Network Working Group J. Lennox Request for Comments: 3880 X. Wu Category: Standards Track H. Schulzrinne Columbia University October 2004
Call Processing Language (CPL): A Language for User Control of Internet Telephony Services
Status of this Memo
This document specifies an Internet standards track protocol for the Internet community, and requests discussion and suggestions for improvements. Please refer to the current edition of the "Internet Official Protocol Standards" (STD 1) for the standardization state and status of this protocol. Distribution of this memo is unlimited.
Copyright (C) The Internet Society (2004).
This document defines the Call Processing Language (CPL), a language to describe and control Internet telephony services. It is designed to be implementable on either network servers or user agents. It is meant to be simple, extensible, easily edited by graphical clients, and independent of operating system or signalling protocol. It is suitable for running on a server where users may not be allowed to execute arbitrary programs, as it has no variables, loops, or ability to run external programs.
The Call Processing Language (CPL) is a language that can be used to describe and control Internet telephony services. It is not tied to any particular signalling architecture or protocol; it is anticipated that it will be used with both the Session Initiation Protocol (SIP)  and H.323 .
CPL is powerful enough to describe a large number of services and features, but it is limited in power so that it can run safely in Internet telephony servers. The intention is to make it impossible for users to do anything more complex (and dangerous) than describe Internet telephony services. The language is not Turing-complete, and provides no way to write loops or recursion.
CPL is also designed to be easily created and edited by graphical tools. It is based on the Extensible Markup Language (XML) , so parsing it is easy and many parsers for it are publicly available.
Lennox, et al. Standards Track [Page 3]
RFC 3880 CPL October 2004
The structure of the language maps closely to its behavior, so an editor can understand any valid script, even ones written by hand. The language is also designed so that a server can easily confirm the validity of a script when the server receives it, rather than discovering problems while a call is being processed.
Implementations of CPL are expected to take place both in Internet telephony servers and in advanced clients; both can usefully process and direct users' calls. This document primarily addresses the usage in servers. A mechanism will be needed to transport scripts between clients and servers; this document does not describe such a mechanism, but related documents will.
The framework and requirements for the CPL architecture are described in RFC 2824, "Call Processing Language Framework and Requirements" .
In this document, the key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" are to be interpreted as described in BCP 14, RFC 2119  and indicate requirement levels for compliant CPL implementations.
Some paragraphs are indented, like this; they give motivations of design choices, advice to implementors, or thoughts on future development of or extensions to CPL. They are not essential to the specification of the language, and are non-normative.
A CPL script consists of two types of information: ancillary information about the script, and call processing actions.
A call processing action is a structured tree that describes the operations and decisions a telephony signalling server performs on a call set-up event. There are two types of call processing actions: top-level actions and subactions. Top-level actions are actions that are triggered by signalling events that arrive at the server. Two top-level actions are defined: "incoming", the action performed when a call arrives whose destination is the owner of the script, and "outgoing", the action performed when a call arrives whose originator is the owner of the script.
Lennox, et al. Standards Track [Page 4]
RFC 3880 CPL October 2004
Subactions are actions which can be called from other actions. CPL forbids subactions from being called recursively: see Section 8.
Ancillary information is information which is necessary for a server to correctly process a script, but which does not directly describe any operations or decisions. Currently, no ancillary information is defined, but the section is reserved for use by extensions.
2.2. Abstract Structure of a Call Processing Action
Abstractly, a call processing action is described by a collection of nodes that describe operations that can be performed or decisions that can be made. A node may have several parameters, which specify the precise behavior of the node; they usually also have outputs, which depend on the result of the decision or action.
For a graphical representation of a CPL action, see Figure 1. Nodes and outputs can be thought of informally as boxes and arrows; CPL is designed so that actions can be conveniently edited graphically using this representation. Nodes are arranged in a tree, starting at a single root node; outputs of nodes are connected to additional nodes. When an action is run, the action or decision described by the action's top-level node is performed; based on the result of that node, the server follows one of the node's outputs, and the subsequent node it points to is performed; this process continues until a node with no specified outputs is reached. Because the graph is acyclic, this will occur after a bounded and predictable number of nodes are visited.
If an output to a node does not point to another node, it indicates that the CPL server should perform a node- or protocol-specific action. Some nodes have specific default behavior associated with them; for others, the default behavior is implicit in the underlying signalling protocol, or can be configured by the administrator of the server. For further details on this, see Section 10.
For flexibility, one piece of information necessary for CPL is not given as node parameters: the set of locations to which a call is to be directed. Instead, this set of locations is stored as an implicit global variable throughout the execution of a processing action (and its subactions). This allows locations to be retrieved from external sources, filtered, and so forth, without requiring general language support for such operations (which could harm the simplicity and tractability of understanding the language). The specific operations which add, retrieve, or filter location sets are given in Section 5.
For the incoming top-level call processing action, the location set is initialized to the empty set. For the outgoing action, it is initialized to the destination address of the call.
Syntactically, CPL scripts are represented by XML documents. XML is thoroughly specified by the XML specification , and implementors of this specification should be familiar with that document. However, as a brief overview, XML consists of a hierarchical structure of tags; each tag can have a number of attributes. It is visually and structurally very similar to HTML , as both languages are simplifications of the earlier and larger standard SGML .
Lennox, et al. Standards Track [Page 6]
RFC 3880 CPL October 2004
See Figure 2 for the XML document corresponding to the graphical representation of the CPL script in Figure 1. Both nodes and outputs in CPL are represented by XML tags; parameters are represented by XML tag attributes. Typically, node tags contain output tags, and vice- versa (with a few exceptions: see Sections 5.1, 5.3, 7.1, and 7.2).
The connection between the output of a node and another node is represented by enclosing the tag representing the pointed-to node inside the tag for the outer node's output. Convergence (several outputs pointing to a single node) is represented by subactions, discussed further in Section 8.
The higher-level structure of a CPL script is represented by tags corresponding to each piece of ancillary information, subactions, and top-level actions, in order. This higher-level information is all enclosed in a special tag "cpl", the outermost tag of the XML document.
A complete XML Schema for CPL is provided in Appendix C. The remainder of the main sections of this document describe the semantics of CPL, while giving its syntax informally. For the formal syntax, please see the appendix.
Tag: "cpl" Parameters: None Sub-tags: "ancillary" See Section 9 "subaction" See Section 8 "outgoing" Top-level actions to take on this user's outgoing calls "incoming" Top-level actions to take on this user's incoming calls
Figure 3: Syntax of the top-level "cpl" tag
Call processing actions, both top-level actions and subactions, consist of a tree of nodes and outputs. Nodes and outputs are both described by XML tags. There are four categories of CPL nodes: switches, which represent choices a CPL script can make, location modifiers, which add or remove locations from the location set, signalling operations, which cause signalling events in the underlying protocol, and non-signalling operations, which trigger behavior which does not effect the underlying protocol.
Switches represent choices a CPL script can make, based on either attributes of the original call request or items independent of the call.
All switches are arranged as a list of conditions that can match a variable. Each condition corresponds to a node output; the output points to the next node that should be executed if the condition is true. The conditions are tried in the order they are presented in the script; the output corresponding to the first node to match is taken.
There are two special switch outputs that apply to every switch type. The output "not-present", which MAY occur anywhere in the list of outputs, is true if the variable the switch was to match was not present in the original call setup request. (In this document, this is sometimes described by saying that the information is "absent".)
Lennox, et al. Standards Track [Page 8]
RFC 3880 CPL October 2004
The output "otherwise", which MUST be the last output specified if it is present, matches if no other condition matched.
If no condition matches and no "otherwise" output was present in the script, the default script behavior is taken. See Section 10 for more information on this.
Switches MAY contain no outputs. They MAY only contain an "otherwise" output.
Such switches are not particularly useful, but might be created by tools which automatically generate CPL scripts.
Address switches allow a CPL script to make decisions based on one of the addresses present in the original call request. They are summarized in Figure 4.
Node: "address-switch" Outputs: "address" Specific addresses to match Parameters: "field" "origin", "destination", or "original-destination" "subfield" "address-type", "user", "host", "port", "tel", or "display" (also: "password" and "alias-type")
Output: "address" Parameters: "is" Exact match "contains" Substring match (for "display" only) "subdomain-of" Sub-domain match (for "host", "tel")
Figure 4: Syntax of the "address-switch" node
Address switches have two node parameters: "field" and "subfield". The mandatory "field" parameter allows the script to specify which address is to be considered for the switch: either the call's origin address (field "origin"), its current destination address (field "destination"), or its original destination (field "original- destination"), the destination the call had before any earlier forwarding was invoked. Servers MAY define additional field values.
