Update fix_single_quotes

hexlib/regex.py

@@ -3,5 +3,5 @@ import re
 LINK_RE = re.compile(r"(https?://[\w\-_.]+\.[a-z]{2,4}([^\s<'\"]*|$))")
 HTML_HREF_RE = re.compile(r"href=\"([^\"]+)\"")
 WHITESPACE_RE = re.compile(r"\s+")
-PUNCTUATION_RE = re.compile(r"[.,;:\"!?/()|*=>]+")
+PUNCTUATION_RE = re.compile(r"[.,;:\"“!?/()|*=>]+")
 XML_ENTITY_RE = re.compile(r"&[a-z]+;")

hexlib/text.py

@@ -51,6 +51,7 @@ def preprocess(text, lowercase=False, clean_html=False, strip=False, remove_punc
 
     if fix_single_quotes:
         text = text.replace("`", "'")
+        text = text.replace("’", "'")
 
     if remove_urls:
         text = LINK_RE.sub(" ", text)

setup.py

@@ -2,7 +2,7 @@ from setuptools import setup
 
 setup(
     name="hexlib",
-    version="1.48",
+    version="1.49",
     description="Misc utility methods",
     author="simon987",
     author_email="me@simon987.net",

test/test_text.py

@@ -130,6 +130,17 @@ class TestText(TestCase):
         expected = "hello world it's it's"
 
         self.assertEqual(cleaned, expected)
+    
+    def test_single_quote(self):
+        text = "it's it`s it’s"
+        cleaned = preprocess(
+            text,
+            lowercase=True,
+            fix_single_quotes=True
+        )
+        expected = "it's it's it's"
+
+        self.assertEqual(cleaned, expected)
 
     def test_html_10(self):
         text = "<div>\n Hello, \t\n<strong>world! it's it`s https://google.ca/test/abc.pdf  </strong>\n\t</div>"