Eric Hawicz

I'm wondering this myself. I just tried building it, with the hopes of submitting a PR to update it to use a newer json-c, and ran into the problems talked...

Dynamically detecting overlap isn't really possible - just because you pick some subnet for docker to use that doesn't conflict with the host's _local_ interface doesn't mean that it isn't...

Copied from comments on 0cc1db6: That seems reasonable, except for the fact that we'll be changing the output somewhat. For instance, with that change the test_parse test now fails because...

Changing the json-c test cases won't help anyone that already depends on the existing behaviour. It sounds like this is a change we need to implement, but I'd like to...

I've been meaning to take a look at http://www.cs.indiana.edu/~dyb/pubs/FP-Printing-PLDI96.pdf to see if it is applicable here, but I just haven't had time.

It looks like the approach from that paper will do what's needed here. Or rather, the approach from a subsequent paper by Florian Loitsch titled "Printing floating-point numbers quickly and...

ew... kind of gross, but effective I suppose. Please rename the flag to JSON_C_TO_STRING_UNSIGNED, and add at least one test, and I'll merge it. However, this will create output that...

It's probably past time to bump the minimum version. Do you know of some system still stuck on cmake

unfortunately past time? or unfortunately you know of some system that can't use a newer cmake? If the latter, why not?

I think at least some (all?) of these warnings are actual problems, and rather than papering over them by casting to int, we need to take adjust the types used....