The optional "subfield" specifies which part of the address is to be considered. The possible subfield values are: "address-type", "user", "host", "port", "tel", and "display". Additional subfield values MAY be defined for protocol-specific values. (The subfield "password" is defined for SIP in Section 4.1.1; the subfield "alias- type" is defined for H.323 in Appendix B.1.) If no subfield is
Lennox, et al. Standards Track [Page 9]
RFC 3880 CPL October 2004
specified, the "entire" address is matched; the precise meaning of this is defined for each underlying signalling protocol. Servers MAY define additional subfield values.
The subfields are defined as follows:
address-type: This indicates the type of the underlying address, i.e., the URI scheme, if the address can be represented by a URI. The types specifically discussed by this document are "sip", "tel", and "h323". The address type is not case- sensitive. It has a value for all defined address types.
user: This subfield of the address indicates, for e-mail style addresses, the user part of the address. For a telephone number style address, it includes the subscriber number. This subfield is case-sensitive; it may be absent.
host: This subfield of the address indicates the Internet host name or IP address corresponding to the address, in host name, IPv4, or IPv6  textual representation format. Host names are compared as strings. IP addresses are compared numerically. (In particular, the presence or location of an IPv6 :: omitted-zero-bits block is not significant for matching purposes.) Host names are never equal to IP addresses -- no DNS resolution is performed. IPv4 addresses are never equal to IPv6 addresses, even if the IPv6 address is a v4-in-v6 embedding. This subfield is not case sensitive, and may be absent.
For host names only, subdomain matching is supported with the "subdomain-of" match operator. The "subdomain-of" operator ignores leading dots in the hostname or match pattern, if any.
port: This subfield indicates the TCP or UDP port number of the address, numerically, in decimal format. It is not case sensitive, as it MUST only contain decimal digits. Leading zeros are ignored.
tel: This subfield indicates a telephone subscriber number, if the address contains such a number. It is not case sensitive (telephone numbers may contain the symbols 'A', 'B', 'C', or 'D'), and may be absent. It may be matched using the "subdomain-of" match operator. Punctuation and separator characters in telephone numbers are discarded.
Lennox, et al. Standards Track [Page 10]
RFC 3880 CPL October 2004
display: This subfield indicates a "display name" or user-visible name corresponding to an address. It is a Unicode string, and is matched using the case-insensitive algorithm described in Section 4.2. The "contains" operator may be applied to it. It may be absent.
For any completely unknown subfield, the server MAY reject the script at the time it is submitted with an indication of the problem; if a script with an unknown subfield is executed, the server MUST consider the "not-present" output to be the valid one.
The "address" output tag may take exactly one of three possible parameters, indicating the kind of matching allowed.
is: An output with this match operator is followed if the subfield being matched in the "address-switch" exactly matches the argument of the operator. It may be used for any subfield, or for the entire address if no subfield was specified.
subdomain-of: This match operator applies only for the subfields "host" and "tel". In the former case, it matches if the hostname being matched is a subdomain of the domain given in the argument of the match operator; thus, subdomain- of="example.com" would match the hostnames "example.com", "research.example.com", and "zaphod.sales.internal.example.com". IP addresses may be given as arguments to this operator; however, they only match exactly. In the case of the "tel" subfield, the output matches if the telephone number being matched has a prefix that matches the argument of the match operator; subdomain-of="1212555" would match the telephone number "1 212 555 1212."
contains: This match operator applies only for the subfield "display". The output matches if the display name being matched contains the argument of the match as a substring.
For SIP, the "origin" address corresponds to the address in the "From" header, "destination" corresponds to the "Request-URI", and "original-destination" corresponds to the "To" header.
The "display" subfield of an address is the display-name part of the address, if it is present. Because of SIP's syntax, the "destination" address field will never have a "display" subfield.
Lennox, et al. Standards Track [Page 11]
RFC 3880 CPL October 2004
The "address-type" subfield of an address is the URI scheme of that address. Other address fields depend on that "address-type".
For SIP URIs, the "user", "host", and "port" subfields correspond to the "user," "host," and "port" elements of the URI syntax. (Note that, following the definitions of RFC 3261 , a SIP URI which does not specify a port is not the same as an explicit port 5060; the former is indicated by an absent port subfield.) The "tel" subfield is defined to be the "user" part of the URI, with visual separators stripped, if the "user=phone" parameter is given to the URI, or if the server is otherwise configured to recognize the user part as a telephone number. An additional subfield, "password", is defined to correspond to the "password" element of the SIP URI, and is case- sensitive. However, use of this field is NOT RECOMMENDED for general security reasons.
For tel URLs, the "tel" and "user" subfields are the subscriber name; in the former case, visual separators are stripped. The "host" and "port" subfields are both not present.
For h323 URLs, subfields MAY be set according to the scheme described in Appendix B.
For other URI schemes, only the "address-type" subfield is defined by this specification; servers MAY set other pre-defined subfields, or MAY support additional subfields.
If no subfield is specified for addresses in SIP messages, the string matched is the URI part of the address. For "is" matches, standard SIP URI matching rules are used; for "contains" matches, the URI is used verbatim.
String switches allow a CPL script to make decisions based on free- form strings present in a call request. They are summarized in Figure 5.
Node: "string-switch" Outputs: "string" Specific string to match Parameters: "field" "subject", "organization", "user-agent", or "display"
Output: "string" Parameters: "is" Exact match "contains" Substring match
Figure 5: Syntax of the "string-switch" node
Lennox, et al. Standards Track [Page 12]
RFC 3880 CPL October 2004
String switches have one node parameter: "field". The mandatory "field" parameter specifies which string is to be matched.
String switches are dependent on the call signalling protocol being used.
Four fields are defined and listed below. The value of each of these fields is a free-form Unicode string with no other structure defined.
subject: The subject of the call.
organization: The organization of the originator of the call.
user-agent: The name of the program or device with which the call request was made.
display: Free-form text associated with the call, intended to be displayed to the recipient, with no other semantics defined by the signalling protocol.
Strings are matched as case-insensitive Unicode strings, in the following manner. First, strings are canonicalized to the "Compatibility Composition" (KC) form, as specified in Unicode Standard Annex #15 . Then, strings are compared using locale- insensitive caseless mapping, as specified in Unicode Standard Annex #21 .
Code to perform the first step, in Java and Perl, is available; see the links from Annex 5 of UAX 15 . The case-insensitive string comparison in the Java standard class libraries already performs the second step; other Unicode-aware libraries should be similar.
The output tag of string matching is named "string", and has a mandatory argument, one of "is" or "contains", indicating whole- string match or substring match, respectively.
Language switches allow a CPL script to make decisions based on the languages in which the originator of the call wishes to communicate. They are summarized in Figure 6.
Node: "language-switch" Outputs: "language" Specific string to match Parameters: None
Output: "language" Parameters: "matches" Match if the given language matches a language-range of the call.
Figure 6: Syntax of the "language-switch" node
Language switches take no parameters.
The "language" output takes one parameter, "matches". The value of the parameter is a language-tag, as defined in RFC 3066 . The caller may have specified a set of language-ranges, also as defined in RFC 3066. The CPL server checks each language-tag specified by the script against the language-ranges specified in the request.
See RFC 3066 for the details of how language-ranges match language- tags. Briefly, a language-range matches a language-tag if it exactly equals the tag, or if it exactly equals a prefix of the tag such that the first character following the prefix is "-".
If the caller specified the special language-range "*", it is ignored for the purpose of matching. Languages with a "q" value of 0 are also ignored.
Time switches allow a CPL script to make decisions based on the time and/or date the script is being executed. They are summarized in Figure 7.
Time switches are independent of the underlying signalling protocol.
Node: "time-switch" Outputs: "time" Specific time to match Parameters: "tzid" RFC 2445 Time Zone Identifier "tzurl" RFC 2445 Time Zone URL
Output: "time" Parameters: "dtstart" Start of interval (RFC 2445 DATE-TIME) "dtend" End of interval (RFC 2445 DATE-TIME) "duration" Length of interval (RFC 2445 DURATION) "freq" Frequency of recurrence ("secondly", "minutely", "hourly", "daily", "weekly", "monthly", or "yearly") "interval" How often the recurrence repeats "until" Bound of recurrence (RFC 2445 DATE-TIME) "count" Number of occurrences of recurrence "bysecond" List of seconds within a minute "byminute" List of minutes within an hour "byhour" List of hours of the day "byday" List of days of the week "bymonthday" List of days of the month "byyearday" List of days of the year "byweekno" List of weeks of the year "bymonth" List of months of the year "wkst" First day of the work week "bysetpos" List of values within set of events specified
Figure 7: Syntax of the "time-switch" node
Time switches are based closely on the specification of recurring intervals of time in the Internet Calendaring and Scheduling Core Object Specification (iCalendar COS), RFC 2445 .
This allows CPL scripts to be generated automatically from calendar books. It also allows us to re-use the extensive existing work specifying time intervals.
If future standards-track documents are published that update or obsolete RFC 2445, any changes or clarifications those documents make to recurrence handling apply to CPL time-switches as well.
