# :title: IMDB plugin for rbot
#
# Author:: Arnaud Cornet <arnaud.cornet@gmail.com>
+# Author:: Giuseppe "Oblomov" Bilotta <giuseppe.bilotta@gmail.com>
+#
# Copyright:: (C) 2005 Arnaud Cornet
-# License:: MIT license
+# Copyright:: (C) 2007 Giuseppe Bilotta
#
-# Notes by Giuseppe Bilotta:
-# TODO return more than one match (configurable)
+# License:: MIT license
require 'uri/common'
class Imdb
+ IMDB = "http://us.imdb.com"
+ TITLE_OR_NAME_MATCH = /<a href="(\/(?:title|name)\/(?:tt|nm)[0-9]+\/?)[^"]*"(?:[^>]*)>([^<]*)<\/a>/
+ TITLE_MATCH = /<a href="(\/title\/tt[0-9]+\/?)[^"]*"(?:[^>]*)>([^<]*)<\/a>/
+ NAME_MATCH = /<a href="(\/name\/nm[0-9]+\/?)[^"]*"(?:[^>]*)>([^<]*)<\/a>/
+ FINAL_ARTICLE_MATCH = /, ([A-Z]\S{0,2})$/
+
def initialize(bot)
@bot = bot
end
def search(rawstr)
- str = URI.escape(rawstr) << ";site=aka"
+ str = URI.escape(rawstr)
+ str << ";site=aka" if @bot.config['imdb.aka']
return do_search(str)
end
def do_search(str)
resp = nil
begin
- resp = @bot.httputil.get_response("http://us.imdb.com/find?q=#{str}",
+ resp = @bot.httputil.get_response(IMDB + "/find?q=#{str}",
:max_redir => -1)
rescue Exception => e
error e.message
end
if resp.code == "200"
- m = /<a href="(\/(?:title|name)\/(?:tt|nm)[0-9]+\/?)[^"]*"(?:[^>]*)>(?:[^<]*)<\/a>/.match(resp.body)
- if m
- url = m[1]
- return url
+ m = []
+ m << TITLE_OR_NAME_MATCH.match(resp.body) if @bot.config['imdb.popular']
+ if resp.body.match(/\(Exact Matches\)<\/b>/) and @bot.config['imdb.exact']
+ m << TITLE_OR_NAME_MATCH.match($')
+ end
+ m.compact!
+ unless m.empty?
+ return m.map { |mm|
+ mm[1]
+ }.uniq
end
elsif resp.code == "302"
- new_loc = resp['location'].gsub(/http:\/\/us.imdb.com/, "")
+ debug "automatic redirection"
+ new_loc = resp['location'].gsub(IMDB, "")
if new_loc.match(/\/find\?q=(.*)/)
return do_search($1)
else
- return new_loc.gsub(/\?.*/, "")
+ return [new_loc.gsub(/\?.*/, "")]
end
end
return nil
end
def info(rawstr)
- sr = search(rawstr)
- if !sr
+ urls = search(rawstr)
+ debug urls
+ if urls.nil_or_empty?
debug "IMDB: search returned NIL"
return nil
end
- type = sr.match(/^\/([^\/]+)\//)[1].downcase.intern rescue nil
- case type
- when :title
- return info_title(sr)
- when :name
- return info_name(sr)
- else
- return "#{sr}"
- end
+ results = []
+ urls.each { |sr|
+ type = sr.match(/^\/([^\/]+)\//)[1].downcase.intern rescue nil
+ case type
+ when :title
+ results << info_title(sr)
+ when :name
+ results << info_name(sr)
+ else
+ results << "#{sr}"
+ end
+ }
+ return results
end
def grab_info(info, body)
/<div class="info">\s+<h5>#{info}:<\/h5>\s+(.*?)<\/div>/mi.match(body)[1] rescue nil
end
+ def fix_article(org_tit)
+ title = org_tit.dup
+ if @bot.config['imdb.fix_article'] and title.gsub!(FINAL_ARTICLE_MATCH, '')
+ art = $1.dup
+ debug art.inspect
+ if art[-1,1].match(/[a-z]/)
+ art << " "
+ end
+ return art + title
+ end
+ return title
+ end
+
def info_title(sr)
resp = nil
begin
- resp = @bot.httputil.get_response('http://us.imdb.com' + sr,
- :max_redir => -1)
+ resp = @bot.httputil.get_response(IMDB + sr, :max_redir => -1)
rescue Exception => e
error e.message
warning e.backtrace.join("\n")
return nil
end
+ info = []
+
if resp.code == "200"
m = /<title>([^<]*)<\/title>/.match(resp.body)
return nil if !m
- title = Utils.decode_html_entities(m[1])
+ title_date = m[1]
+ pre_title, date, extra = title_date.scan(/^(.*)\((\d\d\d\d(?:\/[IV]+)?)\)\s*(.+)?$/).first
+ pre_title.strip!
