ruby-mediawiki 0.1

data/lib/mediawiki/article.rb

@@ -0,0 +1,259 @@
+=begin
+    This file is part of Ruby-MediaWiki.
+
+    Ruby-MediaWiki is free software: you can redistribute it and/or
+    modify it under the terms of the GNU General Public License as
+    published by the Free Software Foundation, either version 3 of the
+    License, or (at your option) any later version.
+
+    Ruby-MediaWiki is distributed in the hope that it will be useful,
+    but WITHOUT ANY WARRANTY; without even the implied warranty of
+    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+    General Public License for more details.
+
+    You should have received a copy of the GNU General Public License
+    along with Ruby-MediaWiki.  If not, see
+    <http://www.gnu.org/licenses/>.
+=end
+
+begin
+  require 'htree'
+rescue LoadError
+  MediaWiki::logger.warn("htree library missing. Cannot sanitize HTML.")
+  require 'rexml/document'
+end
+
+
+module MediaWiki
+  ##
+  # The Article class represents MediaWiki articles.
+  class Article
+    ##
+    # Article name, will be refreshed upon Article#reload
+    attr_accessor :name
+    ##
+    # Article text, will be set by Article#reload
+    attr_accessor :text
+    ##
+    # this Article is read_only
+    attr_accessor :read_only
+
+    ##
+    # Create a new Article instance
+    # wiki:: [Wiki] instance to be used to theive the MiniBrowser
+    # name:: [String] Article name
+    # section:: [Fixnum] Optional article section number
+    # load_text:: [Boolean] Invoke Article#reload to retrieve Article#text
+    def initialize(wiki, name, section = nil, load_text=true)
+      @wiki = wiki
+      @name = name
+      @section = section
+
+      @text = nil
+      @xhtml = nil
+      @xhtml_cached = false
+      @wp_edittoken = nil
+      @wp_edittime = nil
+
+      reload if load_text
+    end
+
+    ##
+    # Return the full article name
+    #
+    # This will only return @name, but may be overriden by descendants
+    # to include namespaces.
+    # result:: [String] Full name
+    def full_name
+      @name
+    end
+
+    ##
+    # Return the URL of the article as configured
+    #
+    # This will return a nice human-readable URL if your MediaWiki
+    # is configured that way, unlike the generic URL returned by
+    # Wiki#full_article_url.
+    # result:: [String] URL
+    def url
+      uri = @wiki.url.dup
+      uri.path, uri.query = xhtml.elements['//li[@id="ca-nstab-main"]//a'].attributes['href'].split(/\?/, 2)
+      uri.to_s
+    end
+    
+    ##
+    # Return the URL of the talk page of the article
+    #
+    # This will return a nice human-readable URL to the talk page
+    # of an article if your MediWiki is configured that way.
+    # For empty talk pages this will return an ugly URL just
+    # as MediaWiki does.
+    # result:: [String] URL
+    def talk_url
+      uri = @wiki.url.dup
+      uri.path, uri.query = xhtml.elements['//li[@id="ca-talk"]//a'].attributes['href'].split(/\?/, 2)
+      uri.to_s
+    end
+
+    ##
+    # Get the XHTML,
+    # will invoke Article#xhtml_reload if not already cached
+    # result:: [REXML::Element] html root element
+    def xhtml
+      unless @xhtml_cached
+        xhtml_reload
+      end
+      @xhtml
+    end
+
+    ##
+    # Reload the xhtml,
+    # will be automatically done by Article#xhtml if not already cached.
+    def xhtml_reload
+      html = @wiki.browser.get_content("#{@wiki.article_url(full_name, @section)}")
+      @xhtml = to_rexml( html )
+      
+      @xhtml_cached = true
+    end
+
+    ##
+    # Reload Article#text,
+    # should be done by Article#initialize.
+    def reload
+      MediaWiki::logger.debug("Loading #{@wiki.article_url(full_name, @section)}&action=edit")
+      parse @wiki.browser.get_content("#{@wiki.article_url(full_name, @section)}&action=edit")
+    end
+
+    class NoEditFormFound < RuntimeError
+    end
+
+    def parse(html)
+      doc = to_rexml( html )
+      # does not work for MediaWiki 1.4.x and is always the same name you ask for under 1.5.x
+      # @name = doc.elements['//span[@class="editHelp"]/a'].attributes['title']
+      if form = doc.elements['//form[@name="editform"]']
+        # we got an editable article
+        @text = form.elements['textarea[@name="wpTextbox1"]'].text
+        begin
+          form.each_element('input') { |e|
