Add headers?

[halloleo]

I have plenty of CSVs without a header line, e.g.:

2023-03-12,16,28.5
2023-03-13,15.1,23
2023-03-14,15.9,26.2
2023-03-15,16.2,29.1

For the moment I generate headers before I look at the file in vd:

 $ csvcut -H data.csv | vd

and then I have in vd:

a	b	c
2023-03-12	16	28.5
2023-03-13	15.1	23
2023-03-14	15.9	26.2
2023-03-15	16.2	29.1

but is there an easy way/command line option to tell vd to generate some default headers?

[ajkerrigan]

I don't know of a way to generate a header row like that. You can use --header 0 to tell vd that there isn't a header in your data, but that'll leave the column names blank. And there are commands to bulk-rename columns (like rename-cols-selected and rename-cols-row), but those use row data to feed into headers.

The closest thing I can think of is doing it with a custom command in your ~/.visidatarc file

import string
from visidata import BaseSheet


@BaseSheet.api
def addDefaultHeader(sheet):
    sheet.options.header = 0
    sheet.reload()
    sheet.vd.sync()
    sheet.updateColNames(string.ascii_lowercase.split(), sheet.visibleCols)


BaseSheet.addCommand(
    "",
    "add-default-header",
    "sheet.addDefaultHeader()",
    "add a default header row",
)

Which would let you run a add-default-header command (optionally with a shortcut key) inside vd to generate that default header line. ...but your csvcut way seems easier :-/. Someone else may have a better idea.

[saulpw]

You could do Cgsgiq.

[ajkerrigan]

Haha that's way less work 😅

[saulpw]

It doesn't make useful Python identifiers though, which is probably what they wanted :)