Hugh McIntyre writes: > James Carlson wrote: > > Do we force future project teams to test in both environments? > > > > If not, then one of those environments will just rot over time. > > Things will fail to work right in one of the environments and, quite > > likely, nobody will care because supporting both is not one of the > > requirements. > > To some extent there are two environments right now, because people with > /usr/ucb first in the path see different versions of commands. Some > things occasionally break, and bugs get filed [*]. The /usr/gnu project > will cause the same type of issue.
The big distinction here is that $PATH is a standard feature of the shell. Reasonably-written scripts _must_ be written to be aware of how it works. Thus, anyone who gets hurt by it was doing something clearly wrong. Any new LINUX_LIKE_ENVIRONMENT=1 variable or zone or other non-standard method to change behavior would _not_ be something that script writers would know about, nor would it be something that's necessarily reasonable for them to accomodate. It wouldn't be a "bug" that they failed to take this new thing into account -- they couldn't have known. It'd be a source of bugs with no plausible escape. > [*] Topically for this discussion, this includes any script or use of > system() that thinks it's portable but uses "df" or "ps" without a full > path. Same bug as above. You get what you ask for in that case. -- James Carlson, Solaris Networking <[EMAIL PROTECTED]> Sun Microsystems / 1 Network Drive 71.232W Vox +1 781 442 2084 MS UBUR02-212 / Burlington MA 01803-2757 42.496N Fax +1 781 442 1677 _______________________________________________ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
