If I want to check for the null string I would do
[ -z $mystr ]
but what if I want to check whether the variable has been defined at all? Or is there no distinction in Bash scripting?
If I want to check for the null string I would do
[ -z $mystr ]
but what if I want to check whether the variable has been defined at all? Or is there no distinction in Bash scripting?
I think the answer you are after is implied (if not stated) by Vinko's answer, though it is not spelled out simply. To distinguish whether VAR is set but empty or not set, you can use:
if [ -z "${VAR+xxx}" ]; then echo "VAR is not set at all"; fi
if [ -z "$VAR" ] && [ "${VAR+xxx}" = "xxx" ]; then echo "VAR is set but empty"; fi
You probably can combine the two tests on the second line into one with:
if [ -z "$VAR" -a "${VAR+xxx}" = "xxx" ]; then echo "VAR is set but empty"; fi
However, if you read the documentation for Autoconf, you'll find that they do not recommend combining terms with '-a
' and do recommend using separate simple tests combined with &&
. I've not encountered a system where there is a problem; that doesn't mean they didn't used to exist (but they are probably extremely rare these days, even if they weren't as rare in the distant past).
You can find the details of these, and other related shell parameter expansions, the test
or [
command and conditional expressions in the Bash manual.
I was recently asked by email about this answer with the question:
You use two tests, and I understand the second one well, but not the first one. More precisely I don't understand the need for variable expansion
if [ -z "${VAR+xxx}" ]; then echo "VAR is not set at all"; fi
Wouldn't this accomplish the same?
if [ -z "${VAR}" ]; then echo "VAR is not set at all"; fi
Fair question - the answer is 'No, your simpler alternative does not do the same thing'.
Suppose I write this before your test:
VAR=
Your test will say "VAR is not set at all", but mine will say (by implication because it echoes nothing) "VAR is set but its value might be empty". Try this script:
(
unset VAR
if [ -z "${VAR+xxx}" ]; then echo "JL:1 VAR is not set at all"; fi
if [ -z "${VAR}" ]; then echo "MP:1 VAR is not set at all"; fi
VAR=
if [ -z "${VAR+xxx}" ]; then echo "JL:2 VAR is not set at all"; fi
if [ -z "${VAR}" ]; then echo "MP:2 VAR is not set at all"; fi
)
The output is:
JL:1 VAR is not set at all MP:1 VAR is not set at all MP:2 VAR is not set at all
In the second pair of tests, the variable is set, but it is set to the empty value. This is the distinction that the ${VAR=value}
and ${VAR:=value}
notations make. Ditto for ${VAR-value}
and ${VAR:-value}
, and ${VAR+value}
and ${VAR:+value}
, and so on.
As Gili points out in his answer, if you run bash
with the set -o nounset
option, then the basic answer above fails with unbound variable
. It is easily remedied:
if [ -z "${VAR+xxx}" ]; then echo "VAR is not set at all"; fi
if [ -z "${VAR-}" ] && [ "${VAR+xxx}" = "xxx" ]; then echo "VAR is set but empty"; fi
Or you could cancel the set -o nounset
option with set +u
(set -u
being equivalent to set -o nounset
).
${+}
and ${-}
it is unclear, but familiarity with those constructs is essential if one is to be a competent user of the shell. –
Haze set -o nounset
enabled. –
Ary [ -v VAR ]
" approach with bash-s v4.2 and beyond. –
Fiesta xxx
can be any non-empty string. It does have to be three characters long. –
Dissection xxx
can be any non-empty string. It does not have to be three characters long (contradicting what was said in the prior comment). The accidental omission of 'not' makes that comment self-inconsistent. –
Dissection ~> if [ -z $FOO ]; then echo "EMPTY"; fi
EMPTY
~> FOO=""
~> if [ -z $FOO ]; then echo "EMPTY"; fi
EMPTY
~> FOO="a"
~> if [ -z $FOO ]; then echo "EMPTY"; fi
~>
-z works for undefined variables too. To distinguish between an undefined and a defined you'd use the things listed here or, with clearer explanations, here.
Cleanest way is using expansion like in these examples. To get all your options check the Parameter Expansion section of the manual.
Alternate word:
~$ unset FOO
~$ if test ${FOO+defined}; then echo "DEFINED"; fi
~$ FOO=""
~$ if test ${FOO+defined}; then echo "DEFINED"; fi
DEFINED
Default value:
~$ FOO=""
~$ if test "${FOO-default value}" ; then echo "UNDEFINED"; fi
~$ unset FOO
~$ if test "${FOO-default value}" ; then echo "UNDEFINED"; fi
UNDEFINED
Of course you'd use one of these differently, putting the value you want instead of 'default value' and using the expansion directly, if appropriate.
Advanced Bash scripting guide, 10.2. Parameter Substitution:
To base your program logic on whether the variable $mystr is defined or not, you can do the following:
isdefined=0
${mystr+ export isdefined=1}
Now, if isdefined=0 then the variable was undefined, and if isdefined=1 the variable was defined.
This way of checking variables is better than the previous answers, because it is more elegant, readable, and if your Bash shell was configured to error on the use of undefined variables (set -u)
, the script will terminate prematurely.
Other useful stuff:
To have a default value of 7 assigned to $mystr if it was undefined, and leave it intact otherwise:
mystr=${mystr- 7}
To print an error message and exit the function if the variable is undefined:
: ${mystr? not defined}
Beware here that I used ':' so as not to have the contents of $mystr executed as a command in case it is defined.
var= ; varname=var ; : ${!varname?not defined}
and this terminates: varname=var ; : ${!varname?not defined}
. But it is a good habit to use set -u
which does the same much easier. –
Autochthon A summary of tests.
