Timeline for Set shell variable then invoke subprocess fails
Current License: CC BY-SA 4.0
4 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jul 4, 2019 at 0:33 | comment | added | cycollins | It also means I'm going to take a good hard look at all the "OPT*" variables and either localize them all or at least give some thought as to how having them changed globally would be a risk. | |
| Jul 4, 2019 at 0:31 | comment | added | cycollins | Holy s... Ahem.... Well, let's just say that was also a good heads up, because that would have confused me mightily. Though after this above discussion, I might have tried to localize OPTIND just as a defensive measure. So, is the more-fundamental problem here that using getopts in a function, as opposed to a script running as a sub shell is just plain fraught with issues and out of its intended usage model. Seems weird because while it is appropriate for stand-alone script "executables", it's also ideal for use in functions. | |
| Jul 3, 2019 at 19:07 | history | edited | ilkkachu | CC BY-SA 4.0 | added 798 characters in body |
| Jul 3, 2019 at 18:38 | history | answered | ilkkachu | CC BY-SA 4.0 |