Re: [dev] wrap: minimalist archiving tool

Hey,

On 8 August 2010 16:02, Szabolcs Nagy <nsz_AT_port70.net> wrote:
> there was a bug in the encoder, it could corrupt your data

It seemed to work during testing, according to cmp, but thanks for the warning.

On 9 August 2010 03:33, Anders Andersson <pipatron_AT_gmail.com> wrote:
> After reading the source it seems a little less general purpose than I
> had hoped for though, but I can see the use of this anyway, for
> archiving things where you don't need to store ownership information,
> permission flags, named pipes, devices etc. More of a source-code
> archiver than a backup archiver.

I think this was the general direction I was going in. I rarely use
archives for anything but source distribution, where tar is overkill.

On 9 August 2010 04:54, David Tweed <david.tweed_AT_gmail.com> wrote:
> The one thing that leaps out at me is that there's no checksumming of
> either the individual files or the whole the archive file performed,
> so if you give it a damaged archive you won't be able to tell or
> isolate the damaged files.

I figure the archive doesn't need to be able to checksum. Many
compression formats and transmission protocols already checksum -
reading a gzipped tarball from the web can result in up to four
checksums. So if you're worried about integrity, just compress. If
you're forced to use raw wraps and you're worried about storage, not
transmission, you can always include checksums as Szabolcs said.

On 9 August 2010 03:33, Anders Andersson <pipatron_AT_gmail.com> wrote:
> A reminder here: The common use case for an archiver is to archive
> quite a lot of files, so reading a list of files from stdin is
> probably necessary for practical use unless you assume that you can
> pass thousands of files as arguments.

True; fixed in the attached patch.

On 9 August 2010 08:18, Szabolcs Nagy <nsz_AT_port70.net> wrote:
> (also it is not checked if a path is truncated or if the same path
> name appears multiple times)

Regarding truncation, true; fixed in the attached patch. As for
repeated paths, "uniq" should suffice.

> (when used for moving files between platforms warnings about path
> name issues might be useful)

Probably true, though I only use Posix systems so could neither use it
nor test it.

Thanks,
cls