HP3000-L Archives

March 1999, Week 3

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Glenn Cole <[log in to unmask]>
Reply To:
Date:
Tue, 16 Mar 1999 14:59:00 -0800
Content-Type:
text/plain
Parts/Attachments:
text/plain (40 lines)
Item Subject: cc:Mail Text
'tar' appears to be wacked in the head, and the associated man pages appear
to be inaccurate.

I tarred and feath^h^h^h^h^hcompressed a bunch of self-describing files,
each with 390-byte records.

When extracting a sample file from the archive, I get the error:

        tar: write error on file "filename": System call error

but AFTER the file has been extracted!
(I think I know the cause of the error; see below.)

Now, the MPE notes on the man page for tar state:

       "File characteristics like file code, record size, and so forth
        are not preserved [by the conversion to byte stream files].
        When you extract files from the archive, they are written
        as byte stream files."

This does not appear to be correct.
"An attempt" appears to have been made to preserve the MPE info.

That is, the file is created with filecode SD, and with fixed-length
records.  The file even has the correct maximum number of user labels.
Unfortunately, no labels are present.  (This is what I suspect as
the cause of the "write error" above -- somehow, it tried to restore
the user labels, but then realized that it doesn't even know what
they are!)

I could probably live with that, but the record size is off by one --
the file is restored with 389-byte records, with the 390th byte being
placed on the next record by itself.

I guess I should have verified the process was reversible before purging
all the original data files. :/  Call this an "exercise for the writer."

--Glenn

ATOM RSS1 RSS2