Check root ids is now skipped by default.

wani.py

@@ -738,7 +738,7 @@ def is_root_id(id_):
     return len(id_.split('.')) == 3
 
 
-def load_corpus(filename):
+def load_corpus(filename, check_ids):
     with open(filename, 'r') as fp:
         xmlstring = re.sub(' xmlns="[^"]+"', '', fp.read(), count=1)
         xmlstring = xmlstring.replace(' xml:', ' ')
@@ -763,7 +763,7 @@ def load_corpus(filename):
             lfrom, dest = l.get('target').replace('#', '').split()
 
         if lfrom in words:
-            if is_root_id(lfrom):
+            if check_ids and is_root_id(lfrom):
                 logging.error("NOO: ", lfrom)
                 sys.exit(1)
 
@@ -896,7 +896,7 @@ def main(input_file, structures_file, args):
         logging.debug(str(s))
 
     logging.info("LOADING TEXT...")
-    words = load_corpus(input_file)
+    words = load_corpus(input_file, args.check_ids)
 
     # useful for faster debugging...
     # import pickle
@@ -934,6 +934,7 @@ if __name__ == '__main__':
     parser.add_argument('structures', help='Structures definitions in xml file')
     parser.add_argument('--output', help='Output file (if none given, then output to stdout)')
 
+    parser.add_argument('--check-ids', help='Checks ids of <w> and <pc> to be in correct structure', action='store_true')
     parser.add_argument('--lemma_only', help='Will not write word ids, forms and msds in output', action='store_true')
     parser.add_argument('--without-rep', help='Will not write representtaions in output', action='store_true')
     parser.add_argument('--group', help='Group collocations with same collocation ID', action='store_true')