feat: Add html escape quotes to cleaning brick (#84)

* feat: Add html escape quotes to cleaning brick

* bump changelog

CHANGELOG.md

@@ -1,9 +1,10 @@
-## 0.3.0-dev4
+## 0.3.0-dev5
 
 * Implement staging brick for Argilla.
 * Removing the local PDF parsing code and any dependencies and tests.
 * Reorganizes the staging bricks in the unstructured.partition module
 * Allow entities to be passed into the Datasaur staging brick
+* Added HTML escapes to the `replace_unicode_quotes` brick
 
 ## 0.2.6
 

test_unstructured/cleaners/test_core.py

@@ -22,6 +22,7 @@ def test_clean_bullets(text, expected):
     [
         ("\x93A lovely quote!\x94", "“A lovely quote!”"),
         ("\x91A lovely quote!\x92", "‘A lovely quote!’"),
+        ("Our dog's bowl.", "Our dog's bowl."),
     ],
 )
 def test_replace_unicode_quotes(text, expected):

unstructured/__version__.py

@@ -1 +1 @@
-__version__ = "0.3.0-dev4"  # pragma: no cover
+__version__ = "0.3.0-dev5"  # pragma: no cover

unstructured/cleaners/core.py

@@ -33,6 +33,7 @@ def replace_unicode_quotes(text) -> str:
     text = text.replace("\x92", "’")
     text = text.replace("\x93", "“")
     text = text.replace("\x94", "”")
+    text = text.replace("'", "'")
     return text