wcid 0.0.1

data/bin/wcid_ex1.rb

@@ -0,0 +1,12 @@
+#!/usr/bin/ruby -w
+require 'wcid'
+include WCID
+
+begin
+  authorities = marc_file_to_auths(ARGV[0])
+rescue
+  puts 'There has been an error! Check that the WorldCat Identities site is up at http://orlabs.oclc.org/Identities/ before complaining to the developer. The site is often down as it is now only in beta.'
+  exit
+end
+
+puts authorities

data/examples/wcid_long_ex1.rb

@@ -0,0 +1,84 @@
+#!/usr/bin/ruby -w
+## Jason Ronallo
+# May 2007
+
+require 'rubygems'
+require "rexml/document"
+require "rexml/xpath"
+require 'net/http'
+require 'uri'
+require 'marc'
+require 'stringio'
+require 'worldcatid'
+
+#file = ''
+if ARGF.filename == '-'
+  print "Enter the full path to a MARC file: "
+  file = (STDIN.gets).chomp  
+else
+  file = ARGF
+end
+
+#wc_search = (WCID::Search.new('Hull, Richard 1945')).search_exact
+x = 0
+begin
+  records = MARC::Reader.new(file).to_a
+rescue MARC::Exception
+  puts "MARC Exception"
+  exit
+rescue Exception => e
+  puts e
+  print "Enter a valid filename: "
+  file = STDIN.gets.chomp
+  retry
+end
+puts "records.length: #{records.length}"
+#STDIN.gets
+records.each do | record |
+  fields = record.name_fields
+  puts "fields.length #{fields.length}"
+  #STDIN.gets
+  fields.each do | field |
+    query = field.pnkey_from_marc
+    rec = (WCID::Search.new(query)).search_exact
+    if rec == nil
+      puts "There's no rec for #{query}"
+      puts "Hit enter to continue"
+      STDIN.gets
+      next
+    end
+    puts "class of rec: #{rec.class}"
+    
+    lc_retries = 0
+    begin
+      auth = rec.get_lc_auth
+    rescue Timeout::Error, Exception => e
+      #puts e; STDIN.gets
+      if e.to_s == "eof"
+        puts "We'll retry in a long time...."
+        sleep 1000
+        retry
+      end
+      if lc_retries < 5
+        puts "retrying...."
+        lc_retries += 1
+        puts
+        puts "retry #{lc_retries}"
+        retry
+      else
+        puts "you're outta luck on the lc_auth lookup"
+        next
+      end
+    end
+    if auth == nil
+      puts "auth nil"
+      STDIN.gets
+      next
+    end
+    puts auth
+    puts auth.inspect
+    STDIN.gets
+  end
+end
+
+puts "this is the end of the line"

data/lib/wcid.rb

@@ -0,0 +1,45 @@
+#wcid is a ruby library for handling WorldCat Identities
+#for more information on WorldCat Identities see #http://orlabs.oclc.org/Identities/
+#
+#USAGE
+#
+#   require 'wcid'
+
+
+require 'rubygems'
+require 'rexml/document'
+require 'net/http'
+require 'uri'
+require 'marc'
+require 'stringio'
+
+require 'wcid/version'
+
+require 'wcid/marc'
+require 'wcid/search'
+require 'wcid/hit'
+require 'wcid/id'
+
+module WCID
+  
+  #given one or more MARC records
+  #it returns an array of associated name authorities
+  #as MARC::Record objects 
+  
+  def marc_file_to_auths(filename)
+    authorities = []
+    records = MARC::Reader.new(filename).to_a
+    records.each do | record |
+      fields = record.name_fields
+      fields.each do | field |
+        query = field.pnkey_from_marc
+        wcid_obj = (WCID::Search.new(query)).search_exact
+        next if wcid_obj.nil?
+        auth = wcid_obj.get_lc_auth
+        next if auth.nil?
+        authorities << auth
+      end
+    end
+    return authorities
+  end
+end

