Regex Module

Iñaki Baz Castillo

Edited by

Iñaki Baz Castillo

Revision History
Revision $Revision: 5901 $$Date$

Table of Contents

1. Admin Guide
1.1. Overview
1.2. Dependencies
1.2.1. OpenSIPS Modules
1.2.2. External Libraries or Applications
1.3. Exported Parameters
1.3.1. file (string)
1.3.2. max_groups (int)
1.3.3. group_max_size (int)
1.3.4. pcre_caseless (int)
1.3.5. pcre_multiline (int)
1.3.6. pcre_dotall (int)
1.3.7. pcre_extended (int)
1.4. Exported Functions
1.4.1. pcre_match (string, pcre_regex)
1.4.2. pcre_match_group (string [, group])
1.5. Exported MI Functions
1.5.1. regex_reload
1.6. Installation and Running
1.6.1. File format

List of Examples

1.1. Set file parameter
1.2. Set max_groups parameter
1.3. Set group_max_size parameter
1.4. Set pcre_caseless parameter
1.5. Set pcre_multiline parameter
1.6. Set pcre_dotall parameter
1.7. Set pcre_extended parameter
1.8. pcre_match usage (forcing case insensitive)
1.9. pcre_match usage (using "end of line" symbol)
1.10. pcre_match_group usage
1.11. regex file
1.12. Using with pua_usrloc
1.13. Incorrect groups file

Chapter 1. Admin Guide

1.1. Overview

This module offers matching operations against regular expressions using the powerful PCRE library.

A text file containing regular expressions categorized in groups is compiled when the module is loaded, storing the compiled PCRE objects in an array. A function to match a string or pseudo-variable against any of these groups is provided. The text file can be modified and reloaded at any time via a MI command. The module also offers a function to perform a PCRE matching operation against a regular expression provided as function parameter.

For a detailed list of PCRE features read the man page of the library.

1.2. Dependencies

1.2.1. OpenSIPS Modules

The following modules must be loaded before this module:

  • No dependencies on other OpenSIPS modules.

1.2.2. External Libraries or Applications

The following libraries or applications must be installed before running OpenSIPS with this module loaded:

  • libpcre-dev - the development libraries of PCRE.

1.3. Exported Parameters

1.3.1. file (string)

Text file containing the regular expression groups. It must be set in order to enable the group matching function.

Default value is NULL.

Example 1.1. Set file parameter

...
modparam("regex", "file", "/etc/opensips/regex_groups")
...

1.3.2. max_groups (int)

Max number of regular expression groups in the text file.

Default value is 20.

Example 1.2. Set max_groups parameter

...
modparam("regex", "max_groups", 40)
...

1.3.3. group_max_size (int)

Max content size of a group in the text file.

Default value is 8192.

Example 1.3. Set group_max_size parameter

...
modparam("regex", "group_max_size", 16384)
...

1.3.4. pcre_caseless (int)

If this options is set, matching is done caseless. It is equivalent to Perl's /i option, and it can be changed within a pattern by a (?i) or (?-i) option setting.

Default value is 0.

Example 1.4. Set pcre_caseless parameter

...
modparam("regex", "pcre_caseless", 1)
...

1.3.5. pcre_multiline (int)

By default, PCRE treats the subject string as consisting of a single line of characters (even if it actually contains newlines). The "start of line" metacharacter (^) matches only at the start of the string, while the "end of line" metacharacter ($) matches only at the end of the string, or before a terminating newline.

When this option is set, the "start of line" and "end of line" constructs match immediately following or immediately before internal newlines in the subject string, respectively, as well as at the very start and end. This is equivalent to Perl's /m option, and it can be changed within a pattern by a (?m) or (?-m) option setting. If there are no newlines in a subject string, or no occurrences of ^ or $ in a pattern, setting this option has no effect.

Default value is 0.

Example 1.5. Set pcre_multiline parameter

...
modparam("regex", "pcre_multiline", 1)
...

1.3.6. pcre_dotall (int)

If this option is set, a dot metacharater in the pattern matches all characters, including those that indicate newline. Without it, a dot does not match when the current position is at a newline. This option is equivalent to Perl's /s option, and it can be changed within a pattern by a (?s) or (?-s) option setting.

Default value is 0.

Example 1.6. Set pcre_dotall parameter

...
modparam("regex", "pcre_dotall", 1)
...

1.3.7. pcre_extended (int)

If this option is set, whitespace data characters in the pattern are totally ignored except when escaped or inside a character class. Whitespace does not include the VT character (code 11). In addition, characters between an unescaped # outside a character class and the next newline, inclusive, are also ignored. This is equivalent to Perl's /x option, and it can be changed within a pattern by a (?x) or (?-x) option setting.

Default value is 0.

