Jeff Mogul <mogul@Shasta> (11/20/84)
The issue is that, in order for the VAX to read S-1 text files
and vice versa, text files must be stored using a different
representation than binary files. There is no reliable way to
determine whether a file should be "text" or "binary" when the
tape is written, and no field in the "tar" header for recording
this information even if the writer could reliably figure it out.
This is one of the things that make Unix hard to use in a heterogeneous
environment (lack of file type tags), but I think the way out of your
dilemma is to assume that files are text files.
How often, after all, are you going to exchange binary files between
the S-1 and a Vax? Tar is a fair standard for exchanging text
files, but if you try to move an 8-bit binary file from the Vax to
the S-1, you'll probably have to massage it anyway. It might be
better to convert binary files to text (e.g., a list of decimal numbers)
than to try to solve the problem in tar.
So, optimize for the dominant case (text files) and hope that
binary files stay rare.Craig Partridge <craig@loki> (11/26/84)
In fact you frequently end up grabbing binaries along with
text files on tar tapes. If you tar a source tree you are likely
to pick up stray binaries. I know, I have spent several hours sometimes
trying to successfully move a source tree from a 10-bit byte machine
to an 8 bit byte machine. Checksum errors keep cropping up because
I missed some binary file somewhere.
Craig Partridge
craig@bbn-loki
...!{decvax,ihnp4,wjh12}!bbncca!craig