>Kagami, I think your fix is basically the same as this abandoned changeset of mine from Bug 1447096. I don't think `colorama.init()` is required, both Clang and Rust just does what it should do without the colorama AnsiToWin32 wrapper. >but if I recall correctly the 'problem' (and reason why this wasn't landed) was that this doesn't work for the output from the Rust build. AFAICT this works with newer MSYS bits. 🚀 >That being said, I don't quite understand why you needed to upgrade your MSYS2 libs for this approach to work. Care to elaborate? It seems updating pacman also updates the core MSYS components ([including mingw32/mingw64/msys](https://www.msys2.org/docs/updating/)). Updating them fixed comment #7 for me, in other words, running msys-backed executables do not turn off the virtual terminal flag anymore. >In regards to adding a mach command to upgrade the MSYS2 libs, I don't think that's a good idea. Agreed. 👍
Bug 1744340 Comment 17 Edit History
Note: The actual edited comment in the bug view page will always show the original commenter’s name and original timestamp.
>Kagami, I think your fix is basically the same as this abandoned changeset of mine from Bug 1447096. I don't think `colorama.init()` is required, both Clang and Rust just does what it should do without the colorama AnsiToWin32 wrapper. >but if I recall correctly the 'problem' (and reason why this wasn't landed) was that this doesn't work for the output from the Rust build. AFAICT this works with newer MSYS bits. 🚀 >That being said, I don't quite understand why you needed to upgrade your MSYS2 libs for this approach to work. Care to elaborate? It seems updating via pacman also updates the core MSYS components ([including mingw32/mingw64/msys](https://www.msys2.org/docs/updating/)). Updating them fixed comment #7 for me, in other words, running msys-backed executables do not turn off the virtual terminal flag anymore. >In regards to adding a mach command to upgrade the MSYS2 libs, I don't think that's a good idea. Agreed. 👍
>Kagami, I think your fix is basically the same as this abandoned changeset of mine from Bug 1447096. ~~I don't think `colorama.init()` is required, both Clang and Rust just does what it should do without the colorama AnsiToWin32 wrapper.~~ Oh, I retried and it seems it's actually needed. >but if I recall correctly the 'problem' (and reason why this wasn't landed) was that this doesn't work for the output from the Rust build. AFAICT this works with newer MSYS bits. 🚀 >That being said, I don't quite understand why you needed to upgrade your MSYS2 libs for this approach to work. Care to elaborate? It seems updating via pacman also updates the core MSYS components ([including mingw32/mingw64/msys](https://www.msys2.org/docs/updating/)). Updating them fixed comment #7 for me, in other words, running msys-backed executables do not turn off the virtual terminal flag anymore. >In regards to adding a mach command to upgrade the MSYS2 libs, I don't think that's a good idea. Agreed. 👍
>Kagami, I think your fix is basically the same as this abandoned changeset of mine from Bug 1447096. I don't think `colorama.init()` is required, both Clang and Rust just does what it should do without the colorama AnsiToWin32 wrapper. ~~Oh, I retried and it seems it's actually needed.~~ (No, I was just confused by the initial `.idl` build which is not done by clang and is never colored. It's not needed on MSYS2 shell nor on mach-outside-mozillabuild with PowerShell.) >but if I recall correctly the 'problem' (and reason why this wasn't landed) was that this doesn't work for the output from the Rust build. AFAICT this works with newer MSYS bits. 🚀 >That being said, I don't quite understand why you needed to upgrade your MSYS2 libs for this approach to work. Care to elaborate? It seems updating via pacman also updates the core MSYS components ([including mingw32/mingw64/msys](https://www.msys2.org/docs/updating/)). Updating them fixed comment #7 for me, in other words, running msys-backed executables do not turn off the virtual terminal flag anymore. >In regards to adding a mach command to upgrade the MSYS2 libs, I don't think that's a good idea. Agreed. 👍
>Kagami, I think your fix is basically the same as this abandoned changeset of mine from Bug 1447096. I don't think `colorama.init()` is required, both Clang and Rust just does what it should do without the colorama AnsiToWin32 wrapper. ~~Oh, I retried and it seems it's actually needed.~~ (No, I was just confused by the initial `.idl` build which is not done by clang and is never colored. It's not needed on MSYS2 shell nor on mach-outside-mozillabuild with PowerShell, both on the Windows 11 default Windows Terminal.) >but if I recall correctly the 'problem' (and reason why this wasn't landed) was that this doesn't work for the output from the Rust build. AFAICT this works with newer MSYS bits. 🚀 >That being said, I don't quite understand why you needed to upgrade your MSYS2 libs for this approach to work. Care to elaborate? It seems updating via pacman also updates the core MSYS components ([including mingw32/mingw64/msys](https://www.msys2.org/docs/updating/)). Updating them fixed comment #7 for me, in other words, running msys-backed executables do not turn off the virtual terminal flag anymore. >In regards to adding a mach command to upgrade the MSYS2 libs, I don't think that's a good idea. Agreed. 👍