[snip] > or > > 2) \:foo means ignore it as a parameter > > or > > 3) Add an attribute to Ingore :foo parameters (but not > :/\d+/ or ?). That > > would be an ODBC private attribute??? or a DBI one??? > > > > The downside of number two is the same as the :: problem, > except it has the > > benefit of being relatively standard, in terms of escape > > sequences...however, the triggers get rewritten, which can be > painful when > > cutting and pasting from a SQL tool such as TOAD... > > > > I actually prefer #3, honestly. > > Me too. > > > The only question I have is, should this be > > a DBI attribute? It seems that since DBD::ODBC is parsing this and not > > letting DBI do the work, I still have to do the work and honor the > > attribute. Is there a chance that this is handy across all DBI? > > Make it a driver-private attribute for now. Maybe > > odbc_ignore_named_placeholders => 1 > > Umm, or maybe something shorter :) Maybe odbc_no_named_ph ? > > The new preparse() stuff the DBI will be getting soon will impact > this kind of thing, so no point adding new DBI attributes just yet.
Ok. I also, at this point, prefer it to be a database option, rather than a statement option. My thinking is that someone who is going to use it will rather set it and forget it than set it for each new STH. Also, C<do> would inherit it... Comments? Jeff
