Re: [PATCH v2 0/3] Really fix the isatty() problem on Windows

Junio C Hamano [off-list ref] writes:

Johannes Schindelin [off-list ref] writes:
...

quoted

- fixed the confusing commit message by using Junio's suggested
  replacement

Sorry, but I didn't mean to "suggest replacement".  I was just
testing my understanding by attempt to rephrase the gist of it.
...
Your "use this opportunity to actually clean up" above suggests that
the answer is the latter, but if you took my "summary of my
understanding", it is likely that that fact is not captured in the
resulting log message.

So using the original log message in v1 and what you wrote in the
message I was responding to as references, let me try a real
"suggested" one as penance.

I need to ask one clarification on what you wrote before completing
that effort, though.

quoted

And incidentally, replacing the previous hack with the clean, new
solution, which specifies explicitly for the file descriptors 0, 1 and 2
whether we detected an MSYS2 pseudo-tty, whether we detected a real
Win32 Console, and whether we had to swap out a real Win32 Console for a
pipe to allow child processes to inherit it.

This has subject but not verb.  I parsed the above like so:

    Replacing the previous hack with the clean, new solution (which
    specifies explicitly for the file descriptors 0, 1 and 2
     - whether we detected an MSYS2 pseudo-tty, 
     - whether we detected a real Win32 Console, and 
     - whether we had to swap out a real Win32 Console for a
       pipe to allow child processes to inherit it
    )

So the entire thing is a noun phrase "replacing with a new patch",
and I take that as the subject of an unfinished sentence.  What did
that subject do?  Replacing with a new patch allows us to do "this
wonderful thing", but what "this wonderful thing" is not clear.

Subject: mingw: replace isatty() hack
From: Jeff Hostetler <redacted>

For over a year, Git for Windows has carried a patch that detects
the MSYS2 pseudo ttys used by Git for Windows' default Git Bash
(i.e. a terminal that is not backed by a Win32 Console), but it did
so by accessing internals of a previous MSVC runtime that is no
longer valid in newer versions.  

Clean up this mess by backporting a patch that was originally done
to compile Git with a recent VC++.  Replacing the previous hack with
the clean, new solution, which specifies explicitly for the file
descriptors 0, 1 and 2 whether we detected an MSYS2 pseudo-tty,
whether we detected a real Win32 Console, and whether we had to swap
out a real Win32 Console for a pipe to allow child processes to
inherit it, lets us do XXXXX.

As a side effect (which was the reason for the back-port), this patch
also fixes the previous misguided attempt to intercept isatty() so that
it handles character devices (such as /dev/null) as Git expects it.