Differences between revisions 1 and 54 (spanning 53 versions)
Revision 1 as of 2007-05-02 23:46:43
Size: 1429
Editor: redondos
Comment:
Revision 54 as of 2022-08-01 10:02:57
Size: 5745
Editor: 89
Comment: fixed quoting, in "else printf ..." below "Bash version 3 and above have..."
Deletions are marked like this. Additions are marked like this.
Line 1: Line 1:
[[Anchor(faq54)]] <<Anchor(faq54)>>
Line 3: Line 3:
First, you have to define what you mean by "number". The most common case when people ask this seems to be "a non-negative integer, with no leading + sign". Or in other words, a string of all digits. Other times, people want to validate a floating-point input, with optional sign and optional decimal point.
Line 4: Line 5:
First, you have to define what you mean by "number". The most common case seems to be that, when people ask this, they actually mean "a non-negative integer, with no leading + sign". === Hand parsing ===
If you're validating a simple "string of digits", you can do it with a [[glob]]:
Line 6: Line 8:
{{{
if [[ $foo = *[^0-9]* ]]; then
   echo "'$foo' has a non-digit somewhere in it"
{{{#!highlight bash
# Bash / Ksh
if [[ -n $foo && $foo != *[!0123456789]* ]]; then
    printf '"%s" is strictly numeric\n' "$foo"
Line 10: Line 13:
   echo "'$foo' is strictly numeric"     printf '"%s" has a non-digit somewhere in it or is empty\n' "$foo"
fi >&2
}}}

Avoid `[0-9]` or `[[:digit:]]` which in some locales and some systems can match characters other than 0123456789.

The same thing can be done in POSIX shells as well, using {{{case}}}:

{{{#!highlight bash
# POSIX
case $var in
    '')
        printf 'var is empty\n';;
    *[!0123456789]*)
        printf '%s has a non-digit somewhere in it\n' "$var";;
    *)
        printf '%s is strictly numeric\n' "$var";;
esac >&2
}}}
Of course, if all you care about is valid vs. invalid, you can combine cases:

{{{#!highlight bash
# POSIX
case $var in
    '' | *[!0123456789]*)
        printf '%s\n' "$0: $var: invalid digit" >&2; exit 1;;
esac
}}}
If you need to allow a leading negative sign, or if want a valid floating-point number or something else more complex, then there are a few possible ways. Standard globs aren't expressive enough to do this, but you can trim off any sign and then compare:

{{{#!highlight bash
# POSIX
case ${var#[-+]} in # notice ${var#prefix} substitution to trim sign
    '')
        printf 'var is empty\n';;
    .)
        printf 'var is just a dot\n';;
    *.*.*)
        printf '"%s" has more than one decimal point in it\n' "$var";;
    *[!0123456789.]*)
        printf '"%s" has a non-digit somewhere in it\n' "$var";;
    *)
        printf '"%s" looks like a valid float\n' "$var";;
esac >&2
}}}
Or in Bash, we can use [[glob|extended globs]]:

{{{#!highlight bash
# Bash -- extended globs must be enabled explicitly in versions prior to 4.1.
# Check whether the variable is all digits.
shopt -s extglob
[[ $var = +([0123456789]) ]]
}}}
A more complex case:

{{{#!highlight bash
# Bash / ksh
shopt -s extglob # not necessary in ksh and bash 4.1 or newer

if [[ $foo = @(*[0123456789]*|!([+-]|)) && $foo = ?([+-])*([0123456789])?(.*([0123456789])) ]]; then
  echo 'foo is a floating-point number'
Line 13: Line 76:
Optionally, `case..esac` may have been used in shells with extended pattern matching. The leading test of {{{$foo}}} is to ensure that it contains at least one digit, isn't empty, and contains more than just + or - by itself.
Line 14: Line 78:
This can be done in Korn and legacy Bourne shells as well, using {{{case}}}: If your definition of "a valid number" is even more complex, or if you need a solution that works in legacy Bourne shells, you might prefer to use an external tool's [[RegularExpression|regular expression]] syntax. Here is a portable version (explained in detail [[http://www.wplug.org/wiki/Meeting-20100612#EXERCISE_TWO|here]]), using {{{awk}}} (not `egrep` which is line-based so would be tricked by variables that contain newline characters):
Line 16: Line 80:
{{{
case "$foo" in
    *[!0-9]*) echo "'$foo' has a non-digit somewhere in it" ;;
    *) echo "'$foo' is strictly numeric" ;;
esac
}}}
{{{#!highlight bash
# Bourne
Line 23: Line 83:
If what you actually mean is "a valid floating-point number" or something else more complex, then you might prefer to use a regular expression. Bash version 3 and above have regular expression support in the [[ command:

