SQL Comparing a table attribute from values from other table - sql

Question

I want to compare values from the first table InputStrings with values in the second table StringConstraints in Sql.
InputStrings
+-----------+------------+-----------+
| Name | Address | City |
+-----------+------------+-----------+
| abcabcabc | xyxyxyxy | qweqweqwe |
| abbcabc | xyxxyxy | qweqwe |
| abccabc | xyxyxyxyxy | qwweqwe |
+-----------+------------+-----------+
StringConstraints
+---------+-----------+-----------+
| colName | minlength | maxlength |
+---------+-----------+-----------+
| Name | 2 | 20 |
| Address | 4 | 10 |
| City | 5 | 10 |
+---------+-----------+-----------+
I want to check if the length of the values in the Name column is between 2 and 20; length of values in the Address column is between 4 and 10; and length of values in the City column is between 5 and 10.
There are 68 rows like this in my table. InputString has 40 columns in it. I can't write for each and every row.
Can anyone help me make a generalized solution to compare the values?
I'm new to the database area.

Answer 1

this will return bad records:
Select *
from InputStrings
where (len(Name) not between (select minlength from StringContraints where colName = 'Name')
and
(select maxlength from StringContraints where colName = 'Name'))
or
(len(Address) not between (select minlength from StringContraints where colName = 'Address')
and
(select maxlength from StringContraints where colName = 'Address'))
or
(len(City) not between (select minlength from StringContraints where colName = 'City')
and
(select maxlength from StringContraints where colName = 'City'))

Answer 2

This is a way to create the same thing dynamically. Please keep in mind that Stack Overflow is not a coding service.
declare #sql varchar(max) =''
,#Name varchar(50)
,#min int
,#max int
,#counter int =1;
declare csr cursor
for
select colName, minlength, [maxlength] from (values ('Name',2,20),('Address',4,10),('City',5,10)) a(colName,minlength,[maxlength])
open csr
fetch next from csr
into #Name,#min,#max
set #sql = 'select * from inputstrings where '
while ##FETCH_STATUS = 0
begin
if(#counter =1 ) set #sql = #sql + '(len('+ #name+') not between ' + cast(#min as varchar(5)) + ' and ' + cast(#max as varchar(5)) + ') '
else set #sql = #sql + 'or (len('+ #name+') not between ' + cast(#min as varchar(5)) + ' and ' + cast(#max as varchar(5)) + ') '
set #counter=#counter +1
fetch next from csr
into #Name,#min,#max
end
close csr
deallocate csr
print #sql
exec(#sql)

Answer 3

You didn't state your DMBS, so this is a solution for Postgres:
select *
from (
select i.id,
t.*,
c.minlength,
c.maxlength,
length(t.value) between c.minlength and c.maxlength as is_valid
from inputstrings i
cross join lateral jsonb_each_text(to_jsonb(i) - 'id') as t(colname, value)
join stringconstraints c on lower(c.colname) = lower(t.colname)
) t
where not is_valid
order by id;
This first turns each row from the table inputstrings into key/value pairs and the result of that is joined to the stringconstraints table. From there it's easy to validate the column values based on the constraints. This is independent of the number of columns in inputstrings. The result will be one row per column value that violates the constraints.
For the following setup:
create table inputstrings (id integer, name text, address text, city text);
insert into inputstrings
values
(1, 'Name OK','Some Address that is too long','City Name OK'),
(2, 'N', 'Address OK', 'Cty'),
(3, 'Good Name', 'Good Address', 'Good City');
create table stringconstraints (colname text, minlength int, maxlength int);
insert into stringconstraints
values
('Name', 2, 20),
('Address', 4, 12),
('City', 5, 15);
The query returns this result:
id | colname | value | minlength | maxlength | is_valid
---+---------+-------------------------------+-----------+-----------+---------
1 | address | Some Address that is too long | 4 | 12 | false
2 | name | N | 2 | 20 | false
2 | city | Cty | 5 | 15 | false
I added the id column so that it is possible to match a invalid column value to the actual source row.
Online example: http://rextester.com/VVG55773
Second example with more columns: http://rextester.com/QCKG53573 (note the query hasn't changed)