check the length of lists given in options before writing them

[yomimono]

/cc @haesbaert - I don't think this is too gory, what do you think?

[yomimono]

I should make this parameter optional since there's only one call where it's anything other than 1, standby for force push...

[haesbaert]

Hmm I think that's doable, but we really need to check them all before default to min=1.

Can you split the bugfixes/whitespace fixes into a different commit, that wrong code number is a good find.

If after we check them all, it might be that the option could be something like emptyok=true instead of giving a min_len, I suspect we will only use min_len=0.

Since we are here, we might consider minimum length checks on single options too, like strings

Hah, only now I saw that the inverse is already done in options_of_buf.

[yomimono]

On March 29, 2017 5:48:04 PM CDT, Christiano Haesbaert ***@***.***> wrote: Hmm I think that's doable, but we really need to check them all before default to min=1. Can you split the bugfixes/whitespace fixes into a different commit, that wrong code number is a good find. If after we check them all, it might be that the option could be something like emptyok=true instead of giving a min_len, I suspect we will only use min_len=0 -- You are receiving this because you authored the thread. Reply to this email directly or view it on GitHub: #48 (comment)

Sure, I can split up the commits. I did check everything that's a list here against the relevant RFCs, just couldn't find any others that had a different minimum - definitely wouldn't hurt for someone else to double check though. -- composed on a touchscreen keyboard; please forgive typos and brevity

[haesbaert]

I think the diff is fine, we're already enforcing min_len for every list option on the other way. So we should enforce when encoding too.

Just split them up and I'll merge :D.

Thanks for this !

[haesbaert]

This actually prompted me to have a look at get_8_list and get_16_list, they don't enforce min_len, I'll write a diff.