🚀 go-pugleaf

How do I (in pure SQL):

1) ensure that, if I have two values, one of which is an autoincremented
integer that provides the primary index for the table, the values in the
second field are unique?  i.e., imagine a table with two fields, x and y,
and y is text while x is an autoincremented integer.  x is guaranteed to be
unique but I also need y to be unique.

2) determine if a given string is present in the table mentioned in item 1,
and if it is not insert it, and, either way, get the value of x
corresponding to the value of y and insert that along with any other
available data for a second table where x from the first table is identified
as a foreign key?  I know how to do the second part of this where the
relevant index is known at compile time (i.e. trivial tests involving
inserting data into the lookup table and then just hard coding the index,
(i.e. the value 1) for the first item inserted into the lookup table, into
the insert statement for the second table, but I don't know the more general
solution when you have no idea whether or not the value to be looked up
inthe lookup table is in it already and, either way, what its index is.

The general problem I am trying to solve using SQL is how to add data to a
pair (or more) of tables linked with one to many relationships.  And, a
constraint is that it must be SQL code I can pass to the database using an
ADODB.Connection!

Thanks,

Ted