#
# 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>/
- CREDIT_NAME_MATCH = /#{NAME_MATCH}<\/td><td[^>]+> \.\.\. <\/td><td[^>]+>(.+?)<\/td>/
+ IMDB = "http://www.imdb.com"
+ TITLE_OR_NAME_MATCH = /<a\s+href="(\/(?:title|name)\/(?:tt|nm)[0-9]+\/?)[^"]*"(?:[^>]*)>([^<]*)<\/a>/
+ TITLE_MATCH = /<a\s+href="(\/title\/tt[0-9]+\/?)[^"]*"(?:[^>]*)>([^<]*)<\/a>/
+ NAME_MATCH = /<a\s+href="(\/name\/nm[0-9]+\/?)[^"]*"(?:[^>]*)>([^<]*)<\/a>/
+ CREDIT_NAME_MATCH = /#{NAME_MATCH}\s*<\/td>\s*<td[^>]+>\s*\.\.\.\s*<\/td>\s*<td[^>]+>\s*(.+?)\s*<\/td>/m
FINAL_ARTICLE_MATCH = /, ([A-Z]\S{0,2})$/
+ DESC_MATCH = /<meta name="description" content="(.*?)\. (.*?)\. (.*?)\."\s*\/>/
MATCHER = {
:title => TITLE_MATCH,
end
def grab_info(info, body)
- /<div class="info">\s+<h5>#{info}:<\/h5>\s+(.*?)<\/div>/mi.match(body)[1] rescue nil
+ /<div (?:id="\S+-info" )?class="(?:txt-block|see-more inline canwrap)">\s*<h[45](?: class="inline")?>\s*#{info}:\s*<\/h[45]>\s*(.*?)<\/div>/mi.match(body)[1] rescue nil
end
def fix_article(org_tit)
if resp.code == "200"
m = /<title>([^<]*)<\/title>/.match(resp.body)
return nil if !m
- title_date = m[1]
- pre_title, date, extra = title_date.scan(/^(.*)\((\d\d\d\d(?:\/[IV]+)?)\)\s*(.+)?$/).first
+ title_date = m[1].ircify_html
+ debug title_date
+ # note that the date dash for series is a - (ndash), not a - (minus sign)
+ pre_title, extra, date, junk = title_date.scan(/^(.*)\((.+?\s+)?(\d\d\d\d(?:–(?:\d\d\d\d)?)?(?:\/[IV]+)?)\)\s*(.+)?$/).first
+ extra.strip! if extra
pre_title.strip!
- title = fix_article(pre_title.ircify_html)
+ title = fix_article(pre_title)
dir = nil
- data = grab_info(/Directors?/, resp.body)
+ data = grab_info(/(?:Director|Creator)s?/, resp.body)
if data
dir = data.scan(NAME_MATCH).map { |url, name|
name.ircify_html
country = data.ircify_html.gsub(' / ','/')
end
- info << [title, "(#{country}, #{date})", extra, dir ? "[#{dir}]" : nil, ": http://us.imdb.com#{sr}"].compact.join(" ")
+ info << [title, "(#{country}, #{date})", extra, dir ? "[#{dir}]" : nil, opts[:nourl] ? nil : ": http://www.imdb.com#{sr}"].compact.join(" ")
return info if opts[:title_only]
if opts[:characters]
info << resp.body.scan(CREDIT_NAME_MATCH).map { |url, name, role|
- "%s: %s" % [name, role]
+ "%s: %s" % [name, role.ircify_html]
}.join('; ')
return info
end
ratings = "no votes"
- m = /<b>([0-9.]+)\/10<\/b>\n?\r?\s+<small>\(<a href="ratings">([0-9,]+) votes?<\/a>\)<\/small>/.match(resp.body)
+ m = resp.body.match(/<b>([0-9.]+)<\/b><span [^>]+>\/10<\/span><\/span>\s*[^<]+<a\s+[^>]*href="ratings"[^>]+>([0-9,]+) votes?<\/a>/m)
if m
ratings = "#{m[1]}/10 (#{m[2]} voters)"
end
genre = Array.new
- resp.body.scan(/<a href="\/Sections\/Genres\/[^\/]+\/">([^<]+)<\/a>/) do |gnr|
+ resp.body.scan(/<a href="\/genre\/[^"]+">([^<]+)<\/a>/) do |gnr|
genre << gnr
end
- plot = nil
- data = grab_info(/Plot (?:Outline|Summary)/, resp.body)
- if data
- plot = "Plot: " + data.ircify_html.gsub(/\s+more$/,'')
- end
