On Thu, Oct 31, 2019 at 11:26:30AM +0100, Johannes Schindelin wrote:

quoted

Another more far-fetched one: IIRC our stdio wrappers on Windows do some
magic to convert ANSI color codes into actual terminal codes. Could that
be a problem here? I think we'd kill off any color codes in the actual
message due to the control-code replacement. In theory the prefix could
have them. I don't think any code does now, but the PUSH_COLOR_ERROR
stuff in builtin/push.c is getting close. I wouldn't be surprised for
that to eventually get folded into error().

A valid concern!

As per eac14f8909d (Win32: Thread-safe windows console output,
2012-01-14), `write()` _implicitly_ gets the ANSI emulation of
`winansi.c`.

So I think we're good.

Thanks for thinking of all kinds of angles,

Good. Thanks for humoring my wild speculation. :)

I remember being confused by this not too long ago, so I did a quick dig
in the archive. And indeed, there was a confusing comment which caught
me in 2016, and which you removed back then (via 3d0a83382f26). I even
reviewed the patch. Maybe this time I will remember the outcome!

-Peff