def unmerge_cells(self, range_string=None, start_row=None, start_column=None, end_row=None, end_column=None):
""" Remove merge on a cell range. Range is a cell range (e.g. A1:E1) """
if not range_string:
if start_row is None or start_column is None or end_row is None or end_column is None:
msg = "You have to provide a value either for "\
"'coordinate' or for 'start_row', 'start_column', 'end_row' *and* 'end_column'"
raise InsufficientCoordinatesException(msg)
else:
range_string = '%s%s:%s%s' % (get_column_letter(start_column + 1), start_row + 1, get_column_letter(end_column + 1), end_row + 1)
elif len(range_string.split(':')) != 2:
msg = "Range must be a cell range (e.g. A1:E1)"
raise InsufficientCoordinatesException(msg)
else:
range_string = range_string.replace('$', '')
if range_string in self._merged_cells:
self._merged_cells.remove(range_string)
min_col, min_row = coordinate_from_string(range_string.split(':')[0])
max_col, max_row = coordinate_from_string(range_string.split(':')[1])
min_col = column_index_from_string(min_col)
max_col = column_index_from_string(max_col)
# Mark cell as unmerged
for col in xrange(min_col,max_col+1):
for row in xrange(min_row,max_row+1):
if not (row == min_row and col == min_col):
self._get_cell('%s%s' % (get_column_letter(col), row)).merged = False
else:
msg = 'Cell range %s not known as merged.' % range_string
raise InsufficientCoordinatesException(msg)
def merge_cells(self, range_string=None, start_row=None, start_column=None, end_row=None, end_column=None):
""" Set merge on a cell range. Range is a cell range (e.g. A1:E1) """
if not range_string:
if start_row is None or start_column is None or end_row is None or end_column is None:
msg = "You have to provide a value either for "\
"'coordinate' or for 'start_row', 'start_column', 'end_row' *and* 'end_column'"
raise InsufficientCoordinatesException(msg)
else:
range_string = '%s%s:%s%s' % (get_column_letter(start_column + 1), start_row + 1, get_column_letter(end_column + 1), end_row + 1)
elif len(range_string.split(':')) != 2:
msg = "Range must be a cell range (e.g. A1:E1)"
raise InsufficientCoordinatesException(msg)
else:
range_string = range_string.replace('$', '')
# Make sure top_left cell exists - is this necessary?
min_col, min_row = coordinate_from_string(range_string.split(':')[0])
max_col, max_row = coordinate_from_string(range_string.split(':')[1])
min_col = column_index_from_string(min_col)
max_col = column_index_from_string(max_col)
# Blank out the rest of the cells in the range
for col in xrange(min_col, max_col + 1):
for row in xrange(min_row, max_row + 1):
if not (row == min_row and col == min_col):
# PHPExcel adds cell and specifically blanks it out if it doesn't exist
self._get_cell('%s%s' % (get_column_letter(col), row)).value = None
self._get_cell('%s%s' % (get_column_letter(col), row)).merged = True
if range_string not in self._merged_cells:
self._merged_cells.append(range_string)
def get_squared_range(self, min_col, min_row, max_col, max_row):
expected_columns = [
get_column_letter(ci) for ci in xrange(min_col, max_col)
]
current_row = min_row
style_table = self._style_table
for row, cells in groupby(
self.get_cells(min_row, min_col, max_row, max_col),
operator.attrgetter('row')):
full_row = []
if current_row < row:
for gap_row in xrange(current_row, row):
dummy_cells = get_missing_cells(gap_row, expected_columns)
yield tuple(
[dummy_cells[column] for column in expected_columns])
current_row = row
temp_cells = list(cells)
retrieved_columns = dict([(c.column, c) for c in temp_cells])
missing_columns = list(
set(expected_columns) - set(retrieved_columns.keys()))
replacement_columns = get_missing_cells(row, missing_columns)
for column in expected_columns:
if column in retrieved_columns:
