3 ms·
Yep, I've done a lot of PL/SQL programming, and we've always used (+)= syntax and joins with all the tables after commas and all the joining conditions after WH
by ajuc 13y ago
Yep, I've done a lot of PL/SQL programming, and we've always used (+)= syntax and joins with all the tables after commas and all the joining conditions after WHERE.
Now I work on different project and we use join syntax, but I could easily imagine people that do joins all day, and not know JOIN .. ON .. syntax.
- jacques_chester 13y agoI came to Oracle after it adopted the ANSI syntax, so that's what I use. So my experience is the opposite of yours -- when I see the (+) I need to look up the syntax to remember if it's left or right outer.
- jfb 13y agoAnd then I ask, hey, where's my BOOLEAN? And then I drink.
- jacques_chester 13y agoOh god. And the lack of a serial/autoincrement/identity type. So. Many. Effing. Triggers. And 32-character identifiers. sigh
- tmzt 13y agoIt's so much better to use a database that only allows one autoincrementing value per record, or one TIMESTAMP and then only allows you to have either a create timestamp or an update timestamp without writing a trigger. I prefer the way Oracle does it, you may have to do more work but it's more explicit and flexible that way.
- jacques_chester 13y agoI don't. I prefer for the common case to be correctly and automatically handled for me.