1 <?xml version="1.0" standalone="no"?>
2 <!DOCTYPE book PUBLIC "-//OASIS//DTD DocBook XML V4.4//EN"
3 "http://www.oasis-open.org/docbook/xml/4.4/docbookx.dtd"
5 <!ENTITY copyright SYSTEM "copyright.xml">
6 <!ENTITY % local SYSTEM "local.ent">
8 <!ENTITY manref SYSTEM "manref.xml">
9 <!ENTITY gpl2 SYSTEM "gpl-2.0.xml">
10 <!ENTITY % idcommon SYSTEM "common/common.ent">
15 <title>Metaproxy - User's Guide and Reference</title>
18 <firstname>Adam</firstname><surname>Dickmeiss</surname>
21 <firstname>Marc</firstname><surname>Cromme</surname>
24 <firstname>Mike</firstname><surname>Taylor</surname>
27 <releaseinfo>&version;</releaseinfo>
29 <year>2005-2014</year>
30 <holder>Index Data</holder>
34 This manual is part of Metaproxy version &version;.
37 Metaproxy is a universal router, proxy and encapsulated
38 metasearcher for information retrieval protocols. It accepts,
39 processes, interprets and redirects requests from IR clients using
40 standard protocols such as the binary
41 <ulink url="&url.z39.50;">ANSI/NISO Z39.50</ulink>
42 and the information search and retrieval
43 web service <ulink url="&url.sru;">SRU</ulink>
44 as well as functioning as a limited
45 <ulink url="&url.http;">HTTP</ulink> server.
48 Metaproxy is configured by an XML file which
49 specifies how the software should function in terms of routes that
50 the request packets can take through the proxy, each step on a
51 route being an instantiation of a filter. Filters come in many
52 types, one for each operation: accepting Z39.50 packets, logging,
53 query transformation, multiplexing, etc. Further filter-types can
54 be added as loadable modules to extend Metaproxy functionality,
58 Metaproxy is covered by the GNU General Public License version 2.
63 <imagedata fileref="common/id.png" format="PNG"/>
66 <imagedata fileref="common/id.eps" format="EPS"/>
73 <chapter id="introduction">
74 <title>Introduction</title>
77 <ulink url="&url.metaproxy;">Metaproxy</ulink>
78 is a stand alone program that acts as a universal router, proxy and
79 encapsulated metasearcher for information retrieval protocols such
80 as <ulink url="&url.z39.50;">Z39.50</ulink> and
81 <ulink url="&url.sru;">SRU</ulink>.
82 To clients, it acts as a server of these protocols: it can be searched,
83 records can be retrieved from it, etc.
84 To servers, it acts as a client: it searches in them,
85 retrieves records from them, etc. it satisfies its clients'
86 requests by transforming them, multiplexing them, forwarding them
87 on to zero or more servers, merging the results, transforming
88 them, and delivering them back to the client. In addition, it
89 acts as a simple <ulink url="&url.http;">HTTP</ulink> server; support
90 for further protocols can be added in a modular fashion, through the
91 creation of new filters.
96 Fish, bananas, cold pyjamas,
97 Mutton, beef and trout!
98 - attributed to Cole Porter.
101 Metaproxy is a more capable alternative to
102 <ulink url="&url.yazproxy;">YAZ Proxy</ulink>,
103 being more powerful, flexible, configurable and extensible. Among
104 its many advantages over the older, more pedestrian work are
105 support for multiplexing (encapsulated metasearching), routing by
106 database name, authentication and authorization and serving local
107 files via HTTP. Equally significant, its modular architecture
108 facilitites the creation of pluggable modules implementing further
112 This manual will describe how to install Metaproxy
113 before giving an overview of its architecture, then discussing the
114 key concept of a filter in some depth and giving an overview of
115 the various filter types, then discussing the configuration file
116 format. After this come several optional chapters which may be
117 freely skipped: a detailed discussion of virtual databases and
118 multi-database searching, some notes on writing extensions
119 (additional filter types) and a high-level description of the
120 source code. Finally comes the reference guide, which contains
121 instructions for invoking the <command>metaproxy</command>
122 program, and detailed information on each type of filter,
127 <chapter id="installation">
128 <title>Installation</title>
130 Metaproxy depends on the following tools/libraries:
132 <varlistentry><term><ulink url="&url.yazplusplus;">YAZ++</ulink></term>
135 This is a C++ library based on <ulink url="&url.yaz;">YAZ</ulink>.
139 <varlistentry><term><ulink url="&url.libxslt;">Libxslt</ulink></term>
141 <para>This is an XSLT processor - based on
142 <ulink url="&url.libxml2;">Libxml2</ulink>. Both Libxml2 and
143 Libxslt must be installed with the development components
144 (header files, etc.) as well as the run-time libraries.
