This is a common error while using pipe over while loops. Consider this shell snippet:<\/p>\n
#!\/bin\/sh\r\n\r\ncat file.txt | while read line\r\ndo\r\n echo \"inside loop\"\r\n exit 1\r\ndone\r\n\r\necho \"outside loop\"\r\nexit 0\r\n<\/pre>\nYou’d expect the script to exit on the first line in file.txt<\/i>. However execute this script and you have:<\/p>\n
inside loop\r\noutside loop\r\n<\/pre>\nIt is as if the
exit 1<\/code> inside the loop is ignored. Another example:<\/p>\n#!\/bin\/sh\r\n\r\na=0\r\ncat file.txt | while read line\r\ndo\r\n echo \"inside loop\"\r\n a=1\r\ndone\r\n\r\necho \"outside loop\"\r\necho \"a=$a\"\r\n<\/pre>\nHere you’d expect the value of a<\/i> to be 1 at the end of the script. Instead, if you execute this you have:<\/p>\n
inside loop\r\noutside loop\r\na=0\r\n<\/pre>\nIt’s as if the variable a<\/i> isn’t even updated. In fact it is, though only inside the loop. So what is happening here?<\/p>\n
The pipe (
|<\/code>) you use to feed the loop creates a subshell. In fact this is really just another process. So theexit 1<\/code> ora=1<\/code> only apply to these piped<\/i> processes.<\/p>\nHow can you fix that?
\nIn the simple case presented above, you can simply use file redirection:<\/p>\nwhile read line\r\ndo\r\n ...\r\ndone < file.txt\r\n<\/pre>\nBut what if you really want to feed the loop with the output of another process. Like you would do with find<\/i> for instance.<\/p>\n
If you use bash you can use process substitution<\/a> as described here<\/a>. But you shouldn’t use bash for scripting anyway. For shell scripting you might be tempted to use a temporary file to store the process output:<\/p>\n
# Use a temporary file.\r\ntmp=$(mktemp)\r\nfind . > $tmp\r\nwhile read line\r\ndo\r\n ...\r\ndone < $tmp \r\nrm $tmp\r\n<\/pre>\nHowever this consumes disk space, and the loop only starts after the find<\/i> process exited. Another option would be to use a named fifo:<\/p>\n
fifo=$(mktemp -u)\r\nmkfifo $fifo\r\nfind . > $fifo &\r\n\r\nwhile read file\r\ndo\r\n ...\r\ndone < $fifo\r\nrm $fifo\r\n<\/pre>\nThis time you create a single file, yet no disk space is used (apart for the fifo inode itself). Also the find<\/i> command is a child process, so the loop reads find<\/i> output as it comes.<\/p>\n
Although the version above already works as it should, you may want to use an anonymous fifo. This way you only need to create a fifo file, although you can delete it immediatly. You can achieve this with a little help from our beloved file descriptor 3.<\/p>\n
fifo=$(mktemp -u)\r\n\r\n# Create fifo\r\nmkfifo $fifo\r\n\r\n# Create fd 3 and unlink fifo file.\r\nexec 3<> $fifo\r\nrm $fifo\r\n\r\n# Redirect find to fd 3.\r\nfind . >&3 &\r\n\r\n# Feed fd 3 to while loop.\r\nwhile read line\r\ndo\r\n ...\r\ndone <&3 # Close fd 3. exec 3>&-\r\n<\/pre>\n","protected":false},"excerpt":{"rendered":"This is a common error while using pipe over while loops. Consider this shell snippet: #!\/bin\/sh cat file.txt | while read line do echo “inside loop” exit 1 done echo “outside loop” exit 0 You’d expect the script to exit … Continue reading