cell = retrieved_columns[column]
if cell.style_id is not None:
style = style_table[int(cell.style_id)]
cell = cell._replace(
number_format=style.number_format.format_code
) #pylint: disable-msg=W0212
if cell.internal_value is not None:
if cell.data_type in Cell.TYPE_STRING:
cell = cell._replace(internal_value=unicode(
self._string_table[int(cell.internal_value)])
) #pylint: disable-msg=W0212
elif cell.data_type == Cell.TYPE_BOOL:
cell = cell._replace(
internal_value=cell.internal_value == '1')
elif cell.is_date:
cell = cell._replace(
internal_value=self._shared_date.from_julian(
float(cell.internal_value)))
elif cell.data_type == Cell.TYPE_NUMERIC:
cell = cell._replace(
internal_value=float(cell.internal_value))
elif cell.data_type in (
Cell.TYPE_INLINE,
Cell.TYPE_FORMULA_CACHE_STRING):
cell = cell._replace(
internal_value=unicode(cell.internal_value))
full_row.append(cell)
else:
full_row.append(replacement_columns[column])
current_row = row + 1
yield tuple(full_row)
def get_squared_range(p, min_col, min_row, max_col, max_row, string_table, style_table, shared_date):
expected_columns = [get_column_letter(ci) for ci in xrange(min_col, max_col)]
current_row = min_row
for row, cells in get_rows(p, min_row=min_row, max_row=max_row, min_column=min_col, max_column=max_col):
full_row = []
if current_row < row:
for gap_row in xrange(current_row, row):
dummy_cells = get_missing_cells(gap_row, expected_columns)
yield tuple([dummy_cells[column] for column in expected_columns])
current_row = row
temp_cells = list(cells)
retrieved_columns = dict([(c.column, c) for c in temp_cells])
missing_columns = list(set(expected_columns) - set(retrieved_columns.keys()))
replacement_columns = get_missing_cells(row, missing_columns)
for column in expected_columns:
if column in retrieved_columns:
cell = retrieved_columns[column]
if cell.style_id is not None:
style = style_table[int(cell.style_id)]
cell = cell._replace(number_format=style.number_format.format_code) #pylint: disable-msg=W0212
if cell.internal_value is not None:
if cell.data_type in Cell.TYPE_STRING:
cell = cell._replace(internal_value=unicode(string_table[int(cell.internal_value)])) #pylint: disable-msg=W0212
elif cell.data_type == Cell.TYPE_BOOL:
cell = cell._replace(internal_value=cell.internal_value == '1')
elif cell.is_date:
cell = cell._replace(internal_value=shared_date.from_julian(float(cell.internal_value)))
elif cell.data_type == Cell.TYPE_NUMERIC:
cell = cell._replace(internal_value=float(cell.internal_value))
elif cell.data_type in(Cell.TYPE_INLINE, Cell.TYPE_FORMULA_CACHE_STRING):
cell = cell._replace(internal_value=unicode(cell.internal_value))
full_row.append(cell)
else:
full_row.append(replacement_columns[column])
current_row = row + 1
yield tuple(full_row)
def test_dump_sheet():
test_filename = _get_test_filename()
wb = Workbook(optimized_write=True)
ws = wb.create_sheet()
letters = [get_column_letter(x + 1) for x in xrange(20)]
expected_rows = []
for row in xrange(20):
expected_rows.append(
['%s%d' % (letter, row + 1) for letter in letters])
for row in xrange(20):
expected_rows.append([(row + 1) for letter in letters])
for row in xrange(10):
expected_rows.append([
datetime(2010, ((x % 12) + 1), row + 1)
for x in range(len(letters))
])
for row in xrange(20):
expected_rows.append(
['=%s%d' % (letter, row + 1) for letter in letters])
for row in expected_rows:
ws.append(row)
wb.save(test_filename)
wb2 = load_workbook(test_filename)
ws = wb2.worksheets[0]
for ex_row, ws_row in zip(expected_rows[:-20], ws.rows):
for ex_cell, ws_cell in zip(ex_row, ws_row):
eq_(ex_cell, ws_cell.value)
os.remove(test_filename)
def test_dump_sheet():
test_filename = _get_test_filename()
wb = Workbook(optimized_write=True)
ws = wb.create_sheet()