Lennox, et al. Standards Track [Page 15]
RFC 3880 CPL October 2004
An algorithm to determine whether an instant falls within a given recurrence is given in Appendix A.
The "time-switch" tag takes two optional parameters, "tzid" and "tzurl", both of which are defined in RFC 2445 (Sections 220.127.116.11 and 18.104.22.168 respectively). The "tzid" is the identifying label by which a time zone definition is referenced. If it begins with a forward slash (solidus), it references a to-be-defined global time zone registry; otherwise it is locally-defined at the server. The "tzurl" gives a network location from which an up-to-date VTIMEZONE definition for the timezone can be retrieved.
While "tzid" labels that do not begin with a forward slash are locally defined, it is RECOMMENDED that servers support at least the naming scheme used by the Olson Time Zone database . Examples of timezone databases that use the Olson scheme are the zoneinfo files on most Unix-like systems, and the standard Java TimeZone class.
Servers SHOULD resolve "tzid" and "tzurl" references to time zone definitions at the time the script is uploaded. They MAY periodically refresh these resolutions to obtain the most up-to-date definition of a time zone. If a "tzurl" becomes invalid, servers SHOULD remember the most recent valid data retrieved from the URL.
If a script is uploaded with a "tzid" and "tzurl" which the CPL server does not recognize or cannot resolve, it SHOULD diagnose and reject this at script upload time. If neither "tzid" nor "tzurl" are present, all non-UTC times within this time switch should be interpreted as being "floating" times, i.e., that they are specified in the local timezone of the CPL server.
Because of daylight-savings-time changes over the course of a year, it is necessary to specify time switches in a given timezone. UTC offsets are not sufficient, or a time-of-day routing rule which held between 9 am and 5 pm in the eastern United States would start holding between 8 am and 4 pm at the end of October.
Authors of CPL servers should be careful to handle correctly the intervals when local time is discontinuous, at the beginning or end of daylight-savings time. Note especially that some times may occur more than once when clocks are set back. The algorithm in Appendix A is believed to handle this correctly.
Time nodes specify a list of periods during which their output should be taken. They have two required parameters: "dtstart", which specifies the beginning of the first period of the list, and exactly one of "dtend" or "duration", which specify the ending time or the
Lennox, et al. Standards Track [Page 16]
RFC 3880 CPL October 2004
duration of the period, respectively. The "dtstart" and "dtend" parameters are formatted as iCalendar COS DATE-TIME values, as specified in Section 4.3.5 of RFC 2445 . Because time zones are specified in the top-level "time-switch" tag, only forms 1 or 2 (floating or UTC times) can be used. The "duration" parameter is given as an iCalendar COS DURATION parameter, as specified in section 4.3.6 of RFC 2445. Both the DATE-TIME and the DURATION syntaxes are subsets of the corresponding syntaxes from ISO 8601 .
For a recurring interval, the "duration" parameter MUST be small enough such that subsequent intervals do not overlap. For non- recurring intervals, durations of any positive length are permitted. Zero-length and negative-length durations are not allowed.
If no other parameters are specified, a time node indicates only a single period of time. More complicated sets of period intervals are constructed as recurrences. A recurrence is specified by including the "freq" parameter, which indicates the type of recurrence rule. Parameters other than "dtstart", "dtend", and "duration" SHOULD NOT be specified unless "freq" is present, though CPL servers SHOULD accept scripts with such parameters present, and ignore the other parameters.
The "freq" parameter takes one of the following values: "secondly", to specify repeating periods based on an interval of a second or more, "minutely", to specify repeating periods based on an interval of a minute or more, "hourly", to specify repeating periods based on an interval of an hour or more, "daily", to specify repeating periods based on an interval of a day or more, "weekly", to specify repeating periods based on an interval of a week or more, "monthly", to specify repeating periods based on an interval of a month or more, and "yearly", to specify repeating periods based on an interval of a year or more. These values are not case-sensitive.
The "interval" parameter contains a positive integer representing how often the recurrence rule repeats. The default value is "1", meaning every second for a "secondly" rule, every minute for a "minutely" rule, every hour for an "hourly" rule, every day for a "daily" rule, every week for a "weekly" rule, every month for a "monthly" rule, and every year for a "yearly" rule.
The "until" parameter defines an iCalendar COS DATE or DATE-TIME value which bounds the recurrence rule in an inclusive manner. If the value specified by "until" is synchronized with the specified recurrence, this date or date-time becomes the last instance of the recurrence. If specified as a date-time value, then it MUST be
Lennox, et al. Standards Track [Page 17]
RFC 3880 CPL October 2004
specified in UTC time format. If not present, and the "count" parameter is not also present, the recurrence is considered to repeat forever.
The "count" parameter defines the number of occurrences at which to range-bound the recurrence. The "dtstart" parameter counts as the first occurrence. The "until" and "count" parameters MUST NOT occur in the same "time" output.
The "bysecond" parameter specifies a comma-separated list of seconds within a minute. Valid values are 0 to 59. The "byminute" parameter specifies a comma-separated list of minutes within an hour. Valid values are 0 to 59. The "byhour" parameter specifies a comma- separated list of hours of the day. Valid values are 0 to 23.
The "byday" parameter specifies a comma-separated list of days of the week. "MO" indicates Monday, "TU" indicates Tuesday, "WE" indicates Wednesday, "TH" indicates Thursday, "FR" indicates Friday, "SA" indicates Saturday, and "SU" indicates Sunday. These values are not case-sensitive.
Each "byday" value can also be preceded by a positive (+n) or negative (-n) integer. If present, this indicates the nth occurrence of the specific day within the "monthly" or "yearly" recurrence. For example, within a "monthly" rule, +1MO (or simply 1MO) represents the first Monday within the month, whereas -1MO represents the last Monday of the month. If an integer modifier is not present, it means all days of this type within the specified frequency. For example, within a "monthly" rule, MO represents all Mondays within the month.
The "bymonthday" parameter specifies a comma-separated list of days of the month. Valid values are 1 to 31 or -31 to -1. For example, -10 represents the tenth to the last day of the month.
The "byyearday" parameter specifies a comma-separated list of days of the year. Valid values are 1 to 366 or -366 to -1. For example, -1 represents the last day of the year (December 31st) and -306 represents the 306th to the last day of the year (March 1st).
The "byweekno" parameter specifies a comma-separated list of ordinals specifying weeks of the year. Valid values are 1 to 53 or -53 to -1. This corresponds to weeks according to week numbering as defined in ISO 8601 . A week is defined as a seven day period, starting on the day of the week defined to be the week start (see "wkst"). Week number one of the calendar year is the first week which contains at least four (4) days in that calendar year. This parameter is only valid for "yearly" rules. For example, 3 represents the third week of the year.
Lennox, et al. Standards Track [Page 18]
RFC 3880 CPL October 2004
Note: Assuming a Monday week start, week 53 can only occur when January 1 is a Thursday or, for leap years, if January 1 is a Wednesday.
The "bymonth" parameter specifies a comma-separated list of months of the year. Valid values are 1 to 12.
The "wkst" parameter specifies the day on which the work week starts. Valid values are "MO", "TU", "WE", "TH", "FR", "SA" and "SU". This is significant when a "weekly" recurrence has an interval greater than 1, and a "byday" parameter is specified. This is also significant in a "yearly" recurrence when a "byweekno" parameter is specified. The default value is "MO", following ISO 8601 .
The "bysetpos" parameter specifies a comma-separated list of values which corresponds to the nth occurrence within the set of events specified by the rule. Valid values are 1 to 366 or -366 to -1. It MUST only be used in conjunction with another byxxx parameter. For example, "the last work day of the month" could be represented as:
Each "bysetpos" value can include a positive (+n) or negative (-n) integer. If present, this indicates the nth occurrence of the specific occurrence within the set of events specified by the rule.
If byxxx parameter values are found which are beyond the available scope (i.e., bymonthday="30" in February), they are simply ignored.
Byxxx parameters modify the recurrence in some manner. Byxxx rule parts for a period of time which is the same or greater than the frequency generally reduce or limit the number of occurrences of the recurrence generated. For example, freq="daily" bymonth="1" reduces the number of recurrence instances from all days (if the "bymonth" parameter is not present) to all days in January. Byxxx parameters for a period of time less than the frequency generally increase or expand the number of occurrences of the recurrence. For example, freq="yearly" bymonth="1,2" increases the number of days within the yearly recurrence set from 1 (if "bymonth" parameter is not present) to 2.
If multiple Byxxx parameters are specified, then after evaluating the specified "freq" and "interval" parameters, the Byxxx parameters are applied to the current set of evaluated occurrences in the following order: "bymonth", "byweekno", "byyearday", "bymonthday", "byday", "byhour", "byminute", "bysecond", and "bysetpos"; then "count" and "until" are evaluated.
Lennox, et al. Standards Track [Page 19]
RFC 3880 CPL October 2004
Here is an example of evaluating multiple Byxxx parameters.
