34.3. Bash, version 4

Chet Ramey announced Version 4 of Bash on the 20th of February, 2009. This release has a number of significant new features, as well as some important bugfixes.

Among the new goodies:

• Associative arrays.

  An associative array can be thought of as a set of two linked arrays -- one holding the data, and the other the keys that index the individual elements of the data array.

  Example 34-5. A simple address database

  #!/bin/bash4 # fetch_address.sh declare -A address # -A option declares associative array. address[Charles]="414 W. 10th Ave., Baltimore, MD 21236" address[John]="202 E. 3rd St., New York, NY 10009" address[Wilma]="1854 Vermont Ave, Los Angeles, CA 90023" echo "Charles's address is ${address[Charles]}." # Charles's address is 414 W. 10th Ave., Baltimore, MD 21236. echo "Wilma's address is ${address[Wilma]}." # Wilma's address is 1854 Vermont Ave, Los Angeles, CA 90023. echo "John's address is ${address[John]}." # John's address is 202 E. 3rd St., New York, NY 10009.

  Example 34-6. A somewhat more elaborate address database

  #!/bin/bash4 # fetch_address-2.sh # A more elaborate version of fetch_address.sh. SUCCESS=0 E_DB=99 # Error code for missing entry. declare -A address # -A option declares associative array. store_address () { address[$1]="$2" return $? } fetch_address () { if [[ -z "${address[$1]}" ]] then echo "$1's address is not in database." return $E_DB fi echo "$1's address is ${address[$1]}." return $? } store_address "Charles Jones" "414 W. 10th Ave., Baltimore, MD 21236" store_address "John Smith" "202 E. 3rd St., New York, NY 10009" store_address "Wilma Wilson" "1854 Vermont Ave, Los Angeles, CA 90023" # Exercise: # Rewrite the above store_address calls to read data from a file, #+ then assign field 1 to name, field 2 to address in the array. # Each line in the file would have a format corresponding to the above. # Use a while-read loop to read from file, sed or awk to parse the fields. fetch_address "Charles Jones" # Charles Jones's address is 414 W. 10th Ave., Baltimore, MD 21236. fetch_address "Wilma Wilson" # Wilma Wilson's address is 1854 Vermont Ave, Los Angeles, CA 90023. fetch_address "John Smith" # John Smith's address is 202 E. 3rd St., New York, NY 10009. fetch_address "Bozo Bozeman" # Bozo Bozeman's address is not in database. exit $? # In this case, exit code = 99, since that is function return.

• Enhancements to the case construct: the ;;& and ;& terminators.

  Example 34-7. Testing characters

  #!/bin/bash4 test_char () { case "$1" in [[:print:]] ) echo "$1 is a printable character.";;& # | # The ;;& terminator continues to the next pattern test. | [[:alnum:]] ) echo "$1 is an alpha/numeric character.";;& # v [[:alpha:]] ) echo "$1 is an alphabetic character.";;& # v [[:lower:]] ) echo "$1 is a lowercase alphabetic character.";;& [[:digit:]] ) echo "$1 is an numeric character.";& # | # The ;& terminator executes the next statement ... # | %%%@@@@@ ) echo "********************************";; # v # ^^^^^^^^ ... even with a dummy pattern. esac } echo test_char 3 # 3 is a printable character. # 3 is an alpha/numeric character. # 3 is an numeric character. # ******************************** echo test_char m # m is a printable character. # m is an alpha/numeric character. # m is an alphabetic character. # m is a lowercase alphabetic character. echo test_char / # / is a printable character. echo # The ;;& terminator can save complex if/then conditions. # The ;& is somewhat less useful.

• The new coproc builtin enables two parallel processes to communicate and interact. As Chet Ramey states in the Bash FAQ [1] , ver. 4.01:

      There is a new 'coproc' reserved word that specifies a coprocess:
      an asynchronous command run with two pipes connected to the creating
      shell. Coprocs can be named. The input and output file descriptors
      and the PID of the coprocess are available to the calling shell in
      variables with coproc-specific names.
        

  Coprocesses use file descriptors. File descriptors enable processes and pipes to communicate.

  #!/bin/bash4 # A coprocess communicates with a while-read loop. coproc cat $0 while read -u ${COPROC[0]} line # ${COPROC[0]} is the do #+ file descriptor of the coprocess. echo "$line" | sed -e 's/line/NOT-ORIGINAL-TEXT/' done kill $COPROC_PID # No longer need the coprocess, #+ so kill its PID.

  But, be careful!

  #!/bin/bash4 echo; echo a=aaa b=bbb c=ccc coproc echo "one two three" while read -u ${COPROC[0]} a b c; # Note that this loop do #+ runs in a subshell. echo "Inside while-read loop: "; echo "a = $a"; echo "b = $b"; echo "c = $c" echo "coproc file descriptor: ${COPROC[0]}" done # a = one # b = two # c = three # So far, so good, but ... echo "-----------------" echo "Outside while-read loop: " echo "a = $a" # a = echo "b = $b" # b = echo "c = $c" # c = echo "coproc file descriptor: ${COPROC[0]}" echo # The coproc is still running, but ... #+ it still doesn't enable the parent process #+ to "inherit" variables from the child process, the while-read loop. # Compare this to the "badread.sh" script.

