Scroll to navigation

Net::SIP::Util(3pm) User Contributed Perl Documentation Net::SIP::Util(3pm)

NAME

Net::SIP::Util - utility functions used by all of Net::SIP

SYNOPSIS

  use Net::SIP::Util qw( create_rtp_sockets );
  my ($port,@socks) = create_rtp_sockets( '192.168.0.10' ) or die;

DESCRIPTION

This package implements various utility function used within various Net::SIP packages and partly usable for the user of Net::SIP too.

Each of this functions is exportable, but none is exported per default. All functions can be exported at once with the import flag ":all".

SUBROUTINES

Invokes callback CALLBACK with additional args @ARGS. CALLBACK can be:
In this case it will be called as $CALLBACK->(@ARGS) and return the return value of this call.
In this case the scalar will be set to $ARGS[0] and the rest of @ARGS will be ignored. If no @ARGS are given the scalar will be set to TRUE. It will return with the value of the scalar.
In this case it will call "$CALLBACK->run(@ARGS)" and return with the return value of this call.
The first element of the array will be interpreted as code reference, while the rest as args, e.g. it will do: 

  my ($coderef,@cb_args) = @$CALLBACK;
  return $coderef->( @cb_args, @ARGS );
In this case it will try to match all @ARGS against the regex. If anything matches it will return TRUE, else FALSE.
This will try to find out which local IP address SRC is used for connections to target DST. This will be done by creating a connected UDP socket to the target and using getsockname to get the local IP address of this socket.
Creates socket with protocol PROTO (default 'udp'). It will use "laddr4dst" to find the appropriate local source IP address.

It will try to bind the socket to port 5060 (default SIP port). If this fails it will try port 5062..5100 and if it cannot bind to any of these ports it will just use any port which gets assigned by the OS.

For multihomed hosts where several addresses are bound to the same interface it will just use one of these addresses. If you need more control about the address the socket is bound to (and which will be used as the local IP in outgoing packets) you need to create the socket yourself.

In scalar context it just returns the newly created socket. In array context it will return the socket and the "ip:port" the created socket is bound to. If the creation of the socket fails it will return "()" and set $!.

Example:

  my ($sock,$ip_port) = create_socket_to ( '192.168.0.1' )
        or die $!;
This tries to allocate sockets for RTP. RTP consists usually of a data socket on an even port number and a control socket (RTCP) and the following port. It will try to create these sockets. MINPORT is the minimal port number to use (default 2000), MAXPORT the highest port (default MINPORT+10000), TRIES is the number of attempts it makes to create such socket pairs and defaults to 1000.

RANGE is the number of consecutive ports it needs to allocate and defaults to 2 (e.g. data and control socket).

Allocation will be done by choosing a random even number between MINPORT and MAXPORT and then trying to allocate all the sockets on this and the following port numbers.

If the allocation fails after TRIES attempts were made it will return "()", otherwise it will return an array with at first the starting port number followed by all the allocated sockets.

Example:

  my ($port,$rtp_sock,$rtcp_sock) = create_rtp_sockets( '192.168.0.10' )
        or die "allocation failed";
Interprets VALUE as a value for the SIP header field KEY and splits it into the parts (prefix, parameter). Because for most keys the delimiter is ";", but for some keys "," the field name KEY need to be known.

KEY needs to be normalized already (lower case, no abbreviation).

Returns array with initial data (up to first delimiter) and the parameters as hash.

Example for key 'to':

  '"Silver; John" <silver@example.com>; tag=...; protocol=TCP'
  -> ( '"Silver; John" <silver@example.com>', { tag => ..., protocol => 'TCP' } )

Example for key 'www-authenticate':

