From cb3187f580a34bbd1c36c3736d31cce8c50f2314 Mon Sep 17 00:00:00 2001
From: Franck Dary <franck.dary@lis-lab.fr>
Date: Thu, 1 Jul 2021 14:53:16 +0200
Subject: [PATCH] Added script mcf2conllu
---
scripts/mcf2conllu.py | 58 +++++++++++++++++++++++++++++++++++++++++++
1 file changed, 58 insertions(+)
create mode 100755 scripts/mcf2conllu.py
diff --git a/scripts/mcf2conllu.py b/scripts/mcf2conllu.py
new file mode 100755
index 0000000..a536a2e
--- /dev/null
+++ b/scripts/mcf2conllu.py
@@ -0,0 +1,58 @@
+#! /usr/bin/env python3
+
+import argparse
+import sys
+
+################################################################################
+if __name__ == "__main__" :
+ parser = argparse.ArgumentParser()
+ parser.add_argument("input", type=str,
+ help="Input mcf file")
+ parser.add_argument("head", type=int,
+ help="Index of the column containing governor relative index.")
+ parser.add_argument("eos", type=int,
+ help="Index of the column containing end of sentence info.")
+ parser.add_argument("--form", type=int, default=None,
+ help="Index of the column containing FORM.")
+ parser.add_argument("--upos", type=int, default=None,
+ help="Index of the column containing UPOS.")
+
+ args = parser.parse_args()
+ mcd = []
+
+ sentence = []
+ sentenceID = 0
+ for line in open(args.input, "r") :
+ line = line.strip()
+ if len(line) == 0 :
+ continue
+ if line[0] == '#' :
+ continue
+ splited = line.split('\t')
+
+ if len(mcd) == 0 :
+ mcd = [str(k) for k in range(1,len(splited)+1)]
+ mcd[args.head] = "HEAD"
+ mcd[args.eos] = "ID"
+ if args.form is not None :
+ mcd[args.form] = "FORM"
+ if args.upos is not None :
+ mcd[args.upos] = "UPOS"
+ print("# global.columns = %s"%" ".join(mcd))
+
+ splited[args.head] = int(splited[args.head])
+ sentence.append(splited)
+ eos = int(splited[args.eos])
+ if eos == 1 :
+ sentenceID += 1
+ print("# sent_id = %d"%sentenceID)
+ for i in range(len(sentence)) :
+ sentence[i][args.eos] = i+1 # Recycling EOS column into ID column
+ if sentence[i][args.head] != 0 :
+ sentence[i][args.head] += i+1 # Transforming relative head into absolute
+ for word in sentence :
+ print('\t'.join(map(str,word)))
+ print("")
+ sentence = []
+################################################################################
+
--
GitLab