+=head2 Delete handler
+
+The argument hash recieved by the delete handler has the following elements:
+
+ $args = {
+ ## Client request:
+ GHANDLE => $obj, ## Global handler specified at creation
+ HANDLE => ref, ## Reference to data structure
+ SETNAME => "id", ## Result set ID
+
+ ## Server response:
+ STATUS => 0 ## Deletion status
+ };
+
+The SETNAME element of the argument hash may or may not be defined.
+If it is, then SETNAME is the name of a result set to be deleted; if
+not, then all result-sets associated with the current session should
+be deleted. In either case, the callback function should report on
+success or failure by setting the STATUS element either to zero, on
+success, or to an integer from 1 to 10, to indicate one of the ten
+possible failure codes described in section 3.2.4.1.4 of the Z39.50
+standard -- see
+http://www.loc.gov/z3950/agency/markup/05.html#Delete-list-statuses1
+
+=head2 Sort handler
+
+The argument hash recieved by the sort handler has the following elements:
+
+ $args = {
+ ## Client request:
+ GHANDLE => $obj, ## Global handler specified at creation
+ HANDLE => ref, ## Reference to data structure
+ INPUT => [ a, b ... ], ## Names of result-sets to sort
+ OUTPUT => "name", ## Name of result-set to sort into
+ SEQUENCE ## Sort specification: see below
+
+ ## Server response:
+ STATUS => 0, ## Success, Partial or Failure
+ ERR_CODE => 0, ## Error code
+ ERR_STR => '', ## Diagnostic message
+
+ };
+
+The SEQUENCE element is a reference to an array, each element of which
+is a hash representing a sort key. Each hash contains the following
+elements:
+
+=over 4
+
+=item RELATION
+
+0 for an ascending sort, 1 for descending, 3 for ascending by
+frequency, or 4 for descending by frequency.
+
+=item CASE
+
+0 for a case-sensitive sort, 1 for case-insensitive
+
+=item MISSING
+
+How to respond if one or more records in the set to be sorted are
+missing the fields indicated in the sort specification. 1 to abort
+the sort, 2 to use a "null value", 3 if a value is provided to use in
+place of the missing data (although in the latter case, the actual
+value to use is currently not made available, so this is useless).
+
+=back
+
+And one or other of the following:
+
+=over 4
+
+=item SORTFIELD
+
+A string indicating the field to be sorted, which the server may
+interpret as it sees fit (presumably by an out-of-band agreement with
+the client).
+
+=item ELEMENTSPEC_TYPE and ELEMENTSPEC_VALUE
+
+I have no idea what this is.
+
+=item ATTRSET and SORT_ATTR
+
+ATTRSET is the attribute set from which the attributes are taken, and
+SORT_ATTR is a reference to an array containing the attributes
+themselves. Each attribute is represented by (are you following this
+carefully?) yet another hash, this one containing the elements
+ATTR_TYPE and ATTR_VALUE: for example, type=1 and value=4 in the BIB-1
+attribute set would indicate access-point 4 which is title, so that a
+sort of title is requested.
+
+=back
+
+Precisely why all of the above is so is not clear, but goes some way
+to explain why, in the Z39.50 world, the developers of the standard
+are not so much worshipped as blamed.
+
+The backend function should set STATUS to 0 on success, 1 for "partial
+success" (don't ask) or 2 on failure, in which case ERR_CODE and
+ERR_STR should be set.
+
+=head2 Support for SRU and SRW
+
+Since release 1.0, SimpleServer includes support for serving the SRU
+and SRW protocols as well as Z39.50. These ``web-friendly'' protocols
+enable similar functionality to that of Z39.50, but by means of rich
+URLs in the case of SRU, and a SOAP-based web-service in the case of
+SRW. These protocols are described at
+http://www.loc.gov/sru
+
+In order to serve these protocols from a SimpleServer-based
+application, it is necessary to launch the application with a YAZ
+Generic Frontend Server (GFS) configuration file, which can be
+specified using the command-line argument C<-f> I<filename>. A
+minimal configuration file looks like this:
+
+ <yazgfs>
+ <server>
+ <cql2rpn>pqf.properties</cql2rpn>
+ </server>
+ </yazgfs>
+
+This file specifies only that C<pqf.properties> should be used to
+translate the CQL queries of SRU and SRW into corresponding Z39.50
+Type-1 queries. For more information about YAZ GFS configuration,
+including how to specify an Explain record, see the I<Virtual Hosts>
+section of the YAZ manual at
+http://indexdata.com/yaz/doc/server.vhosts.tkl
+
+The mapping of CQL queries into Z39.50 Type-1 queries is specified by
+a file that indicates which BIB-1 attributes should be generated for
+each CQL index, relation, modifiers, etc. A typical section of this
+file looks like this:
+
+ index.dc.title = 1=4
+ index.dc.subject = 1=21
+ index.dc.creator = 1=1003
+ relation.< = 2=1
+ relation.le = 2=2
+
+This file specifies the BIB-1 access points (type=1) for the Dublin
+Core indexes C<title>, C<subject> and C<creator>, and the BIB-1
+relations (type=2) corresponding to the CQL relations C<E<lt>> and
+C<E<lt>=>. For more information about the format of this file, see
+the I<CQL> section of the YAZ manual at
+http://indexdata.com/yaz/doc/tools.tkl#tools.cql
+
+The YAZ distribution include a sample CQL-to-PQF mapping configuration
+file called C<pqf.properties>; this is sufficient for many
+applications, and a good base to work from for most others.
+
+If a SimpleServer-based application is run without this SRU-specific
+configuration, it can still serve SRU; however, CQL queries will not
+be translated, but passed straight through to the search-handler
+function, as the C<CQL> member of the parameters hash. It is then the
+responsibility of the back-end application to parse and handle the CQL
+query, which is most easily done using Ed Summers' fine C<CQL::Parser>
+module, available from CPAN at
+http://search.cpan.org/~esummers/CQL-Parser/
+