Regular Expression
About
A list of useful regular expression examples.Regular Expressions are used throughout FreeSWITCH. See Dialplan_XML for complete examples of the regular expressions used in conditional statements in the dialplan.
Capturing Values
Matching items in parentheses are captured in variables sequentially named $1, $2, etc. $0 contains the entire pattern.
Examples
Any toll free Number
Matches any toll free number
^(\+?1)?(8(00|44|55|66|77|88)[2-9]\d{6})$
Example: 8005551212 or 18005551212 or +18005551212 will match and in each case $2 is populated with "8005551212"
Explanation:
- ^ - indicates the beginning of a regular expression (required);
- ( - indicates the beginning of a regular expression block - blocks are important to define inner expressions so they can be referenced by the variables $1, $2, $3, etc;
- \+1|1? - indicates optional digits '+1' or '1' (the ? sign defines the literal as optional);
- ) - closes the block;
- 8 - matches the literal '8';
- ( - another block starting;
- 00|55|66|77|88 - matches 00 or 55 or 66 or 77 or 88 (you guessed right, the | sign is the OR operator);
- ) - closes the inner block;
- [2-9] - matches one digit in the range from 2 to 9 (2, 3, 4, 5, 6, 7, 8 and 9), and as you also guessed the [] pair of brackets encloses a range; other range examples: [0-9] matches 0 to 9; [a-z] matches a, b, c, ...z);
- \d - matches any valid digit (same as [0-9]);
- {6} - defines the number of occurrences for the previous expression, i.e. exactly 6 digits in the range 0-9. This can also contain a variable number of occurrences, for example to match a sequence of 6 to 9 digits: {6,9}; or to match at least 6 with no maximum: {6,};
- ) - closes the other block;
- $ - indicates the end of the regular expression (required);
Any string of exactly seven digits
Matches any string of exactly seven digits
^(\d{7})$
Example: 5551212 will match with $1 having "5551212"
95551212 will not match (string of digits too long)
555121A will not match (no alpha chars allowed)
Any string of exactly 10 digits
Matches any string of exactly 10 digits
^(\d{10})$
Example: 8005551212 will match with $1 having "8005551212"
18005551212 will not match (string of digits too long)
Any string of exactly 11 digits
Matches any string of exactly 11 digits
^(\d{11})$
Example: 18005551212 will match with $1 having "18005551212"
8005551212 will not match (string of digits too short)
A string with prefix of 9 and an additional 11 digits
Matches a string with a prefix of 9 and an additional 11 digits.
^9(\d{11})$
Example: 918005551212 will match with $1 having "18005551212"
8005551212 will not match (string of digits too short)
Any string of exactly 3 or 4 digits
Matches any string of exactly 3 or 4 digits
(^\d{4}$|^\d{3}$)
Example: 1001 will match with $1 having "1001"
102 will match with $1 having "102"
*9999 will not match
*98 will not match
Any string from 3 to 5 digits
Matches any string from 3 to 5 digits
(^\d{3,5}$)
Example: 10010 will match with $1 having "10010"
102 will match with $1 having "102"
*9999 will not match
*98 will not match
Variable number of digits
Matches a variable number of digits after a prefix of '9'
^9(\d+)$
Example: 977 will match ($1 = 77)
877 will not match
966736 will match ($1 = 66736)
9118299983 will match ($1 = 118299983)
9 will not match (+ means one or more)
International country codes
<extension name="international" continue="true">
<!--
^\+(1| # NANPA
2[1-689]\d| # 21X,22X,23X,24X,25X,26X,28X,29X
2[07]| # 20, 27
3[0-469]| # 30,31,32,33,34,36,39
3[578]\d| # 35X,37X,38X
4[0-13-9]| # 40,41,43,44,45,46,47,48,49
42\d| # 42X
5[09]\d| # 50X, 59X
5[1-8]| # 51,52,53,54,55,56,57,58
6[0-6]| # 60,61,62,63,64,65,66
6[7-9]\d| # 67X,68X,69X
7|
8[035789]\d| # 80X,83X,85X,87X,88X,89X
8[1246]| # 81,82,84,86
9[0-58]| # 90,91,92,93,94,95,98
9[679]\d # 96X,97X,99X
)(\d+)
-->
<condition field="destination_number" expression="^\+(1|2[1-689]\d|2[07]|3[0-469]|3[578]\d|4[0-13-9]|42\d|5[09]\d|5[1-8]|6[0-6]|6[7-9]\d|7|8[035789]\d|8[1246]|9[0-58]|9[679]\d)(\d+)">
<action application="set" data="country_code=$1"/>
<action application="set" data="national_number=$2"/>
</condition>
</extension>
NANPA +1NxxNxxXXXX E.164 Dialstring
^(\+1|1)?([2-9]\d\d[2-9]\d{6})$
Example: +13172222222 matches
13172222222 still matches because +1 or 1 are optional
3172222222 still matches because +1 or 1 are optional
3171222222 does not match and is not a valid NANPA number.