{{{
if [[ $foo =~ ^[-+]?[0-9]+\(\.[0-9]+\)?$ ]]; then
    echo "'$foo' looks rather like a number"
if awk -- 'BEGIN {exit !(ARGV[1] ~ /^[-+]?([0123456789]+\.?|[0123456789]*\.[0123456789]+)$/)}' "$foo"; then
    printf '"%s" is a number\n' "$foo"
Line 29: Line 86:
    echo "'$foo' doesn't look particularly numeric to me"     printf '"%s" is not a number\n' "$foo"
Line 32: Line 89:
Bash version 3 and above have regular expression support in the `[[...]]` construct.
Line 33: Line 91:
If you don't have bash version 3, then you would use {{{egrep}}}: {{{#!highlight bash
# Bash
# The regexp must be stored in a var and expanded for backward compatibility with versions < 3.2
Line 35: Line 95:
{{{
if echo "$foo" | egrep '^[-+]?[0-9]+(\.[0-9]+)?$' >/dev/null; then
    echo "'$foo' might be a number"
regexp='^[-+]?[0123456789]*(\.[0123456789]*)?$'
if [[ $foo = *[0123456789]* && $foo =~ $regexp ]]; then
    printf '"%s" looks rather like a number\n' "$foo"
Line 39: Line 99:
    echo "'$foo' might not be a number"     printf '"%s" doesn'\''t look particularly numeric to me.\n' "$foo"
Line 42: Line 102:
=== Using the parsing done by [ and printf (or "using eq") ===
{{{#!highlight bash
# fails with ksh
if [ "$foo" -eq "$foo" ] 2>/dev/null; then
    printf '"%s" is an integer\n' "$foo"
fi
}}}
`[` parses the variable and interprets it a decimal integer because of the `-eq`. If the parsing succeeds the test is trivially true; if it fails `[` prints an error message that `2>/dev/null` hides and sets a status different from 0. However this method fails if the shell is ksh, because ksh evaluates the variable as an arithmetic expression (and that would constitute an arbitrary command injection vulnerability).
Line 43: Line 111:
Note that the parentheses in the {{{egrep}}} regular expression don't require backslashes in front of them, whereas the ones in the bash3 command do. Be careful: the following trick with `printf` (not supported by all shells, and the list of supported float representations varies with the shell as well; not to mention the command injection vulnerability in ksh or zsh)

{{{#!highlight bash
if printf %f "$foo" >/dev/null 2>&1; then
    printf '"%s" is a float\n' "$foo"
fi
}}}
is broken: about the arguments of the {{{a}}}, {{{A}}}, {{{e}}}, {{{E}}}, {{{f}}}, {{{F}}}, {{{g}}}, or {{{G}}} format modifiers, POSIX specifies that ''if the leading character is a single-quote or double-quote, the value shall be the numeric value in the underlying codeset of the character following the single-quote or double-quote.'' Hence this fails when {{{foo}}} expands to a string with a leading single-quote or double-quote: the previous command will happily validate the string as a float.
It also returns 0 when {{{foo}}} expands to a number with a leading {{{0x}}}, which is a valid number in a shell script but may not work elsewhere.

You can use `%d` to parse an integer. Take care that the parsing might be (is supposed to be?) [[locale]]-dependent.

How can I tell whether a variable contains a valid number?

First, you have to define what you mean by "number". The most common case when people ask this seems to be "a non-negative integer, with no leading + sign". Or in other words, a string of all digits. Other times, people want to validate a floating-point input, with optional sign and optional decimal point.

Hand parsing

If you're validating a simple "string of digits", you can do it with a glob:

   1 # Bash / Ksh
   2 if [[ -n $foo && $foo != *[!0123456789]* ]]; then
   3     printf '"%s" is strictly numeric\n' "$foo"
   4 else
   5     printf '"%s" has a non-digit somewhere in it or is empty\n' "$foo"
   6 fi >&2

Avoid [0-9] or [[:digit:]] which in some locales and some systems can match characters other than 0123456789.