First, the interval="2" would be applied to freq="yearly" to arrive at "every other year." Then, bymonth="1" would be applied to arrive at "every January, every other year." Then, byday="SU" would be applied to arrive at "every Sunday in January, every other year." Then, byhour="8,9" would be applied to arrive at "every Sunday in January at 8 AM and 9 AM, every other year." Then, byminute="30" would be applied to arrive at "every Sunday in January at 8:30 AM and 9:30 AM, every other year." Then the second is derived from "dtstart" to end up in "every Sunday in January from 8:30:00 AM to 8:40:00 AM, and from and 9:30:00 AM to 9:40:00 AM, every other year." Similarly, if the "byminute", "byhour", "byday", "bymonthday", or "bymonth" parameter were missing, the appropriate minute, hour, day, or month would have been retrieved from the "dtstart" parameter.
The iCalendar COS RDATE, EXRULE, and EXDATE recurrence rules are not specifically mapped to components of the time-switch node. Equivalent functionality to the exception rules can be attained by using the ordering of switch rules to exclude times using earlier rules; equivalent functionality to the additional-date RDATE rules can be attained by using "sub" nodes (see Section 8) to link multiple outputs to the same subsequent node.
The "not-present" output is never true for a time switch. However, it MAY be included to allow switch processing to be more regular.
4.4.1. iCalendar Differences and Implementation Issues
(This sub-sub-section is non-normative.)
The specification of recurring events in this section is identical (except for syntax and formatting issues) to that of RFC 2445 , with only one additional restriction. That one restriction is that consecutive instances of recurrence intervals may not overlap.
It was a matter of some debate, during the design of CPL, whether the entire iCalendar COS recurrence specification should be included in CPL, or whether only a subset should be included. It was eventually decided that compatibility between the two protocols was of primary importance. This imposes some additional implementation issues on implementors of CPL servers.
Lennox, et al. Standards Track [Page 20]
RFC 3880 CPL October 2004
It does not appear to be possible to determine, in constant time, whether a given instant of time falls within one of the intervals defined by a full iCalendar COS recurrence. The primary concerns are as follows:
o The "count" parameter cannot be checked in constant running time, since it requires that the server enumerate all recurrences from "dtstart" to the present time, in order to determine whether the current recurrence satisfies the parameter. However, a server can expand a "count" parameter once, off-line, to determine the date of the last recurrence. This date can then be treated as a virtual "until" parameter for the server's internal processing.
o Similarly, the "bysetpos" parameter requires that the server enumerate all instances of the occurrence from the start of the current recurrence set until the present time. This requires somewhat more complex pre-processing, but generally, a single recurrence with a "bysetpos" parameter can be split up into several recurrences without them.
o Finally, constant running time of time switches also requires that a candidate starting time for a recurrence can be established quickly and uniquely, to check whether it satisfies the other restrictions. This requires that a recurrence's duration not be longer than its repetition interval, so that a given instant cannot fall within several consecutive potential repetitions of the recurrence. The restriction that consecutive intervals not overlap partially satisfies this condition, but does not fully ensure it. Again, to some extent pre-processing can help resolve this.
The algorithm given in Appendix A runs in constant time after these pre-processing steps.
Servers ought to check that recurrence rules do not create any absurd run-time or memory requirements, and reject those that do, just as they ought to check that CPL scripts in general are not absurdly large.
Priority switches allow a CPL script to make decisions based on the priority specified for the original call. They are summarized in Figure 8. They are dependent on the underlying signalling protocol.
Lennox, et al. Standards Track [Page 21]
RFC 3880 CPL October 2004
Node: "priority-switch" Outputs: "priority" Specific priority to match Parameters: None
Output: "priority" Parameters: "less" Match if priority is less than that specified "greater" Match if priority is greater than that specified "equal" Match if priority is equal to that specified
Figure 8: Syntax of the "priority-switch" node
Priority switches take no parameters.
The "priority" tag takes one of the three parameters "greater", "less", or "equal". The values of these parameters are one of the following priorities: in decreasing order, "emergency", "urgent", "normal", and "non-urgent". These values are matched in a case- insensitive manner. Outputs with the "less" parameter are taken if the priority of the call is less than the priority given in the argument, and so forth.
If no priority is specified in a message, the priority is considered to be "normal". If an unknown priority is specified in the call, it is considered to be equivalent to "normal" for the purposes of "greater" and "less" comparisons, but it is compared literally for "equal" comparisons.
Since every message has a priority, the "not-present" output is never true for a priority switch. However, it MAY be included, to allow switch processing to be more regular.
The abstract location model of CPL is described in Section 2.3. The behavior of several of the signalling operations (defined in Section 6) is dependent on the current location set specified. Location nodes add or remove locations from the location set.
Lennox, et al. Standards Track [Page 22]
RFC 3880 CPL October 2004
There are three types of location nodes defined. Explicit locations add literally-specified locations to the current location set, location lookups obtain locations from some outside source, and location filters remove locations from the set, based on some specified criteria.
Explicit location nodes specify a location literally. Their syntax is described in Figure 9.
Explicit location nodes are dependent on the underlying signalling protocol.
Node: "location" Outputs: None (Next node follows directly) Next node: Any node Parameters: "url" URL of address to add to location set "priority" Priority of this location (0.0-1.0) "clear" Whether to clear the location set before adding the new value
Figure 9: Syntax of the "location" node
Explicit location nodes have three node parameters. The mandatory "url" parameter's value is the URL of the address to add to the location set. Only one address may be specified per location node; multiple locations may be specified by cascading these nodes.
The optional "priority" parameter specifies a priority for the location. Its value is a floating-point number between 0.0 and 1.0. If it is not specified, the server SHOULD assume a default priority of 1.0. The optional "clear" parameter specifies whether the location set should be cleared before adding the new location to it. Its value can be "yes" or "no", with "no" as the default.
Basic location nodes have only one possible result, since there is no way that they can fail. (If a basic location node specifies a location which isn't supported by the underlying signalling protocol, the script server SHOULD detect this and report it to the user at the time the script is submitted.) Therefore, their XML representations do not have explicit output tags; the <location> tag directly contains another node.
Locations can also be specified up through external means, through the use of location lookups. The syntax of these tags is given in Figure 10.
Location lookup is dependent on the underlying signalling protocol.
Node: "lookup" Outputs: "success" Next node if lookup was successful "notfound" Next node if lookup found no addresses "failure" Next node if lookup failed Parameters: "source" Source of the lookup "timeout" Time to try before giving up on the lookup "clear" Whether to clear the location set before adding the new values
Output: "success" Parameters: none
Output: "notfound" Parameters: none
Output: "failure" Parameters: none
Figure 10: Syntax of the "lookup" node
Location lookup nodes have one mandatory parameter and two optional parameters. The mandatory parameter is "source", the source of the lookup. This can either be a URI, or a non-URI value. If the value of "source" is a URI, it indicates a location which the CPL server can query to obtain an object with the text/uri-list media type (see the IANA registration of this type, which also appears in RFC 2483 ). The query is performed verbatim, with no additional information (such as URI parameters) added. The server adds the locations contained in this object to the location set.
CPL servers MAY refuse to allow URI-based sources for location queries for some or all URI schemes. In this case, they SHOULD reject the script at script upload time.
There has been discussion of having CPL servers add URI parameters to the location request, so that (for instance) CGI scripts could be used to resolve them. However, the consensus was that this should be a CPL extension, not a part of the base specification.
Lennox, et al. Standards Track [Page 24]
RFC 3880 CPL October 2004
Non-URL sources indicate a source not specified by a URL which the server can query for addresses to add to the location set. The only non-URL source currently defined is "registration", which specifies all the locations currently registered with the server.
The "lookup" node also has two optional parameters. The "timeout" parameter specifies the time, as a positive integer number of seconds, the script is willing to wait for the lookup to be performed. If this is not specified, its default value is 30. The "clear" parameter specifies whether the location set should be cleared before the new locations are added.
Lookup has three outputs: "success", "notfound", and "failure". Notfound is taken if the lookup process succeeded but did not find any locations; failure is taken if the lookup failed for some reason, including that the specified timeout was exceeded. If a given output is not present, script execution terminates and the default behavior is performed.
A CPL script can also remove locations from the location set, through the use of the "remove-location" node. The syntax of this node is defined in Figure 11.
The meaning of this node is dependent on the underlying signalling Protocol.
Node: "remove-location" Outputs: None (Next node follows directly) Next node: Any node Parameters: "location" Location to remove
Figure 11: Syntax of the "remove-location" node
A "remove-location" node removes locations from the location set. It is primarily useful following a "lookup" node. An example of this is given in Section 12.8.
The "remove-location" node has one optional parameter. The parameter "location" gives the URI of a location to be removed from the set, in a signalling-protocol-dependent manner. If this parameter is not given, all locations are removed from the set.