letters = [get_column_letter(x + 1) for x in xrange(20)]
expected_rows = []
for row in xrange(20):
expected_rows.append(['%s%d' % (letter, row + 1) for letter in letters])
for row in xrange(20):
expected_rows.append([(row + 1) for letter in letters])
for row in xrange(10):
expected_rows.append([datetime(2010, ((x % 12) + 1), row + 1) for x in range(len(letters))])
for row in xrange(20):
expected_rows.append(['=%s%d' % (letter, row + 1) for letter in letters])
for row in expected_rows:
ws.append(row)
wb.save(test_filename)
wb2 = load_workbook(test_filename)
ws = wb2.worksheets[0]
for ex_row, ws_row in zip(expected_rows[:-20], ws.rows):
for ex_cell, ws_cell in zip(ex_row, ws_row):
eq_(ex_cell, ws_cell.value)
os.remove(test_filename)
class TestFloats(TestWorksheet):
sheet_name = 'Sheet2 - Numbers'
query_range = 'K1:L30'
expected = expected = [[(x + 1) / 100.0] for x in xrange(30)]
def test_read_fast_integrated(self):
wb = self._open_wb()
ws = wb.get_sheet_by_name(name = self.sheet_name)
for row, expected_row in zip(ws.iter_rows(self.query_range), self.expected):
row_values = [x.internal_value for x in row]
eq_(row_values, expected_row)
def unmerge_cells(self,
range_string=None,
start_row=None,
start_column=None,
end_row=None,
end_column=None):
""" Remove merge on a cell range. Range is a cell range (e.g. A1:E1) """
if not range_string:
if start_row is None or start_column is None or end_row is None or end_column is None:
msg = "You have to provide a value either for "\
"'coordinate' or for 'start_row', 'start_column', 'end_row' *and* 'end_column'"
raise InsufficientCoordinatesException(msg)
else:
range_string = '%s%s:%s%s' % (
get_column_letter(start_column + 1), start_row + 1,
get_column_letter(end_column + 1), end_row + 1)
elif len(range_string.split(':')) != 2:
msg = "Range must be a cell range (e.g. A1:E1)"
raise InsufficientCoordinatesException(msg)
else:
range_string = range_string.replace('$', '')
if range_string in self._merged_cells:
self._merged_cells.remove(range_string)
min_col, min_row = coordinate_from_string(
range_string.split(':')[0])
max_col, max_row = coordinate_from_string(
range_string.split(':')[1])
min_col = column_index_from_string(min_col)
max_col = column_index_from_string(max_col)
# Mark cell as unmerged
for col in xrange(min_col, max_col + 1):
for row in xrange(min_row, max_row + 1):
if not (row == min_row and col == min_col):
self._get_cell(
'%s%s' %
(get_column_letter(col), row)).merged = False
else:
msg = 'Cell range %s not known as merged.' % range_string
raise InsufficientCoordinatesException(msg)
class TestIntegers(TestWorksheet):
sheet_name = 'Sheet2 - Numbers'
expected = [[x + 1] for x in xrange(30)]
query_range = 'D1:D30'
def test_read_fast_integrated(self):
wb = self._open_wb()
ws = wb.get_sheet_by_name(name=self.sheet_name)
for row, expected_row in zip(ws.iter_rows(self.query_range),
self.expected):
row_values = [x.internal_value for x in row]
assert row_values == expected_row
def merge_cells(self,
range_string=None,
start_row=None,
start_column=None,
end_row=None,
end_column=None):
""" Set merge on a cell range. Range is a cell range (e.g. A1:E1) """
if not range_string:
if start_row is None or start_column is None or end_row is None or end_column is None:
msg = "You have to provide a value either for "\
"'coordinate' or for 'start_row', 'start_column', 'end_row' *and* 'end_column'"
raise InsufficientCoordinatesException(msg)
else:
range_string = '%s%s:%s%s' % (
get_column_letter(start_column + 1), start_row + 1,
get_column_letter(end_column + 1), end_row + 1)
elif len(range_string.split(':')) != 2:
msg = "Range must be a cell range (e.g. A1:E1)"
raise InsufficientCoordinatesException(msg)
else:
range_string = range_string.replace('$', '')