data/lib/wcid/hit.rb

@@ -0,0 +1,34 @@
+module WCID
+  
+  #Conducting a search of the WorldCat Identities site returns a WCID::Hit
+  #object.
+  
+  class Hit
+    
+    
+    
+    attr_reader :score, :established_form, :uri, :citation, :lccn, :pubdates, :name_type 
+
+    #Hit objects contain some potentially useful information, but of 
+    #primary interest to me is the lccn which allows you to grab the 
+    #authority record of a particular hit.
+
+    def initialize m
+      @score = m.attributes['score']
+      @established_form = m.elements["establishedForm"].text if m.elements["establishedForm"]
+      @uri = m.elements['uri'].text if m.elements['uri']
+      @citation = m.elements['citation'].text if m.elements['citation']
+      @lccn = m.elements['lccn'].text if m.elements['lccn']
+      @pubdates = m.elements['pubDates'].text if m.elements['pubDates']
+      @name_type = m.elements['nameType'].text if m.elements['nameType']
+    end 
+
+    #Given a WCID::Hit object it returns a single authority record
+    #as a MARC::Record object.
+    def get_lc_auth
+      src = Net::HTTP.get(URI.parse("http://errol.oclc.org/laf/#{self.lccn}.MarcXML"))
+      record = MARC::XMLReader.new(StringIO.new(string=src)).to_a
+      record = record[0]
+    end
+  end 
+end

data/lib/wcid/id.rb

@@ -0,0 +1,80 @@
+module WCID
+  
+  #A WCID::ID object is a single Identity record in WorldCat Identities.
+  #
+  
+  class ID
+    #lccn = Library of Congress Control Number
+    #type = 
+    #wiki_link = link to Wikipedia for this identity
+    #subfield_a = name
+    #subfield_d = dates associated with the name
+    #pnkey = each WorldCat Identities record is given a unique pnkey made up 
+    #        the name and associated dates (and qualifier?). A pnkey is the most exact way to 
+    #        search WorldCat Identities.
+    
+    attr_reader :lccn, :type, :wiki_link, :subfield_a, :subfield_d, :pnkey
+    def initialize doc
+      @record = doc.elements.to_a("searchRetrieveResponse/records/record")
+      @identity = @record[0].elements['recordData/Identity']
+      @pnkey = @identity.elements['pnkey'].text
+      @lccn_record = @identity.elements['authorityInfo/lccn'].text
+      @lccn = convert_lccn(@lccn_record)
+      @identity = @record[0].elements['recordData/Identity'] 
+      @type = @identity.attributes['type']
+      @wiki_link = "http://en.wikipedia.org/wiki/#{(@identity.elements['authorityInfo/wikiLink']).text}" if @identity.elements['authorityInfo/wikiLink']
+      @subfield_a = @identity.elements['authorityInfo/standardForm/suba'].text if @identity.elements['authorityInfo/standardForm/suba']
+      @subfield_d = @identity.elements['authorityInfo/standardForm/subd'].text if @identity.elements['authorityInfo/standardForm/subd']
+    end
+
+    #A method to convert the Library of Congress Control Number as represented
+    #in the WorldCat Identities database into an LCCN subtable for searching
+    #the Linked Authority File.
+    
+    def convert_lccn(lccn)
+      lccn.gsub!(' ', '')
+      ll = lccn.length
+      if lccn[ll - 6, 1] != '0'
+        lccn = lccn[0, ll - 6] + "-" + lccn[ll - 6, 99]
+      elsif lccn[ll - 5, 1] != '0'
+        lccn = lccn[0, ll - 6] + "-" + lccn[ll - 5, 99]
+      elsif lccn[ll - 4, 1] != '0'
+        lccn = lccn[0, ll - 6] + "-" + lccn[ll - 4, 99]
+      elsif lccn[ll - 3, 1] != '0'
+        lccn = lccn[0, ll - 6] + "-" + lccn[ll - 3, 99]
+      end
+      lccn
+    end
+
+    #Returns a MARC::Record representation of the associated LC authority 
+    #record.
+    
+    def get_lc_auth
+      uri = URI.escape("http://errol.oclc.org/laf/#{self.lccn}.MarcXML")
+      src = ''
+      begin
+        Timeout.timeout 20 do
+          src = Net::HTTP.get(URI.parse("#{uri}"))
+        end
+        record = MARC::XMLReader.new(StringIO.new(string=src)).to_a[0]
+      end
+    rescue Timeout::Error
+      puts "The auth lookup timed out!"
+      raise
+    rescue MARC::Exception => e
+      puts "MARC exception: #{e}"
+      raise
+    rescue REXML::ParseException
+      puts "rexml error"
+      raise
+    rescue EOFError
+      puts "The service seems to be down!"
+      raise "eof"
+    rescue Exception => e
+      puts "some other type of exception"
+      puts e
+      raise
+    end
+  end
+  
+end

