Will, while trying to figure out a parsing error in Rob's safecast gateway, I stumbled across the fact that the dates generated by safecast-air look different than the ones that come in from other devices.
Specifically, both the dates and the times seem to be lacking a leading zero. I am NOT SURE if this is required by any of the date parsers, but I think it might be good practice and helpful if you would change your %d's to %02d's, if you know what I mean.
"2017-3-5T0:5:38Z" is an example of what I see coming out of safecast-air. I've never seen a second that is other than 38 or 56 in my data, but you might want to verify that field also for %02d.
"2017-03-05T00:05:38Z" is an example of what I see coming out of pointcast and out of golang's date libraries.
Will, while trying to figure out a parsing error in Rob's safecast gateway, I stumbled across the fact that the dates generated by safecast-air look different than the ones that come in from other devices.
Specifically, both the dates and the times seem to be lacking a leading zero. I am NOT SURE if this is required by any of the date parsers, but I think it might be good practice and helpful if you would change your %d's to %02d's, if you know what I mean.
"2017-3-5T0:5:38Z" is an example of what I see coming out of safecast-air. I've never seen a second that is other than 38 or 56 in my data, but you might want to verify that field also for %02d.
"2017-03-05T00:05:38Z" is an example of what I see coming out of pointcast and out of golang's date libraries.
Thanks.