Jeff Garzik

Here's a first draft implementation on branch 2017_modify https://github.com/jgarzik/univalue/tree/2017_modify Needs improvement, comments welcome etc. If merged, will there be downstream users?

This seems to fail the well known robustness principle, "be liberal in what you accept" https://en.wikipedia.org/wiki/Robustness_principle Is there a reason why non-minimal UTF8 input should trigger a failure?

Thanks for the additional research and info. The main question is whether the input is *invalid* or merely *not optimal*. The answer to that question should dictate univalue's behavior upon...

See PR #37

Good progress :)

Merged a modified version of your unrelated-to-doxygen changes as ceb1194137421afba01e8c1042bf75f3a2cdddc3

* Yes, `master` is the correct dev branch. * Travis failed, due to `nullptr` changes. * Please either delete the `nullptr` changes, or move them to a separate and more-complete...

Rough concept ACK - will review and test further this week, with an eye towards merging, and pushing out a new version, which is overdue

> Note also that univalue seems to be not yet built with C++11 (at least it looked like that when I last played around). Is that by design (in which...

@domob1812 Sounds good to me. The main difficulty I imagine will be crafting a test that proves the code compiles absent C++11 :)