There are many data types to choose from when defining your table columns. The eighty percent rule tells us we can probably get away with a smaller subset of data types eighty percent of the time.
Here is what the ‘TYPE’ control looks like in the modeler when defining a column:
I was curious to see what data types were used most in my 18.104.22.168 database, so I fired off a quick query:
SELECT COUNT(*), data_type FROM dba_tab_cols WHERE data_type NOT LIKE ('%$%') AND owner NOT IN ('SYS', 'SYSTEM') GROUP BY data_type ORDER BY 1 DESC;
COUNT DATA_TYPE +++++ ++++++++++++++++++++++++++ 30089 VARCHAR2 11562 NUMBER 2553 DATE 2310 RAW 509 CLOB 176 CHAR 170 BLOB 161 XMLTYPE 160 TIMESTAMP(6) 68 XMLTYPEPI 54 NVARCHAR2 50 TIMESTAMP(6) WITH TIME ZONE 39 ROWID 34 XMLTYPEEXTRA 23 SharedValueType754_T 23 SDO_GEOMETRY 22 ANYDATA 21 STRINGLIST 17 MGMT_JOB_VECTOR_PARAMS 15 FLOAT
Instead of seeing all 100+ types when designing your model, what if you could…
Create a ‘Favorites’ list of data types
Yes you can!
In the preferences for v3.1 you can establish your ‘preferred’ data types.
So now when I go to create or modify my attributes, I only see this when the ‘Preferred’ check box is ticked: