Releasing V0.0.8.9 - Minor changes in romanizer rules

Ransaka · Jun 23, 2024 · 80d140f · 80d140f
1 parent 79ea208
commit 80d140f
Show file tree

Hide file tree

Showing 3 changed files with 757 additions and 3 deletions.
diff --git a/pyproject.toml b/pyproject.toml
@@ -1,6 +1,6 @@
 [project]
 name = "sinlib"
-version = "0.0.8.8"
+version = "0.0.8.9"
 description = "Sinhala NLP Toolkit"
 authors = [
     { name = "Ransaka", email = "ransaka.ravihara@gmail.com" }

diff --git a/src/sinlib/romanize.py b/src/sinlib/romanize.py
@@ -34,7 +34,7 @@ def __romanize(self, text: str):
             self.tokenizer.token_id_to_token_map[c] for c in encodings
         ]
         romanized_sinhala = [
-            self.char_mapper.get(ch, ch if ch in NUBERS_AND_PUNKTS.union(" ") else None)
+            self.char_mapper.get(ch, ch if ch in NUBERS_AND_PUNKTS.union(" ") else '')
             for ch in decoded_sinhala_chars
         ]
         romanized_sinhala = "".join(romanized_sinhala)