data/lib/wcid/marc.rb

@@ -0,0 +1,69 @@
+require 'marc'
+
+module MARC
+  class DataField
+    
+    #accept a MARC datafield and create a pnkey for query
+    #really just a bunch of regexps to tur
+    #This should only really work for exact searches if the datafield passed
+    #to the method had proper authority work done on it.
+    
+    def pnkey_from_marc 
+      name = self.name_from_datafield
+      dates = self.dates_from_datafield
+      qualifier = self.qualifier_from_datafield
+
+      query = "\"#{name}"
+      query += "$" + qualifier if qualifier
+      query += dates if dates
+      query += "\""
+      #puts query
+      query
+    end
+    
+    #Should the following be private?
+    def name_from_datafield
+      name = self['a'].downcase
+      name.gsub!('.','')
+      name.gsub!(/,$/,'')
+      name.gsub!('-',' ')
+      name.gsub!('\'','')
+      name
+    end
+    def dates_from_datafield
+      dates = self['d'] if self['d']
+      dates.gsub!('-', ' ') if dates
+      dates.gsub!('.', '') if dates
+      dates = "$" + dates if dates
+      dates
+    end
+    def qualifier_from_datafield
+      qualifier = self['q'] if self['q']
+      qualifier.gsub!('-', ' ') if qualifier
+      qualifier.gsub!(',','') if qualifier
+      qualifier.gsub!('(','') if qualifier
+      qualifier.gsub!(')','') if qualifier
+      qualifier.downcase! if qualifier
+    end
+  end
+  
+  class Record
+    
+    #accept a MARC record and return an array of name fields to be used in queries
+    
+    def name_fields 
+      f700 = self.find_all {|f| f.tag == '700'}
+      name_100 = self['100'] if self['100']
+      name_110 = self['110'] if self['110']
+      names =[]
+      names << name_100 unless name_100.nil? 
+      names << name_110  unless name_110.nil? 
+      names << f700 unless f700.empty? 
+      names.compact!
+      names.flatten!
+      #puts names.inspect
+      names
+    end
+  end
+  
+end

