Add some doc

Author: AngledLuffa

stanza/utils/languages/spanish/__init__.py

@@ -0,0 +1,28 @@
+"""
+A simple, almost trivial script to read sentences from UD and mark any words of 'se' with a delimiter
+
+Should be a useful input format for MLtwist to annotate the words
+"""
+
+from stanza.utils.conll import CoNLL
+
+import sys
+
+filename = sys.argv[1]
+doc = CoNLL.conll2doc(filename, reconstruct_text=True)
+
+def process_se(doc):
+    for sentence in doc.sentences:
+        for idx, token in enumerate(sentence.tokens):
+            if any(word.text == 'se' for word in token.words):
+                yield sentence, idx
+
+for sentence, idx in process_se(doc):
+    token = sentence.tokens[idx]
+    # note that sentences have a start index somewhere in the middle of the doc,
+    # not necessarily at 0
+    start_char = token.start_char - sentence.tokens[0].start_char
+    end_char = token.end_char - sentence.tokens[0].start_char
+    text = sentence.text[:start_char] + "---> " + token.text + " <---" + sentence.text[end_char:]
+    print(text)
+