From ad4b08a7ee90cea8a88a6d79e27ece2cb164cbf4 Mon Sep 17 00:00:00 2001
From: Chris Jaekl <chris@jaekl.net>
Date: Sat, 22 Jun 2024 11:37:41 -0400
Subject: [PATCH] Remove OCLC classification client code

OCLC discontinued their open-access classify service in January 2024.
Thus, there's no point in maintaining this code anymore.
---
 classify/bookclass.rb |  80 ---------------------
 classify/classset.rb  | 122 --------------------------------
 classify/fastset.rb   |  65 -----------------
 classify/lookup.rb    | 158 ------------------------------------------
 classify/queryoclc.rb |  58 ----------------
 5 files changed, 483 deletions(-)
 delete mode 100644 classify/bookclass.rb
 delete mode 100644 classify/classset.rb
 delete mode 100644 classify/fastset.rb
 delete mode 100644 classify/lookup.rb
 delete mode 100644 classify/queryoclc.rb

diff --git a/classify/bookclass.rb b/classify/bookclass.rb
deleted file mode 100644
index 47cad99..0000000
--- a/classify/bookclass.rb
+++ /dev/null
@@ -1,80 +0,0 @@
-# Classification information for a single book
-
-class BookClass
-  def initialize(grouping, title)
-    @author = nil
-    @ddc = nil
-    @grouping = grouping
-    @fast = []
-    @filename = []
-    @lcc = nil
-    @title = title
-  end
-
-  def author 
-    @author 
-  end
-  def author=(value) 
-    @author = value 
-  end
-  def ddc 
-    @ddc 
-  end
-  def ddc=(value) 
-    @ddc = value 
-  end
-  def fast 
-    @fast 
-  end
-  def filename
-    @filename
-  end
-  def filename=(value)
-    @filename = value
-  end
-  def grouping
-    @grouping
-  end
-  def lcc 
-    @lcc 
-  end
-  def lcc=(value) 
-    @lcc = value 
-  end
-  def title 
-    @title 
-  end
-
-  def add_fast(id)
-    @fast.push(id)
-  end
-
-  def inspect 
-    data = []
-
-    if nil != @author_name 
-      data.push('author_name="' + @author_name + '"') 
-    end
-    if nil != @ddc
-      data.push('ddc="' + @ddc + '"') 
-    end
-    if nil != @grouping 
-      data.push('grouping="' + @grouping + '"') 
-    end
-    if nil != @fast 
-      data.push('fast=' + @fast.inspect) 
-    end
-    if nil != @filename
-      data.push('filename=' + @filename.to_s + '"')
-    end
-    if nil != @lcc 
-      data.push('lcc="' + @lcc + '"') 
-    end
-    if nil != @title 
-      data.push('title="' + @title + '"') 
-    end
-
-    return '(BookClass:' + data.join(',') + ')'
-  end
-end
-
diff --git a/classify/classset.rb b/classify/classset.rb
deleted file mode 100644
index 4613fc7..0000000
--- a/classify/classset.rb
+++ /dev/null
@@ -1,122 +0,0 @@
-require 'csv'
-
-require 'bookclass'
-
-class ClassSet
-  @@class_csv_file = 'class.csv'
-
-  def initialize
-    @entries = {}
-    load!(@@class_csv_file)
-  end
-
-  def add!(info)
-    key = construct_key(info.grouping, info.title)
-    @entries[key] = info
-  end
-
-  def construct_key(author_grouping, title)
-    author_grouping.to_s + '|' + title.to_s
-  end
-
-  def get(author_grouping, title)
-    key = construct_key(author_grouping, title)
-    if @entries.has_key?(key)
-      return @entries[key]
-    else
-      return nil
-    end
-  end
-
-  def has_key?(author_grouping, title)
-    @entries.has_key?(construct_key(author_grouping, title))
-  end
-
-  def ensure_contains!(info)
-    if ! has_key?(info.grouping, info.title)
-      add!(info)
-    end
-  end
-
-  def inspect 
-    data = []
-
-    if nil != @entries 
-      data.push('entries=' + @entries.inspect + '') 
-    end
-
-    return '(ClassSet:' + data.join(',') + ')'
-  end
-
-  def load!(file_name)
-    first = true
-    @entries = {}
-
-    if ! File.exist?(file_name)
-      puts 'WARNING:  file "' + file_name + '" not found.'