The same thing can be done in POSIX shells as well, using case:

   1 # POSIX
   2 case $var in
   3     '')
   4         printf 'var is empty\n';;
   5     *[!0123456789]*)
   6         printf '%s has a non-digit somewhere in it\n' "$var";;
   7     *)
   8         printf '%s is strictly numeric\n' "$var";;
   9 esac >&2

Of course, if all you care about is valid vs. invalid, you can combine cases:

   1 # POSIX
   2 case $var in
   3     '' | *[!0123456789]*)
   4         printf '%s\n' "$0: $var: invalid digit" >&2; exit 1;;
   5 esac

If you need to allow a leading negative sign, or if want a valid floating-point number or something else more complex, then there are a few possible ways. Standard globs aren't expressive enough to do this, but you can trim off any sign and then compare:

   1 # POSIX
   2 case ${var#[-+]} in   # notice ${var#prefix} substitution to trim sign
   3     '')
   4         printf 'var is empty\n';;
   5     .)
   6         printf 'var is just a dot\n';;
   7     *.*.*)
   8         printf '"%s" has more than one decimal point in it\n' "$var";;
   9     *[!0123456789.]*)
  10         printf '"%s" has a non-digit somewhere in it\n' "$var";;
  11     *)
  12         printf '"%s" looks like a valid float\n' "$var";;
  13 esac >&2

Or in Bash, we can use extended globs:

   1 # Bash -- extended globs must be enabled explicitly in versions prior to 4.1.
   2 # Check whether the variable is all digits.
   3 shopt -s extglob
   4 [[ $var = +([0123456789]) ]]

A more complex case:

   1 # Bash / ksh
   2 shopt -s extglob # not necessary in ksh and bash 4.1 or newer
   3 
   4 if [[ $foo = @(*[0123456789]*|!([+-]|)) && $foo = ?([+-])*([0123456789])?(.*([0123456789])) ]]; then
   5   echo 'foo is a floating-point number'
   6 fi

Optionally, case..esac may have been used in shells with extended pattern matching. The leading test of $foo is to ensure that it contains at least one digit, isn't empty, and contains more than just + or - by itself.

If your definition of "a valid number" is even more complex, or if you need a solution that works in legacy Bourne shells, you might prefer to use an external tool's regular expression syntax. Here is a portable version (explained in detail here), using awk (not egrep which is line-based so would be tricked by variables that contain newline characters):

   1 # Bourne
   2 
   3 if awk -- 'BEGIN {exit !(ARGV[1] ~ /^[-+]?([0123456789]+\.?|[0123456789]*\.[0123456789]+)$/)}' "$foo"; then
   4     printf '"%s" is a number\n' "$foo"
   5 else
   6     printf '"%s" is not a number\n' "$foo"
   7 fi

Bash version 3 and above have regular expression support in the [[...]] construct.

   1 # Bash
   2 # The regexp must be stored in a var and expanded for backward compatibility with versions < 3.2
   3 
   4 regexp='^[-+]?[0123456789]*(\.[0123456789]*)?$'
   5 if [[ $foo = *[0123456789]* && $foo =~ $regexp ]]; then
   6     printf '"%s" looks rather like a number\n' "$foo"
   7 else
   8     printf '"%s" doesn'\''t look particularly numeric to me.\n' "$foo"
   9 fi

Using the parsing done by [ and printf (or "using eq")

   1 # fails with ksh
   2 if [ "$foo" -eq "$foo" ] 2>/dev/null; then
   3     printf '"%s" is an integer\n' "$foo"
   4 fi

[ parses the variable and interprets it a decimal integer because of the -eq. If the parsing succeeds the test is trivially true; if it fails [ prints an error message that 2>/dev/null hides and sets a status different from 0. However this method fails if the shell is ksh, because ksh evaluates the variable as an arithmetic expression (and that would constitute an arbitrary command injection vulnerability).

Be careful: the following trick with printf (not supported by all shells, and the list of supported float representations varies with the shell as well; not to mention the command injection vulnerability in ksh or zsh)

   1 if printf %f "$foo" >/dev/null 2>&1; then
   2     printf '"%s" is a float\n' "$foo"
   3 fi

is broken: about the arguments of the a, A, e, E, f, F, g, or G format modifiers, POSIX specifies that if the leading character is a single-quote or double-quote, the value shall be the numeric value in the underlying codeset of the character following the single-quote or double-quote. Hence this fails when foo expands to a string with a leading single-quote or double-quote: the previous command will happily validate the string as a float. It also returns 0 when foo expands to a number with a leading 0x, which is a valid number in a shell script but may not work elsewhere.

You can use %d to parse an integer. Take care that the parsing might be (is supposed to be?) locale-dependent.

BashFAQ/054 (last edited 2022-08-01 10:02:57 by 89)