Jari Vetoniemi

Hello, thank you for your response. Regarding graphemes vs codepoints, as long as the intent is clear there is no problem. Normalization is definitely something that should be considered. There...

Unfortunately while I found [issue in my implementation](https://github.com/Cloudef/zig-budoux/commit/dc250edcf3e2da4afb56a12e76747bf6acce67c8) regarding the history bookkeeping it was unrelated to this, here are how the maps are accessed from n-2.. ``` UW1: า =>...

Ah never mind, I just realized that python / javascript / java implementations always return the last chunk regardless of the score, [so that's my dumb mistake](https://github.com/Cloudef/zig-budoux/commit/83d81e60cd8b4d5ff54c423fbe4d56020d86e7ed) :sweat_smile: I think...

Are you giving it input through stdin?

Could this also improve a codegen being a builtin? `**` currently has scenarios where the output is not that nice.

Unless the constant 42 here is 0, zig currently likes to insert the number into the binary N amount of times, worse is that the array here is even never...

Emscripten produces node scripts that has the node hashbang on them. In practice they work as any normal binary.

https://github.com/mesonbuild/meson/blob/1682058decb5f16ce1b78121ee587c024e9cfa17/mesonbuild/compilers/c.py#L140 what currently happens is this. Even if you remove these checks it will later fail at linker stuff. Removing the "emscripten detection" all together makes it work though.

It looks like @jpakkane is the original author of the emscripten support. So before I do any PRs, I'd want to ask what would be the best approach to go...

@jpakkane I created a cross file with contents: ``` [properties] needs_exe_wrapper=false ``` But trying to build pixman gives: ``` Build started at 2020-03-04T04:03:23.998316 Main binary: /usr/bin/python Build Options: -Db_lto=true -Db_pie=true...