fix(in2csv): Recalculate the dimensions of an XLSX file, instead of t…

…rusting the file's metadata, closes #1129
wireservice · Oct 23, 2023 · b2e54cd · b2e54cd
1 parent b4313a9
commit b2e54cd
Show file tree

Hide file tree

Showing 2 changed files with 7 additions and 2 deletions.
diff --git a/CHANGELOG.rst b/CHANGELOG.rst
@@ -1,3 +1,8 @@
+Unreleased
+----------
+
+fix: :doc:`/scripts/in2csv` `recalculates <https://openpyxl.readthedocs.io/en/stable/optimized.html#worksheet-dimensions>`_ the dimensions of XLSX files, instead of trusting the file's metadata.
+
 1.3.0 - October 18, 2023
 ------------------------
 

diff --git a/csvkit/utilities/in2csv.py b/csvkit/utilities/in2csv.py
@@ -155,7 +155,7 @@ def main(self):
                 table = agate.Table.from_xls(self.input_file, sheet=self.args.sheet,
                                              encoding_override=self.args.encoding_xls, **kwargs)
             elif filetype == 'xlsx':
-                table = agate.Table.from_xlsx(self.input_file, sheet=self.args.sheet, **kwargs)
+                table = agate.Table.from_xlsx(self.input_file, sheet=self.args.sheet, reset_dimensions=True, **kwargs)
             elif filetype == 'dbf':
                 if not hasattr(self.input_file, 'name'):
                     raise ValueError('DBF files can not be converted from stdin. You must pass a filename.')
@@ -177,7 +177,7 @@ def main(self):
                 tables = agate.Table.from_xls(self.input_file, sheet=sheets,
                                               encoding_override=self.args.encoding_xls, **kwargs)
             elif filetype == 'xlsx':
-                tables = agate.Table.from_xlsx(self.input_file, sheet=sheets, **kwargs)
+                tables = agate.Table.from_xlsx(self.input_file, sheet=sheets, reset_dimensions=True, **kwargs)
 
             base = splitext(self.input_file.name)[0]
             for i, (sheet_name, table) in enumerate(tables.items()):