Lennox, et al. Standards Track [Page 25]
RFC 3880 CPL October 2004
The "remove-location" node has no explicit output tags. In the XML syntax, the XML "remove-location" tag directly encloses the next node's tag.
The location specified in the "location" parameter of the "remove- location" node is matched against the location set using the standard rules for SIP URI matching (as are used, e.g., to match Contact addresses when refreshing registrations).
Proxy causes the triggering call to be forwarded on to the currently specified set of locations. The syntax of the proxy node is given in Figure 12.
The specific signalling events invoked by the "proxy" node are signalling-protocol-dependent, though the general concept should apply to any signalling protocol.
Lennox, et al. Standards Track [Page 26]
RFC 3880 CPL October 2004
Node: "proxy" Outputs: "busy" Next node if call attempt returned "busy" "noanswer" Next node if call attempt was not answered before timeout "redirection" Next node if call attempt was redirected "failure" Next node if call attempt failed "default" Default next node for unspecified outputs Parameters: "timeout" Time to try before giving up on the call attempt "recurse" Whether to recursively look up redirections "ordering" What order to try the location set in.
Output: "busy" Parameters: none
Output: "noanswer" Parameters: none
Output: "redirection" Parameters: none
Output: "failure" Parameters: none
Output: "default" Parameters: none
Figure 12: Syntax of the "proxy" node
After a proxy operation has completed, the CPL server chooses the "best" response to the call attempt, as defined by the signalling protocol or the server's administrative configuration rules.
If the call attempt was successful, CPL execution terminates and the server proceeds to its default behavior (normally, to allow the call to be set up). Otherwise, the next node corresponding to one of the "proxy" node's outputs is taken. The "busy" output is followed if the call was busy, "noanswer" is followed if the call was not answered before the "timeout" parameter expired, "redirection" is followed if the call was redirected, and "failure" is followed if the call setup failed for any other reason.
If one of the conditions above is true, but the corresponding output was not specified, the "default" output of the "proxy" node is followed instead. If there is also no "default" node specified, CPL execution terminates and the server returns to its default behavior (normally, to forward the best response upstream to the originator).
Lennox, et al. Standards Track [Page 27]
RFC 3880 CPL October 2004
Note: CPL extensions to allow in-call or end-of-call operations will require an additional output, such as "success", to be added.
If no locations were present in the set, or if the only locations in the set were locations to which the server cannot proxy a call (for example, "http" URLs), the "failure" output is taken.
Proxy has three optional parameters. The "timeout" parameter specifies the time, as a positive integer number of seconds, to wait for the call to be completed or rejected; after this time has elapsed, the call attempt is terminated and the "noanswer" branch is taken. If this parameter is not specified, the default value is 20 seconds if the "proxy" node has a "noanswer" or "default" output specified; otherwise the server SHOULD allow the call to ring for a reasonably long period of time (to the maximum extent that server policy allows).
The second optional parameter is "recurse", which can take two values, "yes" or "no". This specifies whether the server should automatically attempt to place further call attempts to telephony addresses in redirection responses that were returned from the initial server. Note that if the value of "recurse" is "yes", the "redirection" output to the script is never taken. In this case this output SHOULD NOT be present. The default value of this parameter is "yes".
The third optional parameter is "ordering". This can have three possible values: "parallel", "sequential", and "first-only". This parameter specifies in what order the locations of the location set should be tried. Parallel asks that they all be tried simultaneously; sequential asks that the one with the highest priority be tried first, the one with the next-highest priority second, and so forth, until one succeeds or the set is exhausted. First-only instructs the server to try only the highest-priority address in the set, and then follow one of the outputs. The priority of locations in a set is determined by server policy, though CPL servers SHOULD honor the "priority" parameter of the "location" tag. The default value of this parameter is "parallel".
Once a proxy operation completes, if control is passed on to other nodes, all locations which have been used are cleared from the location set. That is, the location set is emptied of proxyable locations if the "ordering" was "parallel" or "sequential"; the highest-priority item in the set is removed from the set if "ordering" was "first-only". (In all cases, non-proxyable locations such as "http" URIs remain.) In the case of a "redirection" output, the new addresses to which the call was redirected are then added to the location set.
Redirect causes the server to direct the calling party to attempt to place its call to the currently specified set of locations. The syntax of this node is specified in Figure 13.
The specific behavior the redirect node invokes is dependent on the underlying signalling protocol involved, though its semantics are generally applicable.
Node: "redirect" Outputs: None (No node may follow) Next node: None Parameters: "permanent" Whether the redirection should be considered permanent
Figure 13: Syntax of the "redirect" node
Redirect immediately terminates execution of the CPL script, so this node has no outputs and no next node. It has one parameter, "permanent", which specifies whether the result returned should indicate that this is a permanent redirection. The value of this parameter is either "yes" or "no" and its default value is "no."
The SIP server SHOULD send a 3xx class response to a call request upon executing a "redirect" tag. If "permanent" was "yes", the server SHOULD send the response "301" (Moved permanently), otherwise it SHOULD send "302" (Moved temporarily).
Reject nodes cause the server to reject the call attempt. Their syntax is given in Figure 14. The specific behavior they invoke is dependent on the underlying signalling protocol involved, though their semantics are generally applicable.
Node: "reject" Outputs: None (No node may follow) Next node: None Parameters: "status" Status code to return "reason" Reason phrase to return
Figure 14: Syntax of the "reject" node
A reject node immediately terminates the execution of a CPL script, so this node has no outputs and no next node.
This node has two arguments: "status" and "reason". The "status" argument is required, and can take one of the values "busy", "notfound", "reject", "error", or a signalling-protocol-defined status.
The "reason" argument optionally allows the script to specify a reason for the rejection.
The mail node causes the server to notify a user of the status of the CPL script through electronic mail. Its syntax is given in Figure 15.
Node: "mail" Outputs: None (Next node follows directly) Next node: Any node Parameters: "url" Mailto url to which the mail should be sent
Figure 15: Syntax of the "mail" node
The "mail" node takes one argument: a "mailto" URL giving the address, and any additional desired parameters, of the mail to be sent. The server sends the message containing the content to the given url; it SHOULD also include other status information about the original call request and the CPL script at the time of the notification.
Using a full "mailto" URL rather than just an e-mail address allows additional e-mail headers to be specified, such as <mail url="mailto:firstname.lastname@example.org?subject=Lookup%20failed" />.
A mail node has only one possible result, since failure of e-mail delivery cannot reliably be known in real time. Therefore, its XML representation does not have output tags: the <mail> tag directly contains another node tag.
Note that the syntax of XML requires that ampersand characters, "&", which are used as parameter separators in "mailto" URLs, be quoted as "&" inside parameter values (see Section C.12 of the XML specification ).
This section presents suggested guidelines for the mail sent as a result of the "mail" node, for requests triggered by SIP. The message mailed (triggered by any protocol) SHOULD contain all this information, but servers MAY elect to use a different format.
1. If the "mailto" URI did not specify a subject header, the subject of the e-mail is "[CPL]", followed by the subject header of the SIP request. If the URI specified a subject header, it is used instead.
2. The "From" field of the e-mail is set to a CPL server configured address, overriding any "From" field in the "mailto" URI.
3. Any "Reply-To" header in the URI is honored. If none is given, then an e-mail-ized version of the origin field of the request is used, if possible (e.g., a SIP "From" header with a sip: URI would be converted to an e-mail address by stripping the URI scheme).
4. If the "mailto" URI specifies a body, it is used. If none was specified, the body SHOULD contain at least the identity of the caller (both the caller's display name and address), the date and time of day, the call subject, and if available, the call priority.
The server SHOULD honor the user's requested languages, and send the mail notification using an appropriate language and character set.
The Log node causes the server to log information about the call to non-volatile storage. Its syntax is specified in Figure 16.
Node: "log" Outputs: None (Next node follows directly) Next node: Any node Parameters: "name" Name of the log file to use "comment" Comment to be placed in log file
Figure 16: Syntax of the "log" node
Log takes two arguments, both optional: "name", which specifies the name of the log, and "comment", which gives a comment about the information being logged. Servers SHOULD also include other information in the log, such as the time of the logged event,
Lennox, et al. Standards Track [Page 32]
RFC 3880 CPL October 2004
information that triggered the call to be logged, and so forth. Logs are specific to the owner of the script which logged the event. If the "name" parameter is not given, the event is logged to a standard, server-defined log file for the script owner. This specification does not define how users may retrieve their logs from the server.
The name of a log is a logical name only, and does not necessarily correspond to any physical file on the server. The interpretation of the log file name is server defined, as is a mechanism to access these logs. The CPL server SHOULD NOT directly map log names uninterpreted onto local file names, for security reasons, lest a security-critical file be overwritten.
A correctly operating CPL server SHOULD NOT ever allow the "log" event to fail. As such, log nodes can have only one possible result, and their XML representation does not have explicit output tags. A CPL <log> tag directly contains another node tag.