LNP (Local Number Portability)
^(?:1)?([2-9]\d{2}[2-9]\d{6})(?:;npdi=(?:yes|no))?(?:;rn=([2-9]\d{2}[2-9]\d{6}))?(?:;npdi=(?:yes|no))?
Example:
6045555555;npdi=yes;rn=7785550001
6045555555;rn=7785550001;npdi=yes
16045555555;npdi=yes;rn=7785550001
6045555555;rn=7785550001
Result:
$1=6045555555
$2=7785550001
Example:
6045555555
16045555555
6045555555;npdi=yes
6045555555;npdi=no
Result:
$1=6045555555
$2=empty
Clustering vs Capturing
Sometimes, you need to use parentheses to group a set of choices, but you are not interested in saving what is matched. Place a question mark and colon after the opening parenthesis to accomplish this. The example below matches a leading "+" or leading "00", but the matched information is not saved. The second set of parentheses does save information that is matched.
^(?:\+|00)(\d+)$
Example: +13171222222 matches and captures 13171222222 in $1
0013171222222 matches and captures 13171222222 in $1
The following example matches a number and saves the information in two pieces as leading characters ($1) and the telephone number ($2).
^(\+|00)(\d+)$
Example: +13171222222 matches and captures + in $1 and 13171222222 in $2
0013171222222 matches and captures 00 in $1 and 13171222222 in $2
Parentheses can be nested. When trying to determine which variable holds the matched information, count the opening parentheses from left to right. The first opening parenthesis will store information in $1, the second opening parenthesis will store information in $2 and so on.
^((\+|00)(\d+))$
Example: +13171222222 matches and captures +13171222222 in $1, + in $2 and 13171222222 in $3
0013171222222 matches and captures 0013171222222 in $1, 00 in $2 and 13171222222 in $3
In the following example, the first set of parentheses does not store any information.
^(?:(\+|00)(\d+))$
Example: +13171222222 matches and captures + in $1 and 13171222222 in $2
0013171222222 matches and captures 00 in $1 and 13171222222 in $2
Greediness
^rn=(.*);
Example: rn=1234567890;npdi; matches and captures 1234567890;npdi in $1
If you do .*? it will be un-greedy (or use \d to only match the numbers if you know it will be a number)
^rn=(.*?);
Example: rn=1234567890;npdi; matches and captures 1234567890 in $1
Case Insensitive
To make a match case insensitive, prepend (?i) to your match string.
Example: (?i)restricted matches both Restricted and restricted
Further Reading
Regular Expressions are serious business. They are quite useful in computing in general. Here are some other resources for learning more:
http://regexlib.com/Default.aspx
http://www.regular-expressions.info/
http://www.weitz.de/regex-coach/
http://regexlib.com/CheatSheet.aspx
http://www.zytrax.com/tech/web/regex.htm
http://oreilly.com/catalog/9781565922570/
http://perldoc.perl.org/perlreref.html
Test Your Regex
Use fs_cli
See: mod_commands#regex
Online Tools
Simple Perl Script
You can also use the following perl script to test your regular expression, or make some adjustment of regex based on what you need.
#!/usr/bin/perl
$x=$ARGV[0];
if($x=~/^(\+1|1)?([2-9]\d\d[2-9]\d{6})$/)
{
print "Input: $x\n";
print "\$1 Output: ".$1."\n";
print "\$2 Output: ".$2."\n";
}
Result:
# perl x.pl 12135551212
Input: 12135551212
$1 Output: 1
$2 Output: 2135551212
# perl x.pl +12135551212
Input: +12135551212
$1 Output: +1
$2 Output: 2135551212
# perl x.pl 2135551212
Input: 2135551212
$1 Output:
$2 Output: 2135551212
See Also
regex API command performs pattern matching on the fs_cli
Regular Expression Cheat Sheet: http://overapi.com/regex/