stdweb usage about getrandom HOT 2 CLOSED

This is (probably) going to be addressed in #109

Depending on timing (and dev availability) getrandom 0.2 will come out before rand 0.8. In that case, if you're using getrandom or rand on wasm/stdweb, you will have to set a Custom RNG. The setup should seem familiar if you've ever used a Custom Allocator.

In practice, you would have the following Cargo.toml dependencies:

stdweb = "0.4.20"
getopts = "0.2"
rand = "0.8"
stdweb-getrandom = "0.1"
easy_reader = "0.5.0"
yew = { path = "../.." }

You would also need to have the following code in your crate/binary (to prevent linking issues):

use stdweb_getrandom as _;

We will document all of this as part of getrandom 0.2 and the transition doc or rand 0.8.

from getrandom.

Depending on timing (and dev availability) getrandom 0.2 will come out before rand 0.8

Probably, but I think it doesn't matter? Rand 0.8 will then depend on getrandom 0.2 and must document this, while 0.7 will continue to depend on 0.1.

If getrandom 0.1 gets a shim over 0.2 then the feature flags can opt-in to stdweb_getrandom etc. (only an issue if both stdweb and wasm-bindgen features are used); if not, then both will exist side-by-side.

Summary: no loss of functionality, just a different interface.

(I think there's no reason we can't close this?)

from getrandom.