yahoo_jp_transit 0.1.0

data/README.txt

@@ -0,0 +1,17 @@
+= yahoo_jp_transit.rb
+
+Copyright (c) 2007 SUGAWARA Genki <sgwr_dts@yahoo.co.jp>
+
+== Example
+
+    require 'yahoo_jp_transit'
+
+    searcher = YahooJpTransit::Searcher.new
+    result = searcher.query('大宮(埼玉県)', '東京')
+
+    exit if result.empty?
+
+    begin
+      puts result
+    end while result = result.next_pagerequire
+

data/lib/yahoo_jp_transit.rb

@@ -0,0 +1,350 @@
+#
+# = yahoo_jp_transit.rb
+#
+# Copyright (c) 2007 SUGAWARA Genki <sgwr_dts@yahoo.co.jp>
+#
+# == Example
+#
+#     require 'yahoo_jp_transit'
+#
+#     searcher = YahooJpTransit::Searcher.new
+#     result = searcher.query('大宮(埼玉県)', '東京')
+#
+#     exit if result.empty?
+#
+#     begin
+#       puts result
+#     end while result = result.next_pagerequire
+#
+
+require 'cgi'
+require 'kconv'
+require 'open-uri'
+
+module YahooJpTransit
+  class SearchError < StandardError; end
+
+  class Searcher
+    @@url = 'http://transit.yahoo.co.jp/search'
+
+    DEPARTURE = 'DEP'
+    ARRIVAL   = 'ARR'
+    LAST      = 'LST'
+    NONE      = 'NON'
+
+    attr_reader :kcode
+
+    def initialize(kcode = $KCODE, proxy = nil)
+      @kcode = case kcode
+               when /\Ae/i
+                 Kconv::EUC
+               when /\As/i
+                 Kconv::SJIS
+               when /\Au/i
+                 Kconv::UTF8
+               end
+      @proxy = proxy
+    end
+
+    def query(from, to, options = {})
+      options = options.dup
+      defaults = {
+        :val_htmb          => :result,
+        :from              => from,
+        :p                 => to,
+        :sort              => 0,
+        :num               => 0,
+        :valtimekb         => DEPARTURE,
+        :val_dsmask_charge => :CHARGE,
+        :val_dsmask_air    => :AIR,
+        :val_search        => "\303\265\272\367"
+      }.merge(divide_datetime(Time.now))
+
+      if dt = options.delete(:datetime)
+        options.update(divide_datetime(dt))
+      end
+
+      if timekb = options.delete(:timekb)
+        options[:valtimekb] = timekb
+      end
+
+      if options.delete(:charge) == false or options.delete(:val_dsmask_charge) == false
+        defaults.delete(:val_dsmask_charge)
+      end
+
+      if options.delete(:air) == false or options.delete(:val_dsmask_air) == false
+        defaults.delete(:val_dsmask_air)
+      end
+
+      open("#{@@url}?#{query_string(defaults.merge(options))}", { :proxy => @proxy }) do |f|
+        Parser.new(f, self, from, to, options).parse
+      end
+    end # def query
+
+    private
+    def query_string(params)
+      params.map {|k, v|
+        k = CGI.escape("#{k}")
+        v = CGI.escape(Kconv.kconv("#{v}", Kconv::EUC, @kconv || Kconv::AUTO))
+        "#{k}=#{v}"
+      }.join('&')
+    end
+
+    def divide_datetime(datetime)
+      val_m = datetime.strftime('%M')
+
+      {
+        :val_yymm => datetime.strftime('%Y%m'),
+        :val_dd   => datetime.strftime('%d'),
+        :val_hh   => datetime.strftime('%H'),
+        :val_m1   => val_m.slice(0, 1),
+        :val_m2   => val_m.slice(1, 1)
+      }
+    end
+  end # class Searcher
+
+  class Parser
+    def initialize(input_stream, searcher, from, to, search_options)
+      @input_stream = input_stream
+      @searcher = searcher
+      @from = from
+      @to = to
+      @search_options = search_options
+    end
+
+    def parse
+      rs = []
+      vars = nil
+
+      @input_stream.each do |line|
+        raise(SearchError, parse_error) if line['"error_container"']
+        break if line['"page_control2"']
+
+        # parse page control
+        if line['"page_count"'] and not vars
+          vars = { :searcher => @searcher, :from => @from, :to => @to, :search_options => @search_options }
+          line = @input_stream.gets
+          count, range = line.scan(%r|<em>([^<]+)</em>|e).map {|i| i.first }
+          vars[:count] = count.to_i
+          vars[:head_num] = range.slice(0, 1).to_i
+          vars[:tail_num] = range.slice(-1, 1).to_i
+        end
+
+        next unless line['"result_order"']
+
+        rs << Result.new do |header_class, path_class|
+          header = parse_result_header(header_class)
+          paths = []
+          while parse_result_path(path_class, paths); end
+          [header, paths]
+        end
+      end
+
+      unless rs.empty?
+        rs.instance_variable_set(:@vars, vars)
+        def rs.count; @vars[:count]; end
+        def rs.head_num; @vars[:head_num]; end
+        def rs.tail_num; @vars[:tail_num]; end
+        def rs.page; (head_num.to_f / 3).ceil; end
+        def rs.pages; (count.to_f / 3).ceil; end
+        def rs.next_page;
+          if page < pages
+            options = @vars[:search_options].merge(:num => (3 * page))
+            @vars[:searcher].query(@vars[:from], @vars[:to], options)
+          else
+            nil
+          end
+        end
+        def rs.prev_page;
+          if page > 1
+            options = @vars[:search_options].merge(:num => (3 * (page - 2)))
