How do I get the number of all unique name fields in table collections? In SQL the code would look something like this:
select count(*) from (
select name from table
group by name
)
My first guess was this
aggregated_sums = table.aggregate([{"$group": {"_id": "$name", "count":{"$sum": 1}}}])
print(len(list(aggregated_sums)))
But I'm afraid that in my scenario aggregated_sums might contain tens of millions of records.
This is almost a correct approach as per MongoDB documentation. Distinct has limitations, and might not be applicable in your case - see e.g. discussion under this answer.
The trick here is to use count on a separate step from group:
table.aggregate([
{
"$group": {
"_id": "$name"
}
},
{
"$count": "unique_keys"
}
])
I have tested it on mongoplayground, count clause on the group stage is not mandatory. Of course, the result returned is an aggregated count already, no need to call len afterwards.
