I would like to check if a string begins with "node" e.g. "node001". Something like
if [ $HOST == user* ]
then
echo yes
fi
How can I do it correctly?
I further need to combine expressions to check if HOST is either "user1" or begins with "node"
if [ [[ $HOST == user1 ]] -o [[ $HOST == node* ]] ];
then
echo yes
fi
> > > -bash: [: too many arguments
How can I do it correctly?
This snippet on the Advanced Bash Scripting Guide says:
# The == comparison operator behaves differently within a double-brackets
# test than within single brackets.
[[ $a == z* ]] # True if $a starts with a "z" (wildcard matching).
[[ $a == "z*" ]] # True if $a is equal to z* (literal matching).
So you had it nearly correct; you needed double brackets, not single brackets.
With regards to your second question, you can write it this way:
HOST=user1
if [[ $HOST == user1 ]] || [[ $HOST == node* ]] ;
then
echo yes1
fi
HOST=node001
if [[ $HOST == user1 ]] || [[ $HOST == node* ]] ;
then
echo yes2
fi
Which will echo
yes1
yes2
Bash's if
syntax is hard to get used to (IMO).
If you're using a recent version of Bash (v3+), I suggest the Bash regex comparison operator =~
, for example,
if [[ "$HOST" =~ ^user.* ]]; then
echo "yes"
fi
To match this or that
in a regex, use |
, for example,
if [[ "$HOST" =~ ^user.*|^host1 ]]; then
echo "yes"
fi
Note - this is 'proper' regular expression syntax.
user* means use and zero-or-more occurrences of r, so use and userrrr will match.
user.* means user and zero-or-more occurrences of any character, so user1, userX will match.
^user.* means match the pattern user.* at the begin of $HOST.
If you're not familiar with regular expression syntax, try referring to this resource.
Note that the Bash =~
operator only does regular expression matching when the right hand side is UNQUOTED. If you do quote the right hand side, "any part of the pattern may be quoted to force it to be matched as a string.". You should not quote the right hand side even when doing parameter expansion.
=~
operator only does regular expression matching when the right hand side is UNQUOTED. If you do quote the right hand side "Any part of the pattern may be quoted to force it to be matched as a string." (1.) make sure to always put the regular expressions on the right un-quoted and (2.) if you store your regular expression in a variable, make sure to NOT quote the right hand side when you do parameter expansion.
I always try to stick with POSIX sh
instead of using Bash extensions, since one of the major points of scripting is portability (besides connecting programs, not replacing them).
In sh
, there is an easy way to check for an "is-prefix" condition.
case $HOST in node*)
# Your code here
esac
Given how old, arcane and crufty sh is (and Bash is not the cure: It's more complicated, less consistent and less portable), I'd like to point out a very nice functional aspect: While some syntax elements like case
are built-in, the resulting constructs are no different than any other job. They can be composed in the same way:
if case $HOST in node*) true;; *) false;; esac; then
# Your code here
fi
Or even shorter
if case $HOST in node*) ;; *) false;; esac; then
# Your code here
fi
Or even shorter (just to present !
as a language element -- but this is bad style now)
if ! case $HOST in node*) false;; esac; then
# Your code here
fi
If you like being explicit, build your own language element:
beginswith() { case $2 in "$1"*) true;; *) false;; esac; }
Isn't this actually quite nice?
if beginswith node "$HOST"; then
# Your code here
fi
And since sh
is basically only jobs and string-lists (and internally processes, out of which jobs are composed), we can now even do some light functional programming:
beginswith() { case $2 in "$1"*) true;; *) false;; esac; }
checkresult() { if [ $? = 0 ]; then echo TRUE; else echo FALSE; fi; }
all() {
test=$1; shift
for i in "$@"; do
$test "$i" || return
done
}
all "beginswith x" x xy xyz ; checkresult # Prints TRUE
all "beginswith x" x xy abc ; checkresult # Prints FALSE
This is elegant. Not that I'd advocate using sh
for anything serious -- it breaks all too quickly on real world requirements (no lambdas, so we must use strings. But nesting function calls with strings is not possible, pipes are not possible, etc.)
case $HOST in user01 | node* ) ...
if case $HOST in node*) true;; *) false;; esac; then
I've seen it here and there, to my eye it looks kinda scrunched up.
case
commands are commands, they can go inside if ... then
.
beginswith() { case "$2" in "$1"*) true;; *) false;; esac; }
otherwise if $1
has a literal *
or ?
it might give wrong answer.
You can select just the part of the string you want to check:
if [ "${HOST:0:4}" = user ]
For your follow-up question, you could use an OR:
if [[ "$HOST" == user1 || "$HOST" == node* ]]
${HOST:0:4}
HOST='a b'; if [ ${HOST:0:4} = user ] ; then echo YES ; fi
if [[ ${HOST:0:4} = user ]]
I prefer the other methods already posted, but some people like to use:
case "$HOST" in
user1|node*)
echo "yes";;
*)
echo "no";;
esac
Edit:
I've added your alternates to the case statement above
In your edited version you have too many brackets. It should look like this:
if [[ $HOST == user1 || $HOST == node* ]];
While I find most answers here quite correct, many of them contain unnecessary Bashisms. POSIX parameter expansion gives you all you need:
[ "${host#user}" != "${host}" ]
and
[ "${host#node}" != "${host}" ]
${var#expr}
strips the smallest prefix matching expr
from ${var}
and returns that. Hence if ${host}
does not start with user
(node
), ${host#user}
(${host#node}
) is the same as ${host}
.
expr
allows fnmatch()
wildcards, thus ${host#node??}
and friends also work.
