Fixes for OED parsing

1 files changed, 11 insertions, 7 deletions

diff --git a/bladictionary.py b/bladictionary.py
index f7fcaec..ec5aff9 100755
--- a/bladictionary.py
+++ b/bladictionary.py
@@ -6,7 +6,7 @@ import sys
 import shlex
 from lxml import etree
 
-version = "2.0.5b"
+version = "2.0.7b"
 class Definition(object):
 
 	#ID is relative to the word type, eg noun 1, noun 2, verb 1, verb 2, not to the entire list
@@ -160,6 +160,7 @@ def parse_args():
 def parse_oed(word):
 	types = ["n.", "—n.", "v.", "—v.", "adj.", "—adj.", "adv.", "—adv."]
 	file = open("oed.txt")
+	word_line = ""
 
 	for line in file:
 		word_part = line.split(" ")[0]
@@ -168,11 +169,10 @@ def parse_oed(word):
 			word_line = line
 			break;
 
-	if word_line is None:
+	if word_line == "":
 		return
 
 	print line
-
 	prev_part = ""
 	items = []
 	id = 0
@@ -182,14 +182,14 @@ def parse_oed(word):
 		
 		if part in types:
 			if id > 0:
-				item = Definition(id, word, "oed", word_type, definition, [], [])
+				item = Definition(word, id, "oed", word_type, definition, [], [])
 				items.append(item)
 				definition = ""
 
 			word_type = part
 
-		elif part.isdigit() and prev_part is not None and prev_part[-1] == ".":
-			item = Definition(id, word, "oed", word_type, definition, [], [])
+		elif part.isdigit() and prev_part is not None and (prev_part[-1] == "." or prev_part[-1] == ")"):
+			item = Definition(word, id, "oed", word_type, definition, [], [])
 			items.append(item)
 			definition = ""
 			id = part
@@ -198,6 +198,10 @@ def parse_oed(word):
 			
 		prev_part = part
 
+	if id == 0:
+		id = 1
+	item = Definition(word, id, "oed", word_type, definition, [], [])
+	items.append(item)
 	return items
 
 
@@ -226,7 +230,7 @@ def main():
 			if suppress_print:
 				num_more+=1
 			else:
-				definition = item.word_type + " " + item.id + ": " + item.definition
+				definition = item.word_type + " " + str(item.id) + ": " + item.definition
 				if line_length + len(definition) > 455:
 					suppress_print = True
 					num_more+= 1