On 8 Aug 2007, at 00:52, Timothy Bates wrote:

It would be handy if there was a code for language grammar capture numbering, so that these are more resilient and self-healing when the regexp underlying their definition changes.

I believe the plan for TM2 is to use the named captures functionality of Oniguruma, so you can instead give captured groups identifiers of your choice and then use those identifiers to refer to them for naming.