Tags: patch
The Moby Thesaurus dictionary package would have much
greater utility as a dictionary per se if it were inverted;
that is, if every word currently used as part of the
"definition" were instead a term defined (using the current
defined terms as the words in the definition). The attached
extended conv.py includes the inverted thesaurus in the same
dict file.
import sys, dictdlib
# Conversion script by John Goerzen
itextfile = open("introduction.txt", "rt")
dw = dictdlib.DictWriter("moby-thesaurus",
'http://www.ibiblio.org/pub/docs/books/gutenberg/etext02/mthes10.zip',
'Moby Thesaurus II by Grady Ward, 1.0',
itextfile.read(-1))
imt = { }
while 1:
line = sys.stdin.readline()
if not line:
break
line.strip()
base, d = line.split(',', 1)
defstr = ""
defline = " "
count = 0
for defword in d.split(','):
if defword not in imt:
imt[defword] = [ ]
imt[defword].append(base)
count += 1
if len(defline) + len(defword) + 3 > 70:
defstr += defline + "," + "\n"
defline = " " + defword
elif defline == " ":
defline += defword
else:
defline += ", " + defword
if defline != " ":
defstr += defline + "\n"
dw.writeentry("%s Moby Thesaurus words for \"%s\":\n" % (count, base) + \
defstr, [base])
for base in imt.keys():
defstr = ""
defline = " "
count = 0
for defword in imt[base]:
count += 1
if len(defline) + len(defword) + 3 > 70:
defstr += defline + "," + "\n"
defline = " " + defword
elif defline == " ":
defline += defword
else:
defline += ", " + defword
if defline != " ":
defstr += defline + "\n"
dw.writeentry("%s Moby Thesaurus terms refer to \"%s\":\n" % (count, base) \
+ defstr, [base])
dw.finish()