+            @wp_edittoken = e.attributes['value'] if e.attributes['name'] == 'wpEditToken'
+            @wp_edittime = e.attributes['value'] if e.attributes['name'] == 'wpEdittime'
+          }
+          @read_only = false
+        rescue NoMethodError
+          # wpEditToken might be missing, that's ok
+        end
+      else
+        if doc.elements['//textarea']
+          # the article is probably locked and you do not have sufficient privileges
+          @text = doc.elements['//textarea'].text
+          @read_only = true
+        else
+          raise NoEditFormFound, "Error while parsing result, no edit form found"
+        end
+      end
+    end
+
+    ##
+    # Push the *Submit* button
+    #
+    # Send the modified Article#text to the MediaWiki.
+    # summary:: [String] Change summary
+    # minor_edit:: [Boolean] This is a Minor Edit
+    # watch_this:: [Boolean] Watch this article
+    def submit(summary, minor_edit=false, watch_this=false, retries=10)
+      raise "This Article is read-only." if read_only
+      MediaWiki::logger.debug("Posting to #{@wiki.article_url(full_name, @section)}&action=submit with wpEditToken=#{@wp_edittoken} wpEdittime=#{@wp_edittime}")
+      data = {'wpTextbox1' => @text, 'wpSummary' => summary, 'wpSave' => 1, 'wpEditToken' => @wp_edittoken, 'wpEdittime' => @wp_edittime}
+      data['wpMinoredit'] = 1 if minor_edit
+      data['wpWatchthis'] = 'on' if watch_this
+      begin
+        parse @wiki.browser.post_content("#{@wiki.article_url(full_name, @section)}&action=submit", data)
+      rescue NoEditFormFound
+        # This means, we havn't got the preview page, but the posted article
+        # So everything is Ok, but we must reload the edit page here, to get
+        # a new wpEditToken and wpEdittime
+        reload
+        return
+      rescue Net::HTTPInternalServerError
+      end
+
+      unless @wp_edittoken.to_s == '' and @wp_edittime.to_s == ''
+        if (data['wpEditToken'] != @wp_edittoken) or (data['wpEdittime'] != @wp_edittime)
+          if retries > 0
+            submit(summary, minor_edit, watch_this, retries - 1)
+          else
+            raise "Re-submit limit reached"
+          end
+        end
+      end
+    end
+
+    ##
+    # Delete this article
+    # reason:: [String] Delete reason
+    def delete(reason)
+      data = {'wpReason' => reason, 'wpEditToken' => @wp_edittoken, 'wpConfirmB' => 'Delete Page'}
+      result = @wiki.browser.post_content("#{@wiki.article_url(full_name)}&action=delete", data)
+    end
+
+    ##
+    # Protect this article
+    # reason:: [String] Protect reason
+    def protect(reason, moves_only=false)
+      data = {'wpReasonProtect' => reason, 'wpEditToken' => @wp_edittoken, 'wpConfirmProtectB' => 'Protect Page'}
+      data['wpMoveOnly'] = 1 if moves_only
+      result = @wiki.browser.post_content("#{@wiki.article_url(full_name)}&action=protect", data)
+    end
+
+    ##
+    # Unprotect this article
+    # reason:: [String] Unprotect reason
+    def unprotect(reason)
+      data = {'wpReasonProtect' => reason, 'wpEditToken' => @wp_edittoken, 'wpConfirmProtectB' => 'Protect Page'}
+      result = @wiki.browser.post_content("#{@wiki.article_url(full_name)}&action=unprotect", data)
+    end
+
+    ##
+    # "what links here" url for this article
+    def what_links_here_url(count = nil)
+      url = @wiki.article_url("Special:Whatlinkshere/#{full_name}")
+      url << "&limit=#{count}" if count
+    end
+    
+
+    ##
+    # What articles link to this article?
+    # result:: [Array] of [String] Article names
+    def what_links_here(count = nil)
+      res = []
+      url = what_links_here_url(count)
+      links = to_rexml(@wiki.browser.get_content(url))
+      links.each_element('//div[@id="bodyContent"]//ul/li/a') { |a|
+        res << a.attributes['title']
+      }
+      res
+    end
+    
+    def fast_what_links_here(count = nil)
+      res = []
+      url = what_links_here_url(count)
+      content = @wiki.browser.get_content(url)
+      content.scan(%r{<li><a href=".+?" title="(.+?)">.+?</a>.+?</li>}).flatten.map { |title|
+        REXML::Text.unnormalize(title)
+      }
+    end
+
+  protected
+    def to_rexml( html )
+      if Class.constants.member?( 'HTree' )
+        rexml = HTree( html ).to_rexml
+      else
+        rexml = REXML::Document.new( html )
+      end
+      rexml.root
+    end
+
+  end
+
+end

