Carlos E. R. wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
The Sunday 2008-04-06 at 12:39 -0400, Sam Clemens wrote:
It is not as simple as that.
If a different set of options corrected the problem they found, they would have said so and write up the correct encantation. But no, they said it was a bug.
It has to be some bug, then, probably corrected since, or maybe not. I can't say unless they say what the bug was.
If it was an actual bug, then tar would be completely broken, and not fit for use under any circumstance.
If he says "bug" then I have to trust him and think it was a bug. A bug doesn't have to affect all uses of a program.
But what would be broken that ONLY hurts that particular command pipeline? writing to stdout instead of a regular file??? That's merely a change of a file descriptor, which would mean that all of tar is broken. Same goes for reading from stdin instead of a regular file. That's one of the big breakthroughs of the Unix concept, totally unheard of at the time...to use the same semantics for writing to a disk file or to print on the screen or to feed to another program... or to send to some other thing... Either tar reads and writes acceptable data, or it doesn't.
I know the current kernel we use has bugs... I know some of them. Is the kernel unfit for use under any circumstance, then?
Tar is such a basic resource that it's either trustworthy, or it is not.
The editor was just lazy... or, giving him the benefit of the doubt, didn't realize that these flags had been created and were now available.
I'll grant him laziness, for not specifying the bug.
-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org