Sometimes, a command is stalled attempting to write to a FIFO/pipe that no other process is currently reading from, but typing Ctrl-Z to put the process to the background by sending it the SIGTSTP signal (i.e. the *suspend* non-printing control character ^Z) does work. Example: $ mkfifo p $ ll p prw-rw-r-- 1 me me 0 Jul 30 16:27 p| $ echo "Go!" >p # Here & has been forgotten after the >p redirection (stalled) [Ctrl-Z] ^Z # Pressing Ctrl-Z just prints "^Z" on the terminal, nothing else happens [Ctrl-D] # Attempting to send the EOF character (nothing) # Doesn't print "^D" and does nothing and this is the behavior even though SIGTSTP (or susp) is reported to be attached to ^Z: $ stty -a speed 38400 baud; rows 24; columns 91; line = 0; intr = ^C; quit = ^\; erase = ^?; kill = ^U; eof = ^D; eol = ; eol2 = ; swtch = ; start = ^Q; stop = ^S; susp = ^Z; rprnt = ^R; werase = ^W; lnext = ^V; discard = ^O; min = 1; time = 0; -parenb -parodd -cmspar cs8 -hupcl -cstopb cread -clocal -crtscts -ignbrk -brkint -ignpar -parmrk -inpck -istrip -inlcr -igncr icrnl -ixon -ixoff -iuclc -ixany -imaxbel iutf8 opost -olcuc -ocrnl onlcr -onocr -onlret -ofill -ofdel nl0 cr0 tab0 bs0 vt0 ff0 isig icanon iexten echo echoe echok -echonl -noflsh -xcase -tostop -echoprt echoctl echoke -flusho -extproc How comes that in this case the terminal does not catch and transmit to the process the SIGTSTP signal generated by the Ctrl-Z? Is it because redirecting with > to a FIFO/pipe puts the terminal in raw mode (cf. The TTY demystified )? But if so, why does "^Z" get printed on the screen, and why Ctrl-D (EOF) doesn't print anything and yet doesn't have any effect either? Incidentally, is there an alternative way to send the stalled process to the background in such cases (instead of just terminate the process with Ctrl-C (i.e. ^C / SIGINT)?

Apologies if this is a basic question - I'm stuck trying to solve a larger problem, and it's come down to how a shell script is invoked - directly (shellScript.sh) or using sh shellScript.sh. Here's a model for the problem: When I execute on bash: cat <(echo 'Hello') I see the output Hello But when I use: sh -c "cat <(echo 'Hello')" I see errors: sh: -c: line 0: syntax error near unexpected token `(' sh: -c: line 0: `cat <(echo 'Hello')' I've tried escaping the <, ( and ) in various combinations, but I don't see the output anywhere. What am I missing here? My actual problem is that I'm passing a <() as an input argument to a python script within a shell script, and while it works fine when I invoke the shell script using just the name, if I use sh to invoke it, I get errors similar to what I've shown above. Thank you!

I have this command, which works fine for bash: sh <(curl -sSl https://appi.sh/launch) (launches [this interactive script](https://github.com/Appish/appi.sh/blob/master/launch.sh)) I'd like to get this 1-liner working for ksh, too. Currently, it errors with:

syntax error: (' unexpected

Running the script from ksh as sh launch.sh works fine, so I'm hoping there's a way to have a singular 1-liner that will work in both bash/ksh without exponentially increasing the length of the command.

I'm trying to compare the output of my fizzbuzz.js program: for (let i = 1; i <= 100; i++) { let out = ""; if (i % 3 == 0) out += "Fizz"; if (i % 5 == 0) out += "Buzz"; if (i % 3 && i % 5) out = i; console.log(out); } with the contents of a expected-output.txt file which contains: 1 2 Fizz 4 Buzz Fizz 7 8 Fizz Buzz 11 Fizz 13 14 [...] I can't get process substitution to work: diff -u expected-output.txt <(node fizzbuzz.js) seems to hang indefinitely. What could be causing this? My environment is macOS Mojave, bash 3.2.57, node v12.5.0. EDIT: this seems to be a macOS issue. Everything works fine on a Ubuntu 18.04 machine with newer bash 4.4.19

On occasion process substitution will not work as expected. Here is an example: Input: gcc <(echo 'int main(){return 0;}') Output: /dev/fd/63: file not recognized: Illegal seek collect2: error: ld returned 1 exit status Input: But it works as expected when used with a different command: grep main <(echo 'int main(){return 0;}') Output: int main(){return 0;} I have noticed similar failures with other commands (i.e. the command expecting the file from the process substitution can't use /dev/fd/63 or similar). This failure with gcc is just the most recent. Is there some general rule that I should be aware of to determine when process substitution will fail in this way and should not be used? I am using this BASH version on Ubuntu 12.04 (I've also seen this in arch and debian): GNU bash, version 4.3.11(1)-release (i686-pc-linux-gnu)

