[python] openpyxl - adjust column width size

I have following script which is converting a CSV file to an XLSX file, but my column size is very narrow. Each time I have to drag them with mouse to read data. Does anybody know how to set column width in openpyxl?

Here is the code I am using.

#!/usr/bin/python2.6
import csv
from openpyxl import Workbook
from openpyxl.cell import get_column_letter

f = open('users_info_cvs.txt', "rU")

csv.register_dialect('colons', delimiter=':')

reader = csv.reader(f, dialect='colons')

wb = Workbook()
dest_filename = r"account_info.xlsx"

ws = wb.worksheets[0]
ws.title = "Users Account Information"

for row_index, row in enumerate(reader):
    for column_index, cell in enumerate(row):
        column_letter = get_column_letter((column_index + 1))
        ws.cell('%s%s'%(column_letter, (row_index + 1))).value = cell

wb.save(filename = dest_filename)

We can convert numbers to their ASCII values and give it to column_dimension parameter

import openpyxl as xl

work_book = xl.load_workbook('file_location')
sheet = work_book['Sheet1']
column_number = 2
column = str(chr(64 + column_number))
sheet.column_dimensions[column].width = 20
work_book.save('file_location')

With openpyxl 3.0.3 the best way to modify the columns is with the DimensionHolder object, which is a dictionary that maps each column to a ColumnDimension object. ColumnDimension can get parameters as bestFit, auto_size (which is an alias of bestFit) and width. Personally, auto_size doesn't work as expected and I had to use width and fogured out that the best width for the column is len(cell_value) * 1.23.

To get the value of each cell it's necessary to iterate over each one, but I personally didn't use it because in my project I just had to write worksheets, so I got the longest string in each column directly on my data.

The example below just shows how to modify the column dimensions:

import openpyxl
from openpyxl.worksheet.dimensions import ColumnDimension, DimensionHolder
from openpyxl.utils import get_column_letter

wb = openpyxl.load_workbook("Example.xslx")
ws = wb["Sheet1"]

dim_holder = DimensionHolder(worksheet=ws)

for col in range(ws.min_column, ws.max_column + 1):
    dim_holder[get_column_letter(col)] = ColumnDimension(ws, min=col, max=col, width=20)

ws.column_dimensions = dim_holder

Even more pythonic way to set the width of all columns that works at least in openpyxl version 2.4.0:

for column_cells in worksheet.columns:
    length = max(len(as_text(cell.value)) for cell in column_cells)
    worksheet.column_dimensions[column_cells[0].column].width = length

The as_text function should be something that converts the value to a proper length string, like for Python 3:

def as_text(value):
    if value is None:
        return ""
    return str(value)

Here is an answer for Python 3.8 and OpenPyXL 3.0.0.

I tried to avoid using the get_column_letter function but failed.

This solution uses the newly introduced assignment expressions aka "walrus operator":

import openpyxl
from openpyxl.utils import get_column_letter

workbook = openpyxl.load_workbook("myxlfile.xlsx")

worksheet = workbook["Sheet1"]

MIN_WIDTH = 10
for i, column_cells in enumerate(worksheet.columns, start=1):
    width = (
        length
        if (length := max(len(str(cell_value) if (cell_value := cell.value) is not None else "")
                          for cell in column_cells)) >= MIN_WIDTH
        else MIN_WIDTH
    )
    worksheet.column_dimensions[get_column_letter(i)].width = width

My variation of Bufke's answer. Avoids a bit of branching with the array and ignores empty cells / columns.

Now fixed for non-string cell values.

ws = your current worksheet
dims = {}
for row in ws.rows:
    for cell in row:
        if cell.value:
            dims[cell.column] = max((dims.get(cell.column, 0), len(str(cell.value))))    
for col, value in dims.items():
    ws.column_dimensions[col].width = value

As of openpyxl version 3.0.3 you need to use

 dims[cell.column_letter] = max((dims.get(cell.column_letter, 0), len(str(cell.value))))

as the openpyxl library will raise a TypeError if you pass column_dimensions a number instead of a column letter, everything else can stay the same.

