Initial commit

scripts/extract_from_grammalecte.py

@@ -0,0 +1,20 @@
+with open("grammalecte.txt", "r") as f:
+    lines = f.read().splitlines()
+
+lines = lines[16:]
+
+for line in lines:
+    (id_, fid, flexion, lemme, etiquettes, metagraphe, metaphone,
+     notes, semantique, etymologie, sous_dictionnaire, google_1_grams,
+     wikipedia, wikisource, litterature, total, doublons, multiples,
+     frequence, indice) = line.split("\t")
+
+    etiquettes = etiquettes.split()
+    if "nom" in etiquettes:
+        print(flexion)
+    elif "adj" in etiquettes:
+        print(flexion)
+    elif "adv" in etiquettes:
+        print(flexion)
+    elif "infi" in etiquettes:
+        print(flexion)

scripts/wordlize.py

@@ -0,0 +1,23 @@
+import sys
+import re
+from unidecode import unidecode
+
+file = sys.argv[1]
+
+if file == "-":
+    lines = sys.stdin
+else:
+    with open(file, "r") as f:
+        lines = f.read().splitlines()
+
+output = set()
+for line in lines:
+    wordlized = unidecode(line).strip().upper()
+    if not re.match(r"^[A-Z]*$", wordlized): # ignore words with dashes, apostrophes...
+        continue
+    output.add(wordlized)
+
+output = sorted(list(output))
+
+for line in output:
+    print(line)