  'Digest method="md5", qop="auth"'
  -> ( 'Digest', { method => 'md5', qop => 'auth' } )
Inverse function to sip_hdrval2parts, e.g constructs header value for KEY from PREFIX and %PARAMETER and returns value.
Returns parts from URI. If called in scalar context it returns only the domain part. In array context it returns an array with the following values:
The lower cased domain part as given in the URI, i.e. "host", "host:port", "[ipv6]:port" etc.
The optional user part of the SIP address.
The protocol, e.g. "sip" or "sips". If not explicitly given it will default to "sip".
A hash reference to any parameters following "data" as returned by sip_hdrval2parts.
The original full part before any parameter, i.e. the part containing the domain, optional user and optional proto.
Returns URI from parts, i.e. the reverse to sip_uri2parts. For the meaning of the parameters see there.
This extracts information from URI which are needed for creating the socket: HOST is the IP address or host name, PORT the port (undef if not given) and FAMILY the family, i.e. AF_INET, AF_INET6 or undef if HOST is not an IP address. PROTO will be set based on the SIP protocol and parameters, i.e. "tls" for "sips" URI, "tcp" or "udp" for "sip" URI with explicit "transport" parameter and "undef" otherwise.

If OPAQUE the host name is not required to be a valid name. See OPAQUE in "ip_string2parts" for more information.

This is the reverse to sip_uri2sockinfo. See there for the meaning of the parameters.

If the first argument is a hash ref it will expected to provided the arguments in the keys "proto", "host", "port" and "family".

Returns true if both URIs point to the same SIP address. This compares user part case sensitive, domain part case insensitive (does no DNS resolution) protocol and ports in domain (assumes default ports for protocol if no port is given).
This will parse the given string STR and split it into the parts as follows: 

  IPv4, [IPv4]               -> (IPv4, undef, AF_INET)
  IPv4:port, [IPv4]:port     -> (IPv4, port,  AF_INET)
  IPv6, [IPv6]               -> (IPv6, undef, AF_INET6)
  [IPv6]:port                -> (IPv6, port,  AF_INET6)
  host, [host]               -> (host, undef, undef)
  host:port, [host]:port     -> (host, port,  undef)

The IP address and host will be returned in a canonicalized way.

If this function is used to parse strings where the host part is not a real hostname but some identifier with more allowed characters than a hostname then OPAQUE should be set and in this way no strict checking and no canonicalization is done.

If the function is called with a scalar context it will return the result as a hash ref with the keys "host", "addr", "port", "family" where "addr" is only set if it is IP address.

This is the reverse to ip_string2parts. If family is not given it will be determined by checking if ip_or_host contains ':' (i.e IPv6 address): 

  (ip_or_host)              -> ip_or_host
  (ipv4_or_host,port)       -> ipv4_or_host:port
  (ipv6,port)               -> [ipv6]:port
  (ipv6,undef,*,true)       -> [ipv6]

If the first argument is a hash ref it will be treated as a hash ref as returned by ip_string2parts. In this case also "use_host" can be given which prefers "host" to "addr" for stringification. Also, "default_port" can be used to give a port number which will be treated as default and omitted from string.

This will return the IP, PORT and FAMILY from a sockaddr_in or sockaddr_in6. If FAMILY is not given it will be determined based on the size of SOCKADDR.

If the function is called with a scalar context it will return the result as a hash ref as described for ip_string2parts.

This will create a sockaddr_in or sockaddr_in6 from IP and PORT. FAMILY will be determined by checking the IP if not given.

If the first argument is a hash ref it will be treated as a hash ref as returned by ip_sockaddr2parts.

This will return the STR from a sockaddr_in or sockaddr_in6, i.e. like combining "ip_sockaddr2parts" with "ip_parts2string". If FAMILY is not given it will be determined based on the size of SOCKADDR.
This will create the hostname used in reverse lookup of an IP address, i.e. "*.in-addr.arpa" or "*.ip6.arpa". If FAMILY is not given it will be determined based on the syntax of IP.
This will convert the given IP address into a canonical form suitable for comparison.
This will lookup the given name using getaddrinfo and return the IP addresses. If FAMILY is given only addresses matching the family will be returned.
This checks if the given IP address is a valid IPv4 address.
This checks if the given IP address is a valid IPv6 address.
This checks if the given IP address is a valid IPv4 or IPv6 address and returns the address family if this is an IP address.
2024-09-08 perl v5.38.2