Get Excel-Style Column Names from Column Number

Question

This is the code for providing the COLUMN name when the row and col ID is provided but when I give values like row = 1 and col = 104, it should return CZ<

Answer 1

Here's another way to get excel column names without a loop, using a pandas multiindex dataframe. This is a modification of a generalized base-to-base converter, so the code is a little lengthier than some of the other options, but I think it's effective:

def xlcolumn(num):
    base = ["A","B","C","D","E","F","G","H","I","J","K","L","M","N","O","P","Q","R","S","T","U","V","W","X","Y","Z"]
    baseln = len(base)
    idx = [""]+base
    num = num-1

    # create pandas multiindex using idx, base
    # current excel version has 16384 columns (A --> XFD), so multiindex needs to have a minimum of 3 levels:
    #   (26x26x26 = 17576 > 16384 columns) 

    index = pd.MultiIndex.from_product([idx, idx, idx],names=['level 1', 'level 2', 'level 3'])
    df = pd.DataFrame(index = index)
    df = df.drop("",level = 'level 3')
    df = df.iloc[:baseln].append(df.drop("",level = 'level 2'))
    df['val']=1

    if num < baseln:
        xlcol = str(df.iloc[num].name[2])
    elif num >= baseln and num < baseln**2:
        xlcol = str(df.iloc[num].name[1])+str(df.iloc[num].name[2])
    else:
        xlcol = str(df.iloc[num].name[0])+str(df.iloc[num].name[1])+str(df.iloc[num].name[2])

    return xlcol

Answer 2

I think it is something like this :

def get_col(col):
    """Get excel-style column names"""
    (div, mod) = divmod(col, 26)
    if div == 0:
        return str(unichr(mod+64))
    elif mod == 0:
        return str(unichr(div+64-1)+'Z')
    else:
        return str(unichr(div+64)+unichr(mod+64))

Some tests :

>>> def get_col(col):
...     (div, mod) = divmod(col, 26)
...     if div == 0:
...         return str(unichr(mod+64))
...     elif mod == 0:
...         return str(unichr(div+64-1)+'Z')
...     else:
...         return str(unichr(div+64)+unichr(mod+64))
... 
>>> get_col(105)
'DA'
>>> get_col(104)
'CZ'
>>> get_col(1)
'A'
>>> get_col(55)
'BC'

Answer 3

I think i figured it out. divmod(104,26) gives mod=0 which makes chr(0+64) = 64 ie '@'.

if i add this line before column_label "mod=26 if mod==0 else mod" i think it should work fine

column_label=''
div=104
while div:
    (div, mod) = divmod(div, 26)
    mod=26 if mod==0 else mod
    column_label = chr(mod + 64) + column_label

print column_label

Answer 4

use this code:

def xlscol(colnum):
    a = []
    while colnum:
        colnum, remainder = divmod(colnum - 1, 26)
        a.append(remainder)
    a.reverse()
    return ''.join([chr(n + ord('A')) for n in a])

Answer 5

I love maritineau's answer since its code looks plain and easy to follow. But it can't handle the column number which is greater than 26**2 + 26. So I modify part of it.

def excel_col(col):
    """Covert 1-relative column number to excel-style column label."""
    quot, rem = divmod(col-1,26)
    return excel_col(quot) + chr(rem+ord('A')) if col!=0 else ''



if __name__=='__main__':
    for i in [1, 26, 27, 26**3+26**2+26]:
        print 'excel_col({0}) -> {1}'.format(i, excel_col(i))

Results

excel_col(1) -> A
excel_col(26) -> Z
excel_col(27) -> AA
excel_col(18278) -> ZZZ

Answer 6

def ColNum2ColName(n):
   convertString = "ABCDEFGHIJKLMNOPQRSTUVWXYZ"
   base = 26
   i = n - 1

   if i < base:
      return convertString[i]
   else:
      return ColNum2ColName(i//base) + convertString[i%base]

---

EDIT: Right, right zondo.

I just approached A, B, .. AA, AB, ... as a numeric base with digits A-Z.

A = 1
B = 2
 .
 .
X = 24
Y = 25
Z = 26
 .
 .
 .

It's an easy way without any while loop etc. and works for any number > 0.