songleash

commit 322ee37350f05fcd7edda86a31ad555ddd64dbfd
parent 467322a8972cef28ab6790e2a3044269d7e3efb0
Author: Antoine Amarilli <a3nm@a3nm.net>
Date:   Thu, 27 Aug 2015 02:16:54 +0200

only accept lexique words with correct split

Diffstat:
only3.py
 | 
11
++++++++---

1 file changed, 8 insertions(+), 3 deletions(-)
diff --git a/only3.py b/only3.py
@@ -1,6 +1,6 @@
 #!/usr/bin/python3
 
-from common import vowels
+from common import vowels, fem
 import re
 import sys
 
@@ -8,8 +8,13 @@ for l in sys.stdin.readlines():
     f = l.split('\t')
     vowels_regexp = re.compile('([' + vowels + '])')
     f[-1] = f[-1].strip()
-    if len(f) <= 3:
-        f.append("")
+    if len(f) < 3:
+        # f.append("")
+        # only take words with complete syllabification in lexique
+        continue
+    if len(f[2].split('-')) != (4 if fem(f[0]) else 3):
+        # garbled syllabification from lexique
+        continue
     chunks = f[1].split('-')
     # words cannot start with a vowel
     # as last consonant before last vowel will be kept