[ILUG] Perl / MySQL oddity

[Niall]

On Sat, Apr 22, 2000 at 07:31:06PM +0100, Wesley Darlington wrote:

> > is used as an rvalue BEFORE I call DBI->quote it gets the value fine, so
> > DBI->quote is obviously not stomping on it (even if I hadn't looked at the code to
> > make sure).
> 
> <Sound of lighbulb appearing above head> 
> Ah yes, got you now. Very interesting. *Very* interesting. 

Isn't it ? 

> Still don't like the $2 being passed to the /second/ call to quote in the
> one line

Not relevant whatsoever. That's in my real code - in my demo code, there's
only the one call. And anyway. the 'fix' of using $2 as an rvalue works in
the real code or the demo code.

> Hmmmmm, this is going to annoy me now. :-)

It's annoying the hell out of me, I can tell you. Particularly when I start
to look into DBI.pm. I've been using Perl for a few years now, quite
successfully, as in what I do works and I get paid for it, but I look at the
innards of DBI.pm and I realise that I haven't got a clue what's going on in
there :-( I tried to use the debugger to singlestep into the call to
DBI->quote but it acts as if I pressed n rather than s. Does this mean that
DBI->quote is not a perl subroutine at all ? Why ? If I fire up emacs on my
copy of DBI.pm and remove the quote subroutine from it completely, calling
DBI->quote still works fine. Why ? I imagine it's because quote is
overridden somewhere, but I'm fscked if I can find where. (I suspect the
answers to these two whys are closely related). To remove quote, I need to
remove the line


quote           => { U =>[2,3, '$string [, $data_type ]' ], O=>0x30 }

in the Database Session Class Interface. What on earth does this line mean ?
I've a horrible feeling that I'm dipping my toes into the XS waters here,
and that a stonking great crab is going to bite on them soon :-) 


Yours in bemusement,


Niall