On Jul 31, 2011, at 08:28, Joshua Root wrote: > On 2011-7-31 18:21 , Ryan Schmidt wrote: >> >> I'd like to get the ball rolling on this ticket, the goal of which is to >> print a warning if a reinplace call doesn't end up changing anything in the >> file. >> >> https://trac.macports.org/ticket/15514 >> >> I wrote the patch three years ago and have been running it on my MacPorts >> installs since then, and it does help locate reinplaces that should be fixed >> or removed. >> >> I'd like to commit just the first half of the patch, which introduces the -q >> flag to reinplace, and release that with 2.0.1 (without the flag (or absence >> of it) actually doing anything). Then in a later release, we'll add the >> actual warning message when the flag is not used. The reason is that in some >> cases the author of a reinplace expects it to not do anything, and this >> would cause superfluous warnings -- most annoyingly right now, thousands of >> them (one per file in the distfile) for each kde4 portgroup port, and >> similarly for documentation files in the mysql5 port. If we commit the flag >> now without it doing anything, then later when we make the flag do >> something, we can simultaneously update those few >> designed-to-not-do-anything reinplaces to use the flag so they don't spew >> warnings. > > Why is this suddenly so urgent that it needs to go straight into a > release without ever having been in trunk?
The sooner the user population has a version of MacPorts with the placeholder -q switch, the sooner the other changes can be made. But it's not urgent. I'll put it in trunk first. _______________________________________________ macports-dev mailing list [email protected] http://lists.macosforge.org/mailman/listinfo.cgi/macports-dev