  The coprocess is asynchronous, and this might cause a problem. It may terminate before another process has finished communicating with it. #!/bin/bash4 coproc cpname { for i in {0..10}; do echo "index = $i"; done; } # ^^^^^^ This is a *named* coprocess. read -u ${cpname[0]} echo $REPLY # index = 0 echo ${COPROC[0]} #+ No output ... the coprocess timed out # after the first loop iteration.

• The new mapfile builtin makes it possible to load an array with the contents of a text file without using a loop or command substitution.

  #!/bin/bash4 mapfile Arr1 < $0 # Same result as Arr1=( $(cat $0) ) echo "${Arr1[@]}" # Copies this entire script out to stdout. echo "--"; echo # But, not the same as read -a !!! read -a Arr2 < $0 echo "${Arr2[@]}" # Reads only first line of script into the array. exit

• The read builtin got a minor facelift. The -t timeout option now accepts (decimal) fractional values [2] and the -i option permits preloading the edit buffer. [3] Unfortunately, these enhancements are still a work in progress and not (yet) usable in scripts.

• Parameter substitution gets case-modification operators.

  #!/bin/bash4 var=veryMixedUpVariable echo ${var} # veryMixedUpVariable echo ${var^} # VeryMixedUpVariable # * First char --> uppercase. echo ${var^^} # VERYMIXEDUPVARIABLE # ** All chars --> uppercase. echo ${var,} # veryMixedUpVariable # * First char --> lowercase. echo ${var,,} # verymixedupvariable # ** All chars --> lowercase.

• The declare builtin now accepts the -l lowercase and -c capitalize options.

  #!/bin/bash4 declare -l var1 # Will change to lowercase var1=MixedCaseVARIABLE echo "$var1" # mixedcasevariable # Same effect as echo $var1 | tr A-Z a-z declare -c var2 # Changes only initial char to uppercase. var2=originally_lowercase echo "$var2" # Originally_lowercase # NOT the same effect as echo $var2 | tr a-z A-Z

• Brace expansion has more options.

  Increment/decrement, specified in the final term within braces.

  #!/bin/bash4 echo {40..60..2} # 40 42 44 46 48 50 52 54 56 58 60 # All the even numbers, between 40 and 60. echo {60..40..2} # 60 58 56 54 52 50 48 46 44 42 40 # All the even numbers, between 40 and 60, counting backwards. # In effect, a decrement. echo {60..40..-2} # The same output. The minus sign is not necessary. # But, what about letters and symbols? echo {X..d} # X Y Z [ ] ^ _ ` a b c d echo {X..d..2} # X Z ^ ` b d # It seems to work for upper/lowercase letters, #+ but the increment is a bit inconsistent on symbols.

  Zero-padding, specified in the first term within braces, prefixes each term in the output with the same number of zeroes.

  bash4$ echo {010..15} 010 011 012 013 014 015 bash4$ echo {000..10} 000 001 002 003 004 005 006 007 008 009 010

• Substring extraction on positional parameters now starts with $0 as the zero-index. (This corrects an inconsistency in the treatment of positional parameters.)

  #!/bin/bash4 # show-params.bash4 # Invoke this scripts with at least one positional parameter. E_BADPARAMS=99 if [ -z "$1" ] then echo "Usage $0 param1 ..." exit $E_BADPARAMS fi echo ${@:0} # bash3 show-params.bash4 one two three # one two three # bash4 show-params.bash4 one two three # show-params.bash4 one two three # $0 $1 $2 $3

• The new ** globbing operator matches filenames and directories recursively.

  #!/bin/bash4 # filelist.bash4 shopt -s globstar # Must enable globstar, otherwise ** doesn't work. # The globstar shell option is new to version 4 of Bash. echo "Using *"; echo for filename in * do echo "$filename" done # Lists only files in current directory ($PWD). echo; echo "--------------"; echo echo "Using **" for filename in ** do echo "$filename" done # Lists complete file tree, recursively. exit Using * allmyfiles filelist.bash4 -------------- Using ** allmyfiles allmyfiles/file.index.txt allmyfiles/my_music allmyfiles/my_music/me-singing-60s-folksongs.ogg allmyfiles/my_music/me-singing-opera.ogg allmyfiles/my_music/piano-lesson.1.ogg allmyfiles/my_pictures allmyfiles/my_pictures/at-beach-with-Jade.png allmyfiles/my_pictures/picnic-with-Melissa.png filelist.bash4

• The new $BASHPID internal variable.

• There is a new builtin error-handling function named command_not_found_handle.

  #!/bin/bash4 command_not_found_handle () { # Accepts implicit parameters. echo "The following command is not valid: \""$1\""" echo "With the following argument(s): \""$2\"" \""$3\""" # $4, $5 ... } # $1, $2, etc. are not explicitly passed to the function. bad_command arg1 arg2 # The following command is not valid: "bad_command" # With the following argument(s): "arg1" "arg2"