[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Carry tape
Has anyone modified Read Carry Tape so that it doesn't ask so many
questions? We'd like to be able to write a carry tape containing a
hierarchy with lots of subdirectories, and then when we read it at the
destination we would like to be able to specify the new directory that
corresponds to the original top-level directory, rather than getting
queried (three times -- once to answer "O", a second time to allow the
directory to be created, and again to answer "A" for the remaining files
in the directory) for each directory.
For example, currently a tape containing a directory and its
subdirectory results in the following:
Command: Write Carry Tape (pathnames of files) OCCAM:>barmar>test-dir>**>*.*.newest
(keywords) :Tape Spec (a tape spec) "AQUINAS: cart, den=6250, reclen=27648"
To be dumped:
OCCAM:>barmar>test-dir>**>*.*.newest: 5 files
Is this right? (Y or N) Yes.
Tape mounted on drive CART.
Dumping OCCAM:>barmar>test-dir>login.shell.1 (8-bit bytes)
Dumping OCCAM:>barmar>test-dir>test-file.lisp.1 (8-bit bytes)
Dumping OCCAM:>barmar>test-dir>test-sub>foo.bar.1 (8-bit bytes)
Dumping OCCAM:>barmar>test-dir>test-sub>foo.baz.1 (8-bit bytes)
Dumping OCCAM:>barmar>test-dir>test-sub>foo.lisp.2 (8-bit bytes)
Dump complete.
Occam: Read Carry Tape
Type tape host or spec [default AQUINAS: cart, den=6250, reclen=8192]: AQUINAS: cart, den=6250, reclen=8192
Tape mounted on drive CART.
Carry dump made by Barmar.
Dump taken at 2/11/89 14:30:53.
Dumped on machine OCCAM.
Dumped: OCCAM:>barmar>test-dir>**>*.*.newest
Load OCCAM:>barmar>test-dir>login.shell.1 into G:/u2/barmar/login.shell? (Y, N, O, H, or A) Other place
New pathname for OCCAM:>barmar>test-dir>login.shell.1
[default G:/u2/barmar/login.shell]: PLUGH:/data/u9/barmar/test-dir/login.shell
No such directory. The directory specified does not exist.
For PLUGH:/data/u9/barmar/test-dir/
Shall I attempt to create the directory PLUGH:/data/u9/barmar/test-dir? (Y or N) Yes.
Created PLUGH:/data/u9/barmar/test-dir.
Loading OCCAM:>barmar>test-dir>login.shell.1 into PLUGH:/data/u9/barmar/test-dir/login.shell (8-bit bytes) ... Loaded.
Load OCCAM:>barmar>test-dir>test-file.lisp.1 into PLUGH:/data/u9/barmar/test-dir/test-file.lisp? (Y, N, O, H, or A) Accept this de
fault for this directory
Loading OCCAM:>barmar>test-dir>test-file.lisp.1 into PLUGH:/data/u9/barmar/test-dir/test-file.lisp (8-bit bytes) ... Loaded.
Load OCCAM:>barmar>test-dir>test-sub>foo.bar.1 into G:/u2/barmar/foo.bar? (Y, N, O, H, or A) Other place
New pathname for OCCAM:>barmar>test-dir>test-sub>foo.bar.1
[default G:/u2/barmar/foo.bar]: PLUGH:/data/u9/barmar/test-dir/test-sub/login.shell
No such directory. The directory specified does not exist.
For PLUGH:/data/u9/barmar/test-dir/test-sub/
Shall I attempt to create the directory PLUGH:/data/u9/barmar/test-dir/test-sub? (Y or N) Yes.
Created PLUGH:/data/u9/barmar/test-dir/test-sub.
Loading OCCAM:>barmar>test-dir>test-sub>foo.bar.1 into PLUGH:/data/u9/barmar/test-dir/test-sub/login.shell (8-bit bytes) ... Load
ed.
Load OCCAM:>barmar>test-dir>test-sub>foo.baz.1 into PLUGH:/data/u9/barmar/test-dir/test-sub/foo.baz? (Y, N, O, H, or A) Accept thi
s default for this directory
Loading OCCAM:>barmar>test-dir>test-sub>foo.baz.1 into PLUGH:/data/u9/barmar/test-dir/test-sub/foo.baz (8-bit bytes) ... Loaded.
Loading OCCAM:>barmar>test-dir>test-sub>foo.lisp.2 into PLUGH:/data/u9/barmar/test-dir/test-sub/foo.lisp (8-bit bytes) ... Loaded
.
End of data on tape.
I'd like to be able to answer one question about the translation of the
directory name, perhaps a single question about creating the new
top-level directory, and then not be bothered until it encounters a file
that wasn't in that hierarchy.
If anyone can think of an existing mechanism in Genera that allows this
kind of pathname translation, but uses a different tape format, that
would be good, too. We're trying to use this for our software
distribution tape (our software isn't organized using DEFSYSTEM,
although in the past we've used a DEFSYSTEM-based kludge for the tapes),
and it would be preferable if the customer didn't need to load something
just to be able to read the tape.
barmar
- Follow-Ups:
- Carry tape
- From: David Vinayak Wallace <Gumby@MCC.COM>
- Carry tape
- From: Kalman Reti <Reti@SCRC-STONY-BROOK.ARPA>
- Carry tape
- From: Stephen Robbins <Stever@SCRC-YUKON.ARPA>