XML syntax defines a tree. To allow more general call flow diagrams, and to allow script re-use and modularity, we define subactions.
Two tags are defined for subactions: subaction definitions and subaction references. Their syntax is given in Figure 17.
Tag: "subaction" Subtags: Any node Parameters: "id" Name of this subaction
Pseudo-node: "sub" Outputs: None in XML tree Parameters: "ref" Name of subaction to execute
Figure 17: Syntax of subactions and "sub" pseudo-nodes
Subactions are defined through "subaction" tags. These tags are placed in the CPL script after any ancillary information (see Section 9), but before any top-level tags. They take one argument: "id", a token indicating a script-chosen name for the subaction. The "id" value for every "subaction" tag in a script MUST be unique within that script.
Subactions are called from "sub" tags. The "sub" tag is a "pseudo- node", and can be used anyplace in a CPL action that a true node could be used. It takes one parameter, "ref", the name of the subaction to be called. The "sub" tag contains no outputs of its own, instead control passes to the subaction.
Lennox, et al. Standards Track [Page 33]
RFC 3880 CPL October 2004
References to subactions MUST refer to subactions defined before the current action. A "sub" tag MUST NOT refer to the action it appears in, or to any action defined later in the CPL script. Top-level actions cannot be called from "sub" tags, or through any other means. Script servers MUST verify at the time the script is submitted that no "sub" node refers to any subaction that is not its proper predecessor.
Allowing only back-references of subs forbids any sort of recursion. Recursion would introduce the possibility of non- terminating or non-decidable CPL scripts, a possibility our requirements specifically excluded.
Every sub MUST refer to a subaction ID defined within the same CPL script. No external links are permitted.
Subaction IDs are case sensitive.
If any subsequent version or extension defines external linkages, it should probably use a different tag, perhaps XLink . Ensuring termination in the presence of external links is a difficult problem.
No ancillary information is defined in the base CPL specification. If ancillary information, not part of any operation, is found to be necessary for a CPL extension, it SHOULD be placed within this tag.
The (trivial) definition of the ancillary information tag is given in Figure 18.
It may be useful to include timezone definitions inside CPL scripts directly, rather than referencing them externally with "tzid" and "tzurl" parameters. If it is, an extension could be defined to include them here.
When a CPL node reaches an unspecified output, either because the output tag is not present, or because the tag is present but does not contain a node, the CPL server's behavior is dependent on the current state of script execution. This section gives the operations that should be taken in each case.
no location modifications or signalling operations performed, location set empty: Look up the user's location through whatever mechanism the server would use if no CPL script were in effect. Proxy, redirect, or send a rejection message, using whatever policy the server would use in the absence of a CPL script.
no location modifications or signalling operations performed, location set non-empty: (This can only happen for outgoing calls.) Proxy the call to the addresses in the location set.
location modifications performed, no signalling operations: Proxy or redirect the call, whichever is the server's standard policy, to the addresses in the current location set. If the location set is empty, return a "notfound" rejection.
noanswer output of proxy, no timeout given: (This is a special case.) If the "noanswer" output of a proxy node is unspecified, and no timeout parameter was given to the proxy node, the call should be allowed to ring for the maximum length of time allowed by the server (or the request, if the request specified a timeout).
proxy operation previously taken: Return whatever the "best" response is of all accumulated responses to the call to this point, according to the rules of the underlying signalling protocol.
Servers MAY support additional CPL features beyond those listed in this document. Some of the extensions which have been suggested are a means of querying how a call has been authenticated, richer control over H.323 addressing, end-system or administrator-specific features, regular-expression matching for strings and addresses, and mid-call or end-of-call controls.
CPL extensions are indicated by XML namespaces . Every extension MUST have an appropriate XML namespace assigned to it. The XML namespace of the extension MUST be different from the XML namespace
Lennox, et al. Standards Track [Page 35]
RFC 3880 CPL October 2004
defined in Section 14. The extension MUST NOT change the syntax or semantics of the CPL schema defined in this document. All XML tags and attributes that are part of the extension MUST be appropriately qualified so as to place them within that namespace.
Tags or attributes in a CPL script which are in the global namespace (i.e., not associated with any namespace) are equivalent to tags and attributes in the CPL namespace "urn:ietf:params:xml:ns:cpl".
A CPL script SHOULD NOT specify any namespaces it does not use. For compatibility with non-namespace-aware parsers, a CPL script MAY omit the base CPL namespace for a script which does not use any extensions.
A CPL server MUST reject any script containing a reference to a namespace it does not understand. It MUST reject any script containing an extension tag or attribute that is not qualified to be in an appropriate namespace.
was suggested as an alternate way of handling extensions. This would allow scripts to be uploaded to a server without requiring a script author to somehow determine which extensions a server supports. However, experience developing other languages, notably Sieve , was that this added excessive complexity to languages. The "extension-switch" tag could, of course, itself be defined in a CPL extension.
In the XML schema of CPL, we introduce three abstract elements, namely 'toplevelaction', 'switch', and 'action', which accordingly have the abstract type 'TopLevelActionType', 'SwitchType', and 'ActionType'. Any top-level action in a CPL extension MUST be defined as the substitutionGroup of the abstract 'toplevelaction' element, and have the type extended from the 'TopLevelActionType'. Any switch in a CPL extension MUST be defined as the substitutionGroup of the abstract 'switch' element, and have the type
Lennox, et al. Standards Track [Page 36]
RFC 3880 CPL October 2004
extended from the 'SwitchType'. Any action in a CPL extension MUST be defined as the substitutionGroup of the abstract 'action' element, and have the type extended from the 'ActionType'.
The script in Figure 20 illustrates some more complex behavior. We see an initial proxy attempt to one address, with further operations if that fails. We also see how several outputs take the same action subtree, through the use of subactions.
The script in Figure 21 illustrates further proxy behavior. The server initially tries to proxy to a single address. If this attempt is redirected, a new redirection is generated using the locations returned. In all other failure cases for the proxy node, a default operation -- forwarding to voicemail -- is performed.
The script in Figure 22 illustrates address switches and call rejection, in the form of a call screening script. Note also that because the address-switch lacks an "otherwise" clause, if the initial pattern does not match, the script does not define any operations. The server therefore proceeds with its default behavior, which would presumably be to contact the user.
The script in Figure 23 illustrates service selection based on a call's priority value and language settings. If the call request had a priority of "urgent" or higher, the default script behavior is performed. Otherwise, the language field is checked for the language "es" (Spanish). If it is present, the call is proxied to a Spanish- speaking operator; other calls are proxied to an English-speaking operator.
Figure 26 illustrates filtering operations on the location set. In this example, we assume that version 0.9beta2 of the "Inadequate Software SIP User Agent" mis-implements some features, and so we must work around its problems. We know that it cannot talk successfully to one particular mobile device we may have registered, so we remove that location from the location set. Once this operation has been completed, call setup is allowed to proceed normally.
Figure 27 illustrates non-signalling operations; in particular, alerting a user by electronic mail if the lookup server failed. The primary motivation for having the "mail" node is to allow this sort of out-of-band notification of error conditions, as the user might otherwise be unaware of any problem.
Finally, Figure 30 is a complex example which shows the sort of sophisticated behavior that can be achieved by combining CPL nodes. In this case, the user attempts to have his calls reach his desk; if he does not answer within a small amount of time, calls from his boss are forwarded to his mobile phone, and all other calls are directed to voicemail. If the call setup failed, no operation is specified, so the server's default behavior is performed.
CPL is designed to allow services to be specified in a manner which prevents potentially hostile or mis-configured scripts from launching security attacks, including denial-of-service attacks. Because script runtime is strictly bounded by acyclicity, and because the number of possible script operations are strictly limited, scripts should not be able to inflict damage upon a CPL server.
Because scripts can direct users' telephone calls, the method by which scripts are transmitted from a client to a server MUST be strongly authenticated. Such a method is not specified in this document.
Script servers SHOULD allow server administrators to control the details of what CPL operations are permitted.
This document registers a new MIME type, application/cpl+xml, and a new URN per RFC 2141 , RFC 2648 , and RFC 3688 .
The XML namespace urn:ietf:params:xml:ns:cpl will only refer to the version of CPL in this document and will not change. Any CPL enhancements MUST be made by extensions and MUST have different namespaces.
14.1. URN Sub-Namespace Registration for urn:ietf:params:xml:ns:cpl
Interoperability considerations: Different CPL servers may use incompatible address types. However, all potential interoperability issues should be resolvable at the time a script is uploaded; there should be no interoperability issues which cannot be detected until runtime.
Published specification: This document.
Lennox, et al. Standards Track [Page 50]
RFC 3880 CPL October 2004
Applications which use this media type: SIP proxy servers and other telephony servers, and client software to control their behavior.