# Make sure top_left cell exists - is this necessary?
min_col, min_row = coordinate_from_string(range_string.split(':')[0])
max_col, max_row = coordinate_from_string(range_string.split(':')[1])
min_col = column_index_from_string(min_col)
max_col = column_index_from_string(max_col)
# Blank out the rest of the cells in the range
for col in xrange(min_col, max_col + 1):
for row in xrange(min_row, max_row + 1):
if not (row == min_row and col == min_col):
# PHPExcel adds cell and specifically blanks it out if it doesn't exist
self._get_cell('%s%s' %
(get_column_letter(col), row)).value = None
self._get_cell('%s%s' %
(get_column_letter(col), row)).merged = True
if range_string not in self._merged_cells:
self._merged_cells.append(range_string)
from openpyxl.reader.worksheet import read_dimension
from openpyxl.shared.compat import unicode
from openpyxl.shared.ooxml import (
PACKAGE_WORKSHEETS,
ARC_STYLE,
SHEET_MAIN_NS
)
TYPE_NULL = Cell.TYPE_NULL
MISSING_VALUE = None
RE_COORDINATE = re.compile('^([A-Z]+)([0-9]+)$')
SHARED_DATE = SharedDate()
_COL_CONVERSION_CACHE = dict((get_column_letter(i), i) for i in xrange(1, 18279))
def column_index_from_string(str_col, _col_conversion_cache=_COL_CONVERSION_CACHE):
# we use a function argument to get indexed name lookup
return _col_conversion_cache[str_col]
del _COL_CONVERSION_CACHE
RAW_ATTRIBUTES = ['row', 'column', 'coordinate', 'internal_value',
'data_type', 'style_id', 'number_format']
BaseRawCell = namedtuple('RawCell', RAW_ATTRIBUTES)
class RawCell(BaseRawCell):
"""Optimized version of the :class:`openpyxl.cell.Cell`, using named tuples.
def range(self, range_string, row=0, column=0):
"""Returns a 2D array of cells, with optional row and column offsets.
