Looping through rows and columns

How do I loop through rows and columns without converting the worksheet to csv or json before doing so?

Is it not possible to do something like

var worksheet = workbook.Sheets['Sheet1'];

for (var row in worksheet) {
  for (var col in row) {
    console.log(worksheet[row][col]);
  }
}

[reviewher]

@gregpinero the JSON conversion has a little bit of “intelligence” and is poorly documented 😕

For example consider this worksheet:

|foo|bar|baz|
| 1 | 2 | 3 |
| 4 | 5 | 6 |

Let’s say we are in node (the relevant parts work the same way in browser). Read the workbook:

var XLSX = require('xlsx');
var wb = XLSX.readFile('Workbook4.xlsx');
var ws = wb.Sheets.Sheet1;

So by default, the conversion will read the first row and use them as keys in the row object:

> XLSX.utils.sheet_to_json(wb.Sheets.Sheet1)
[ { foo: '1', bar: '2', baz: '3' },
  { foo: '4', bar: '5', baz: '6' } ]

If a field is missing, the data won’t show up. If there’s data in a cell with no header (say, data in cell D2 but not D1), data won’t show up.

passing an options argument with header controls the output:

Setting header to 1 will generate arrays:

> XLSX.utils.sheet_to_json(wb.Sheets.Sheet1, {header:1})
[ [ 'foo', 'bar', 'baz' ], [ '1', '2', '3' ], [ '4', '5', '6' ] ]

Setting header to A will use the column names:

> XLSX.utils.sheet_to_json(wb.Sheets.Sheet1, {header:'A'})
[ { A: 'foo', B: 'bar', C: 'baz' },
  { A: '1', B: '2', C: '3' },
  { A: '4', B: '5', C: '6' } ]

You can even give custom labels by passing an array:

> XLSX.utils.sheet_to_json(wb.Sheets.Sheet1, {header:["what", "the", "flip"]})
[ { what: 'foo', the: 'bar', flip: 'baz' },
  { what: '1', the: '2', flip: '3' },
  { what: '4', the: '5', flip: '6' } ]

The easiest way to control the range is to change the !ref value on the sheet. It’s expected to be an A1-style range like “A1:C4”. Changing the value will change the range that the utils look at:

> wb.Sheets.Sheet1['!ref']
'A1:C3'
> XLSX.utils.sheet_to_json(wb.Sheets.Sheet1)
[ { foo: '1', bar: '2', baz: '3' },
  { foo: '4', bar: '5', baz: '6' } ]

// now change to A1:A3
> wb.Sheets.Sheet1['!ref'] = 'A1:A3';
'A1:A3'
// the output will change
> XLSX.utils.sheet_to_json(wb.Sheets.Sheet1)
[ { foo: '1' }, { foo: '4' } ]

The same parameter also affects writing workbooks – cells outside of that range are not included:

> wb.Sheets.Sheet1['!ref'] = 'A1:C3';
'A1:C3'
> XLSX.read(XLSX.write(wb, {type:'binary'}), {type:'binary'}).Sheets.Sheet1
{ '!ref': 'A1:C3',
  A1: { t: 's', v: 'foo', h: 'foo', w: 'foo' },
  B1: { t: 's', v: 'bar', h: 'bar', w: 'bar' },
  C1: { t: 's', v: 'baz', h: 'baz', w: 'baz' },
  A2: { t: 'n', v: 1, w: '1' },
  B2: { t: 'n', v: 2, w: '2' },
  C2: { t: 'n', v: 3, w: '3' },
  A3: { t: 'n', v: 4, w: '4' },
  B3: { t: 'n', v: 5, w: '5' },
  C3: { t: 'n', v: 6, w: '6' } }
> wb.Sheets.Sheet1['!ref'] = 'A1:A3';
'A1:A3'
> XLSX.read(XLSX.write(wb, {type:'binary'}), {type:'binary'}).Sheets.Sheet1
{ '!ref': 'A1:A3',
  A1: { t: 's', v: 'foo', h: 'foo', w: 'foo' },
  A2: { t: 'n', v: 1, w: '1' },
  A3: { t: 'n', v: 4, w: '4' } }

[gregpinero]

Thanks. That does work.

I would like something just like sheet2arr above in the utils. That what I thought XLSX.utils.sheet_to_row_object_array would do but it seemed to give me strange results missing most of the data.