Hi. I have a feature request.
I have a bunch of data being created on the i/o stream. I want to add
this data into an existing, materialized tar archive as a file.
Traditionally, I would do something like this:
% create_data > big_file.dat
% tar -rf existing_tar_archive.tar big_file.dat
% rm big_file.dat
Of course, the big_file.dat that I'm creating can be big. I'd rather not
use the above approach because there will exist a time where I'm storing
that data on my disk twice -- once to materialize the data, and once as
part of the tar archive.
I would rather be able to add it directly to the tar archive from the
stream:
% create_data | tar -rf existing_tar_archive.tar -
Of course, I would need to be able to name the file somehow. Pehaps an
argument for that.
% create_data | \
tar -r --stdin-file-name=big_file.dat -f existing_tar_archive.tar -
Did I miss something in the docs? Is this actually possible to do and I
just missed it?
If not, would this be a friendly addition to tar? Is this as easy as I
think, or are there technological limitations making this difficult?
Okay, thanks.
--Ryan