148 <varlistentry><term><ulink url="&url.boost;">Boost</ulink></term>
151 The popular C++ library. Initial versions of Metaproxy
152 was built with 1.32 but this is no longer supported.
153 Metaproxy is known to work with Boost version 1.33 through 1.55.
160 In order to compile Metaproxy a modern C++ compiler is
161 required. Boost, in particular, requires the C++ compiler
162 to facilitate the newest features. Refer to Boost
163 <ulink url="&url.boost.compilers.status;">Compiler Status</ulink>
164 for more information.
167 We have successfully built Metaproxy using the compilers
168 <ulink url="&url.gcc;">GCC</ulink> and
169 <ulink url="&url.vstudio;">Microsoft Visual Studio</ulink>.
173 As an option, Metaproxy may also be compiled with
174 <ulink url="&url.usemarcon;">USEMARCON</ulink> support which allows for
175 MARC conversions for the <xref linkend="ref-record_transform"/> filter.
177 <section id="installation.unix">
178 <title>Installation on Unix (from Source)</title>
180 Here is a quick step-by-step guide on how to compile all the
181 tools that Metaproxy uses. Only few systems have none of the required
182 tools binary packages. If, for example, Libxml2/libxslt are already
183 installed as development packages use those (and omit compilation).
188 <ulink url="&url.usemarcon;">USEMARCON</ulink> is not available
189 as a package at the moment, so Metaproxy must be built from source
190 if that is to be used.
194 <section id="libxml2.fromsource">
195 <title>Libxml2/libxslt</title>
200 gunzip -c libxml2-version.tar.gz|tar xf -
208 gunzip -c libxslt-version.tar.gz|tar xf -
216 <section id="usemarcon">
217 <title>USEMARCON (optional)</title>
219 gunzip -c usemarcon317.tar.gz|tar xf -
228 <section id="yaz.fromsource">
229 <title>YAZ/YAZ++</title>
231 gunzip -c yaz-version.tar.gz|tar xf -
239 gunzip -c yazpp-version.tar.gz|tar xf -
248 <title id="boost.fromsource">Boost</title>
250 Metaproxy needs components thread and test from
254 gunzip -c boost-version.tar.gz|tar xf -
256 ./configure --with-libraries=thread,test,regex --with-toolset=gcc
262 However, under the hood bjam is used. You can invoke that with
265 ./bjam --toolset=gcc --with-thread --with-test --with-regex stage
268 Replace <literal>stage</literal> with <literal>clean</literal> /
269 <literal>install</literal> to perform clean and install respectively.
272 Add <literal>--prefix=DIR</literal> to install Boost in other
273 prefix than <literal>/usr/local</literal>.
276 <section id="metaproxy.fromsource">
277 <title>Metaproxy</title>
279 gunzip -c metaproxy-version.tar.gz|tar xf -
287 You may have to tell configure where Boost is installed by supplying
288 options <literal>--with-boost</literal> and <literal>--with-boost-toolset</literal>.
289 The former sets the PREFIX for Boost (same as --prefix for Boost above).
290 The latter the compiler toolset (eg. gcc34).
293 Pass <literal>--help</literal> to configure to get a list of
299 <section id="installation.debian">
300 <title>Installation on Debian GNU/Linux</title>
302 All dependencies for Metaproxy are available as
303 <ulink url="&url.debian;">Debian</ulink> packages.
306 The procedures for Debian based systems, such as
307 <ulink url="&url.ubuntu;">Ubuntu</ulink> is probably similar
310 There is currently no official Debian package for YAZ++.
311 And the official Debian package for YAZ is probably too old.
312 But Index Data builds "new" versions of those for Debian (i386, amd64 only).
315 Update the <filename>/etc/apt/sources.list</filename>
316 to include the Index Data repository.
317 See YAZ' <ulink url="&url.yaz.download.debian;">Download Debian</ulink>
318 for more information.
321 apt-get install libxslt1-dev
322 apt-get install libyazpp6-dev
323 apt-get install libboost-dev
324 apt-get install libboost-system-dev
325 apt-get install libboost-thread-dev
326 apt-get install libboost-test-dev
327 apt-get install libboost-regex-dev
330 With these packages installed, the usual configure + make
331 procedure can be used for Metaproxy as outlined in
332 <xref linkend="installation.unix"/>.
