On Wed, Jan 23, 2008 at 3:23 PM, James Dennett <[EMAIL PROTECTED]> wrote: > The POSIX/Single Unix Spec documentation for fnmatch might be a good > source, but I agree with the idea that SQLite should just document what > it does rather than assuming that there's a universal standard for > globbing.
Seems to me that GLOB is a poor substitute for REGEXP. At the shell level, I think an argument can be made that it is more concise, but I don't think that really holds for this use. Rather than extending GLOB towards a full REGEXP implementation, maybe it would be more reasonable to find (or write) a REGEXP implementation which was slight enough to be bundled into the SQLite core? No, I'm not volunteering :-). I think it would be pretty reasonable for SQLite to strictly match the TCL glob command's operation, for obvious reasons. I'd probably miss not being able to say [^x], but, *shrug*, what are you going to do? I could also see matching the glob provided by some other popular scripting language which might have SQLite embedded, say Python or Perl. -scott ----------------------------------------------------------------------------- To unsubscribe, send email to [EMAIL PROTECTED] -----------------------------------------------------------------------------
