On Sat, Mar 13, 2021 at 1:01 AM Rob Landley <[email protected]> wrote:
> On 3/12/21 2:15 PM, enh wrote: > > But could I get a second opinion on that since I obviously flubbed > allowing this > > patch in last time? My blog suggests I was hip-deep in doing the > first round of > > modern shell stuff. > > Which I was also doing when I got this email, and it was the end of a long > day, > and none of _my_ test cases ever hit this so I wanted to see if you had > one that > did. (That's why I wanted the second opinion.) > > > > it seems from the history like the EXDEV assignment was for the > later test of > > > `errno != EXDEV` and is really meant to be a "did we successfully > move?". i > > > don't think that's still valid now the force/clobber code has been > added? > > > > I think the rest of it's ok? > > > > yeah, i _think_ so, but i worry that it's past the level of complexity > that > > humans can cope with. sure you don't just want an explicit `moved` > boolean? > > I'll add a comment on why we're setting errno = EXDEV, and rename "rc" > to... I > dunno, "send"? (rc = 0 means "don't try to send this file to the > destination > anymore, whether by cp or mv" and it consistently means that through its > lifetime. Splitting that role into two variables doesn't seem clearer to > me?) > yeah, i think the use of _just_ `rc` was okay too. it was really the "am i following `errno`, `rc`, or both here?" that was particularly confusing. > Rob >
_______________________________________________ Toybox mailing list [email protected] http://lists.landley.net/listinfo.cgi/toybox-landley.net
