| Message ID | 20201007072052.28200-1-rasmus.villemoes@prevas.dk |
|---|---|
| Headers | show |
| Series | allow positional arguments with "run" | expand |
On 07/10/2020 09.20, Rasmus Villemoes wrote: > This enables one to use positional arguments $1..$9 in functions > defined in the environment, Tom, can I ask for a decision on these? I know Wolfgang is opposed, and if that counts as a veto, fine, I'd just like to know so these are at least not kept hanging indefinitely. Thanks, Rasmus
On Thu, Nov 05, 2020 at 08:25:31AM +0100, Rasmus Villemoes wrote: > On 07/10/2020 09.20, Rasmus Villemoes wrote: > > This enables one to use positional arguments $1..$9 in functions > > defined in the environment, > > Tom, can I ask for a decision on these? I know Wolfgang is opposed, and > if that counts as a veto, fine, I'd just like to know so these are at > least not kept hanging indefinitely. Sorry for the lack of feedback. I guess, I just don't know. There's at least two series now (this and Simon's setexp) where part of the feedback has been "our hush is ancient, and we should replace and keep it in sync, and add features _upstream_". That position isn't wrong. But it's not easy to do, either. I know you already said you didn't have time and wouldn't step up to do that.
Dear Tom, In message <20201106205245.GH5340@bill-the-cat> you wrote: > > Sorry for the lack of feedback. I guess, I just don't know. There's at > least two series now (this and Simon's setexp) where part of the > feedback has been "our hush is ancient, and we should replace and keep > it in sync, and add features _upstream_". That position isn't wrong. > But it's not easy to do, either. I know you already said you didn't > have time and wouldn't step up to do that. This argument is not new, and I can fully understand this position, too. I'mm all to often myself in the position where the right Thing (TM) requires more efforts and/or time than what is available in the given project. And from the maintainer's point of view, this has always been the argument to sneak in code which is a workaround at best, and which itself cements the original problem even more and makes it more difficult to solve it at the roots. Best regards, Wolfgang Denk
This enables one to use positional arguments $1..$9 in functions defined in the environment, rather than having to use global variables. I.e., whereas run foo requires one to set the (shell or environment) variables referenced from foo beforehand, with this one can instead do run foo -- arg1 arg2 arg3 and use $1... up to $9 in the definition of foo. $# is set so foo can make decisions based on that, and ${3:-default} works as expected. In v2, at Heinrich's suggestion, make this available as an extension of the existing run command rather than being a separate 'call' command. Rasmus Villemoes (3): cli_hush.c: refactor handle_dollar() to prepare for "run with arguments" allow positional arguments with "run" command ut: add small hush tests cmd/Kconfig | 10 ++++ cmd/nvedit.c | 7 ++- common/cli.c | 44 +++++++++++++++--- common/cli_hush.c | 50 ++++++++++++++++---- configs/sandbox64_defconfig | 1 + configs/sandbox_defconfig | 1 + include/cli_hush.h | 9 ++++ include/test/suites.h | 1 + test/cmd/Makefile | 1 + test/cmd/hush.c | 91 +++++++++++++++++++++++++++++++++++++ test/cmd_ut.c | 6 +++ 11 files changed, 204 insertions(+), 17 deletions(-) create mode 100644 test/cmd/hush.c