I have very complicated set of commands:

command | ... | ... | tee >(grep -c '[^3]$') >(grep -c '[^35]$') 1>/dev/null

I don't want to have a temporary file to save up output, as it is pretty huge. I tried doing >(grep -c '[^3]$' | read variable) and >(grep -c '[^3]$' | read variable2) but I guess bit doesn't work because of process substitution's sub-shell invocation. What can I do to pipe output directly to multiple variables? Is it even possible? Right now I have this workaround:

var=$(command ... | ... | ... | tee >(grep -c '[^3]$') >(grep -c '[^35]$') 1>/dev/null)
var1=$(tail -1 <<< $var)
var2=$(head -1 <<< $var)

but I think it is clumsy and doesn't look nice. I am aware I can grep into another file but I don't like it either. Thanks in advance.

In bash, I want to assign my current working directory to a variable. Using a subshell, I can do this. var=$(pwd) echo $var /home/user.name If I use process substitution like so: var=<(pwd) echo $var /dev/fd/63 I have understood that process substitution is primarily used when a program does not accept STDIN. It is unclear to me what a process substitution exactly does and why it assigns /dev/fd/63 to var.

I have some vagrant virtual machines. To log into them I issue the vagrant ssh command. I want to log into them using regular ssh command. The vagrant ssh-config outputs the suitable config file $ vagrant ssh-config Host default HostName 127.0.0.1 User vagrant Port 2201 UserKnownHostsFile /dev/null StrictHostKeyChecking no PasswordAuthentication no IdentityFile /home/cbliard/.vagrant.d/insecure_private_key IdentitiesOnly yes LogLevel FATAL When outputing this config in a file and using with ssh -F, everything works fine: $ vagrant ssh-config > /tmp/config $ ssh -F /tmp/config default => logged successfully When using process substitution operator <(cmd) to prevent the creation of the temporary config file, it fails: $ ssh -F <(vagrant ssh-config) default Can't open user config file /proc/self/fd/11: No such file or directory Same error happens when using <(cat /tmp/config) $ ssh -F <(cat /tmp/config) default Can't open user config file /proc/self/fd/11: No such file or directory I am using zsh and I observe the same behavior with bash. What am I doing wrong here?

I have both BSD (default) and GNU Awk implementations installed on a Mac (the latter is envoked using gawk instead of just awk), and I have a shell script with contents like this:

none
awk -v maxLen=72 '
    .
    .
    .
' "$1" > "$2"

Now the tricky part. To work correctly, the script must use GNU Awk. Which means I have at least two options: 1)To modify the first line of the script so that there will be gawk instead of awk, 2) To make a system-wide alias for GNU Awk so that it will be envoked using awk instead gawk. But I would prefer to avoid either option, and instead to execute the script using a wrapper. That is, instead of

none
./script.sh input.txt output.txt

I would prefer something like

none
sed -e 's/awk/gawk/' script.sh

and then to pass the output to Zsh somehow. I was told this is possible using Zsh process substitution feature, but how?

If you have Pandoc installed, the following script will retrieve its HTML template, write it into a temporary file, make some adjustments, and then pass this file as a parameter:

template=$(mktemp default.html5.XXXXXX)
sed -z 's/.*\s*//' > $template
pandoc input.md -s --template $template -o output.htm
rm $template

It works, but I learn how to use Zsh effectively, and for this reason I want to change it in such a way that creating a temporary file manually won't be required. As far as I know, Zsh process substitution feature creates temporary files automatically. For example:

${template::==(pandoc -D html)}

But how (and whether) it is possible to re-write the original script using this technique?

According to this process substitution can be used to do a SIMPLE diff between local and remote machine. However, I want to use the -qr option in the diff command, i.e., it must be recursive and brief (reports only files that differ). If I try to do

`
diff -qr <(ls -R ~/Documents) <(ssh myusername@192.168.1.150 ls -R /pathinmyremote/My_remote_Documents)

` I get Files /dev/fd/63 and /dev/fd/62 differ which is obviously not what I am looking for. There are suggestions of using rsync with the dry-run option, but that does not include the brief option. The reason for -qr is that I want to know 1) which files are in my local machine and missing in my remote backup, and 2) which files that have been changed/updated in my local machine with respect to the ones in the backup remote. **Update** The suggestion to use rsync -n -ai --delete ~/Documents myusername@192.168.1.150:/pathinmyremote/My_remote_Documents results in:

which is also not right (due to the existing files/directories)