+            @vars[:searcher].query(@vars[:from], @vars[:to], options)
+          else
+            nil
+          end
+        end
+      else
+        def rs.next_page; nil; end
+        def rs.prev_page; nil; end
+      end
+
+      return rs
+    end
+
+    private
+    def parse_result_header(header_class)
+      kwargs = {}
+      kwargs[:start] = search('"start') {|l| l.match(/>([^<]+)</e)[1] }
+      kwargs[:reach] = search('"reach') {|l| l.match(/>([^<]+)</e)[1] }
+      kwargs[:lapse] = search('"result_lapse') do |l|
+        l.match(%r|>(.+)</div>|e)[1].gsub(%r|</?span[^>]*>|e, '')
+      end
+      kwargs[:distance] = search('"result_distance') {|l| l.match(/>([^<]+)</e)[1] }
+      kwargs[:passage] = search('"result_passage') do |l|
+        l.match(%r|>(.+)</div>|e)[1].gsub(%r|</?span[^>]*>|e, '')
+      end
+      kwargs[:connection] = search('"result_connection') do |l|
+        l.match(%r|>(.+)</div>|e)[1].gsub(%r|</?span[^>]*>|e, '')
+      end
+      kwargs[:pass] = search('"result_pass') do |l|
+        l.match(%r|>(.+)</div>|e)[1].gsub(%r|</?em>|e, '')
+      end
+
+      header = header_class.new(kwargs)
+
+      if @searcher.kcode
+        header.kconv(@searcher.kcode)
+      end
+
+      return header
+    end
+
+    def parse_result_path(path_class, out)
+      kwargs = {}
+      kwargs[:station] = search(['[', "\241\316"]) {|l| l.match(%r|<em>(.+)</em>|e)[1] }
+
+      line = ''
+      line = @input_stream.gets until line["\265\376"]
+      line = @input_stream.gets until line =~ /\A<tr>/e || line =~ %r|\A</div>|e
+
+      if has_next = (line =~ /\A<tr>/e)
+        kwargs[:time] = search('<small>') {|l| l.match(%r|<small>([^<]+)</small>|e)[1] }
+        kwargs[:course], kwargs[:fare] = search('<small>') do |l|
+          desc = l.match(%r|>([^<]+)</font>|e)[1]
+          fare = (m = l.match(%r|>([^<]+)</td>|e)) ? m[1] : nil
+          [desc, fare]
+        end
+      end
+
+      path = path_class.new(kwargs)
+
+      if @searcher.kcode
+        path.kconv(@searcher.kcode)
+      end
+
+      out << path
+      return has_next
+    end
+
+    def parse_error
+      errs = []
+
+      @input_stream.each do |line|
+        break if line['</div>']
+
+        if line["\241\246"]
+          errs << line.gsub("\241\246", '').gsub(%r|</?[^>]+>|e, '').strip
+        end
+      end
+
+      if (errmsg = errs.join(', ')).empty?
+        errmsg = 'search error'
+      elsif @searcher.kcode
+        errmsg = Kconv.kconv(errmsg, @searcher.kcode, Kconv::EUC)
+      end
+
+      return errmsg
+    end
+
+    def search(strs, limit = 64)
+      if strs.kind_of?(String)
+        strs = [strs]
+      end
+
+      limit.times do
+        line = @input_stream.gets
+
+        if strs.any? {|str| line[str] }
+          return yield(line)
+        end
+      end
+
+      raise "cannot find strings: #{strs.join(', ')}"
+    end
+  end # class Parser
+
+  class Result
+    class Header
+      @@attrs = [ :start, :reach, :lapse, :distance, :passage, :connection, :pass ]
+      attr_reader *@@attrs
+
+      def initialize(kwargs)
+        @@attrs.each {|kw| self.instance_variable_set("@#{kw}", kwargs[kw]) }
+      end
+
+      def to_s
+        @@attrs.map {|kw|
+          val = self.instance_variable_get("@#{kw}")
+          "#{kw}=#{val}"
+        }.join(', ')
+      end
+
+      def kconv(out_code)
+        @@attrs.map do |kw|
+          val = self.instance_variable_get("@#{kw}") or next
+          val = Kconv.kconv(val, out_code, Kconv::EUC)
+          self.instance_variable_set("@#{kw}", val)
+        end
+      end
+
+      def to_hash
+        vals = @@attrs.map {|kw| self.instance_variable_get("@#{kw}") }
+        Hash[*(@@attrs.zip(vals).flatten)]
+      end
+    end
+
+    class Path
+      @@attrs = [ :station, :time, :course, :fare ]
+      attr_reader *@@attrs
+
+      def initialize(kwargs)
+        @@attrs.each {|kw| self.instance_variable_set("@#{kw}", kwargs[kw]) }
+      end
+
+      def has_next?
+        not @time.nil?
+      end
+
+      def kconv(out_code)
+        @@attrs.map do |kw|
+          val = self.instance_variable_get("@#{kw}") or next
+          val = Kconv.kconv(val, out_code, Kconv::EUC)
+          self.instance_variable_set("@#{kw}", val)
+        end
+      end
+
+      def to_hash
+        hash = {}
+        @@attrs.map do |kw|
+          val = self.instance_variable_get("@#{kw}")
+          hash[kw] = val if val
+        end
+        return hash
+      end
+    end
+
+    attr_reader :header, :paths
+
+    def initialize
+      @header, @paths = yield(Header, Path)
+    end
+
+    def to_yaml
+      yaml = ['---']
+
+      yaml << 'header:'
+      @header.to_hash.each {|k, v| yaml << "  #{k}: #{v}" }
+
+      yaml << 'paths:'
+      @paths.each do |path|
+        yaml << '  - ' + path.to_hash.map {|k, v| "#{k}: #{v}" }.join("#{$/}    ")
+      end
+
+      yaml.join($/)
+    end
+
+    def to_s
+      to_yaml
+    end
+  end # class Result
+end # module YahooJpTransit

