Re: [PATCH] compat: add a getpass() compatibility function

From: Junio C Hamano <gitster@pobox.com>
Date: May 19, 2011, 16:30 UTC
Message-ID: <7vmxii8xgv.fsf@alter.siamese.dyndns.org>
In-Reply-To: <BANLkTinPHeSfZXRb7pqt7-XWkR5fH=wAjg@mail.gmail.com>

Erik Faye-Lund <kusmabite@gmail.com> writes:

> Do we really need two implementations? Wouldn't it be better to factor
> out the mingw-version to a separate source file, and then improve it?

Thanks for reminding.

Show 6 quoted lines

> Windows doesn't have /dev/tty, but the logic in this version handles
> that by using stdin/stderr instead. The signal-stuff has a comment
> that indicates it might not even be correct. tcgetattr/tcsetattr isn't
> supported on Windows, but it's not needed if we use getch (as the
> version in compat/mingw.c does). POSIX/curses getch respects the
> echo-setting, while Windows getch never echo.

I forgot to check how generic/reusable mingw version would look. As you said, it does feel a lot more attractive approach if it is trivial to add the "noecho" bit to it, tweak the tty-ness and deal with signals.

I however suspect the result would end up on the borderline to be called "trivial", rough guestimate being 3-4 times as large as the 14-line mingw version. But it certainly feels that it is worth a try.