Count distinct rows (in powershell)

By far my most commonly run query is this:

Select Column, Count(*)
from SomeTable
Group by Column
order by 2 desc

When I have plain text and I want to perform that same check, I usually use a little tool I wrote called Grouply (or NimbleGroup)

But to do it in plain powershell you use a command like this:

cat .\22889_.txt | group | % { ($_.Name + ',' + $_.Count) }

(Assuming 22889_.txt is a file containing the rows you wish to group.)

Note: If all you want to do is count the number of lines, use Measure-Object (alias measure) instead.

e.g. to count all the date entries (Date & Time Long Format) in a file called journal.md you can do this

cat journal.md | ? { $_ -match '^[0-9]+:[0-9]+ (A|P)M [A-Z]+day, [0-9]+ [A-Z]+ [0-9]+' } | measure | % count