metadata

@@ -0,0 +1,48 @@
+--- !ruby/object:Gem::Specification 
+rubygems_version: 0.9.2
+specification_version: 1
+name: yahoo_jp_transit
+version: !ruby/object:Gem::Version 
+  version: 0.1.0
+date: 2007-10-22 00:00:00 +09:00
+summary: Scraping library for Yahoo! route information.(http://transit.yahoo.co.jp/)
+require_paths: 
+- lib
+email: sgwr_dts@yahoo.co.jp
+homepage: http://yahoo_jp_transit.rubyforge.org
+rubyforge_project: yahoo_jp_transit
+description: Scraping library for Yahoo! route information.(http://transit.yahoo.co.jp/)
+autorequire: 
+default_executable: 
+bindir: bin
+has_rdoc: true
+required_ruby_version: !ruby/object:Gem::Version::Requirement 
+  requirements: 
+  - - ">"
+    - !ruby/object:Gem::Version 
+      version: 0.0.0
+  version: 
+platform: ruby
+signing_key: 
+cert_chain: 
+post_install_message: 
+authors: 
+- winebarrel
+files: 
+- README.txt
+- lib/yahoo_jp_transit.rb
+test_files: []
+
+rdoc_options: 
+- --main
+- README.txt
+extra_rdoc_files: 
+- README.txt
+executables: []
+
+extensions: []
+
+requirements: []
+
+dependencies: []
+