Comments for The Oracle Sponge

Comment on Back in the Saddle by Red94

Red94 — Fri, 23 Oct 2009 08:30:51 +0000

They have a laid-back humorous approach to cars, car repair, cup holders, pets, lawyers, car repair mechanics, SUVs, and almost everything else. ,

Comment on Back in the Saddle by BadGirl24

BadGirl24 — Thu, 22 Oct 2009 08:52:02 +0000

This is either the person who actually created the work or, if the work was made for hire, the employer or other person for whom the work was prepared. ,

Comment on v$filestat And Multiblock Read Statistics by Peter

Peter — Tue, 20 Oct 2009 13:16:22 +0000

Hi,
Should the last selection not be reversed:

Decode(READTIM-SINGLEBLKRDTIM,0,null,READTIM-SINGLEBLKRDTIM)/
(PHYRDS-SINGLEBLKRDS)

Comment on Back in the Saddle by Coder83

Coder83 — Sat, 10 Oct 2009 21:47:49 +0000

Encourage Peer-to-Peer Support and Mentoring Leaders need support, too. ,

Comment on Metacode Gone Wrong by joel garry

joel garry — Fri, 25 Sep 2009 23:19:55 +0000

I think I am the last one on earth who still uses @(#).

Comment on Metacode Gone Wrong by David Aldridge

David Aldridge — Fri, 18 Sep 2009 06:50:34 +0000

I haven’t, no. I’ll look at them when I have a chance.

Comment on Metacode Gone Wrong by dombrooks

dombrooks — Wed, 16 Sep 2009 11:47:59 +0000

Have you see either of these? Might also be nice to tie in such comments to something that can be extracted to a javadoc style resource as well as being inline in the code.
http://pldoc.sourceforge.net/
http://plsqldoc.dev.java.net/

Comment on Metacode Gone Wrong by Mark Brady

Mark Brady — Tue, 15 Sep 2009 17:08:35 +0000

Don’t get me wrong, I’m not arguing against your format. I love it. But in a Code Comment Hierarchy of Needs, why is the most important, what is second. I don’t know — Third Base!

Comment on Metacode Gone Wrong by David Aldridge

David Aldridge — Tue, 15 Sep 2009 17:02:47 +0000

Yes, if you know what you're looking for I think that most of those are pretty straightforward. But there are lots of less straightforward cases to consider, such as multitable insert, line-breaks in odd places ... etc.. One issue that has always bothered me a little with PL/SQL is breaking out the code into procedures and functions and getting an overview of the code. I think that tagging the beginning of procedures and functions and the important functionality in them would let you get a reasonable overview with a query against user_source ... || :META: Package pile_of_stuff || :META: section private || :META: Procedure change_employee_address || :META: selects employee || :META: inserts employee || :META: updates department || :META: updates address || :META: section public || :META: Function log_operation || :META: calls is_number || :META: calls log_entry || :META: calls dbms_application_info || :META: changes global_counter I'm eating my own dog food on this at the moment, so I'll see how it works out. Maybe I'll be able to post a "metacode dump" and we'll see if it makes sense.

Comment on Metacode Gone Wrong by Mark Brady

Mark Brady — Tue, 15 Sep 2009 15:47:30 +0000

/* || :META: selects employee || :META: inserts employee || :META: updates department || :META: updates address || :META: calls is_number || :META: calls log_entry || :META: calls dbms_application_info || :META: changes global_counter */ With a little bit of string parsing, this stuff *could* be automated. I'm of the opinion that what can't be answered with automation is WHY. You can discover all the servers on my network, and get CPU and memory and what processes it runs, but you can't tell why it runs them or how important it is to my business. In your example, somewhere in that proc, global_counter either exists on the left of := or after an INTO. Code can parse that out. WHY did you increment global_counter is something the next person may not understand. But yeh, no one does code reviews anymore.