Question

I need to devise a subquery to select the rows having matching values in a column.

Example

Select * 
from person 
where first_name in ('Java','SQL','Oracle');

However this list inside the parenthesis can be big, upto 30,000 values. I will be reading the values from a file and will be passing it inside this the parenthesis. However I thought there might be a limitation to the number of values that I can provide inside the parenthesis. Is there an optimal solution to address this scenario without creating and loading a new table with data?

EDIT: Thanks for your responses. Is the below query an option to be considered –

Select * 
from person 
where first_name like 'Java' 
   or first_name like 'SQL' 
   or first_name like 'Oracle';

Thanks.

Answer 1

I agree with Jonathan, concetanating 30000 values is not the way to go here, and his solution is a good way. But here is another idea, Oracle has a nice feature called External Tables.

With External Tables, you can treat a file as if it is a table. If your file is well structured (like CSV) you can easily use this feature.

Google “Oracle External Tables” and you’ll find a lot of information and examples, but for starters read this:

Basically, you’ll need a directory object to tell Oracle where your file is located, then you will create the External Table, perhaps something like below:

create directory DIR_MYFILE as '/my_source_dir/';

create table EXT_MY_FILE
    ( 
      some_id    number(8),
      some_value varchar2(100)
    )
   organization external
   ( default directory DIR_MYFILE 
     access parameters
     ( records delimited by newline
       fields terminated by ','
     )
     location ('my_file.csv')  
 );

Using this table in your query:

select * from person where first_name in (select some_value from EXT_MY_FILE);

EDIT: See APC’s comment, he has a good point.