data/lib/wcid/search.rb

@@ -0,0 +1,78 @@
+module WCID
+  
+  #WCID::Search objects are for creating and running queries
+  #against WorldCat Identities.
+  
+  class Search
+    attr_reader :query
+    
+    #pass it a string
+    # name_search = WCID::Search.new('Twain, Mark')
+    
+    def initialize query
+      @query = query
+    end
+
+    
+    #Actually performs the search returns array of Hit objects
+    #  hits = name_search.search
+    #           or
+    #  hits =  (WCID::Search.new('Twain, Mark')).search
+    
+    def search 
+      uri = URI.escape("http://orlabs.oclc.org/Identities/find?fullName=#{@query}")
+      doc = do_search uri
+      m = doc.elements.to_a("nameAuthorities/match")
+      hits = []
+      m.each do | m |
+        hit_object = WCID::Hit.new(m)
+      hits << hit_object
+      end
+      hits
+    end
+
+    #returns single ID object (the first hit) or nil
+    # hit = name_search.search_exact
+    #       or
+    # hit = (WCID::Search.new('Twain, Mark')).search_exact
+    
+    def search_exact 
+      uri = URI.escape("http://orlabs.oclc.org/SRW/search/Identities?query=local.pnkey+exact+#{@query}")
+      #puts uri
+      begin
+        doc = do_search uri
+	#puts doc
+        if doc.elements['searchRetrieveResponse/numberOfRecords'].text == '0'
+          #puts "no record"
+          return nil
+        else
+          #puts 'a record here'
+          record = WCID::ID.new(doc)
+        end
+      rescue
+        #puts 'um no text for node?'
+        raise
+      end
+    end
+
+    #This should maybe be private or somesuch?
+    def do_search uri
+      begin
+        file = ''
+        Timeout.timeout 20 do
+          file = Net::HTTP.get(URI.parse("#{uri}"))
+          #puts file
+        end
+        doc = REXML::Document.new file
+      rescue Timeout::Error
+        puts "timeout error!"
+        raise 
+      rescue Errno::ECONNRESET
+        puts "errno!"
+        raise
+      end
+    end
+  end
+  
+  
+end

data/lib/wcid/version.rb

@@ -0,0 +1,9 @@
+module Wcid #:nodoc:
+  module VERSION #:nodoc:
+    MAJOR = 0
+    MINOR = 0
+    TINY  = 1
+
+    STRING = [MAJOR, MINOR, TINY].join('.')
+  end
+end

data/scripts/txt2html

@@ -0,0 +1,67 @@
+#!/usr/bin/env ruby
+
+require 'rubygems'
+require 'redcloth'
+require 'syntax/convertors/html'
+require 'erb'
+require File.dirname(__FILE__) + '/../lib/wcid/version.rb'
+
+version  = Wcid::VERSION::STRING
+download = 'http://rubyforge.org/projects/wcid'
+
+class Fixnum
+  def ordinal
+    # teens
+    return 'th' if (10..19).include?(self % 100)
+    # others
+    case self % 10
+    when 1: return 'st'
+    when 2: return 'nd'
+    when 3: return 'rd'
+    else    return 'th'
+    end
+  end
+end
+
+class Time
+  def pretty
+    return "#{mday}#{mday.ordinal} #{strftime('%B')} #{year}"
+  end
+end
+
+def convert_syntax(syntax, source)
+  return Syntax::Convertors::HTML.for_syntax(syntax).convert(source).gsub(%r!^<pre>|</pre>$!,'')
+end
+
+if ARGV.length >= 1
+  src, template = ARGV
+  template ||= File.dirname(__FILE__) + '/../website/template.rhtml'
+  
+else
+  puts("Usage: #{File.split($0).last} source.txt [template.rhtml] > output.html")
+  exit!
+end
+
+template = ERB.new(File.open(template).read)
+
+title = nil
+body = nil
+File.open(src) do |fsrc|
+  title_text = fsrc.readline
+  body_text = fsrc.read
+  syntax_items = []
+  body_text.gsub!(%r!<(pre|code)[^>]*?syntax=['"]([^'"]+)[^>]*>(.*?)</
>!m){
+    ident = syntax_items.length
+    element, syntax, source = $1, $2, $3
+    syntax_items << "<#{element} class='syntax'>#{convert_syntax(syntax, source)}</#{element}>"
+    "syntax-temp-#{ident}"
+  }
+  title = RedCloth.new(title_text).to_html.gsub(%r!<.*?>!,'').strip
+  body = RedCloth.new(body_text).to_html
+  body.gsub!(%r!(?:<pre><code>)?syntax-temp-(d+)(?:</code></pre>)?!){ syntax_items[$1.to_i] }
+end
+stat = File.stat(src)
+created = stat.ctime
+modified = stat.mtime
+
+$stdout << template.result(binding)