Bash – How to handle raw binary data in a bash pipe

bash can't hold binary data in its variables. It's already bad enough to process text with shell loops, it would be even worse for processing binary data. The shell is the tool to run other tools.

Also note that the read buit-in command reads characters, not bytes.

Also, dd does one read system call, so a dd bs=77 count=1 won't necessarily read 77 bytes especially if stdin is a pipe (the GNU implementation of dd has iflag=fullblock for that).

Here, you want to use a data processing programming language like perl:

In perl:

perl -ne 'BEGIN{$/=\77}
  print "Do something with the 77 byte long <$_> record\n"'

With GNU awk:

LC_ALL=C awk -vRS='.{,77}' '{print "the record is in <" RT ">"}'

If you want to use a shell, your best option is probably zsh which is the only one that can store binary data in its variables:

while LC_ALL=C IFS= read -ru0 -k77 record; do
  print -r -- "you may only call builtins with $record
    anyway since you can't pass NUL bytes in arguments
    to an external command"
done

If all you want to do is pass each chunk as stdin to a new invocation of some command, then you can use GNU split and its --filter option:

split -b 77 --filter='some command'

--filter starts a new shell to evaluate some command for each chunk. Unless your sh does the optimisation already by itself, you can do:

split -b 77 --filter='exec some command'

To save a fork.

Using dd, you could parse its stderr output to find out the end of input. You'd need the GNU specific iflag=fullblock as well:

while
  {
    report=$({
      LC_ALL=C dd bs=77 iflag=fullblock count=1 2>&3 |
        some command >&4 3>&- 4>&-
    } 3>&1)
  } 4>&1
  [ "${report%%+*}" -eq 1 ]
do
  : nothing
done

If the input size is multiple of 77 though, some command will be run an extra time with an empty input.

First, the use of yet another cat doesn't really make much difference, and you shouldn't bother about it.

Second, the commands that make up a pipeline are executed in separate processes anyway, no matter if they're external commands or built-ins:

$ a=0
$ a=1 | a=2 | a=3
$ echo $a
0

As to your exact problem, it's not possible to simply connect 'stdin' to 'stdout'; even if a shell had some nop builtin which would collapse when used in a pipeline (eg | nop | -> |), the shell has no way to know in advance, at the time it sets up the pipeline, that your "switchboard" will switch to nop instead of awk or sort.

You can also achieve the same effect as you "switchboards" by building the pipeline yourself, and then calling eval to run it. Example:

$ cat test.sh
type=`file -zi "$1"`
case $type in
*application/gzip*)     mycat='zcat "$1"';;
*)                      mycat='cat "$1"';;
esac
case $type in
*charset=utf-16le*)     mycat="$mycat | iconv -f utf16le";;
esac
# highlight comments in blue
esc=`printf '\033'`;
mycat="$mycat | sed 's/^#.*/$esc[34m&$esc[m/'"
echo >&2 "$mycat"    # show the built pipeline
eval "$mycat"   # ... and run it
$ iconv -t utf16 test.sh > test16.sh; gzip test16.sh
$ sh test.sh test16.sh.gz

That's a bit off-topic, but on linux there is a faster way to copy the stdin to stdout (if any of them is a pipe) -- the splice(2) syscall, which doesn't involve moving the data to and from the userland:

$ cat splice_cat.c
#define _GNU_SOURCE
#include <fcntl.h>
#include <stdlib.h>
#include <err.h>

int main(int ac, char **av){
    ssize_t r;
    size_t block = ac > 1 ? strtoul(av[1], 0, 0) : 0x20000;
    for(;;)
            if((r = splice(0, NULL, 1, NULL, block, 0)) < 1){
                    if(r < 0) err(1, "splice");
                    return 0;
            }
}
$ cc -Wall splice_cat.c -o splice_cat
$ dd if=/dev/zero bs=1M count=100 status=none | (time cat >/dev/null)
real    0m0.153s
user    0m0.012s
sys     0m0.056s
$ dd if=/dev/zero bs=1M count=100 status=none | (time ./splice_cat >/dev/null)
real    0m0.100s
user    0m0.004s
sys     0m0.020s

However (afaik), that's not used by either the shell or cat, dd, etc.