fix: use encoding in context class (#4030)

Given the fact that the `_CsvPartitioningContext` defines an `_encoding`
property, this property was meant to be used. Behaviorally this change
should be a no-op, but supports future efforts where the partitioning
context applies internal logic.

Author: qued

CHANGELOG.md

@@ -1,17 +1,18 @@
-## 0.18.0
+## 0.18.1-dev0
 
 ### Enhancements
 
 ### Features
-- Upgraded Python version to 3.12
 
 ### Fixes
+- The `encoding` property of the `_CsvPartitioningContext` is now properly used.
 
-## 0.17.11-dev3
+## 0.18.0
 
 ### Enhancements
 
 ### Features
+- Upgraded Python version to 3.12
 
 ### Fixes
 - Fix type error when `result_file_type` is expected to be a `FileType` but is `None`

unstructured/__version__.py

@@ -1 +1 @@
-__version__ = "0.18.0"  # pragma: no cover
+__version__ = "0.18.1-dev0"  # pragma: no cover

unstructured/partition/csv.py

@@ -55,7 +55,7 @@ def partition_csv(
     )
 
     with ctx.open() as file:
-        dataframe = pd.read_csv(file, header=ctx.header, sep=ctx.delimiter, encoding=encoding)
+        dataframe = pd.read_csv(file, header=ctx.header, sep=ctx.delimiter, encoding=ctx.encoding)
 
     html_table = HtmlTable.from_html_text(
         dataframe.to_html(index=False, header=include_header, na_rep="")
@@ -135,6 +135,11 @@ def header(self) -> int | None:
         """Identifies the header row, if any, to Pandas, by idx."""
         return 0 if self._include_header else None
 
+    @lazyproperty
+    def encoding(self) -> str | None:
+        """The encoding to use for reading the file."""
+        return self._encoding
+
     @lazyproperty
     def last_modified(self) -> str | None:
         """The best last-modified date available, None if no sources are available."""