Fork me on GitHub
#yada
<
2016-04-13
>
mccraigmccraig07:04:55

@malcolmsparks: while porting to 1.1 we've come across a couple of places where we would like custom coercion... would you be happy with a pr to allow a custom coercer-fn to be specified on a resource ?

malcolmsparks07:04:24

@mccraigmccraig: thanks for those reflogs, I'm looking into it now but will have more time this afternoon

malcolmsparks07:04:11

on the custom coercer front, did you know you can provide your own defmethods? But that's app global - if you need on a resource-by-resource basis that makes sense too - yes, please send a PR

malcolmsparks07:04:32

sorry for the brevity yesterday, was busy with stuff

mccraigmccraig08:04:01

np, brevity is functional simple_smile

malcolmsparks08:04:22

there's been some fixes to multipart recently, can you let me know the version/sha1 you're using?

mccraigmccraig08:04:33

we are now on 1.1.5

mccraigmccraig08:04:09

we were on 1.1.2, but i upgraded at some point last night and was still seeing the error on bigger files

mccraigmccraig08:04:29

yeah, that looks like it

malcolmsparks08:04:45

the multipart upload code is a state machine, so it's pretty easy for fix once I find which state is wonky

mccraigmccraig08:04:08

i have a file that reliably causes it (given my browser etc) if that would be useful

malcolmsparks08:04:13

having a repeatable test with data is very handy, in the past I've had a couple of reports that haven't been repeatable

malcolmsparks08:04:28

yes - it would - very much - if you don't mind sending it to me