This is a dirty fix. But openpyxl actually supports auto_fit. But there is no method to access the property.

import openpyxl
from openpyxl.utils import get_column_letter

wb = openpyxl.load_workbook("Example.xslx")
ws = wb["Sheet1"]
for i in range(1, ws.max_column+1):
    ws.column_dimensions[get_column_letter(i)].bestFit = True
    ws.column_dimensions[get_column_letter(i)].auto_size = True

A slight improvement of the above accepted answer, that I think is more pythonic (asking for forgiveness is better than asking for permission)

column_widths = []
for row in workSheet.iter_rows():
    for i, cell in enumerate(row):
        try:
            column_widths[i] = max(column_widths[i], len(str(cell.value)))
        except IndexError:
            column_widths.append(len(str(cell.value)))

for i, column_width in enumerate(column_widths):
    workSheet.column_dimensions[get_column_letter(i + 1)].width = column_width

Since in openpyxl 2.6.1, it requires the column letter, not the column number, when setting the width.

 for column in sheet.columns:
    length = max(len(str(cell.value)) for cell in column)
    length = length if length <= 16 else 16
    sheet.column_dimensions[column[0].column_letter].width = length

All the above answers are generating an issue which is that col[0].column is returning number while worksheet.column_dimensions[column] accepts only character such as 'A', 'B', 'C' in place of column. I've modified @Virako's code and it is working fine now.

import re
import openpyxl
..
for col in _ws.columns:
    max_lenght = 0
    print(col[0])
    col_name = re.findall('\w\d', str(col[0]))
    col_name = col_name[0]
    col_name = re.findall('\w', str(col_name))[0]
    print(col_name)
    for cell in col:
        try:
            if len(str(cell.value)) > max_lenght:
                max_lenght = len(cell.value)
        except:
            pass
    adjusted_width = (max_lenght+2)
    _ws.column_dimensions[col_name].width = adjusted_width

I have a problem with merged_cells and autosize not work correctly, if you have the same problem, you can solve with the next code:

for col in worksheet.columns:
    max_length = 0
    column = col[0].column # Get the column name
    for cell in col:
        if cell.coordinate in worksheet.merged_cells: # not check merge_cells
            continue
        try: # Necessary to avoid error on empty cells
            if len(str(cell.value)) > max_length:
                max_length = len(cell.value)
        except:
            pass
    adjusted_width = (max_length + 2) * 1.2
    worksheet.column_dimensions[column].width = adjusted_width

This is my version referring @Virako 's code snippet

def adjust_column_width_from_col(ws, min_row, min_col, max_col):

        column_widths = []

        for i, col in \
                enumerate(
                    ws.iter_cols(min_col=min_col, max_col=max_col, min_row=min_row)
                ):

            for cell in col:
                value = cell.value
                if value is not None:

                    if isinstance(value, str) is False:
                        value = str(value)

                    try:
                        column_widths[i] = max(column_widths[i], len(value))
                    except IndexError:
                        column_widths.append(len(value))

        for i, width in enumerate(column_widths):

            col_name = get_column_letter(min_col + i)
            value = column_widths[i] + 2
            ws.column_dimensions[col_name].width = value

And how to use is as follows,

adjust_column_width_from_col(ws, 1,1, ws.max_column)

I had to change @User3759685 above answer to this when the openpxyl updated. I was getting an error. Well @phihag reported this in the comments as well

for column_cells in ws.columns:
    new_column_length = max(len(as_text(cell.value)) for cell in column_cells)
    new_column_letter = (openpyxl.utils.get_column_letter(column_cells[0].column))
    if new_column_length > 0:
        ws.column_dimensions[new_column_letter].width = new_column_length + 1

After update from openpyxl2.5.2a to latest 2.6.4 (final version for python 2.x support), I got same issue in configuring the width of a column.

Basically I always calculate the width for a column (dims is a dict maintaining each column width):

dims[cell.column] = max((dims.get(cell.column, 0), len(str(cell.value))))

Afterwards I am modifying the scale to something shortly bigger than original size, but now you have to give the "Letter" value of a column and not anymore a int value (col below is the value and is translated to the right letter):

worksheet.column_dimensions[get_column_letter(col)].width = value +1 

This will fix the visible error and assigning the right width to your column ;) Hope this help.