+ title = fix_article(pre_title)
- m = /<b>([0-9.]+)\/10<\/b>\n?\r?\s+<small>\(<a href="ratings">([0-9,]+) votes?<\/a>\)<\/small>/.match(resp.body)
- return nil if !m
- score = m[1]
- votes = m[2]
+ dir = nil
+ data = grab_info(/Directors?/, resp.body)
+ if data
+ dir = data.scan(NAME_MATCH).map { |url, name|
+ name
+ }.join(', ')
+ end
- plot = nil
- data = grab_info(/Plot (?:Outline|Summary)/, resp.body)
+ country = nil
+ data = grab_info(/Country/, resp.body)
if data
- plot = "Plot: #{data.ircify_html.gsub(/\s+more$/,'')}"
+ country = data.ircify_html
+ end
+
+ info << [title, "(#{country}, #{date})", extra, dir ? "[#{dir}]" : nil, ": http://us.imdb.com#{sr}"].compact.join(" ")
+
+ ratings = "no votes"
+ m = /<b>([0-9.]+)\/10<\/b>\n?\r?\s+<small>\(<a href="ratings">([0-9,]+) votes?<\/a>\)<\/small>/.match(resp.body)
+ if m
+ ratings = "#{m[1]}/10 (#{m[2]} voters)"
end
genre = Array.new
resp.body.scan(/<a href="\/Sections\/Genres\/[^\/]+\/">([^<]+)<\/a>/) do |gnr|
genre << gnr
end
- info = "#{title} : http://us.imdb.com#{sr}\n"
- info << "Ratings: #{score}/10 (#{votes} voters). Genre: #{genre.join('/')}\n"
- info << plot if plot
+
+ plot = nil
+ data = grab_info(/Plot (?:Outline|Summary)/, resp.body)
+ if data
+ plot = "Plot: " + data.ircify_html.gsub(/\s+more$/,'')
+ end
+
+ info << ["Ratings: " << ratings, "Genre: " << genre.join('/') , plot].compact.join(". ")
+
return info
end
return nil
def info_name(sr)
resp = nil
begin
- resp = @bot.httputil.get_response('http://us.imdb.com' + sr,
- :max_redir => -1)
+ resp = @bot.httputil.get_response(IMDB + sr, :max_redir => -1)
rescue Exception => e
error e.message
warning e.backtrace.join("\n")
return nil
end
+ info = []
+
if resp.code == "200"
m = /<title>([^<]*)<\/title>/.match(resp.body)
return nil if !m
- name = Utils.decode_html_entities(m[1])
+ name = m[1]
+
+ info << "#{name} : http://us.imdb.com#{sr}"
birth = nil
data = grab_info("Date of Birth", resp.body)
death = "Death: #{data.ircify_html.gsub(/\s+more$/,'')}"
end
+ info << [birth, death].compact.join('. ') if birth or death
+
movies = {}
filmorate = nil
begin
- filmorate = @bot.httputil.get("http://us.imdb.com" + sr + "filmorate")
+ filmorate = @bot.httputil.get(IMDB + sr + "filmorate")
rescue Exception
end
if filmorate
filmorate.scan(/<div class="filmo">.*?<a href="\/title.*?<\/div>/m) { |str|
what = str.match(/<a name="[^"]+">([^<]+)<\/a>/)[1] rescue nil
- # next unless what
- next unless ['Actor', 'Director'].include?(what)
- movies[what] = str.scan(/<a href="\/title\/[^"]+">([^<]+)<\/a>/)[0..2].map { |tit|
- Utils.decode_html_entities(tit)
+ next unless what
+ movies[what] = str.scan(TITLE_MATCH)[0..2].map { |url, tit|
+ fix_article(tit)
}
}
end
- debug movies.inspect
- info = "#{name} : http://us.imdb.com#{sr}\n"
- info << [birth, death].compact.join('. ') << "\n"
+ preferred = ['Actor', 'Director']
+ if resp.body.match(/Jump to filmography as: (.*?)<\/div>/)
+ txt = $1
+ preferred = txt.scan(/<a[^>]+>([^<]+)<\/a>/)[0..2].map { |pref|
+ pref.first
+ }
+ end
+
unless movies.empty?
- info << "Top Movies:: "
+ all_keys = movies.keys.sort
+ debug all_keys.inspect
+ keys = []
+ preferred.each { |key|
+ keys << key if all_keys.include? key
+ }
+ keys = all_keys if keys.empty?
ar = []
- movies.keys.sort.each { |key|
+ keys.each { |key|
ar << key.dup
- ar.last << ": " + movies[key].join(', ')
+ ar.last << ": " + movies[key].join('; ')
}
info << ar.join('. ')
end
end
class ImdbPlugin < Plugin
+ BotConfig.register BotConfigBooleanValue.new('imdb.aka',
+ :default => true,
+ :desc => "Look for IMDB matches also in translated titles and other 'also known as' information")
+ BotConfig.register BotConfigBooleanValue.new('imdb.popular',
+ :default => true,
+ :desc => "Display info on popular IMDB entries matching the request closely")
+ BotConfig.register BotConfigBooleanValue.new('imdb.exact',
+ :default => true,
+ :desc => "Display info on IMDB entries matching the request exactly")
+ BotConfig.register BotConfigBooleanValue.new('imdb.fix_article',
+ :default => false,
+ :desc => "Try to detect an article placed at the end and move it in front of the title")
+
def help(plugin, topic="")
"imdb <string> => search http://www.imdb.org for <string>"
end
m.reply "Nothing found for #{what}"
return nil
end
- m.reply info
+ if info.length == 1
+ m.reply Utils.decode_html_entities info.first.join("\n")
+ else
+ m.reply info.map { |i|
+ Utils.decode_html_entities i.join(" | ")
+ }.join("\n")
+ end
end
end
projects / user / henk / code / ruby / rbot.git / blobdiff