But seriously, although echo is typically a shell built-in and distinctly faster than /usr/bin/echo, it’s still much slower than <<<, presumably because it still has to set up a pipe and an extra… shall we say pseudoprocess.
Comparing behaviours for feeding text into `true` (typically a shell built-in, so that process spawn times doesn’t drown the signal):
try() {
echo -e "\e[32;1m$1\e[m"
for (( run = 0; run < 5; run++ )); do
time (for (( i = 0; i < 1000; i++ )); do
$2
done)
done
}
a() { /usr/bin/echo .dump | true; }; try /usr/bin/echo a
b() { echo .dump | true; }; try echo b
c() { <<<.dump true; }; try '<<<' c
d() { true; }; try "no piping" d
• /usr/bin/echo: 750μs/845μs
• echo: 469μs/371μs
• <<<: 11μs/31μs
• No piping: 3μs/10μs
So… yeah, on a very slightly older or slower machine than mine, using <<< may save you more than half a millisecond. That’s a much bigger difference than I expected—I was expecting it to be well under 200μs, maybe under 100μs, though the more I think about it the more I realise my expectation may have been unreasonable.
[Just noting that the implementation here isn't equivalent for nerdsnipe-ery, not arguing for real attempts at optimisation of simple pipelines.]
¹ Always with older versions, only for large strings with newer versions.
python <<<'import subprocess; subprocess.run(["ls", "-la", "/proc/self/fd"])'
(I know <(…) is substituted in zsh with /proc/self/fd/… for a fd corresponding to a pipe:[…], and regular | piping makes 0 be a pipe:[…].)
I know some common configurations use concrete /tmp. Mine is tmpfs.
Thanks for the info! I like sharing these kinds of things because I have found them interesting and expect a few others will too, and people often add to them details I hadn’t known and like to know!
$ bash -c 'realpath /dev/stdin <<< small'
/proc/79631/fd/pipe:[5282632]
$ bash -c 'realpath /dev/stdin <<< $(printf "%65536s")'
/tmp/sh-thd.yZ2L6p (deleted)
In case you weren't aware, you can also force a "real" file with zsh process substitution by using =(…)¹. It can be useful when the tool you're using doesn't behave correctly with <(…), if it wishes to seek across it for example. Sadly, =(…) isn't supported in bash [yet?].
¹ In your case it still wouldn't hit a disk as it is in /tmp, but it at least becomes seekable.
try() {
echo -e "\e[32;1m$1\e[m"
for (( run = 0; run < 5; run++ )); do
time (for (( i = 0; i < 1000; i++ )); do
eval $1
done)
done
}
try "/usr/bin/echo .dump | true"
try "echo .dump | true"
try "<<<.dump true"
try "true" sqlite3 db.sqlite3 .dump
This uses two things that shell users are commonly unfamiliar with:
• <<<: as `> filename` redirects stdout to the named file and `< filename` pipes the named file into stdin, `<<< word` pipes the word into stdin. (The whitespace is optional, and word follows normal argument splitting so you can do things like `<<<"Hello, world!"`.)
• Ordering: although most commonly written at the end of commands, redirections can be placed at the start as well. So if you do things like `2>/dev/null >&2` to suppress all output¹, that bit can go at the start or the end. I’ve become increasingly partial to putting redirections at the start of the line, especially in cases where I’m appending to a file for a while, because a leading `>>filename` supports visual alignment better than a trailing.²
And since we’re talking about different ways: as given, this is feeding the SQL to sqlite3 via stdin, but you can also pass it on the command line, after the filename. But things like this in Git (aliases, diff textconv, probably more) work by taking the string and appending the filename to the end, so you need a way of reordering the arguments. The solution is an immediately-invoked function:
So when you diff mydb.sqlite, it runs `sh -c 'f() { sqlite3 "$@" .dump }; f mydb.sqlite'` or equivalent, which winds up executing `sqlite3 mydb.sqlite .dump`, as desired.I use this technique a number of times in my Git aliases, saving the bother of putting them in separate shell scripts somewhere where path management is a bother, at the cost of maintaining a one-liner with sometimes too many semicolons.³
—⁂—
¹ “Take stderr (2) and redirect it (>) to /dev/null, then take stdout (default/implicit, could also write 1 explicitly) and redirect it (>) to stderr (&2).” There are plenty of other ways of writing this!
² Lists can be a better solution for this specific case, allowing you to redirect to the file only once for a whole bunch of commands:
³ My longest is thirteen lines, though half of them barely count as lines. The line from my ~/.config/git/config, within [alias]: