On Sun, Mar 05, 2017 at 08:19:42PM +0100, Timo Buhrmester wrote:
> > > In the error case
> > Fine by me (although I find the semicolon sufficient to communicate
> > that it's still referring to the error case).
> Ping?
Paul just fixed it.
Martin
> > In the error case
> Fine by me (although I find the semicolon sufficient to communicate
> that it's still referring to the error case).
Ping?
(forgot to cc the list)
> In the error case
Fine by me (although I find the semicolon sufficient to communicate
that it's still referring to the error case).
On Mon, Feb 20, 2017 at 03:17:54PM +0100, Timo Buhrmester wrote:
> On Mon, Feb 20, 2017 at 03:05:40PM +0100, Martin Husemann wrote:
> > You mean "is" ?
> That would give "[...] is resolvedname was not allocated by realpath()".
>
> I really did mean 'if' because IF resolvedname is caller-allocated,
On Mon, Feb 20, 2017 at 03:05:40PM +0100, Martin Husemann wrote:
> You mean "is" ?
That would give "[...] is resolvedname was not allocated by realpath()".
I really did mean 'if' because IF resolvedname is caller-allocated,
realpath(3) will store information about the offending pathname there.
On Mon, Feb 20, 2017 at 03:02:49PM +0100, Timo Buhrmester wrote:
> > If an error occurs, realpath() returns NULL,
> > and resolvedname was not allocated by realpath(),
> > it will contain the pathname which caused the problem.
> Had to read it thrice before it made sense. Maybe this is slightly be
> If an error occurs, realpath() returns NULL,
> and resolvedname was not allocated by realpath(),
> it will contain the pathname which caused the problem.
Had to read it thrice before it made sense. Maybe this is slightly better:
--- lib/libc/gen/realpath.3.orig
+++ lib/libc/gen/realpath.3
@@ -8