data/lib/mediawiki/category.rb

@@ -0,0 +1,54 @@
+=begin
+    This file is part of Ruby-MediaWiki.
+
+    Ruby-MediaWiki is free software: you can redistribute it and/or
+    modify it under the terms of the GNU General Public License as
+    published by the Free Software Foundation, either version 3 of the
+    License, or (at your option) any later version.
+
+    Ruby-MediaWiki is distributed in the hope that it will be useful,
+    but WITHOUT ANY WARRANTY; without even the implied warranty of
+    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+    General Public License for more details.
+
+    You should have received a copy of the GNU General Public License
+    along with Ruby-MediaWiki.  If not, see
+    <http://www.gnu.org/licenses/>.
+=end
+
+require 'rexml/document'
+require 'mediawiki/article'
+
+module MediaWiki
+  ##
+  # The Category class represents MediaWiki categories.
+  class Category < Article
+    ##
+    # This returns the full article name prefixed with "Category:"
+    # instead of the name, which should not carry a prefix.
+    def full_name
+      "Category:#{@name}"
+    end
+
+    ##
+    # Calls the reload function of the super-class (Article#reload)
+    # but removes the prefix (namespace) then.
+    #
+    # Use to full_name to obtain the name with namespace.
+    def reload
+      super
+      @name.sub!(/^.+?:/, '')
+    end
+    
+    ##
+    # Which articles belong to this category?
+    # result:: [Array] of [String] Article names
+    def articles
+      res = []
+      xhtml.each_element('//div[@id="bodyContent"]//ul/li/a') { |a,|
+        res << a.attributes['title']
+      }
+      res
+    end
+  end
+end

data/lib/mediawiki/dotfile.rb

@@ -0,0 +1,51 @@
+=begin
+    This file is part of Ruby-MediaWiki.
+
+    Ruby-MediaWiki is free software: you can redistribute it and/or
+    modify it under the terms of the GNU General Public License as
+    published by the Free Software Foundation, either version 3 of the
+    License, or (at your option) any later version.
+
+    Ruby-MediaWiki is distributed in the hope that it will be useful,
+    but WITHOUT ANY WARRANTY; without even the implied warranty of
+    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+    General Public License for more details.
+
+    You should have received a copy of the GNU General Public License
+    along with Ruby-MediaWiki.  If not, see
+    <http://www.gnu.org/licenses/>.
+=end
+
+require 'yaml'
+require 'mediawiki'
+
+module MediaWiki
+  ##
+  # dotfile function reads the user's MediaWiki config and
+  # creates a Wiki instance.
+  #
+  # The filename is determined by the environment variable
+  # MEDIAWIKI_RC or defaults to ~/.mediawikirc .
+  #
+  # A configured wiki can be chosen with the MEDIAWIKI_WIKI
+  # environment variable, by the option mywiki or defaults 
+  # to the wiki pointed by default.
+  #
+  # A robot may set [myrealm] to retrieve a second result
+  # output: a section with this name in the current wiki's
+  # configuration file for configuration of specific robot
+  # tasks.
+  def MediaWiki.dotfile(myrealm=nil,mywiki=nil)
+    filename = ENV['MEDIAWIKI_RC'] || "#{ENV['HOME']}/.mediawikirc"
+    dotfile = YAML::load(File.new(filename))
+
+    wikiconf = dotfile[mywiki] || dotfile[ENV['MEDIAWIKI_WIKI'] || dotfile['default']]
+    wiki = Wiki.new(wikiconf['url'], wikiconf['user'], wikiconf['password'])
+
+    if myrealm
+      [wiki, wikiconf[myrealm]]
+    else
+      wiki
+    end
+  end
+end

