Re: MinGW vs. c-api.test

[Eli Zaretskii]

> From: address@hidden (Ludovic Courtès)
> Cc: guile-devel <address@hidden>
> Date: Tue, 10 Jun 2014 13:44:17 +0200
> 
> Eli, I noticed there are many other occurrences of /dev/null in the test
> suite.  Do they all need to be patched to use NUL, or is /dev/null
> somehow interpreted correctly in some contexts?

When we send /dev/null to some Guile primitive, which calls open or
fopen internally, gnulib replaces that by nul, so it transparently
works.  So only strings that are passed verbatim to the system shell
need to be fixed.