-      return 
-    end
-
-    File.open(file_name, 'r:UTF-8') do |fd|
-      csv = CSV.new(fd)
-      csv.to_a.each do |row|
-        if first
-          first = false
-        elsif row.length >= 6
-          ddc = row[0]
-          lcc = row[1]
-          grouping = row[2]
-          author = row[3]
-          filename = row[4]
-          title = row[5]
-          fast = []
-          if nil != row[6]
-            fast = row[6].split(';')
-          end
-  
-          bookclass = BookClass.new(grouping, title)
-          bookclass.ddc = ddc
-          bookclass.lcc = lcc
-          bookclass.author = author
-          bookclass.filename = filename
-  
-          fast.each do |id|
-            bookclass.add_fast(id)
-          end
-  
-          key = construct_key(grouping, title)
-          @entries[key] = bookclass
-
-          #puts 'LOADED[' + key.inspect + ']: ' + bookclass.inspect
-        end
-      end
-    end
-  end
-
-  def save(file_name)
-    CSV.open(file_name, 'w:UTF-8') do |csv|
-      csv << ['Dewey', 'LCC', 'Grouping', 'Author', 'Filename', 'Title', 'FAST']
-
-      @entries.keys.sort.each do |key|
-        info = @entries[key]
-
-        #puts 'SAVING[' + key.inspect + ']: ' + info.inspect
-
-        ddc = info.ddc
-        lcc = info.lcc
-        grouping = info.grouping
-        author = info.author
-        filename = info.filename
-        title = info.title
-        fast = info.fast.join(';')
-        
-        csv << [ ddc, lcc, grouping, author, filename, title, fast ]
-      end
-    end
-  end
-
-  def save_state
-    save(@@class_csv_file)
-  end
-end
-
diff --git a/classify/fastset.rb b/classify/fastset.rb
deleted file mode 100644
index 0d97aeb..0000000
--- a/classify/fastset.rb
+++ /dev/null
@@ -1,65 +0,0 @@
-
-class FastSet
-  @@fast_csv_file = 'fast.csv'
-
-  def initialize
-    @entries = {}
-    load!(@@fast_csv_file)
-  end
-
-  def add(id, descr)
-    @entries[id] = descr
-  end
-
-  def has_key?(value)
-    @entries.has_key?(value)
-  end
-
-  def inspect 
-    data = []
-
-    if nil != @entries 
-      data.push('entries=' + @entries.inspect + '') 
-    end
-
-    return '(FastSet:' + data.join(',') + ')'
-  end
-
-  def load!(file_name)
-    first = true
-    @entries = {}
-
-    if ! File.exist?(file_name)
-      puts 'WARNING:  file "' + file_name + '" not found.'
-      return 
-    end
-
-    first = true
-    CSV.open(file_name, 'r:UTF-8') do |csv|
-      csv.to_a.each do |row|
-        if first
-          first = false
-        elsif row.length >= 2
-          id = row[0]
-          descr = row[1]
-          @entries[id] = descr
-        end
-      end
-    end
-  end
-
-  def save(file_name)
-    CSV.open(file_name, 'w:UTF-8') do |csv|
-      csv << [ 'Code', 'Description' ]
-      
-      @entries.keys.sort.each do |key|
-        csv << [ key.to_s, @entries[key].to_s ]
-      end
-    end
-  end
-
-  def save_state
-    save(@@fast_csv_file)
-  end
-end
-
diff --git a/classify/lookup.rb b/classify/lookup.rb
deleted file mode 100644
index 889db51..0000000
--- a/classify/lookup.rb
+++ /dev/null
@@ -1,158 +0,0 @@
-require 'erb'
-require 'net/http'
-require 'nokogiri'
-
-require 'bookclass'
-require 'classset'
-require 'fastset'
-
-class Lookup
-  def initialize
-    @class_set = ClassSet.new()
-    @fast_set = FastSet.new()
-  end
- 
-  def construct_url(params)
-    first = true
-    cmd = 'http://classify.oclc.org/classify2/Classify'
-  
-    params += [ ['summary', 'false' ] ]
-  
-    params.each do |tuple|
-      name, value = tuple
-      if (first)
-        cmd += '?'
