On 23 March 2016 at 01:09, Mart Frauenlob <mart.frauenlob@xxxxxxxxx> wrote:
>
> I've not looked up the code calling this, but:
> First: bash only?
> Second: It's not granted to be in /bin.
> Third: May not be the wanted version.
>
> So a shebang like:
> #!/usr/bin/env bash
> or
> #!/urs/bin/env sh
> should be more compatible and fail proof.
>
>> +
>> +set -e
>> +
>> +MAX_JUMPS=16
>> +
>> +$NFT add table t
>
> Unquoted variable, may fail if, unlikely but possible, the name contains
> i.e. spaces.
>
>> +
>> +for i in $(seq 1 $MAX_JUMPS)
>> +do
>> + $NFT add chain t c${i}
>> +done
>
> Requires `seq' binary.
> I think for ((i=1; i<=$MAX_JUMPS; i++)) is more portable.
>
>> +
>> +for i in $(seq 1 $((MAX_JUMPS - 1)))
>> +do
>> + $NFT add rule t c${i} jump c$((i + 1))
>> +done
>
>
> Why not add functions? i.e.
>
Hi Mart,
yes, bash only :-) I think it's a very good shell. I'm not using any
bash operand which depends on the version... I don't expect anybody to
run this testsuite in bash v1 (20 years old already).
I didn't really care about portability... this is an internal script
to perform internal tests, this is going to be run in very few
systems.
Also, no functions, because I want the nft commands to be really
really clear, so debugging is easier.
Is this testsuite failing for you? Have you run it?
I would really like to review patches to incrementally improve the
testsuite, which has been deliberately simplified to focus on what
matter for us: testing nftables :-)
thanks, best regards.
--
Arturo Borrero González
--
To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