336 <section id="installation.rpm">
337 <title>Installation on RPM based Linux Systems</title>
339 All external dependencies for Metaproxy are available as
340 RPM packages, either from your distribution site, or from the
341 <ulink url="http://fr.rpmfind.net/">RPMfind</ulink> site.
344 For example, an installation of the requires Boost C++ development
345 libraries on RedHat Fedora C4 and C5 can be done like this:
347 wget ftp://fr.rpmfind.net/wlinux/fedora/core/updates/testing/4/SRPMS/boost-1.33.0-3.fc4.src.rpm
348 sudo rpmbuild --buildroot src/ --rebuild -p fc4/boost-1.33.0-3.fc4.src.rpm
349 sudo rpm -U /usr/src/redhat/RPMS/i386/boost-*rpm
353 The <ulink url="&url.yaz;">YAZ</ulink> library is needed to
354 compile &metaproxy;, see there
355 for more information on available RPM packages.
358 There is currently no official RPM package for YAZ++.
359 See the <ulink url="&url.yazplusplus;">YAZ++</ulink> pages
360 for more information on a Unix tarball install.
363 With these packages installed, the usual configure + make
364 procedure can be used for Metaproxy as outlined in
365 <xref linkend="installation.unix"/>.
369 <section id="installation.windows">
370 <title>Installation on Windows</title>
372 Metaproxy has been tested Microsoft
373 <ulink url="&url.vstudio;">Visual Studio</ulink>.
376 <section id="installation.windows.boost">
379 For Windows, it's easiest to get the precompiled Boost
380 package from <ulink url="&url.boost.windows.download;">here</ulink>.
381 Several versions of the Boost libraries may be selected when
382 installing Boost for windows. Please choose at least the
383 <emphasis>multithreaded</emphasis> (non-DLL) version because
384 the Metaproxy makefile uses that.
387 For more information about installing Boost refer to the
388 <ulink url="&url.boost.getting.started;">getting started</ulink>
393 <section id="installation.windows.libxslt">
394 <title>Libxslt</title>
396 <ulink url="&url.libxslt;">Libxslt</ulink> can be downloaded
398 <ulink url="&url.libxml2.download.windows;">here</ulink>.
401 Libxslt also requires libxml2 to operate.
405 <section id="installation.windows.yaz">
408 <ulink url="&url.yaz;">YAZ</ulink> can be downloaded
410 <ulink url="&url.yaz.download.win32;">here</ulink>.
414 <section id="installation.windows.yazplusplus">
417 Get <ulink url="&url.yazplusplus;">YAZ++</ulink> as well.
418 Version 1.6.0 or later is required.
421 YAZ++ includes NMAKE makefiles, similar to those found in the
426 <section id="installation.windows.metaproxy">
427 <title>Metaproxy</title>
429 Metaproxy is shipped with NMAKE makefiles as well - similar
430 to those found in the YAZ++/YAZ packages. Adjust this Makefile
431 to point to the proper locations of Boost, Libxslt, Libxml2,
432 zlib, iconv, yaz and yazpp.
436 <varlistentry><term><literal>DEBUG</literal></term>
438 If set to 1, the software is
439 compiled with debugging libraries (code generation is
440 multi-threaded debug DLL).
441 If set to 0, the software is compiled with release libraries
442 (code generation is multi-threaded DLL).
447 <term><literal>BOOST</literal></term>
450 Boost install location
456 <term><literal>BOOST_VERSION</literal></term>
459 Boost version (replace . with _).
465 <term><literal>BOOST_TOOLSET</literal></term>
474 <term><literal>LIBXSLT_DIR</literal>,
475 <literal>LIBXML2_DIR</literal> ..</term>
478 Specify the locations of Libxslt, libiconv, libxml2 and
487 After successful compilation you'll find
488 <literal>metaproxy.exe</literal> in the
489 <literal>bin</literal> directory.
497 <chapter id="yazproxy-comparison">
498 <title>YAZ Proxy Comparison</title>
500 The table below lists facilities either supported by either
501 <ulink url="&url.yazproxy;">YAZ Proxy</ulink> or Metaproxy.
