updating all columns

I have a 5.1 database where one table has many columns.
Is there a way to update all the columns in a row without specifying
every column identifier? I've looked at "UPDATE" and "INSERT ... ON
DUPLICATE KEY UPDATE", but neither seem to be quite what I'm looking for.


Robert Huff

Robert Huff <roberthuff@rcn.com> wrote:

> I have a 5.1 database where one table has many columns.
> Is there a way to update all the columns in a row without specifying
> every column identifier? I've looked at "UPDATE" and "INSERT ... ON
> DUPLICATE KEY UPDATE", but neither seem to be quite what I'm looking for.

Nasty, nasty: REPLACE INTO... If you're using a primary ID, this will
actually delete the row with that ID, and insert a new one with that ID.
Any constraints may be checked and/or cascaded AFAIK. I'm actually very
against this kind of 'update'. Also, specifying column names is a good
thing. It will save you loads of debugging when for instance it's decided
there should be a field more or less in a table.

--
Rik Wasmus
Posted on Usenet, not any forum you might see this in.
Ask Smart Questions: http://tinyurl.com/anel

Rik wrote:

> Also, specifying column names is a good
> thing. It will save you loads of debugging when for instance it's
> decided there should be a field more or less in a table.

Fully aware of that. However:
a) this is a one-shot circumstance where it's not an unreasonable
technical choice, and
b) it's not entirely my call.


Robert Huff