data/lib/mediawiki/minibrowser.rb

@@ -0,0 +1,140 @@
+=begin
+    This file is part of Ruby-MediaWiki.
+
+    Ruby-MediaWiki is free software: you can redistribute it and/or
+    modify it under the terms of the GNU General Public License as
+    published by the Free Software Foundation, either version 3 of the
+    License, or (at your option) any later version.
+
+    Ruby-MediaWiki is distributed in the hope that it will be useful,
+    but WITHOUT ANY WARRANTY; without even the implied warranty of
+    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+    General Public License for more details.
+
+    You should have received a copy of the GNU General Public License
+    along with Ruby-MediaWiki.  If not, see
+    <http://www.gnu.org/licenses/>.
+=end
+
+require 'net/http'
+require 'net/https'
+require 'cgi'
+
+module MediaWiki
+  ##
+  # The MiniBrowser is used to perform GET and POST requests
+  # over HTTP and HTTPS, supporting:
+  # * HTTP-Auth encoding in URLs (proto://user:password@host/...)
+  # * Cookie support
+  # * HTTP Redirection (max. 10 in a row)
+  #
+  # All interaction with MiniBrowser is normally done by
+  # MediaWiki::Wiki.
+  class MiniBrowser
+    ##
+    # Initialize a MiniBrowser instance
+    # url:: [URI::HTTP] or [URI::HTTPS]
+    def initialize(url)
+      @url = url
+      @http = Net::HTTP.new( @url.host, @url.port )
+      @http.use_ssl = true if @url.class == URI::HTTPS
+      @user_agent = 'WikiBot'
+      @cookies = {}
+    end
+
+    ##
+    # Add cookies to the volatile cookie cache
+    # cookies:: [Array]
+    def add_cookie(cookies)
+      cookies.each do | c |
+        c.gsub!(/;.*$/, '')
+        if match = c.match(/([^=]+)=(.*)/)
+          @cookies[match[1]] = match[2]
+        end
+      end
+    end
+
+    ##
+    # Get the cookie cache in a serialized form ready for HTTP.
+    # result:: [String]
+    def cookies
+      c = @cookies.collect do | key, value | "#{key}=#{value}" end
+      c.join(";")
+    end
+
+    ##
+    # Perform a GET request
+    #
+    # This method accepts 10 HTTP redirects at max.
+    # url:: [String]
+    # result:: [String] Document
+    def get_content(url)
+      retries = 10
+
+      @http.start { |http|
+        loop {
+          raise "too many redirects" if retries < 1
+
+          request = Net::HTTP::Get.new(url, {'Content-Type' => 'application/x-www-form-urlencoded',
+                                              'User-Agent' => @user_agent,
+                                              'Cookie' => cookies})
+          request.basic_auth(@url.user, @url.password) if @url.user
+          response = http.request(request)
+
+          case response 
+            when Net::HTTPSuccess, Net::HTTPNotFound then 
+              return response.body
+            when Net::HTTPRedirection then
+              MediaWiki::logger.debug("Redirecting to #{response['Location']}")
+              retries -= 1
+              url = response['Location']
+            else
+              raise "Unknown Response: #{response.inspect}"
+          end
+        }
+      }
+    end
+
+    ##
+    # Perform a POST request
+    #
+    # Will switch to MiniBrowser#get_content upon HTTP redirect.
+    # url:: [String]
+    # data:: [Hash] POST data
+    # result:: [String] Document
+    def post_content(url, data)
+      post_data = data.collect { | key, value | "#{CGI::escape(key.to_s)}=#{CGI::escape(value.to_s)}" }.join('&')
+      response = nil
+
+      @http.start { |http|
+        request = Net::HTTP::Post.new(url, {'Content-Type' => 'application/x-www-form-urlencoded',
+                                             'User-Agent' => @user_agent,
+                                             'Cookie' => cookies})
+        request.basic_auth(@url.user, @url.password) if @url.user
+        response = http.request(request, post_data)
+      }
+
+      case response 
+        when Net::HTTPSuccess 
+        then
+          begin
+            add_cookie( response.get_fields('Set-Cookie') ) if response['Set-Cookie']
+          rescue NoMethodError
+            add_cookie( response['Set-Cookie'] ) if response['Set-Cookie']
+          end
+          return response.body
+        when Net::HTTPRedirection
+        then
+          MediaWiki::logger.debug("Redirecting to #{response['Location']}")
+          begin
+            add_cookie( response.get_fields('Set-Cookie') ) if response['Set-Cookie']
+          rescue NoMethodError
+            add_cookie( response['Set-Cookie'] ) if response['Set-Cookie']
+          end
+          return get_content(response['Location'])
+        else
+          raise "Unknown Response on #{url}: #{response.inspect}"
+      end
+    end
+  end
+end