503 <table id="yazproxy-comparison-table">
504 <title>Metaproxy / YAZ Proxy comparison</title>
508 <entry>Facility</entry>
509 <entry>Metaproxy</entry>
510 <entry>YAZ Proxy</entry>
515 <entry>Z39.50 server</entry>
516 <entry>Using filter <xref linkend="ref-frontend_net"/></entry>
517 <entry>Supported</entry>
520 <entry>SRU server</entry>
521 <entry>Supported with filter <xref linkend="ref-sru_z3950"/></entry>
522 <entry>Supported</entry>
525 <entry>Z39.50 client</entry>
526 <entry>Supported with filter <xref linkend="ref-z3950_client"/></entry>
527 <entry>Supported</entry>
530 <entry>SRU client</entry>
531 <entry>Supported with filter <xref linkend="ref-zoom"/></entry>
532 <entry>Unsupported</entry>
535 <entry>Connection reuse</entry>
536 <entry>Supported with filter <literal>session_shared</literal></entry>
537 <entry>Supported</entry>
540 <entry>Connection share</entry>
541 <entry>Supported with filter <literal>session_shared</literal></entry>
542 <entry>Unsupported</entry>
545 <entry>Result set reuse</entry>
546 <entry>Supported with filter <literal>session_shared</literal></entry>
547 <entry>Within one Z39.50 session / HTTP keep-alive</entry>
550 <entry>Record cache</entry>
551 <entry>Supported by filter <literal>session_shared</literal></entry>
552 <entry>Supported for last result set within one Z39.50/HTTP-keep alive session</entry>
555 <entry>Z39.50 Virtual database, i.e. select any Z39.50 target for database</entry>
556 <entry>Supported with filter <literal>virt_db</literal></entry>
557 <entry>Unsupported</entry>
560 <entry>SRU Virtual database, i.e. select any Z39.50 target for path</entry>
561 <entry>Supported with filter <literal>virt_db</literal>,
562 <literal>sru_z3950</literal></entry>
563 <entry>Supported</entry>
566 <entry>Multi target search</entry>
567 <entry>Supported with filter <literal>multi</literal> (round-robin)</entry>
568 <entry>Unsupported</entry>
571 <entry>Retrieval and search limits</entry>
572 <entry>Supported using filter <literal>limit</literal></entry>
573 <entry>Supported</entry>
576 <entry>Bandwidth limits</entry>
577 <entry>Supported using filter <literal>limit</literal></entry>
578 <entry>Supported</entry>
581 <entry>Connect limits</entry>
582 <entry>Supported by filter <literal>frontend_net</literal> (connect-max)</entry>
583 <entry>Supported</entry>
586 <entry>Retrieval sanity check and conversions</entry>
587 <entry>Supported using filter <literal>record_transform</literal></entry>
588 <entry>Supported</entry>
591 <entry>Query check</entry>
593 Supported by <literal>query_rewrite</literal> which may be check
594 a query and throw diagnostics (errors)
596 <entry>Supported</entry>
599 <entry>Query rewrite</entry>
600 <entry>Supported with <literal>query_rewrite</literal></entry>
601 <entry>Unsupported</entry>
604 <entry>Session invalidate for -1 hits</entry>
605 <entry>Unsupported</entry>
606 <entry>Supported</entry>
609 <entry>Architecture</entry>
610 <entry>Multi-threaded + select for networked modules such as
611 <literal>frontend_net</literal>)</entry>
612 <entry>Single-threaded using select</entry>
616 <entry>Extensability</entry>
617 <entry>Most functionality implemented as loadable modules</entry>
618 <entry>Unsupported and experimental</entry>
622 <entry><ulink url="&url.usemarcon;">USEMARCON</ulink></entry>
623 <entry>Supported with <literal>record_transform</literal></entry>
624 <entry>Supported</entry>
628 <entry>Portability</entry>
630 Requires YAZ, YAZ++ and modern C++ compiler supporting
631 <ulink url="&url.boost;">Boost</ulink>.
634 Requires YAZ and YAZ++.
635 STL is not required so pretty much any C++ compiler out there should work.
644 <chapter id="architecture">
645 <title>The Metaproxy Architecture</title>
647 The Metaproxy architecture is based on three concepts:
648 the <emphasis>package</emphasis>,
649 the <emphasis>route</emphasis>
650 and the <emphasis>filter</emphasis>.
654 <term>Packages</term>
657 A package is request or response, encoded in some protocol,
658 issued by a client, making its way through Metaproxy, send to or
659 received from a server, or sent back to the client.
662 The core of a package is the protocol unit - for example, a
663 Z39.50 Init Request or Search Response, or an SRU searchRetrieve
664 URL or Explain Response. In addition to this core, a package
665 also carries some extra information added and used by Metaproxy
669 In general, packages are doctored as they pass through
670 Metaproxy. For example, when the proxy performs authentication
671 and authorization on a Z39.50 Init request, it removes the
672 authentication credentials from the package so that they are not
673 passed onto the back-end server; and when search-response
674 packages are obtained from multiple servers, they are merged
675 into a single unified package that makes its way back to the
684 Packages make their way through routes, which can be thought of
685 as programs that operate on the package data-type. Each
686 incoming package initially makes its way through a default
687 route, but may be switched to a different route based on various
688 considerations. Routes are made up of sequences of filters (see
697 Filters provide the individual instructions within a route, and
698 effect the necessary transformations on packages. A particular
699 configuration of Metaproxy is essentially a set of filters,
700 described by configuration details and arranged in order in one
701 or more routes. There are many kinds of filter - about a dozen
702 at the time of writing with more appearing all the time - each
703 performing a specific function and configured by different
707 The word ``filter'' is sometimes used rather loosely, in two
708 different ways: it may be used to mean a particular
709 <emphasis>type</emphasis> of filter, as when we speak of ``the
710 auth_simple filter'' or ``the multi filter''; or it may be used
711 to be a specific <emphasis>instance</emphasis> of a filter
712 within a Metaproxy configuration. For example, a single
713 configuration will often contain multiple instances of the
714 <literal>z3950_client</literal> filter. In
715 operational terms, of these is a separate filter. In practice,
716 context always make it clear which sense of the word ``filter''
720 Extensibility of Metaproxy is primarily through the creation of
721 plugins that provide new filters. The filter API is small and
722 conceptually simple, but there are many details to master. See
724 <link linkend="filters">Filters</link>.
730 Since packages are created and handled by the system itself, and
731 routes are conceptually simple, most of the remainder of this
732 document concentrates on filters. After a brief overview of the
733 filter types follows, along with some thoughts on possible future
740 <chapter id="filters">
741 <title>Filters</title>
744 <section id="filters-introductory-notes">
745 <title>Introductory notes</title>
747 It's useful to think of Metaproxy as an interpreter providing a small
748 number of primitives and operations, but operating on a very
749 complex data type, namely the ``package''.
752 A package represents a Z39.50 or SRU/W request (whether for Init,
753 Search, Scan, etc.) together with information about where it came
754 from. Packages are created by front-end filters such as
755 <literal>frontend_net</literal> (see below), which reads them from
756 the network; other front-end filters are possible. They then pass
757 along a route consisting of a sequence of filters, each of which
758 transforms the package and may also have side-effects such as
759 generating logging. Eventually, the route will yield a response,
760 which is sent back to the origin.
763 There are many kinds of filter: some that are defined statically
764 as part of Metaproxy, and others may be provided by third parties
765 and dynamically loaded. They all conform to the same simple API
766 of essentially two methods: <function>configure()</function> is
767 called at startup time, and is passed an XML DOM tree representing that
768 part of the configuration file that pertains to this filter
769 instance: it is expected to walk that tree extracting relevant
770 information; and <function>process()</function> is called every
771 time the filter has to processes a package.
774 While all filters provide the same API, there are different modes
775 of functionality. Some filters are sources: they create
777 (<literal>frontend_net</literal>);
778 others are sinks: they consume packages and return a result
779 (<literal>backend_test</literal>,
780 <literal>bounce</literal>,
781 <literal>http_file</literal>,
782 <literal>z3950_client</literal>);
783 the others are true filters, that read, process and pass on the
784 packages they are fed
785 (<literal>auth_simple</literal>,
786 <literal>log</literal>,
787 <literal>multi</literal>,
788 <literal>query_rewrite</literal>,
789 <literal>record_transform</literal>,
790 <literal>session_shared</literal>,
791 <literal>sru_z3950</literal>,
792 <literal>template</literal>,
793 <literal>virt_db</literal>).
798 <section id="overview.filter.types">
799 <title>Overview of filter types</title>
801 We now briefly consider each of the types of filter supported by
802 the core Metaproxy binary. This overview is intended to give a
803 flavor of the available functionality; more detailed information
804 about each type of filter is included below in
805 <xref linkend="reference"/>.
