Using argparse

wani.py

@@ -4,6 +4,7 @@ from enum import Enum
 from collections import defaultdict
 import sys
 import logging
+import argparse
 
 from msd_translate import MSD_TRANSLATE
 
@@ -11,10 +12,6 @@ from msd_translate import MSD_TRANSLATE
 MAX_NUM_COMPONENTS = 5
 
 
-STAVKI = sys.argv[1]
-STRUKTURE = sys.argv[2]
-FILE_OUT = sys.argv[3]
-
 CODES = {
     "Noun": "N",
     "Verb": "V",
@@ -787,11 +784,12 @@ def main():
     import time
     t = time.time()
 
-    structures = build_structures(STRUKTURE)
+    structures = build_structures(structures_file)
     for s in structures:
         logging.debug(str(s))
 
-    words = load_corpus(STAVKI)
+    logging.info("LOADING TEXT...")
+    words = load_corpus(input_file)
     # import pickle
     # with open("words.p", "wb") as fp:
     #     pickle.dump(words, fp)
@@ -872,10 +870,18 @@ def main():
         # print(s.id)
         # print(groups)
 
-
-
 if __name__ == '__main__':
-    logging.basicConfig(level=logging.INFO)
-    main()
+    parser = argparse.ArgumentParser(description='Extract structures from a parsed corpus.')
+    parser.add_argument('input', help='input xml file in `ssj500k form`')
+    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('--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')
+    parser.add_argument('--verbose', help='Enable verbose output to stderr', choices=["warning", "info", "debug"], default="info")
+
+    args = parser.parse_args()
+    logging.basicConfig(stream=sys.stderr, level=args.verbose.upper())
 
 # 6, 7 primeri laznih zadetkov?