+ plot = resp.body.match(DESC_MATCH)[3] rescue nil
+ # TODO option to extract the long storyline
+ # data = resp.body.match(/<h2>Storyline<\/h2>\s+/m).post_match.match(/<\/p>/).pre_match rescue nil
+ # if data
+ # data.sub!(/<em class="nobr">Written by.*$/m, '')
+ # plot = data.ircify_html.gsub(/\s+more\s*$/,'').gsub(/\s+Full summary » \| Full synopsis »\s*$/,'')
+ # end
+ plot = "Plot: #{plot}" if plot
info << ["Ratings: " << ratings, "Genre: " << genre.join('/') , plot].compact.join(". ")
if resp.code == "200"
m = /<title>([^<]*)<\/title>/.match(resp.body)
return nil if !m
- name = m[1]
+ name = m[1].sub(/ - IMDb/, '')
- info << "#{name} : http://us.imdb.com#{sr}"
+ info << name
+ info.last << " : http://www.imdb.com#{sr}" unless opts[:nourl]
return info if opts[:name_only]
- if year = opts[:movies_in_year]
+ if opts[:movies_by_year]
filmoyear = @bot.httputil.get(IMDB + sr + "filmoyear")
if filmoyear
- info << filmoyear.scan(/#{TITLE_MATCH} \(#{year}\)[^\[\n]*((?:\s+\[[^\]]+\](?:\s+\([^\[<]+\))*)+)\s+</)
+ info << filmoyear.scan(/#{TITLE_MATCH} \((\d\d\d\d)\)[^\[\n]*((?:\s+\[[^\]]+\](?:\s+\([^\[<]+\))*)+)\s+</)
end
return info
end
return nil
end
- def year_movies(urls, year)
+ def year_movies(urls, years_txt_org, role_req)
+ years_txt = years_txt_org.dup
+ years_txt.sub!(/^'/,'')
+ years_txt = "9#{years_txt}" if years_txt.match(/^\d\ds?$/)
+ years_txt = "1#{years_txt}" if years_txt.match(/^\d\d\ds?$/)
+
+ years = []
+ case years_txt
+ when /^\d\d\d\d$/
+ years << years_txt
+ when /^(\d\d\d\d)s$/
+ base = $1.to_i
+ base.upto(base+9) { |year|
+ years << year.to_s
+ }
+ end
+
urls.map { |url|
- info = info_name(url, :movies_in_year => year)
+ info = info_name(url, :movies_by_year => true)
debug info.inspect
bclip = b[1][0,5]
quot = '"'
(aclip == quot ? 1 : -1) <=> (bclip == quot ? 1 : -1)
- }.each { |url, pre_title, pre_roles|
+ }.each { |url, pre_title, year, pre_roles|
+ next unless years.include?(year)
title = fix_article(pre_title.ircify_html)
if title[0] == ?" and not @bot.config['imdb.tv_series_in_movies']
next
end
+ title << " (#{year})" unless years.length == 1
role_array = []
pre_roles.strip.scan(/\[([^\]]+)\]((?:\s+\([^\[]+\))+)?/) { |txt, comm|
+ role = nil
+ extra = nil
if txt.match(/^(.*)\s+\.\.\.\.\s+(.*)$/)
- role_array << "#{$1} (#{$2})"
+ role = $1
+ extra = "(#{$2.ircify_html})"
else
- role_array << txt
+ role = txt
end
- role_array.last << " " + comm.ircify_html if comm
+ next if role_req and not role.match(/^#{role_req}/i)
+ if comm
+ extra ||= ""
+ extra += comm.ircify_html if comm
+ end
+ role_array << [role, extra]
}
+ next if role_req and role_array.empty?
- roles = role_array.join(', ')
- movies << [roles, title].join(": ")
+ roles = role_array.map { |ar|
+ if role_req
+ ar[1] # works for us both if it's nil and if it's something
+ else
+ ar.compact.join(" ")
+ end
+ }.compact.join(', ')
+ roles = nil if roles.empty?
+ movies << [roles, title].compact.join(": ")
}
if movies.empty?
