----- On 19 Feb, 2015, at 21:42, René J.V. Bertin [email protected] wrote:
Does the procedure leading up to a port's configure step test all > dependencies to see if they don't behave unexpectedly? Expected answer: no, > because there's no universal definition of such behaviour. No. > 2) What could reinstalling the binary package possibly have changed, except > for > some information in the registry or other metadata? (Reinstalling => same > version as was installed already.) If it was just a re-installation, nothing. I'm talking about a case where it isn't. > Not saying you're wrong, just that I don't see what my use of -n could have > borked that caused this glitch. The problem with -n is that it might work at the time when you use it, but it can come back later and bite you. > Fine. Than at least give us a "hold" feature to clamp a given port to its > current version ... O:-) That would require versioned dependencies, which MacPorts does not support. -- Clemens Lang _______________________________________________ macports-dev mailing list [email protected] https://lists.macosforge.org/mailman/listinfo/macports-dev
