I am working with a spreadsheet where I have to combine all possible combinations of comma-separated values into their own lines. Below is the main logic for the current implementation.

sheet_data = []
for sheetname, sheetdata in excel_content.items():
    for irow, row in enumerate(sheetdata.values):
        for a in str(row[5]).split(','):
            for b in str(row[17]).split(','):
                for c in str(row[21]).split(','):
                    for d in str(row[24]).split(','):
                        for e in str(row[27]).split(','):
                            for f in str(row[33]).split(','):
                                sheet_data.append(
                                    [row[0], row[1], row[2], row[3], row[4], a, row[6], row[7], row[8], row[9], row[10],row[11], row[12], row[13], row[14], row[15], row[16], b, row[18], row[19], row[20],c, row[22], row[23], d, row[25], row[26], e, row[28], row[29], row[30], row[31],row[32], f, row[34], row[35], row[36], row[37], row[38], row[39], row[40], row[41],row[42], row[43], row[44], row[45], row[46], row[47], row[48], row[49], row[50],row[51], row[52], row[53], row[54], row[55], row[56], row[57], row[58]])

Is there any way I could re-write this to be more re-usable and dynamic? What if I need to add a new column to also combine? That'd mean that I would need to create a new for-loop.

CodePudding user response：

Simple example of traversing this way for understanding:

import itertools

a = [['1,2', '3', '4,5,6', '7,8']]
for val in a:
    product_items = itertools.product(
        val[0].split(','),
        val[2].split(','),
        val[3].split(','),
    )
    product = list(itertools.product(product_items))

    for ss in product:
        ss0 = ss[0]
        print(ss0[0], val[1], ss0[1], ss0[2])

Returns:

1 3 4 7
1 3 4 8
1 3 5 7
1 3 5 8
1 3 6 7
1 3 6 8
2 3 4 7
2 3 4 8
2 3 5 7
2 3 5 8
2 3 6 7
2 3 6 8

So in your case you can do pretty much the same:

import itertools

sheet_data = []
for sheetname, sheetdata in excel_content.items():
    for irow, row in enumerate(sheetdata.values):
        product_items = itertools.product(
            str(row[5]).split(','),
            str(row[17]).split(','),
            str(row[21]).split(','),
            str(row[24]).split(','),
            str(row[27]).split(','),
            str(row[33]).split(','),
        )
        product = list(itertools.product(product_items))

        for ss in product:
            a, b, c, d, e, f = ss[0]
            sheet_data.append([
                row[0], row[1], row[2], row[3], row[4], a, row[6], row[7], 
                row[8], row[9], row[10], row[11], row[12], row[13], row[14], 
                row[15], row[16], b, row[18], row[19], row[20], c, row[22], 
                row[23], d, row[25], row[26], e, row[28], row[29], row[30], 
                row[31], row[32], f, row[34], row[35], row[36], row[37], 
                row[38], row[39], row[40], row[41], row[42], row[43], 
                row[44], row[45], row[46], row[47], row[48], row[49], 
                row[50], row[51], row[52], row[53], row[54], row[55], 
                row[56], row[57], row[58]]
            )

And it also could be improved in terms of readability.

• Block str(row[5]).split(','), str(row[17]).split(','), ... could be replaced with [str(row[i]).split(',') for i in [5, 17, ...]]
• Block row[0], row[1], row[2], row[3], row[4], a, row[6], ... could be replaced with *[row[x] for x in range(0, 5)], a, [row[x] for x in range(6, ...)]

CodePudding user response：

you can setup a list of column numbers to expand and use product to obtain the combinations:

import itertools

comb = [5,17,21,24,27,33] # columns to expand / combine
cols = 59                 # row width

for row in sheetdata.values:
    cells = (cell.split(',') if c in comb else [cell] for cell in row[:cols])
    for newRow in itertools.product(*cells):
        sheet_data.append(list(newRow))