end
class ImdbPlugin < Plugin
- BotConfig.register BotConfigBooleanValue.new('imdb.aka',
+ Config.register Config::BooleanValue.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',
+ Config.register Config::BooleanValue.new('imdb.popular',
:default => true,
:desc => "Display info on popular IMDB entries matching the request closely")
- BotConfig.register BotConfigBooleanValue.new('imdb.exact',
+ Config.register Config::BooleanValue.new('imdb.exact',
:default => true,
:desc => "Display info on IMDB entries matching the request exactly")
- BotConfig.register BotConfigBooleanValue.new('imdb.fix_article',
+ Config.register Config::BooleanValue.new('imdb.fix_article',
:default => false,
:desc => "Try to detect an article placed at the end and move it in front of the title")
- BotConfig.register BotConfigBooleanValue.new('imdb.tv_series_in_movies',
+ Config.register Config::BooleanValue.new('imdb.tv_series_in_movies',
:default => false,
:desc => "Whether searching movies by person/year should also return TV series")
def help(plugin, topic="")
- "imdb <string> => search http://www.imdb.org for <string>: prefix <string> with 'name' or 'title' if you only want to search for people or films respectively, e.g.: imdb name ed wood"
+ case plugin
+ when "movies"
+ "movies by <who> in <years> [as <role>] => display the movies in the <years> where which <who> was <role>; <role> can be one of actor, actress, director or anything: if it's omitted, the role is defined by the prefix: \"movies by ...\" implies director, \"movies with ...\" implies actor or actress; the years can be specified as \"in the 60s\" or as \"in 1953\""
+ when /characters?/
+ "character played by <who> in <movie> => show the character played by <who> in movie <movie>. characters in <movie> => show the actors and characters in movie <movie>"
+ else
+ "imdb <string> => search http://www.imdb.org for <string>: prefix <string> with 'name' or 'title' if you only want to search for people or films respectively, e.g.: imdb name ed wood. see also movies and characters"
+ end
end
attr_reader :i
+ TITLE_URL = %r{^http://(?:[^.]+\.)?imdb.com(/title/tt\d+/)}
+ NAME_URL = %r{^http://(?:[^.]+\.)?imdb.com(/name/nm\d+/)}
+ def imdb_filter(s)
+ loc = Utils.check_location(s, TITLE_URL)
+ if loc
+ sr = loc.first.match(TITLE_URL)[1]
+ extra = $2 # nothign for the time being, could be fullcredits or whatever
+ res = i.info_title(sr, :nourl => true, :characters => (extra == 'fullcredits'))
+ debug res
+ if res
+ return {:title => res.first, :content => res.last}
+ else
+ return nil
+ end
+ end
+ loc = Utils.check_location(s, NAME_URL)
+ if loc
+ sr = loc.first.match(NAME_URL)[1]
+ extra = $2 # nothing for the time being, could be filmoyear or whatever
+ res = i.info_name(sr, :nourl => true, :movies_by_year => (extra == 'filmoyear'))
+ debug res
+ if res
+ name = res.shift
+ return {:title => name, :content => res.join(". ")}
+ else
+ return nil
+ end
+ end
+ return nil
+ end
+
def initialize
super
@i = Imdb.new(@bot)
+ @bot.register_filter(:imdb, :htmlinfo) { |s| imdb_filter(s) }
end
# Find a person or movie on IMDB. A :type (name/title, default both) can be
end
end
if info.length == 1
- m.reply Utils.decode_html_entities info.first.join("\n")
+ m.reply Utils.decode_html_entities(info.first.join("\n"))
else
m.reply info.map { |si|
Utils.decode_html_entities si.join(" | ")
end
# Find the movies with a participation of :who in the year :year
- # TODO: allow year to be either a year or a decade ('[in the] 1960s')
+ # TODO: allow year to be either a year or a decade ('[in the] 1960s')
#
def movies(m, params)
who = params[:who].to_s
- year = params[:year]
+ years = params[:years]
+ role = params[:role]
+ if role and role.downcase == 'anything'
+ role = nil
+ elsif not role
+ case params[:prefix].intern
+ when :with
+ role = /actor|actress/i
+ when :by
+ role = 'director'
+ end
+ end
name_urls = i.search(who, :type => :name)
unless name_urls
return
end
- movie_urls = i.year_movies(name_urls, year)
+ movie_urls = i.year_movies(name_urls, years, role)
debug movie_urls.inspect
debug movie_urls[0][1]
m.reply movie_urls.join("\n")
else
m.reply movie_urls.map { |si|
- si[1] = "no movies in #{year}" unless si[1]
+ si[1] = "no movies in #{years}" unless si[1]
Utils.decode_html_entities si.join(" | ")
}.join("\n")
end
plugin = ImdbPlugin.new
plugin.map "imdb [:type] *what", :requirements => { :type => /name|title/ }, :defaults => { :type => 'both' }
-plugin.map "movies :prefix *who in :year", :requirements => { :prefix => /with|by|from/, :year => /\d+/ }
+plugin.map "movies :prefix *who in [the] :years [as :role]", :requirements => { :prefix => /with|by|from/, :years => /'?\d+s?/ }
plugin.map "character [played] by *who in *movie"
plugin.map "character of *who in *movie"
plugin.map "characters in *movie", :action => :imdb
projects / user / henk / code / ruby / rbot.git / blobdiff