GitLab

From IRC:

13:33 <+armadev> cohosh / irl: speaking of the bridgestrap dataset output, we 
                 currently have it at '1' or '0' right? would it be smart to 
                 include the error message in a 0?
13:34 < irl[m]> if the error message is safe to share
13:34 <+armadev> like, if there are a bunch of 0's, why are they 0. that's the 
                 first question we'll be asking ourselves.
13:34 <+armadev> yeah, i think many of the error messages are things like "no 
                 route to host" which sounds safe
13:34 <+armadev> maybe some parsing is needed for some. i'm not sure.
13:35 < irl[m]> if there was an error message then we could display that in relay 
                search
13:35 < irl[m]> we can always add extra fields on the end of each test line

More info:

13:38 <+cohosh> irl[m]: so would this need coordination with you before we 
                implement/deploy it or can we just go ahead?
13:39 < irl[m]> as long as you only add new fields to the existing 
                bridgestrap-test lines, you don't need to coordinate
13:40 < irl[m]> the fields are currently space separated so ideally there would be 
                some codes for different errors rather than a human-readable 
                string containing spaces
13:40 < irl[m]> something like destination-host-unreachable would be fine but 
                "destination host unreachable" would present issues if we wanted 
                to add new fields in the future
13:40 <+armadev> is there any way to group elements, like putting quotes around 
                 them?
13:41 <+armadev> or is our format disallowing any elements with spaces in them?
13:41 < irl[m]> the parser would have to be extended to handle quoted strings 
                rather than just string.split(" ")
13:42 < irl[m]> in dir-spec anything that has freeform text strings is usually on 
                its own line