[[ $host == user* ]]
might be necessary, since it's far more readable than [ "${host#user}" != "${host}" ]
. As such granted that you control the environment where the script is executed (target the latest versions of bash
), the former is preferable.
has_prefix()
function and never look at it again.
Since #
has a meaning in Bash, I got to the following solution.
In addition I like better to pack strings with "" to overcome spaces, etc.
A="#sdfs"
if [[ "$A" == "#"* ]];then
echo "Skip comment line"
fi
blah:
, looks like this is the answer!
case $A in "#"*) echo "Skip comment line";; esac
is both shorter and more portable.
Adding a tiny bit more syntax detail to Mark Rushakoff's highest rank answer.
The expression
$HOST == node*
Can also be written as
$HOST == "node"*
The effect is the same. Just make sure the wildcard is outside the quoted text. If the wildcard is inside the quotes it will be interpreted literally (i.e. not as a wildcard).
@OP, for both your questions you can use case/esac:
string="node001"
case "$string" in
node*) echo "found";;
* ) echo "no node";;
esac
Second question
case "$HOST" in
node*) echo "ok";;
user) echo "ok";;
esac
case "$HOST" in
node*|user) echo "ok";;
esac
Or Bash 4.0
case "$HOST" in
user) ;&
node*) echo "ok";;
esac
;&
is only available in Bash >= 4.
if [ [[ $HOST == user1 ]] -o [[ $HOST == node* ]] ];
then
echo yes
fi
doesn't work, because all of [
, [[
, and test
recognize the same nonrecursive grammar. See section CONDITIONAL EXPRESSIONS on your Bash man page.
As an aside, the SUSv3 says
The KornShell-derived conditional command (double bracket [[]]) was removed from the shell command language description in an early proposal. Objections were raised that the real problem is misuse of the test command ([), and putting it into the shell is the wrong way to fix the problem. Instead, proper documentation and a new shell reserved word (!) are sufficient. Tests that require multiple test operations can be done at the shell level using individual invocations of the test command and shell logicals, rather than using the error-prone -o flag of test.
You'd need to write it this way, but test doesn't support it:
if [ $HOST == user1 -o $HOST == node* ];
then
echo yes
fi
test uses = for string equality, and more importantly it doesn't support pattern matching.
case
/ esac
has good support for pattern matching:
case $HOST in
user1|node*) echo yes ;;
esac
It has the added benefit that it doesn't depend on Bash, and the syntax is portable. From the Single Unix Specification, The Shell Command Language:
case word in
[(]pattern1) compound-list;;
[[(]pattern[ | pattern] ... ) compound-list;;] ...
[[(]pattern[ | pattern] ... ) compound-list]
esac
[
and test
are Bash builtins as well as external programs. Try type -a [
.
if [ -z $aa -or -z $bb ]
; ... gives "bash: [: -or: binary operator expected" ; however if [ -z "$aa" -o -z "$bb" ] ; ...
passes.
grep
Forgetting performance, this is POSIX and looks nicer than case
solutions:
mystr="abcd"
if printf '%s' "$mystr" | grep -Eq '^ab'; then
echo matches
fi
Explanation:
printf '%s' to prevent printf from expanding backslash escapes: Bash printf literal verbatim string
grep -q prevents echo of matches to stdout: How to check if a file contains a specific string using Bash
grep -E enables extended regular expressions, which we need for the ^
I tweaked @markrushakoff's answer to make it a callable function:
function yesNo {
# Prompts user with $1, returns true if response starts with y or Y or is empty string
read -e -p "
$1 [Y/n] " YN
[[ "$YN" == y* || "$YN" == Y* || "$YN" == "" ]]
}
Use it like this:
$ if yesNo "asfd"; then echo "true"; else echo "false"; fi
asfd [Y/n] y
true
$ if yesNo "asfd"; then echo "true"; else echo "false"; fi
asfd [Y/n] Y
true
$ if yesNo "asfd"; then echo "true"; else echo "false"; fi
asfd [Y/n] yes
true
$ if yesNo "asfd"; then echo "true"; else echo "false"; fi
asfd [Y/n]
true
$ if yesNo "asfd"; then echo "true"; else echo "false"; fi
asfd [Y/n] n
false
$ if yesNo "asfd"; then echo "true"; else echo "false"; fi
asfd [Y/n] ddddd
false
Here is a more complex version that provides for a specified default value:
function toLowerCase {
echo "$1" | tr '[:upper:]' '[:lower:]'
}
function yesNo {
# $1: user prompt
# $2: default value (assumed to be Y if not specified)
# Prompts user with $1, using default value of $2, returns true if response starts with y or Y or is empty string
local DEFAULT=yes
if [ "$2" ]; then local DEFAULT="$( toLowerCase "$2" )"; fi
if [[ "$DEFAULT" == y* ]]; then
local PROMPT="[Y/n]"
else
local PROMPT="[y/N]"
fi
read -e -p "
$1 $PROMPT " YN
YN="$( toLowerCase "$YN" )"
{ [ "$YN" == "" ] && [[ "$PROMPT" = *Y* ]]; } || [[ "$YN" = y* ]]
}
Use it like this:
$ if yesNo "asfd" n; then echo "true"; else echo "false"; fi
asfd [y/N]
false
$ if yesNo "asfd" n; then echo "true"; else echo "false"; fi
asfd [y/N] y
true
$ if yesNo "asfd" y; then echo "true"; else echo "false"; fi
asfd [Y/n] n
false
Keep it simple
word="appel"
if [[ $word = a* ]]
then
echo "Starts with a"
else
echo "No match"
fi
Success story sharing
[[ $a == z* ]]
and[[ $a == "z*" ]]
? In other words: do they work differently? And what specifically do you mean when you say "$a is equal to z*"?[[ $a == *com ]]