Richard Lowe <richlowe at richlowe.net> writes: [Sorry for the spam Steve, need more caffeine :\]
> Richard Lowe <richlowe at richlowe.net> writes: > >> Hey, >> >> I could have sworn I filed a bug in bugster about class-action >> scripts, but I'll be damned if I can find it... >> Actually, looking through them again, it seems /etc/system maybe the only real problem in this regard, most of the others only update the ident in the conditional chunk (or don't conditionalize such a thing). It's unclear to me whether they'd misbehave if #ident wasn't expanded (my thinking here is, acr would have exploded when bfu'ing from the hg gate, if so...) i.etcsystem seems to be the only victim here (it does everything in the condition body, to include deciding whether to clobber the installed version, or land in /etc/system.new) I don't suppose anyone feels like double checking my thinking here? -- Rich