Magic number: None
File extension: .cpl or .xml
Macintosh file type code: "TEXT"
Person and e-mail address for further information: Jonathan Lennox <email@example.com> Xiaotao Wu <firstname.lastname@example.org> Henning Schulzrinne <email@example.com>
This document was reviewed and commented upon by the IETF IP Telephony Working Group. We specifically acknowledge the following people for their help:
The outgoing call screening script was written by Kenny Hom.
Paul E. Jones contributed greatly to the mappings of H.323 addresses.
The text of the time-switch section was taken (lightly modified) from RFC 2445 , by Frank Dawson and Derik Stenerson.
We drew a good deal of inspiration, notably the language's lack of Turing-completeness and the syntax of string matching, from the specification of Sieve , a language for user filtering of electronic mail messages.
Thomas F. La Porta and Jonathan Rosenberg had many useful discussions, contributions, and suggestions.
Richard Gumpertz performed a very useful last-minute technical and editorial review of the specification.
The following algorithm determines whether a given instant falls within a repetition of a "time-switch" recurrence. If the pre- processing described in Section 4.4.1 has been done, it operates in constant time. Open-source Java code implementing this algorithm is available at http://www.cs.columbia.edu/~lennox/Cal-Code/ on the world wide web.
This algorithm is believed to be correct, but this section is non- normative. Section 4.4, and RFC 2445 , are the definitive definitions of recurrences.
1. Compute the time of the call, in the timezone of the time switch.
2. If the call time is earlier than "dtstart", fail NOMATCH.
3. If the call time is less than "duration" after dtstart, succeed MATCH.
4. Determine the smallest unit specified in a "byxxx" rule or by the "freq." Call this the Minimum Unit. Determine the previous instant (before or equal to the call time) when all the time units smaller than the minimum unit are the same as those of "dtstart." If the minimum unit is a second, this time is the same as the instant. If the minimum unit is a minute or an hour, the minutes or the minutes and hours, respectively, must be the same as "dtstart". For all other minimum units, the time-of-day must be the same as "dtstart." If the minimum unit is a week, the day-of-the-week must be the same as "dtstart." If the minimum unit is a month, the day-of-the- month must be the same as "dtstart." If the minimum unit is a year, the month and day-of-month must both be the same as "dtstart." (Note that this means it may be necessary to roll back more than one minimum unit -- if the minimum unit is a month, then some months do not have a 31st (or 30th or 29th) day; if the minimum unit is a year, then some years do not have a February 29th. In the Gregorian calendar, it is never necessary to roll back more than two months if the minimum unit is a month, or eight years if the minimum unit is a year. Between 1904 and 2096, it is never necessary to roll back more than four years -- the eight-year rollback can only occur when the Gregorian calendar "skips" a leap year.
Call this instant the Candidate Start Time.
Lennox, et al. Standards Track [Page 52]
RFC 3880 CPL October 2004
5. If the time between the candidate start time and the call time is more than the duration, fail NOMATCH.
6. If the candidate start time is later than the "until" parameter of the recurrence (or the virtual "until" computed off-line from "count"), fail NOMATCH.
7. Call the unit of the "freq" parameter of the recurrence the Frequency Unit. Determine the frequency unit enclosing the Candidate Start Time, and that enclosing "dtstart". Calculate the number of frequency units that have passed between these two times. If this is not a multiple of the "interval" parameter, fail NOMATCH.
8. For every "byxxx" rule, confirm that the candidate start time matches one of the options specified by that "byxxx" rule. If so, succeed MATCH.
9. Calculate a previous candidate start time. Repeat until the difference between the candidate start time and the call time is more than the duration. If no candidate start time has been validated, fail NOMATCH.
This appendix gives a suggested usage of CPL with H.323 . Study Group 16 of the ITU, which developed H.323, is proposing to work on official CPL mappings for that protocol. This section is therefore not normative.
Address switches are specified in Section 4.1. This section specifies the mapping between H.323 messages and the fields and subfields of address-switches.
For H.323, the "origin" address corresponds to the alias addresses in the "sourceAddress" field of the "Setup-UUIE" user-user information element, and to the Q.931  information element "Calling party number." If both fields are present, or if multiple alias addresses for "sourceAddress" are present, which one has priority is a matter of local server policy; the server SHOULD use the same resolution as it would use for routing decisions in this case. Similarly, the "destination" address corresponds to the alias addresses of the "destinationAddress" field, and to the Q.931 information element "Called party number."
Lennox, et al. Standards Track [Page 53]
RFC 3880 CPL October 2004
The "original-destination" address corresponds to the "Redirecting number" Q.931 information element, if it is present; otherwise it is the same as the "destination" address.
The mapping of H.323 addresses into subfields depends on the type of the alias address. An additional subfield type, "alias-type", is defined for H.323 servers, corresponding to the type of the address. Possible values are "dialedDigits", "h323-ID", "url-ID", "transportID", "email-ID", "partyNumber", "mobileUIM", and "Q.931IE". If future versions of the H.323 specification define additional types of alias addresses, those names MAY also be used.
In versions of H.323 prior to version 4, "dialedDigits" was known as "e164". The two names SHOULD be treated as synonyms.
The value of the "address-type" subfield for H.323 messages is "h323" unless the alias type is "url-ID" and the URL scheme is something other than h323; in this case the address-type is the URL scheme, as specified in Section 4.1.1 for SIP.
An H.323-aware CPL server SHOULD map the address subfields from the primary alias used for routing. It MAY also map subfields from other aliases, if subfields in the primary address are not present.
The following mappings are used for H.323 alias types:
dialedDigits, partyNumber, mobileUIM, and Q.931IE: the "tel" and "user" subfields are the string of digits, as is the "entire-address" form. The "host" and "port" subfields are not present.
h323-ID: the "user" field is the string of characters, as is the "entire-address" form. All other subfields are not present.
email-ID: the "user" and "host" subfields are set to the corresponding parts of the e-mail address. The "port" and "tel" subfields are not present. The "entire-address" form corresponds to the entire e-mail address.
transportID: if the TransportAddress is of type "ipAddress," "ipSourceRoute," or "ip6Address," the "host" subfield is set to the "ip" element of the sequence, translated into the standard IPv4 or IPv6 textual representation, and the "port" subfield is set to the "port" element of the sequence represented in decimal. The "tel" and "user" fields are not present. The "entire-address" form is not defined. The
Lennox, et al. Standards Track [Page 54]
RFC 3880 CPL October 2004
representation and mapping of transport addresses is not defined for non-IP addresses.
H.323  defines an "h323" URI scheme. This appendix defines a mapping for these URIs onto the CPL "address-switch" subfields, as given in Section 4.1. This definition is also available as RFC 3508 , which is an excerpt from the H.323 specification.
For h323 URIs, the "user", "host", and "port" subfields are set to the corresponding parts of the H.323 URL. The "tel" subfield is not present. The "entire-address" form corresponds to the entire URI.
This mapping MAY be used both for h323 URIs in an h323 "url-ID" address alias, and for h323 URIs in SIP messages.
For H.323, the "string-switch" node (see Section 4.2) is used as follows. The field "display" corresponds to the Q.931 information element of the same name, copied verbatim. The fields "subject", "organization", and "user-agent" are not used and are never present.
The "display" IE is conventionally used for Caller-ID purposes, so arguably it should be mapped to the "display" subfield of an "address-match" with the field "originator". However, since a) it is a message-level information element, not an address-level one, and b) the Q.931 specification  says only that "[t]he purpose of the Display information element is to supply display information that may be displayed by the user," it seems to be more appropriate to allow it to be matched in a "string-switch" instead.
Location removal nodes (Section 5.3) remove addresses with the alias type "url-ID" using verbatim string matching on the URLs. If a "tel" URL is specified as the location, matching addresses (ignoring visual separators) with the alias types "dialedDigits" ("e164"), "partyNumber", "mobileUIM", or "Q.931IE" are also removed. No mechanism is provided to remove other alias types.
This section includes a full XML Schema describing the XML syntax of CPL. Every script submitted to a CPL server SHOULD comply with this XML Schema. When parsing scripts comply with the CPL DTD in earlier documents, the DOCTYPE lines in the scripts should be ignored. Note that compliance with this schema is not a sufficient condition for correctness of a CPL script, as many of the conditions described in this specification are not expressible in schema syntax. Figure 31 shows the structure of the schema. 'incoming' and 'outgoing' are defined as the substitutionGroup of the 'toplevelaction'. All the switches are defined as the substitutionGroup of the 'switch' element. All the actions are defined as the substitutionGroup of the 'action' element.