Example 1.7. Set pcre_extended parameter

...
modparam("regex", "pcre_extended", 1)
...

1.4. Exported Functions

1.4.1.  pcre_match (string, pcre_regex)

Matches the given string parameter against the regular expression pcre_regex, which is compiled into a PCRE object. Returns TRUE if it matches, FALSE otherwise.

Meaning of the parameters is as follows:

  • string - String or pseudo-variable to compare.

  • pcre_regex - Regular expression to be compiled in a PCRE object. It can be a string or pseudo-variable.

NOTE: To use the "end of line" symbol '$' in the pcre_regex parameter use '$$'.

This function can be used from REQUEST_ROUTE, FAILURE_ROUTE, ONREPLY_ROUTE, BRANCH_ROUTE and LOCAL_ROUTE.

Example 1.8.  pcre_match usage (forcing case insensitive)

...
if (pcre_match("$ua", "(?i)^twinkle")) {
    xlog("L_INFO", "User-Agent matches\n");
}
...

Example 1.9.  pcre_match usage (using "end of line" symbol)

...
if (pcre_match("$rU", "^user[1234]$$")) {  # Will be converted to "^user[1234]$"
    xlog("L_INFO", "RURI username matches\n");
}
...

1.4.2.  pcre_match_group (string [, group])

It uses the groups readed from the text file (see Section 1.6.1, “File format”) to match the given string parameter against the compiled regular expression in group number group. Returns TRUE if it matches, FALSE otherwise.

Meaning of the parameters is as follows:

  • string - String or pseudo-variable to compare.

  • group - Number of group to use in the operation. If not specified then 0 (the first group) is used.

This function can be used from REQUEST_ROUTE, FAILURE_ROUTE, ONREPLY_ROUTE, BRANCH_ROUTE and LOCAL_ROUTE.

Example 1.10.  pcre_match_group usage

...
if (pcre_match_group("$rU", "2")) {
    xlog("L_INFO", "RURI username matches group 2\n");
}
...

1.5. Exported MI Functions

1.5.1.  regex_reload

Causes regex module to re-read the content of the text file and re-compile the regular expressions. The number of groups in the file can be modified safely.

Name: regex_reload

Parameters: none

MI FIFO Command Format:

:regex_reload:_reply_fifo_file_
_empty_line_

1.6. Installation and Running

1.6.1. File format

The file contains regular expressions categorized in groups. Each group starts with "[number]" line. Lines starting by space, tab, CR, LF or # (comments) are ignored. Each regular expression must take up just one line, this means that a regular expression can't be splitted in various lines.

An example of the file format would be the following:

Example 1.11. regex file

### List of User-Agents publishing presence status
[0]

# Softphones
^Twinkle/1
^X-Lite
^eyeBeam
^Bria
^SIP Communicator
^Linphone

# Deskphones
^Snom

# Others
^SIPp
^PJSUA


### Blacklisted source IP's
[1]

^190\.232\.250\.226$
^122\.5\.27\.125$
^86\.92\.112\.


### Free PSTN destinations in Spain
[2]

^1\d{3}$
^((\+|00)34)?900\d{6}$

The module compiles the text above to the following regular expressions:

group 0: ((^Twinkle/1)|(^X-Lite)|(^eyeBeam)|(^Bria)|(^SIP Communicator)|
          (^Linphone)|(^Snom)|(^SIPp)|(^PJSUA))
group 1: ((^190\.232\.250\.226$)|(^122\.5\.27\.125$)|(^86\.92\.112\.))
group 2: ((^1\d{3}$)|(^((\+|00)34)?900\d{6}$))

The first group can be used to avoid auto-generated PUBLISH (pua_usrloc module) for UA's already supporting presence:

Example 1.12. Using with pua_usrloc

route[REGISTER] {
    if (! pcre_match_group("$ua", 0)) {
        xlog("L_INFO", "Auto-generated PUBLISH for $fu ($ua)\n");
        pua_set_publish();
    }
    save("location");
    exit;
}

NOTE: It's important to understand that the numbers in each group header ([number]) must start by 0. If not, the real group number will not match the number appearing in the file. For example, the following text file:

Example 1.13. Incorrect groups file

[1]
^aaa
^bbb

[2]
^ccc
^ddd

will generate the following regular expressions:

group 0: ((^aaa)|(^bbb))
group 1: ((^ccc)|(^ddd))

Note that the real index doesn't match the group number in the file. This is, compiled group 0 always points to the first group in the file, regardless of its number in the file. In fact, the group number appearing in the file is used for nothing but for delimiting different groups.

NOTE: A line containing a regular expression cannot start by '[' since it would be treated as a new group. The same for lines starting by space, tab, or '#' (they would be ignored by the parser). As a workaround, using brackets would work:

[0]
([0-9]{9})
( #abcde)
( qwerty)