X-Git-Url: https://git.netwichtig.de/gitweb/?a=blobdiff_plain;f=data%2Frbot%2Fplugins%2Fimdb.rb;h=0f5c6024b6306ddcd73079500208dde7a02dbf55;hb=a5ba341ceac6a73b690c2d84001766b75bcb249f;hp=33e13b637295e8d5535337ed33e8b21897bdf3b5;hpb=2448d5a41b452f8e4e29e8f50d9624a3a6f62ed7;p=user%2Fhenk%2Fcode%2Fruby%2Frbot.git diff --git a/data/rbot/plugins/imdb.rb b/data/rbot/plugins/imdb.rb index 33e13b63..0f5c6024 100644 --- a/data/rbot/plugins/imdb.rb +++ b/data/rbot/plugins/imdb.rb @@ -12,14 +12,17 @@ # License:: MIT license class Imdb - IMDB = "http://us.imdb.com" - TITLE_OR_NAME_MATCH = /]*)>([^<]*)<\/a>/ - TITLE_MATCH = /]*)>([^<]*)<\/a>/ - NAME_MATCH = /]*)>([^<]*)<\/a>/ - CREDIT_NAME_MATCH = /#{NAME_MATCH}<\/td>]+> \.\.\. <\/td>]+>(.+?)<\/td>/ + IMDB = "https://www.imdb.com" + TITLE_OR_NAME_MATCH = /]*)>([^<]*)<\/a>/ + TITLE_MATCH = /]*)>([^<]*)<\/a>/ + NAME_MATCH = /]*)>([^<]*)<\/a>/ + CHAR_MATCH = /]*)>([^<]*)<\/a>/ + CREDIT_NAME_MATCH = /#{NAME_MATCH}\s*<\/td>\s*]+>\s*\.\.\.\s*<\/td>\s*]+>\s*(.+?)\s*<\/td>/m FINAL_ARTICLE_MATCH = /, ([A-Z]\S{0,2})$/ + DESC_MATCH = // MATCHER = { + :character => CHAR_MATCH, :title => TITLE_MATCH, :name => NAME_MATCH, :both => TITLE_OR_NAME_MATCH @@ -30,8 +33,12 @@ class Imdb end def search(rawstr, rawopts={}) + # allow the user to search directly for (movie) IDs + if rawstr.match /$tt\d+^/ then + return ["/movie/#{rawstr}/"] + end str = CGI.escape(rawstr) - str << ";site=aka" if @bot.config['imdb.aka'] + str << "&site=aka" if @bot.config['imdb.aka'] opts = rawopts.dup opts[:type] = :both unless opts[:type] return do_search(str, opts) @@ -52,6 +59,9 @@ class Imdb matcher = MATCHER[opts[:type]] if resp.code == "200" + if opts[:all] + return resp.body.scan(matcher).map { |m| m.first }.compact.uniq + end m = [] m << matcher.match(resp.body) if @bot.config['imdb.popular'] if resp.body.match(/\(Exact Matches\)<\/b>/) and @bot.config['imdb.exact'] @@ -99,7 +109,7 @@ class Imdb end def grab_info(info, body) - /
\s+
#{info}:<\/h5>\s+(.*?)<\/div>/mi.match(body)[1] rescue nil + /
\s*\s*#{info}:\s*<\/h[45]>\s*(.*?)<\/div>/mi.match(body)[1] rescue nil end def fix_article(org_tit) @@ -122,6 +132,8 @@ class Imdb def info_title(sr, opts={}) resp = nil begin + # movie urls without tailing / trigger a redirect + sr += '/' if sr[-1,1] != '/' resp = @bot.httputil.get_response(IMDB + sr, :max_redir => -1) rescue Exception => e error e.message @@ -134,13 +146,17 @@ class Imdb if resp.code == "200" m = /([^<]*)<\/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) + # also, the second date, if missing, is an no-break space + 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 @@ -165,21 +181,28 @@ class Imdb end ratings = "no votes" - m = /<b>([0-9.]+)\/10<\/b>\n?\r?\s+[^<]+<a href="ratings"[^>]+>([0-9,]+) votes?<\/a>/.match(resp.body) - if m - ratings = "#{m[1]}/10 (#{m[2]} voters)" + # parse imdb rating value: + if resp.body.match(/itemprop="ratingValue">([^<]+)</) + ratings = "#{$1}/10" + end + # parse imdb rating count: + if resp.body.match(/itemprop="ratingCount">([^<]+)</) + ratings += " (#{$1} voters)" end genre = Array.new - resp.body.scan(/<a href="\/Sections\/Genres\/[^\/]+\/">([^<]+)<\/a>/) do |gnr| - genre << gnr + resp.body.scan(/<a\s+href="\/genre\/[^\?]+\?[^"]+"\s+>([^<]+)<\/a>/) do |gnr| + genre << gnr.first.strip end - plot = nil - data = grab_info(/Plot (?:Outline|Summary)/, resp.body) - if data - plot = "Plot: " + data.ircify_html.gsub(/\s+more$/,'') - 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(". ") @@ -203,9 +226,9 @@ class Imdb if resp.code == "200" m = /<title>([^<]*)<\/title>/.match(resp.body) return nil if !m - name = m[1] + name = m[1].sub(/ - IMDb/, '') - info << "#{name}" + info << name info.last << " : http://www.imdb.com#{sr}" unless opts[:nourl] return info if opts[:name_only] @@ -296,8 +319,8 @@ class Imdb } end - urls.map { |url| - info = info_name(url, :movies_by_year => true) + urls.map { |u| + info = info_name(u, :movies_by_year => true) debug info.inspect @@ -357,14 +380,18 @@ class Imdb end def name_in_movie(name_urls, movie_urls) + debug name_urls info = [] movie_urls.each { |movie| title_info = info_title(movie, :title_only => true) valid = [] data = @bot.httputil.get(IMDB + movie + "fullcredits") - data.scan(CREDIT_NAME_MATCH).each { |url, name, role| - valid << [url, name.ircify_html, role.ircify_html] if name_urls.include?(url) + data.scan(CREDIT_NAME_MATCH).each { |url, name, role_data| + ch_url, role = role_data.scan(CHAR_MATCH).first + debug [ch_url, role] + wanted = name_urls.include?(url) || name_urls.include?(ch_url) + valid << [url, name.ircify_html, role.ircify_html] if wanted } valid.each { |url, name, role| info << "%s : %s was %s in %s" % [name, IMDB + url, role, title_info] @@ -509,17 +536,23 @@ class ImdbPlugin < Plugin # Find the character played by :who in :movie # def character(m, params) - who = params[:who].to_s movie = params[:movie].to_s - - name_urls = i.search(who, :type => :name) - unless name_urls - m.reply "nothing found about #{who}, sorry" + movie_urls = i.search(movie, :type => :title) + unless movie_urls + m.reply "movie #{who} not found, sorry" return end - movie_urls = i.search(movie, :type => :title) - unless movie_urls + if params[:actor] + who = params[:actor].to_s + type = :name + else + who = params[:character].to_s + type = :character + end + + name_urls = i.search(who, :type => type, :all => true) + unless name_urls m.reply "nothing found about #{who}, sorry" return end @@ -550,7 +583,7 @@ plugin = ImdbPlugin.new plugin.map "imdb [:type] *what", :requirements => { :type => /name|title/ }, :defaults => { :type => 'both' } 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 "character [played] by *actor in *movie" +plugin.map "character of *character in *movie" plugin.map "characters in *movie", :action => :imdb