unique synset numbers

tools/eqsemrel.cnf

+eq_be_in_state
+eq_causes
+eq_generalization
+eq_has_holonym
+eq_has_hyperonym
+eq_has_hyponym
+eq_has_meronym
+eq_has_subevent
+eq_involved
+eq_is_caused_by
+eq_is_state_of
+eq_is_subevent_of
+eq_metonym
+eq_near_synonym
+eq_role
+eq_synonym

tools/make-indexes.sh

@@ -21,6 +21,9 @@ EX_EXT=`echo $EXPORTFILE |sed -r "s/^[.]*[^.]+[.]([^.]+)$/\1/1"`
 echo 'Do we have exportfile extension?'
 echo $EX_EXT
 
+# Do we have unique Synset numbers?
+NO_OF_ORIGSYNSETS=( cat $1 |egrep '^0'|wc -l )
+
 # all operations are done on NORMFILE!
 # extensions of index files
 NORMEXT='norm'    # normfile

tools/make-indexes.sh-mac

@@ -26,6 +26,9 @@ else
     echo "Synset numbers ARE NOT unique"
     cat $1 |grep '^0'|sed -E 's/(0 @)([0-9]+)(@.+$)/\2/g'|sort|uniq > all.numbers
     cat $1 |grep '^0'|sed -E 's/(0 @)([0-9]+)(@.+$)/\2/g'|sort|uniq -c|sort -nr|egrep '^ {3}2'|sed -E 's/^ {3}2 //g' > doubled.numbers
+
+    # rm all.numbers
+    # rm doubled.numbers
 fi
 echo "$NO_OF_ORIGSYNSETS"
 echo "$NO_OF_UNIQESYNSETS"

tools/semrel.cnf

+antonym
+be_in_state
+belongs_to_class
+causes
+fuzzynym
+has_holo_location
+has_holo_madeof
+has_holo_member
+has_holo_part
+has_holo_portion
+has_holonym
+has_hyperonym
+has_hyponym
+has_instance
+has_mero_location
+has_mero_madeof
+has_mero_member
+has_mero_part
+has_mero_portion
+has_meronym
+has_subevent
+has_xpos_hyperonym
+has_xpos_hyponym
+involved
+involved_agent
+involved_instrument
+involved_location
+involved_patient
+involved_target_direction
+is_caused_by
+is_subevent_of
+near_antonym
+near_synonym
+role
+role_agent
+role_instrument
+role_location
+role_patient
+role_target_direction
+state_of
+xpos_fuzzynym
+xpos_near_antonym
+xpos_near_synonym

tools/unique_synsets.py

@@ -3,8 +3,9 @@
 
 __author__ = 'Neeme Kahusk <neeme.kahusk@ut.ee>'
 __version__ = '1.0'
-__date__ = 'T nov   20 15:24:18 EET 2012'
+__date__ = 'Thu Jun 25 22:59:28 EEST 2015'
 
+import eurown
 
 from optparse import OptionParser
 
@@ -18,53 +19,29 @@ parser = OptionParser(usage="%prog [options]",
 
 
 (options, args) = parser.parse_args()
-if len(args) != 2:
+if len(args) != 3:
     parser.error("incorrect number of arguments")
 
-RAW_INDEX_FN = args[0]
-LITERAL_INDEX_FN = args[-1]
+ALL_NUMBERS_FN = args[0]
+DOUBLED_NUMBERS_FN = args[1]
+WN_FN = args[2]
 
-def read_raw_index(filename):
+def read_raw(filename):
     f = open(filename)
-    lines = map(lambda x: x.strip().split(':'),f.readlines())
+    lines = map(lambda x: int(x.strip()), f.readlines())
     f.close()
     return lines
 
-def make_literal_index(iList):
-    """iList should be list of lists with 2 members
-
-    """
-    oDict = dict()
-
-    for i in iList:
-        # print i
-        try:
-            if oDict and i[0] in oDict:
-                oDict[i[0]].append(i[1])
-            else:
-                oDict.update({i[0]:[i[1]]})
-        except KeyError:
-            print 'võtme viga',i
-
-    return oDict
-
-ri = read_raw_index(RAW_INDEX_FN)
-
-def write_index(iDict,filename):
-    """Write index (dict) to file 
-
-    """
-    f = open(filename,'a')
-    for i in iDict:
-        try:
-            oStr = u'%s:%s\n' % (i.decode('utf8'),
-                                 u' '.join(iDict[i])
-                                 )
-        except UnicodeDecodeError:
-            print iDict[i]
-        f.write(oStr.encode('utf-8'))
-    f.close()
 
-litindex = make_literal_index(ri)
+al = read_raw(ALL_NUMBERS_FN)
+doub = read_raw(DOUBLED_NUMBERS_FN)
+
+print al
+print doub
+
+print len(al)
+print len(doub)
 
-write_index(litindex,LITERAL_INDEX_FN)
+p = eurown.Parser(WN_FN)
+s = p.parse_synset()
+print s.polarisText