non-obvious formatting of floating point numbers

[Roman-Koshelev]

1. If the value representation type is not specified, then it is not g. A mandatory zero is added after the period. The number does not match the shortest representation.
2. Fixed point has a fixed precision of 6 digits. Only significant digits are expected (it doesn't matter if there are more than 6 or less)
   https://godbolt.org/z/EEEedePaf

Somewhere it was written about compatibility with std. Will this be fixed?

[vitaut]

std::to_chars is a completely different API and we don't provide any compatibility guarantees with it.

[Roman-Koshelev]

But std::format is built on top of it and is compatible. fmt::format won't be fully compatible with std ?
https://www.open-std.org/jtc1/sc22/wg21/docs/papers/2019/p0645r10.html

[vitaut]

std::format is specified in terms of std::to_chars but for your examples to make sense you need to correctly form invocations to the latter. It doesn't make a lot of sense to talk about abstract compatibility of the two APIs. fmt::format is compatible with std::format on numeric types with one exception related to precision already tracked in #3649.