While you do need to get the text value from different table, so there's a little bit more work to do when fetching data, IMHO it's better to save ID.
I tend to get base data first, then loop over it and get all unique IDs for separate tables I might need to query, then get only used rows from these tables and loop over original data again and assign text values.
Very basic and dirty example:
PHP Code:
$q = $this->db->where('a', 'b')->get('table_a');
$tableBids = [];
$tableCids = [];
$data = [];
foreach ($q->result() as $row) {
// only keep unique IDs
$tableBids[$row->table_b_id] = $row->table_b_id;
$tableCids[$row->table_c_id] = $row->table_c_id;
$data[] = $row;
}
$q = $this->db->where_in('id', $tableBids)->limit(count($tableBids))->get('table_b');
$tableBids = [];
if ($q->num_rows()) {
foreach ($q->result() as $row) {
$tableBids[$row->id] = $row->text_label;
}
}
// ...
foreach ($data as $row) {
$row->table_b_label = isset($tableBids[$row->table_b_id]) ? $tableBids[$row->table_b_id] : false;
// ...
}
Might be overkill for one or two joins, but I never forget my first day at my current job, where I had to fix someone's silly query that basically joined together all the tables in our database. It crashed DB server on most calls, so that was definitely jumping in the deep end
