Standardizing Object and Collection Names
After implementing table functions in several applications, I was alarmed to see the number of duplicate object and collection types I had implemented. For example, I had created these two objects:
SQL>DESC experiment_results_oName Null? Type ----------------------------------------- -------- ------ SAMPLE_AMT NUMBER SQL>DESC research_tallies_oName Null? Type ----------------------------------------- -------- ------ TALLY_TOTAL NUMBER
And then rolled them up into collections with similar names—replacing the “_o” (underscore o) suffix with “_t”. This is just the simplest example of the clutter I created because I hadn’t focused on the database as a whole. I’ve since gone back and replaced the two objects with a single one like this:
SQL> DESC number_o
Name Null? Type
----------------------------------------- -------- ------
COL1 NUMBERI also have similar generic objects for other datatypes including several standard lengths of VARCHAR2 fields.
Another standard I follow is that my object names have “_o” appended while my collections (or tables) have “_t” appended. This allows me to discern quickly what type they are.