fix(#544): modernize xlsx handling

docs/changelog.rst

@@ -13,7 +13,8 @@ NEXT RELEASE
 TBD
 -------------------
 
-* Minor nitpick to remove unused line (`#559`_ by `@KyleKing`_)
+* Minor nitpick to remove unused line (`#562`_ by `@KyleKing`_)
+* Utilize openpyxl for xlsx after xlrd dropped support (`#559`_ by `@KyleKing`_)
 
 2.0.0
 -------------------

pyproject.toml

@@ -22,7 +22,8 @@ dependencies = [
     "extract-msg>=0.40.0",
     "pdfminer.six>=20221105",
     "python-pptx>=0.6.18",
-    "xlrd>=1.2.0,<2.0.0",  # v2 drops support for xlsx. Requires openpyxl or alternative to support xlsx if upgrading xlrd. See: https://github.com/deanmalmgren/textract/pull/543#issuecomment-2684619988
+    "openpyxl>=3.0.0",  # Replaces xlrd for Excel file support (xlrd 2.0+ dropped xlsx support)
+    "xlrd>=2.0.0",  # Supports legacy .xls files (openpyxl does not support .xls)
 ]
 
 [project.optional-dependencies]

textract/parsers/xls_parser.py

@@ -1,3 +1,28 @@
-from .xlsx_parser import Parser
+import xlrd
 
-__all__ = ["Parser"]
+from .utils import BaseParser
+
+
+class Parser(BaseParser):
+    """Extract text from legacy Excel files (.xls)."""
+
+    def extract(self, filename, **kwargs):
+        workbook = xlrd.open_workbook(filename)
+        sheets_name = workbook.sheet_names()
+        output = "\n"
+        for names in sheets_name:
+            worksheet = workbook.sheet_by_name(names)
+            num_rows = worksheet.nrows
+            num_cells = worksheet.ncols
+
+            for curr_row in range(num_rows):
+                new_output = []
+                for index_col in range(num_cells):
+                    value = worksheet.cell_value(curr_row, index_col)
+                    if value:
+                        if isinstance(value, (int, float)):
+                            value = str(value)
+                        new_output.append(value)
+                if new_output:
+                    output += " ".join(new_output) + "\n"
+        return output

textract/parsers/xlsx_parser.py

@@ -1,28 +1,31 @@
-import xlrd
+import openpyxl
 
 from .utils import BaseParser
 
 
 class Parser(BaseParser):
-    """Extract text from Excel files (.xls/xlsx)."""
+    """Extract text from Excel files (.xlsx)."""
 
     def extract(self, filename, **kwargs):
-        workbook = xlrd.open_workbook(filename)
-        sheets_name = workbook.sheet_names()
+        workbook = openpyxl.load_workbook(filename, data_only=True)
         output = "\n"
-        for names in sheets_name:
-            worksheet = workbook.sheet_by_name(names)
-            num_rows = worksheet.nrows
-            num_cells = worksheet.ncols
+        for sheet_name in workbook.sheetnames:
+            worksheet = workbook[sheet_name]
+
+            for row in worksheet.iter_rows(values_only=True):
+                non_empty_values = []
+                for value in row:
+                    if value is None or value is False:
+                        continue
+                    if isinstance(value, bool):
+                        value = "1"
+                    elif isinstance(value, (int, float)):
+                        # Convert to float to preserve decimal format (e.g., 83 -> 83.0)
+                        value = str(float(value))
+                    else:
+                        value = str(value)
+                    non_empty_values.append(value)
+                if non_empty_values:
+                    output += " ".join(non_empty_values) + "\n"
 
-            for curr_row in range(num_rows):
-                new_output = []
-                for index_col in range(num_cells):
-                    value = worksheet.cell_value(curr_row, index_col)
-                    if value:
-                        if isinstance(value, (int, float)):
-                            value = str(value)
-                        new_output.append(value)
-                if new_output:
-                    output += " ".join(new_output) + "\n"
         return output