[Bug 268479] lib/libc/stdlib/getenv.c may have a problem with putenv()

https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=268479

--- Comment #3 from Dennis Clarke <dclarke@blastwave.org> ---
(In reply to Mark Millard from comment #2)

I must agree fully. There does not seem to be a reason to disallow
a name=value pair to exist where the value is nothing at all. I see
the words "the value shall be composed of characters from the portable
character set" but no where does the POSIX spec claim that we *must*
have some characters after the equal '=' to form a "value". So I guess
putenv() and friends are doing what they claim to do. Correctly.

That leads to the other problem wherein env var names such as UNAME_s
can override the uname(3) struct members within FreeBSD. If we have an
env var name=value such as "UNAME_r=" then how shall a process ever
determine the uname(3) release data? It seems that I need to write my
own hackary to walk through extern char **environ and mangle it myself?

-- 
You are receiving this mail because:
You are the assignee for the bug.