-        first = false
-      else
-        cmd += '&'
-      end
-    cmd += name + '='
-      cmd += ERB::Util.url_encode(value)
-    end
-  
-    return cmd
-  end
-  
-  def isUpper?(c)
-      return /[[:upper:]]/.match(c)
-  end
-  
-  def lookup(author_grouping, pathname)
-    params = [ 
-        ['author', massage_author(author_grouping)],
-        ['title', massage_title(pathname)]
-      ]
-  
-    cmd = construct_url(params)
-    res = submit_request(cmd)
-  
-    doc = Nokogiri::XML(res.body)
-
-    if "4" == response_code(doc)
-      # Multiple matches; pick the first one and re-query
-      owi = doc.css("works work")[0]["owi"]
-  
-      params = [
-          ['owi', owi]
-        ]
-      cmd = construct_url(params)
-      res = submit_request(cmd)
-    
-      #puts res.body
-    
-      doc = Nokogiri::XML(res.body)
-    end
-  
-    if "2" != response_code(doc)
-      # Lookup failed.  Let's try shortening the title, if it's in multiple parts
-
-      #TODO TODO
-      puts "Lookup failed"
-      return nil
-    end
-
-    title = doc.css("classify editions edition")[0]["title"]
-  
-    info = BookClass.new(author_grouping, title)
-
-    author = doc.css("classify editions edition")[0]["author"]
-    info.author = author
-  
-    nodes = doc.css("classify recommendations ddc mostPopular")
-    if nil != nodes && nodes.length > 0
-      ddc = nodes[0]["sfa"]
-      info.ddc = ddc
-    end
-  
-    nodes = doc.css("classify recommendations lcc mostPopular")
-    if nil != nodes && nodes.length > 0
-      lcc = nodes[0]["sfa"]
-    end
-    info.lcc = lcc
-
-    headings = doc.css("classify recommendations fast headings heading")
-    headings.each do |heading|
-      #puts heading.inspect
-      id = heading['ident']
-      #puts 'ID: ' + id
-      descr = heading.content
-      #puts 'DESCR: ' + descr
-      info.add_fast(id)
-      @fast_set.add(id, descr)
-    end
-
-    info.filename = title
-
-    @class_set.ensure_contains!(info)
-
-    return info
-  end
-
-  def massage_author(input)
-      if nil == input
-        return nil
-      end
-  
-      reading_order = ""
-      input.each_char do |c|
-        if isUpper?(c) and (reading_order.length > 0)
-          reading_order += " "
-       end
-        reading_order += c
-      end
-  
-      return reading_order
-  end
-
-  def massage_title(pathname)
-    basename = File.basename(pathname, '.*')
-
-    basename.gsub!('_', ' ')
-    basename.gsub!('--', ': ')
-    basename.gsub!('-s ', "'s ")
-    basename.gsub!('s- ', "s' ")
-
-    return basename
-  end
-
-  def response_code(doc)
-    return doc.css("classify response")[0]["code"]
-  end
-
-  def save_state
-    @class_set.save_state()
-    @fast_set.save_state()
-  end
-
-  def submit_request(cmd)
-    puts ('GET ' + cmd)
-  
-    url = URI.parse(cmd)
-    req = Net::HTTP::Get.new(url.to_s)
-    res = Net::HTTP.start(url.host, url.port) {|http|
-      http.request(req)
-    }
-    return res
-  end
-end
-
diff --git a/classify/queryoclc.rb b/classify/queryoclc.rb
deleted file mode 100644
index a451290..0000000
--- a/classify/queryoclc.rb
+++ /dev/null
@@ -1,58 +0,0 @@
-require 'classset'
-require 'lookup'
-
-#if ARGV.length != 2
-#  puts 'Usage:  ruby classify.rb author title'
-#  exit 1
-#end
-#
-#author_grouping = ARGV[0]
-#title = ARGV[1]
-
-classset = ClassSet.new()
-lookup = Lookup.new()
-
-CSV.open('unclassified.csv', 'r:UTF-8') do |csv|
-  query_count = 0
-  first = true
-  csv.to_a.each do |row|
-    if first
-      first = false
-    elsif row.length >= 2
-      author_grouping = row[0]
-      pathname = row[1]
-
-      info = classset.get(author_grouping, pathname)
-      if nil == info
-        query_count += 1
-        info = lookup.lookup(author_grouping, pathname)
-
-        puts info.inspect()
-
-        puts 'Saving state...'
-        classset.save_state()
-        lookup.save_state()
-
-        sleep_time = 10 + rand(10)
-        puts 'Pausing for ' + sleep_time.to_s + ' seconds...'
-        sleep(sleep_time)  # Pause between lookup requests, to be polite to the server
-      end
-
-      if nil != info
-        classset.ensure_contains!(info)
-      else
-        puts 'WARNING:  lookup of ' + author_grouping + ', "' + pathname + '" failed.'
-        File.open('failed.log', 'a:UTF-8') do |fd|
-          fd.puts(author_grouping.to_s + ',' + pathname.to_s)
-        end
-      end
-
-#      if query_count > 5
-#        break
-#      end
-    end
-  end
-end
-
-classset.save_state()
-lookup.save_state()
-- 
2.39.2