</xs:simpleType> <xs:simpleType name="DayType"> <xs:restriction base="xs:NMTOKEN"> <xs:pattern value="[m|M][o|O]"/> <xs:pattern value="[t|T][u|U]"/> <xs:pattern value="[w|W][e|E]"/> <xs:pattern value="[t|T][h|H]"/> <xs:pattern value="[f|F][r|R]"/> <xs:pattern value="[s|S][a|A]"/> <xs:pattern value="[s|S][u|U]"/> </xs:restriction> </xs:simpleType> <xs:complexType name="TimeType"> <xs:annotation> <xs:documentation>Exactly one of the two attributes "dtend" and "duration" must occur. None of the attributes following freq are meaningful unless freq appears. </xs:documentation> </xs:annotation> <xs:group ref="Node"/> <xs:attribute name="dtstart" type="xs:string" use="required"> <xs:annotation> <xs:documentation>RFC 2445 DATE-TIME</xs:documentation> </xs:annotation> </xs:attribute> <xs:attribute name="dtend" type="xs:string" use="optional"> <xs:annotation> <xs:documentation>RFC 2445 DATE-TIME</xs:documentation> </xs:annotation> </xs:attribute> <xs:attribute name="duration" type="xs:string" use="optional"> <xs:annotation> <xs:documentation>RFC 2445 DURATION</xs:documentation> </xs:annotation> </xs:attribute> <xs:attribute name="freq" type="FreqType" use="optional"/> <xs:attribute name="interval" type="xs:positiveInteger" default="1"/> <xs:attribute name="until" type="xs:string" use="optional"> <xs:annotation> <xs:documentation>RFC 2445 DATE-TIME</xs:documentation> </xs:annotation> </xs:attribute> <xs:attribute name="count" type="xs:positiveInteger" use="optional"/> <xs:attribute name="bysecond" type="xs:string" use="optional"> <xs:annotation> <xs:documentation>Comma-separated list of seconds within a
Lennox, et al. Standards Track [Page 63]
RFC 3880 CPL October 2004
minute. Valid values are 0 to 59.</xs:documentation> </xs:annotation> </xs:attribute> <xs:attribute name="byminute" type="xs:string" use="optional"> <xs:annotation> <xs:documentation>Comma-separated list of minutes within an hour. Valid values are 0 to 59.</xs:documentation> </xs:annotation> </xs:attribute> <xs:attribute name="byhour" type="xs:string" use="optional"> <xs:annotation> <xs:documentation>Comma-separated list of hours of the day. Valid values are 0 to 23.</xs:documentation> </xs:annotation> </xs:attribute> <xs:attribute name="byday" type="xs:string" use="optional"> <xs:annotation> <xs:documentation>Comma-separated list of days of the week. Valid values are "MO", "TU", "WE", "TH", "FR", "SA" and "SU". These values are not case-sensitive. Each can be preceded by a positive (+n) or negative (-n) integer.</xs:documentation> </xs:annotation> </xs:attribute> <xs:attribute name="bymonthday" type="xs:string" use="optional"> <xs:annotation> <xs:documentation>Comma-separated list of days of the month. Valid values are 1 to 31 or -31 to -1.</xs:documentation> </xs:annotation> </xs:attribute> <xs:attribute name="byyearday" type="xs:string" use="optional"> <xs:annotation> <xs:documentation>Comma-separated list of days of the year. Valid values are 1 to 366 or -366 to -1.</xs:documentation> </xs:annotation> </xs:attribute> <xs:attribute name="byweekno" type="xs:string" use="optional"> <xs:annotation> <xs:documentation>Comma-separated list of ordinals specifying weeks of the year. Valid values are 1 to 53 or -53 to -1.</xs:documentation> </xs:annotation> </xs:attribute> <xs:attribute name="bymonth" type="xs:string" use="optional"> <xs:annotation> <xs:documentation>Comma-separated list of months of the year.
<xs:sequence> <xs:element name="ancillary" type="AncillaryType" minOccurs="0" maxOccurs="1"/> <xs:element name="subaction" type="SubactionType" minOccurs="0" maxOccurs="unbounded"/> <xs:element ref="toplevelaction" minOccurs="0" maxOccurs="unbounded"> <xs:annotation> <xs:documentation>Any toplevel action MUST NOT appear more than once.</xs:documentation> </xs:annotation> </xs:element> </xs:sequence> </xs:complexType> <xs:element name="cpl" type="CPLType"/> </xs:schema> END
 Rosenberg, J., Schulzrinne, H., Camarillo, G., Johnston, A., Peterson, J., Sparks, R., Handley, M., and E. Schooler, "SIP: Session Initiation Protocol", RFC 3261, June 2002.
 Bray, T., Paoli, J., Sperberg-McQueen, C. M., Maler, E., and F. Yergeau, "Extensible Markup Language (XML) 1.0 (Third Edition)", W3C Recommendation REC-xml-20040204, World Wide Web Consortium (W3C), February 2004. Available at http://www.w3.org/XML/.
 Bradner, S., "Key words for use in RFCs to Indicate Requirement Levels", BCP 14, RFC 2119, March 1997.
 Hinden, R. and S. Deering, "Internet Protocol Version 6 (IPv6) Addressing Architecture", RFC 3513, April 2003.
 Mealling, M. and R. Daniel, "URI Resolution Services Necessary for URN Resolution", RFC 2483, January 1999.
 Bray, T., Hollander, D., and A. Layman, "Namespaces in XML", W3C Recommendation REC-xml-names-19990114, World Wide Web Consortium (W3C), January 1999. Available at http://www.w3.org/TR/REC- xml-names/.
 Moats, R., "A URN Namespace for IETF Documents", RFC 2648, August 1999.
 Mealling, M., "The IETF XML Registry", BCP 81, RFC 3688, January 2004.
 Murata, M., St.Laurent, S., and D. Kohn, "XML Media Types", RFC 3023, January 2001.
 International Telecommunication Union, "Packet-based multimedia communication systems", Recommendation H.323, Telecommunication Standardization Sector of ITU, Geneva, Switzerland, July 2003.
 Lennox, J. and H. Schulzrinne, "Call Processing Language Framework and Requirements", RFC 2824, May 2000.
 Raggett, D., Le Hors, A., and I. Jacobs, "HTML 4.01 Specification", W3C Recommendation REC-html401-19991224, World Wide Web Consortium (W3C), December 1999. Available at http://www.w3.org/TR/html4/.
 ISO (International Organization for Standardization), "Information processing -- Text and office systems -- Standard Generalized Markup Language (SGML)", ISO Standard ISO 8879:1986(E), International Organization for Standardization, Geneva, Switzerland, October 1986.
Lennox, et al. Standards Track [Page 71]
RFC 3880 CPL October 2004
 ISO (International Organization for Standardization), "Data elements and interchange formats -- Information interchange -- Representation of dates and times", ISO Standard ISO 8601:2000(E), International Organization for Standardization, Geneva, Switzerland, December 2000.
 DeRose, S., Maler, E., Orchard, D., and B. Trafford, "XML Linking Language (XLink) Version 1.0", W3C Recommendation REC- xlink-20010627, World Wide Web Consortium (W3C), June 2001. Available at http://www.w3.org/TR/xlink/.
 Showalter, T., "Sieve: A Mail Filtering Language", RFC 3028, January 2001.
 International Telecommunication Union, "Digital Subscriber Signalling System No. 1 (DSS 1) - ISDN user-network interface layer 3 specification for basic call control", Recommendation Q.931, International Telecommunication Union, Geneva, Switzerland, March 1993.
Jonathan Lennox Dept. of Computer Science Columbia University 1214 Amsterdam Avenue, MC 0401 New York, NY 10027 USA
Xiaotao Wu Dept. of Computer Science Columbia University 1214 Amsterdam Avenue, MC 0401 New York, NY 10027 USA
Henning Schulzrinne Dept. of Computer Science Columbia University 1214 Amsterdam Avenue, MC 0401 New York, NY 10027 USA
Lennox, et al. Standards Track [Page 73]
RFC 3880 CPL October 2004
Full Copyright Statement
Copyright (C) The Internet Society (2004).
This document is subject to the rights, licenses and restrictions contained in BCP 78, and except as set forth therein, the authors retain all their rights.
This document and the information contained herein are provided on an "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY AND THE INTERNET ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
The IETF takes no position regarding the validity or scope of any Intellectual Property Rights or other rights that might be claimed to pertain to the implementation or use of the technology described in this document or the extent to which any license under such rights might or might not be available; nor does it represent that it has made any independent effort to identify any such rights. Information on the IETF's procedures with respect to rights in IETF Documents can be found in BCP 78 and BCP 79.
Copies of IPR disclosures made to the IETF Secretariat and any assurances of licenses to be made available, or the result of an attempt made to obtain a general license or permission for the use of such proprietary rights by implementers or users of this specification can be obtained from the IETF on-line IPR repository at http://www.ietf.org/ipr.
The IETF invites any interested party to bring to its attention any copyrights, patents or patent applications, or other proprietary rights that may cover technology that may be required to implement this standard. Please address the information to the IETF at ietf- firstname.lastname@example.org.
Funding for the RFC Editor function is currently provided by the Internet Society.