univalue VS RapidJSON

nlohmann's json is ergonomic but is slow. I use this lib when I need speed.

Not a huge fan of it. But it does the job. I prefer either the lib I maintain: UniValue or the ubiquitous nlohmann.

Ok, updated. I added some examples to the readme. https://github.com/cculianu/univalue

Here's what I did: - I bought a theme that I liked, and used mstch as template engine (matched the theme's template format) - Imported rapidjson for parsing JSON, and used mstch for generating JSON (the JSON I generate is simple enough that I can do this) - Wrote an HTTP handler class for each page that handles the GET and POST requests, generating HTML or JSON depending on the request

And documentation in most cases is more user-friendly if you will use something like MkDocs(based on Markdown), example http://rapidjson.org/

Use RapidJSON https://rapidjson.org/

It seems not super widely used compared to other famous libraries like RapidJSON, nlohmann-json, or simdjson. But it seems the author is very active in developing this project which can mitigate this "lack of community" issue.

have a look at rapidjson for a a nice simple json framework https://rapidjson.org/ I use it and it's quite fast and as good as anything I used in python.

So I just cloned https://github.com/Tencent/rapidjson to see what is the current status. At compile time they check if the C++ compiler supports the "noexcept" keyword, and if so they define RAPIDJSON_NOEXCEPT to be "noexcept". Throughout the implementation, RAPIDJSON_NOEXCEPT is added to quite a few function declarations (about a hundred of them).

RapidJSON is also excellent, like fantastic when performance matters. Not sure there's a much faster JSON implementation anywhere, and I've written my own SAX-style parser as a fun, hobby project for C++.