Add -- to cp, mv and rm #1820
Add -- to cp, mv and rm #1820
Conversation
|
Thank you! |
| @@ -743,8 +743,8 @@ static const char * const envs[] = { | |||
|
|
|||
| #define PROGRESS_CP "cpg -giRp" | |||
| #define PROGRESS_MV "mvg -gi" | |||
| static char cp[sizeof PROGRESS_CP] = "cp -iRp"; | |||
| static char mv[sizeof PROGRESS_MV] = "mv -i"; | |||
| static char cp[sizeof PROGRESS_CP] = "cp -iRp --"; | |||
There was a problem hiding this comment.
I'm pretty sure this is broken now since the string exceeds the size of the buffer. IIRC the intention here was to have a buffer that is big enough to hold both the strings @KlzXS. And since the "progress" string was longer, it was used as a size.
Hopefully that made sense. If not feel free to ask questions.
There was a problem hiding this comment.
Ah I hadn't noticed the other PR #1821. That should solve it.
There was a problem hiding this comment.
Yup! IMO this is a pretty hacky way to size the buffer, but since these strings ideally don't change often, I guess it's good enough.
If you notice anything else that stopped working or started working in a weird way point it out so that I can include it in the other PR. This one got merged a bit sooner than I expected.
I think these are all the instances where we use
cp,mvandrmon user supplied paths, but I want someone else to confirm that as well.Can someone who regularly uses the more "exotic" shells confirm if they support this syntax and if it's even necessary? They should, but let's make sure. Particularly in regard to quitcd.
I don't foresee any regressions, but I haven't done any testing as of yet.
Fixes #1818