Added scripts.

scripts/dbm2syn.rb

@@ -0,0 +1,77 @@
+#! /usr/bin/ruby
+
+#--
+###############################################################################
+#                                                                             #
+# dbm2syn -- Convert DBM to Lingo dictionary for perseus-a                    #
+#                                                                             #
+# Copyright (C) 2009 Cologne University of Applied Sciences,                  #
+#                    Claudiusstr. 1,                                          #
+#                    50678 Cologne, Germany                                   #
+#                                                                             #
+# Authors:                                                                    #
+#     Jens Wille <[email protected]>                                    #
+#                                                                             #
+# dbm2syn 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.                                                          #
+#                                                                             #
+# dbm2syn 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 dbm2syn. If not, see <http://www.gnu.org/licenses/>.                   #
+#                                                                             #
+###############################################################################
+#++
+
+abort "Usage: #{$0} <dbm> <syn> [<key>]" unless [2, 3].include?(ARGV.size)
+
+STDOUT.sync = true
+
+rec, syn = {}, Hash.new { |h, k| h[k] = [] }
+
+ID_RE  = %r{\AID:(.*)}
+KEY_RE = %r{\A(#{ARGV[2] || '.*?'}):(.*)}
+REC_RE = %r{\A&&&\z}
+SEP_RE = %r{\*}
+
+File.foreach(ARGV[0]) { |line|
+  print '.' if $. % 10_000 == 0
+
+  case line.chomp
+    when ID_RE
+      rec[:id] = $1
+    when KEY_RE
+      (rec[$1] ||= []) << $2
+    when REC_RE
+      if id = rec.delete(:id)
+        rec.values.each { |value|
+          warn "#{id}: #{value}" if value =~ SEP_RE
+
+          syn[value] << id
+        }
+      end
+
+      rec = {}
+  end
+}
+
+puts
+
+syn.sort!
+
+puts
+
+File.open(ARGV[1], 'w') { |f|
+  syn.each_with_index { |(value, ids), index|
+    print '.' if index % 1_000 == 0
+
+    f.puts "#{value}*#{ids.sort.uniq.join('|')}"
+  }
+}
+
+puts

scripts/merge_syn.rb

@@ -0,0 +1,88 @@
+#! /usr/bin/ruby
+
+#--
+###############################################################################
+#                                                                             #
+# merge_syn -- Intersection of Lingo results for perseus-a                    #
+#                                                                             #
+# Copyright (C) 2009 Cologne University of Applied Sciences,                  #
+#                    Claudiusstr. 1,                                          #
+#                    50678 Cologne, Germany                                   #
+#                                                                             #
+# Authors:                                                                    #
+#     Jens Wille <[email protected]>                                    #
+#                                                                             #
+# merge_syn 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.                                                          #
+#                                                                             #
+# merge_syn 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 merge_syn. If not, see <http://www.gnu.org/licenses/>.                 #
+#                                                                             #
+###############################################################################
+#++
+
+abort "Usage: #{$0} <in.syn>... <out.syn>" unless ARGV.size >= 2
+
+STDOUT.sync = true
+
+KEY_SEPARATOR   = '*'.freeze
+VALUE_SEPARATOR = '|'.freeze
+
+merge = Hash.new { |h, k| h[k] = [] }
+outfile = ARGV.pop
+
+# first run sets the basis!
+File.foreach(ARGV.shift) { |line|
+  print '.' if $. % 1_000 == 0
+
+  line.chomp!
+
+  key, values = line.split(KEY_SEPARATOR, 2)
+  merge[key] = values.split(VALUE_SEPARATOR)
+}
+
+keys = merge.keys
+
+puts
+
+ARGV.each { |syn|
+  _keys = []
+
+  File.foreach(syn) { |line|
+    print '.' if $. % 1_000 == 0
+
+    line.chomp!
+
+    key, values = line.split(KEY_SEPARATOR, 2)
+    merge[key] &= values.split(VALUE_SEPARATOR)
+
+    _keys << key
+  }
+
+  keys &= _keys
+
+  puts
+}
+
+merge.delete_if { |key, values|
+  values.empty? || !keys.include?(key)
+}
+
+puts
+
+File.open(outfile, 'w') { |f|
+  merge.sort.each_with_index { |(key, values), i|
+    print '.' if i % 1_000 == 0
+
+    f.puts "#{key}#{KEY_SEPARATOR}#{values.sort.join(VALUE_SEPARATOR)}"
+  }
+}
+
+puts