   sockwrite -n $sockname Host: $sock($sockname).addr
   sockwrite -n $sockname $crlf
 }
 
 ; socket events for youtube - sockread
 on *:sockread:yttitle.*:{ 
   if $sockerr { return }
 
   var %data | sockread %data
 
   while ($sockbr) {
     if (*<title>*</title>* iswm %data) { 
       ; get the title 
       var %title = $regsubex(%data,/(.*<title>|</title>.*)/g,)
 
       ; send the title to titleOutput with other info
       titleOutput youtube $gettok($sock($sockname).mark,1,32) $gettok($sock($sockname).mark,2,32) __new %title
 
       ; if we set the linenumber info to the sockmark, we need to write the title into the DB file on that line
       if ($gettok($sock($sockname).mark,3,32)) {
         var %oldurl = $hget(links.hash,$gettok($sock($sockname).mark,1,32))
         write $+(-l,$v1) $db %oldurl %title
         hadd links.hash $gettok($sock($sockname).mark,1,32) %oldurl %title 
       }
 
       sockclose $sockname
       return
     }
     sockread %data
   }
 } 
 
 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
 ;
 ; the rest of the script is a part of my [[how to make your own theme]], and is only for an example usage of $urlc()
 ; basically you just need to run a line through $urlc() to color the links and log them
 ; you can add this into whatever event you like, ie. var %line = $urlc($1-) | echo %line
 ; this example also holds the if ($ctrlenter) { set %inputSendLink 1 } part, which is used in the title fetcher
 ;
 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
 
 on *:INPUT:*: {
   ; we will check that ctrl wasn't pressed while pressing enter, and that it's not a command
   if ($left($1,1) === $readini(mirc.ini,text,commandchar)) && (!$ctrlenter) { return } 
 
   ; check if ctrl was pressed while pressing enter, and store it in a variable (for sending title to the channel in case of a youtube link)
   unset %inputSendLink
   if ($ctrlenter) { set %inputSendLink 1 } 
 
   ; halt the default actions
   haltdef
 
   ; we'll put the line into a variable, using $urlc() to store & color links if there are any
   var %line = $urlc($1-)
 
   ; here is the part what we send as msg, we want to send $1- and not the modified variable
   .msg $target $1-
 
   ; then we have the echo
   echo $color(own text) -at $+(<,$iif($chan,$nick($chan,$me).pnick,$me),>) %line 
 }
 
 on ^*:TEXT:*:#: {
   ; halt the default actions and assign $1- into a variable, and run it through $urlc() to color the links and log them
   haltdef
   var %line = $urlc($1-) 
 
   ; echo the text into a right channel, -m highlights the switchbar button with "message color", -t adds timestamp
   echo $color(normal) -mt $chan $+(<,$nick($chan,$nick).pnick,>) %line
 } 
 
 on ^*:TEXT:*:?: {
   ; halt default actions, and set $1- into a variable, and run it through $urlc()
   haltdef
   var %line = $urlc($1-) 
 
   ; echo the text into a right query 
   echo $color(normal) -mt $nick $+(<,$nick,>) %line
 }