Excel cell level validation data exposed

README.rst

@@ -43,15 +43,27 @@ Worksheet instance.
 Tip: A ``sheets`` property containing the sheet names is available on
 the Workbook instance.
 
-The ``rows()`` method will hand out an iterator to read the worksheet
-rows.
+
+Cell Metadata and Data Validation
+---------------------------------
+
+Each cell now exposes metadata fields: validation, cell_type, format, and constraints.
+This allows downstream libraries (like pandas) to assign datatypes based on Excel's intended validation/type.
 
 .. code:: python
 
-   # You can use .rows(sparse=True) to skip empty rows
    for row in sheet.rows():
-       print(row)
-   # [Cell(r=0, c=0, v='TEXT'), Cell(r=0, c=1, v=42.1337)]
+       for cell in row:
+           print(cell)
+           # Cell(r=0, c=0, v='TEXT', validation='Text', cell_type='str', format='General', constraints=None)
+
+You can retrieve validation info for a specific cell:
+
+.. code:: python
+
+   info = sheet.get_cell_validation(0, 0)
+   print(info)
+   # {'validation': 'Text', 'cell_type': 'str', 'format': 'General', 'constraints': None}
 
 Do note that dates will appear as floats. You must use the
 ``convert_date(date)`` method from the ``pyxlsb`` module to turn them

pyxlsb/worksheet.py

@@ -8,7 +8,7 @@
 if sys.version_info > (3,):
   xrange = range
 
-Cell = namedtuple('Cell', ['r', 'c', 'v'])
+Cell = namedtuple('Cell', ['r', 'c', 'v', 'validation', 'cell_type', 'format', 'constraints'])
 
 class Worksheet(object):
   def __init__(self, name, fp, rels_fp=None, stringtable=None, debug=False):
@@ -23,6 +23,7 @@ def __init__(self, name, fp, rels_fp=None, stringtable=None, debug=False):
     self.cols = []
     self.rels = {}
     self.hyperlinks = {}
+    self._cell_validations = {}  # {(row, col): {validation info}}
     self._parse()
 
   def __enter__(self):
@@ -39,6 +40,18 @@ def _parse(self):
       for el in self._rels:
         self.rels[el.attrib['Id']] = el.attrib['Target']
 
+    # Parse validation records (placeholder, actual parsing needed)
+    for item in self._reader:
+      # Example: if item[0] == biff12.DATAVALIDATION:
+      #   for each cell in validation range:
+      #     self._cell_validations[(row, col)] = {
+      #       'validation': item[1].type,
+      #       'cell_type': item[1].cell_type,
+      #       'format': item[1].format,
+      #       'constraints': item[1].constraints
+      #     }
+      pass  # TODO: Implement actual validation parsing
+
     for item in self._reader:
       if item[0] == biff12.DIMENSION:
         self.dimension = item[1]
@@ -64,19 +77,27 @@ def rows(self, sparse=False):
         if not sparse:
           while row_num < item[1].r - 1:
             row_num += 1
-            yield [Cell(row_num, i, None) for i in xrange(self.dimension.c + self.dimension.w)]
+            yield [Cell(row_num, i, None, None, None, None, None) for i in xrange(self.dimension.c + self.dimension.w)]
         row_num = item[1].r
-        row = [Cell(row_num, i, None) for i in xrange(self.dimension.c + self.dimension.w)]
+        row = [Cell(row_num, i, None, None, None, None, None) for i in xrange(self.dimension.c + self.dimension.w)]
       elif item[0] >= biff12.BLANK and item[0] <= biff12.FORMULA_BOOLERR:
         if item[0] == biff12.STRING and self._stringtable is not None:
-          row[item[1].c] = Cell(row_num, item[1].c, self._stringtable[item[1].v])
+          meta = self._cell_validations.get((row_num, item[1].c), {})
+          row[item[1].c] = Cell(row_num, item[1].c, self._stringtable[item[1].v],
+                                meta.get('validation'), meta.get('cell_type'), meta.get('format'), meta.get('constraints'))
         else:
-          row[item[1].c] = Cell(row_num, item[1].c, item[1].v)
+          meta = self._cell_validations.get((row_num, item[1].c), {})
+          row[item[1].c] = Cell(row_num, item[1].c, item[1].v,
+                                meta.get('validation'), meta.get('cell_type'), meta.get('format'), meta.get('constraints'))
       elif item[0] == biff12.SHEETDATA_END:
         if row is not None:
           yield row
         break
 
+  def get_cell_validation(self, row, col):
+    """Return validation info for a given cell (row, col)."""
+    return self._cell_validations.get((row, col), None)
+
   def close(self):
     self._reader.close()
     if self._rels_fp is not None: