[Jim wrote me his message was rejected by the mailing list manager,
as he is not subscribed to it. Let me forward it in his name.]
From: Jim Meyering
Date: 30 Dec 1999 23:59:17 +0100
Message-ID:
François Pinard writes:
| "Paul W. Abrahams" écrit:
|
| > Shows that there's more than one way to skin a cat <groan>.
|
| I've read the following, somewhere. An old Unixer was chatting with an
| adolescent about their working habits. The youth was explaining how it
| could not imagine working without the mouse, and qualified himself as
| a mouse-type. The Unixer replied that himself was more of a cat-type! :-)
|
| > - cp copies a set of files from one place to another but does not in
| > any way modify them in transit.
|
| For the record, GNU `cp' did not always copy a file without modification.
| There was a time it was automatically sparsifying files (that is, replacing
| long sequences of zeroes by "holes"). According to current documentation,
| it does not do that anymore, unless the original file is already sparse,
| or `--sparse=always' is used. I would guess that sparsification is never
| done when the output is a device. (Jim, if easy, would you confirm?)
Hi, François!
If --sparse=auto is in effect (the default), then holes are introduced
IFF the input file is determined to be sparse and the output file (which
has been opened, and possibly created by the time this test is performed)
is a `regular' file.
If --sparse=always is specified, then GNU cp makes holes
in the output file, regardless of its type.
Happy holidays!
Jim
--
François Pinard http://www.iro.umontreal.ca/~pinard
--
To unsubscribe send e-mail to suse-linux-e-unsubscribe@suse.com
For additional commands send e-mail to suse-linux-e-help@suse.com
Also check the FAQ at http://www.suse.com/Support/Doku/FAQ/