808 The filters are here named by the string that is used as the
809 <literal>type</literal> attribute of a
810 <literal><filter></literal> element in the configuration
811 file to request them, with the name of the class that implements
812 them in parentheses. (The classname is not needed for normal
813 configuration and use of Metaproxy; it is useful only to
817 The filters are here listed in alphabetical order:
824 New virt_db-alike that does inteligent peer choice, explain merging,
825 adds FD&N to explain. Keeps init responses (like "virt_db Classic"),
826 makes routing choices based on local explain knowledge. Ref IDDI
829 Filter to convert Explain Classic to ZeeRex.
831 CQL2PQF (which needs augmented ZeeRex) - MARC for Talis.
835 Figure out what additional information we need in:
836 ZeeRex (check against D3.1)
837 Init request (e.g. loop detection)
838 Query package (e.g. number of hops)
839 Query response (e.g. record source)
843 <section id="auth_simple">
844 <title><literal>auth_simple</literal>
845 (mp::filter::AuthSimple)</title>
847 Simple authentication and authorization. The configuration
848 specifies the name of a file that is the user register, which
849 lists <varname>username</varname>:<varname>password</varname>
850 pairs, one per line, colon separated. When a session begins, it
851 is rejected unless username and passsword are supplied, and match
852 a pair in the register. The configuration file may also specific
853 the name of another file that is the target register: this lists
854 lists <varname>username</varname>:<varname>dbname</varname>,<varname>dbname</varname>...
855 sets, one per line, with multiple database names separated by
856 commas. When a search is processed, it is rejected unless the
857 database to be searched is one of those listed as available to
862 <section id="backend_test">
863 <title><literal>backend_test</literal>
864 (mp::filter::Backend_test)</title>
866 A partial sink that provides dummy responses in the manner of the
867 <literal>yaz-ztest</literal> Z39.50 server. This is useful only
868 for testing. Seriously, you don't need this. Pretend you didn't
869 even read this section.
873 <section id="bounce">
874 <title><literal>bounce</literal>
875 (mp::filter::Bounce)</title>
877 A sink that swallows <emphasis>all packages</emphasis>,
878 and returns them almost unprocessed.
879 It never sends any package of any type further down the row, but
880 sets Z39.50 packages to Z_Close, and HTTP_Request packages to
881 HTTP_Response err code 400 packages, and adds a suitable bounce
883 The bounce filter is usually added at end of each filter chain route
884 to prevent infinite hanging of for example HTTP
885 requests packages when only the Z39.50 client partial sink
886 filter is found in the
891 <section id="cql_rpn">
892 <title><literal>cql_rpn</literal>
893 (mp::filter::CQLtoRPN)</title>
895 A query language transforming filter which catches Z39.50
896 <literal>searchRequest</literal>
897 packages containing <literal>CQL</literal> queries, transforms
898 those to <literal>RPN</literal> queries,
899 and sends the <literal>searchRequests</literal> on to the next
900 filters. It is among other things useful in a SRU context.
904 <section id="frontend_net">
905 <title><literal>frontend_net</literal>
906 (mp::filter::FrontendNet)</title>
908 A source that accepts Z39.50 connections from a port
909 specified in the configuration, reads protocol units, and
910 feeds them into the next filter in the route. When the result is
911 received, it is returned to the original origin.
915 <section id="http_file">
916 <title><literal>http_file</literal>
917 (mp::filter::HttpFile)</title>
919 A partial sink which swallows only
920 <literal>HTTP_Request</literal> packages, and
921 returns the contents of files from the local
922 filesystem in response to HTTP requests.
923 It lets Z39.50 packages and all other forthcoming package types
926 does mean that Metaproxy is also a Web-server in its spare time. So
927 far it does not contain either an email-reader or a Lisp
928 interpreter, but that day is surely coming.)
932 <section id="load_balance">
933 <title><literal>load_balance</literal>
934 (mp::filter::LoadBalance)</title>
936 Performs load balancing for incoming Z39.50 init requests.
937 It is used together with the <literal>virt_db</literal> filter,
938 but unlike the <literal>multi</literal> filter it does send an
939 entire session to only one of the virtual backends. The
940 <literal>load_balance</literal> filter is assuming that
941 all backend targets have equal content, and chooses the backend
942 with least load cost for a new session.
945 This filter is experimental and yet not mature for heavy load
953 <title><literal>log</literal>
954 (mp::filter::Log)</title>
956 Writes logging information to standard output, and passes on
957 the package unchanged. A log file name can be specified, as well
958 as multiple different logging formats.
963 <title><literal>multi</literal>
964 (mp::filter::Multi)</title>
966 Performs multi-database searching.
968 <link linkend="multidb">the extended discussion</link>
969 of virtual databases and multi-database searching below.