[ -n "$var" ] && echo "var is set and not empty"
[ -z "$var" ] && echo "var is unset or empty"
[ "${var+x}" = "x" ] && echo "var is set" # may or may not be empty
[ -n "${var+x}" ] && echo "var is set" # may or may not be empty
[ -z "${var+x}" ] && echo "var is unset"
[ -z "${var-x}" ] && echo "var is set and empty"
${var+x}
it is not necessary except if variable names are allowed to have spaces in them. –
Lobell [
to get correct results. If var
is unset or empty, [ -n $var ]
reduces to [ -n ]
which has exit status 0, while [ -n "$var" ]
has the expected (and correct) exit status of 1. –
Blackamoor The explicit way to check for a variable being defined would be:
[ -v mystr ]
test
operator in the tail end of the section Bourne Shell Builtins. I don't know when it was added, but it is not in the Apple version of bash
(based on bash
3.2.51), so it is probably a 4.x feature. –
Dissection GNU bash, version 4.1.2(1)-release (x86_64-redhat-linux-gnu)
. The error is -bash: [: -v: unary operator expected
. Per a comment here, it requires at minimum bash 4.2 to work. –
Prader Test if a variable is set in bash when using "set -o nounset" contains a better answer (one that is more readable and works with set -o nounset
enabled). It works roughly like this:
if [ -n "${VAR-}" ]; then
echo "VAR is set and is not empty"
elif [ "${VAR+DEFINED_BUT_EMPTY}" = "DEFINED_BUT_EMPTY" ]; then
echo "VAR is set, but empty"
else
echo "VAR is not set"
fi
Another option: the "list array indices" expansion:
$ unset foo
$ foo=
$ echo ${!foo[*]}
0
$ foo=bar
$ echo ${!foo[*]}
0
$ foo=(bar baz)
$ echo ${!foo[*]}
0 1
The only time this expands to the empty string is when foo
is unset, so you can check it with the string conditional:
$ unset foo
$ [[ ${!foo[*]} ]]; echo $?
1
$ foo=
$ [[ ${!foo[*]} ]]; echo $?
0
$ foo=bar
$ [[ ${!foo[*]} ]]; echo $?
0
$ foo=(bar baz)
$ [[ ${!foo[*]} ]]; echo $?
0
should be available in any Bash version 3.0 or greater.
Not to shed this bike even further, but wanted to add
shopt -s -o nounset
is something you could add to the top of a script, which will error if variables aren't declared anywhere in the script.
The message you'd see is unbound variable, but as others mention, it won't catch an empty string or null value.
To make sure any individual value isn't empty, we can test a variable as it's expanded with ${mystr:?}
, also known as dollar sign expansion, which would error with parameter null or not set
.
The Bash Reference Manual is an authoritative source of information about Bash.
Here's an example of testing a variable to see if it exists:
if [ -z "$PS1" ]; then
echo This shell is not interactive
else
echo This shell is interactive
fi
(From section 6.3.2.)
Note that the whitespace after the open [
and before the ]
is not optional.
Tips for Vim users
I had a script that had several declarations as follows:
export VARIABLE_NAME="$SOME_OTHER_VARIABLE/path-part"
But I wanted them to defer to any existing values. So I rewrote them to look like this:
if [ -z "$VARIABLE_NAME" ]; then
export VARIABLE_NAME="$SOME_OTHER_VARIABLE/path-part"
fi
I was able to automate this in Vim using a quick regex:
s/\vexport ([A-Z_]+)\=("[^"]+")\n/if [ -z "$\1" ]; then\r export \1=\2\rfi\r/gc
This can be applied by selecting the relevant lines visually, then typing :
. The command bar pre-populates with :'<,'>
. Paste the above command and hit Enter.
It was tested on this version of Vim:
VIM - Vi IMproved 7.3 (2010 Aug 15, compiled Aug 22 2015 15:38:58)
Compiled by [email protected]
Windows users may want different line endings.
Here is what I think is a much clearer way to check if a variable is defined:
var_defined() {
local var_name=$1
set | grep "^${var_name}=" 1>/dev/null
return $?
}
Use it as follows:
if var_defined foo; then
echo "foo is defined"
else
echo "foo is not defined"
fi
grep
is awful. Note that bash
supports ${!var_name}
as a way of getting the value of a variable whose name is specified in $var_name
, so name=value; value=1; echo ${name} ${!name}
yields value 1
. –
Dissection A shorter version to test an undefined variable can simply be:
test -z ${mystr} && echo "mystr is not defined"
Call set without any arguments... it outputs all the defined variables.
The last ones on the list would be the ones defined in your script.
So you could pipe its output to something that could figure out what things are defined and what’s not.
© 2022 - 2024 — McMap. All rights reserved.
[ -n "${VAR+x}"] && echo not null
– Parodist[ -z ]
instead of[ -z "" ]
; if it has spaces, it becomes[ -z "my" "test" ]
instead of[ -z my test ]
; and if it's[ -z * ]
, then the*
is replaced with the names of files in your directory. – Valeric[[ -z $mystr ]]
no splitting or globbing will be done. (I leave the argument about whether you should still quote for this or that reason to someone else, but I believe splitting and globbing would no longer be an issue.) – Emission[[
in the manner you show. (There are other cases where quoting can be needed in[[
, but they're in combination with operators like=
or=~
) – Valeric