:param range_string: cell range string or `named range` name
:type range_string: string
:param row: number of rows to offset
:type row: int
:param column: number of columns to offset
:type column: int
:rtype: tuples of tuples of :class:`openpyxl.cell.Cell`
"""
if ':' in range_string:
# R1C1 range
result = []
min_range, max_range = range_string.split(':')
min_col, min_row = coordinate_from_string(min_range)
max_col, max_row = coordinate_from_string(max_range)
if column:
min_col = get_column_letter(
column_index_from_string(min_col) + column)
max_col = get_column_letter(
column_index_from_string(max_col) + column)
min_col = column_index_from_string(min_col)
max_col = column_index_from_string(max_col)
cache_cols = {}
for col in xrange(min_col, max_col + 1):
cache_cols[col] = get_column_letter(col)
rows = xrange(min_row + row, max_row + row + 1)
cols = xrange(min_col, max_col + 1)
for row in rows:
new_row = []
for col in cols:
new_row.append(self.cell('%s%s' % (cache_cols[col], row)))
result.append(tuple(new_row))
return tuple(result)
else:
try:
return self.cell(coordinate=range_string, row=row,
column=column)
except CellCoordinatesException:
pass
# named range
named_range = self._parent.get_named_range(range_string)
if named_range is None:
msg = '%s is not a valid range name' % range_string
raise NamedRangeException(msg)
if isinstance(named_range, NamedRangeContainingValue):
msg = '%s refers to a value, not a range' % range_string
raise NamedRangeException(msg)
result = []
for destination in named_range.destinations:
worksheet, cells_range = destination
if worksheet is not self:
msg = 'Range %s is not defined on worksheet %s' % \
(cells_range, self.title)
raise NamedRangeException(msg)
content = self.range(cells_range)
if isinstance(content, tuple):
for cells in content:
result.extend(cells)
else:
result.append(content)
if len(result) == 1:
return result[0]
else:
return tuple(result)
from zipfile import ZipFile
import re
import tempfile
import zlib
import zipfile
import struct
TYPE_NULL = Cell.TYPE_NULL
MISSING_VALUE = None
RE_COORDINATE = re.compile('^([A-Z]+)([0-9]+)$')
SHARED_DATE = SharedDate()
_COL_CONVERSION_CACHE = dict(
(get_column_letter(i), i) for i in xrange(1, 18279))
def column_index_from_string(str_col,
_col_conversion_cache=_COL_CONVERSION_CACHE):
# we use a function argument to get indexed name lookup
return _col_conversion_cache[str_col]
del _COL_CONVERSION_CACHE
RAW_ATTRIBUTES = [
'row', 'column', 'coordinate', 'internal_value', 'data_type', 'style_id',
'number_format'
]
class TestFloats(TestWorksheet):
sheet_name = 'Sheet2 - Numbers'
query_range = 'K1:L30'
expected = expected = [[(x + 1) / 100.0] for x in xrange(30)]
def range(self, range_string, row=0, column=0):
"""Returns a 2D array of cells, with optional row and column offsets.
:param range_string: cell range string or `named range` name
:type range_string: string
:param row: number of rows to offset
:type row: int
:param column: number of columns to offset
:type column: int
:rtype: tuples of tuples of :class:`openpyxl.cell.Cell`
"""
if ':' in range_string:
# R1C1 range
result = []
min_range, max_range = range_string.split(':')
min_col, min_row = coordinate_from_string(min_range)
max_col, max_row = coordinate_from_string(max_range)
if column:
min_col = get_column_letter(
column_index_from_string(min_col) + column)
max_col = get_column_letter(
column_index_from_string(max_col) + column)
min_col = column_index_from_string(min_col)
max_col = column_index_from_string(max_col)
cache_cols = {}
for col in xrange(min_col, max_col + 1):
cache_cols[col] = get_column_letter(col)
rows = xrange(min_row + row, max_row + row + 1)
cols = xrange(min_col, max_col + 1)
for row in rows:
new_row = []
for col in cols:
new_row.append(self.cell('%s%s' % (cache_cols[col], row)))
result.append(tuple(new_row))
return tuple(result)
else:
try:
return self.cell(coordinate=range_string,
row=row,
column=column)
except CellCoordinatesException:
pass
# named range
named_range = self._parent.get_named_range(range_string)
if named_range is None:
msg = '%s is not a valid range name' % range_string
raise NamedRangeException(msg)
if isinstance(named_range, NamedRangeContainingValue):
msg = '%s refers to a value, not a range' % range_string
raise NamedRangeException(msg)
result = []
for destination in named_range.destinations:
worksheet, cells_range = destination
if worksheet is not self:
msg = 'Range %s is not defined on worksheet %s' % \
(cells_range, self.title)
raise NamedRangeException(msg)
content = self.range(cells_range)
if isinstance(content, tuple):
for cells in content:
result.extend(cells)
else:
result.append(content)
if len(result) == 1:
return result[0]
else:
return tuple(result)