973 <section id="query_rewrite">
974 <title><literal>query_rewrite</literal>
975 (mp::filter::QueryRewrite)</title>
977 Rewrites Z39.50 <literal>Type-1</literal>
978 and <literal>Type-101</literal> (``<literal>RPN</literal>'')
980 three-step process: the query is transliterated from Z39.50
981 packet structures into an XML representation; that XML
982 representation is transformed by an XSLT stylesheet; and the
983 resulting XML is transliterated back into the Z39.50 packet
989 <section id="record_transform">
990 <title><literal>record_transform</literal>
991 (mp::filter::RecordTransform)</title>
993 This filter acts only on Z3950 present requests, and let all
994 other types of packages and requests pass untouched. It's use is
995 twofold: blocking Z3950 present requests, which the backend
996 server does not understand and can not honor, and transforming
997 the present syntax and elementset name according to the rules
998 specified, to fetch only existing record formats, and transform
999 them on the fly to requested record syntaxes.
1003 <section id="session_shared">
1004 <title><literal>session_shared</literal>
1005 (mp::filter::SessionShared)</title>
1007 This filter implements global sharing of
1008 result sets (i.e. between threads and therefore between
1009 clients), yielding performance improvements by clever resource
1014 <section id="sru_z3950">
1015 <title><literal>sru_z3950</literal>
1016 (mp::filter::SRUtoZ3950)</title>
1018 This filter transforms valid
1019 SRU GET/POST/SOAP searchRetrieve requests to Z3950 init, search,
1020 and present requests, and wraps the
1021 received hit counts and XML records into suitable SRU response
1023 The <literal>sru_z3950</literal> filter processes also SRU
1024 GET/POST/SOAP explain requests, returning
1025 either the absolute minimum required by the standard, or a full
1026 pre-defined ZeeReX explain record.
1028 <ulink url="&url.zeerex.explain;">ZeeReX Explain</ulink>
1029 standard pages and the
1030 <ulink url="&url.sru.explain;">SRU Explain</ulink> pages
1031 for more information on the correct explain syntax.
1032 SRU scan requests are not supported yet.
1036 <section id="template">
1037 <title><literal>template</literal>
1038 (mp::filter::Template)</title>
1040 Does nothing at all, merely passing the packet on. (Maybe it
1041 should be called <literal>nop</literal> or
1042 <literal>passthrough</literal>?) This exists not to be used, but
1043 to be copied - to become the skeleton of new filters as they are
1044 written. As with <literal>backend_test</literal>, this is not
1045 intended for civilians.
1049 <section id="virt_db">
1050 <title><literal>virt_db</literal>
1051 (mp::filter::VirtualDB)</title>
1053 Performs virtual database selection: based on the name of the
1054 database in the search request, a server is selected, and its
1055 address added to the request in a <literal>VAL_PROXY</literal>
1056 otherInfo packet. It will subsequently be used by a
1057 <literal>z3950_client</literal> filter.
1059 <link linkend="multidb">the extended discussion</link>
1060 of virtual databases and multi-database searching below.
1064 <section id="z3950_client">
1065 <title><literal>z3950_client</literal>
1066 (mp::filter::Z3950Client)</title>
1068 A partial sink which swallows only Z39.50 packages.
1069 It performs Z39.50 searching and retrieval by proxying the
1070 packages that are passed to it. Init requests are sent to the
1071 address specified in the <literal>VAL_PROXY</literal> otherInfo
1072 attached to the request: this may have been specified by client,
1073 or generated by a <literal>virt_db</literal> filter earlier in
1074 the route. Subsequent requests are sent to the same address,
1075 which is remembered at Init time in a Session object.
1076 HTTP_Request packages and all other forthcoming package types
1077 are passed untouched.
1082 <section id="zeerex_explain">
1083 <title><literal>zeerex_explain</literal>
1084 (mp::filter::ZeerexExplain)</title>
1086 This filter acts as a sink for
1087 Z39.50 explain requests, returning a static ZeeReX
1088 Explain XML record from the config section. All other packages
1091 <ulink url="&url.zeerex.explain;">ZeeReX Explain</ulink>
1093 for more information on the correct explain syntax.
1097 This filter is not yet completed.
1106 <section id="future.directions">
1107 <title>Future directions</title>
1109 Some other filters that do not yet exist, but which would be
1110 useful, are briefly described. These may be added in future
1111 releases (or may be created by third parties, as loadable
1117 <term><literal>frontend_cli</literal> (source)</term>
1120 Command-line interface for generating requests.
1125 <term><literal>sru_client</literal> (sink)</term>
1128 SRU/GET and SRU/SOAP searching and retrieval.
1133 <term><literal>opensearch_client</literal> (sink)</term>
1136 A9 OpenSearch searching and retrieval.
1146 <chapter id="configuration">
1147 <title>Configuration: the Metaproxy configuration file format</title>
1150 <section id="configuration-introductory-notes">
1151 <title>Introductory notes</title>
1153 If Metaproxy is an interpreter providing operations on packages, then
1154 its configuration file can be thought of as a program for that
1155 interpreter. Configuration is by means of a single XML file, the name
1156 of which is supplied as the sole command-line argument to the
1157 <command>metaproxy</command> program. (See
1158 <xref linkend="reference"/> below for more information on invoking
1163 <section id="overview.xml.structure">
1164 <title>Overview of the config file XML structure</title>
1166 All elements and attributes are in the namespace
1167 <ulink url="http://indexdata.com/metaproxy"/>.
1168 This is most easily achieved by setting the default namespace on
1169 the top-level element, as here:
1172 <metaproxy xmlns="http://indexdata.com/metaproxy" version="1.0">
1175 The top-level element is <metaproxy>. This contains
1176 a <dlpath> element,
1177 a <start> element,
1178 a <filters> element and
1179 a <routes> element, in that order. <dlpath> and
1180 <filters> are optional; the other two are mandatory.
1181 All four are non-repeatable.
1184 The <dlpath;> element contains a text element which
1185 specifies the location of filter modules. This is only needed
1186 if Metaproxy must load 3rd party filters (most filters with Metaproxy
1187 are built into the Metaproxy application).
1190 The <start> element is empty, but carries a
1191 <literal>route</literal> attribute, whose value is the name of
1192 route at which to start running - analogous to the name of the
1193 start production in a formal grammar.
1196 If present, <filters> contains zero or more <filter>
1197 elements. Each filter carries a <literal>type</literal> attribute
1198 which specifies what kind of filter is being defined
1199 (<literal>frontend_net</literal>, <literal>log</literal>, etc.)
1200 and contain various elements that provide suitable configuration
1201 for a filter of its type. The filter-specific elements are
1203 <xref linkend="reference"/>.
1204 Filters defined in this part of the file must carry an
1205 <literal>id</literal> attribute so that they can be referenced
1209 <routes> contains one or more <route> elements, each
1210 of which must carry an <literal>id</literal> element. One of the
1211 routes must have the ID value that was specified as the start
1212 route in the <start> element's <literal>route</literal>
1213 attribute. Each route contains zero or more <filter>
1214 elements. These are of two types. They may be empty, but carry a
1215 <literal>refid</literal> attribute whose value is the same as the
1216 <literal>id</literal> of a filter previously defined in the
1217 <filters> section. Alternatively, a route within a filter
1218 may omit the <literal>refid</literal> attribute, but contain
1219 configuration elements similar to those used for filters defined
1220 in the <filters> section. (In other words, each filter in a
1221 route may be included either by reference or by physical
1227 <section id="example.configuration">
1228 <title>An example configuration</title>
1230 The following is a small, but complete, Metaproxy configuration
1231 file (included in the distribution as
1232 <literal>metaproxy/etc/config1.xml</literal>).
1233 This file defines a very simple configuration that simply proxies
1234 to whatever back-end server the client requests, but logs each
1235 request and response. This can be useful for debugging complex
1236 client-server dialogues.
1238 <screen><![CDATA[<?xml version="1.0"?>
1239 <metaproxy xmlns="http://indexdata.com/metaproxy" version="1.0">
1240 <dlpath>/usr/lib/metaproxy/modules</dlpath>
1241 <start route="start"/>
1243 <filter id="frontend" type="frontend_net">
1246 <filter id="backend" type="z3950_client">
1251 <filter refid="frontend"/>
1252 <filter type="log"/>
1253 <filter refid="backend"/>
1254 <filter type="bounce"/>
1260 It works by defining a single route, called
1261 <literal>start</literal>, which consists of a sequence of four
1262 filters. The first and last of these are included by reference:
1263 their <literal><filter></literal> elements have
1264 <literal>refid</literal> attributes that refer to filters defined
1265 within the prior <literal><filters></literal> section. The
1266 middle filter is included inline in the route.
1269 The four filters in the route are as follows: first, a
1270 <literal>frontend_net</literal> filter accepts Z39.50 requests
1271 from any host on port 9000; then these requests are passed through
1272 a <literal>log</literal> filter that emits a message for each
1273 request; they are then fed into a <literal>z3950_client</literal>
1274 filter, which forwards all Z39.50 requests to the client-specified
1275 back-end Z39.509 server. Those Z39.50 packages are returned by the
1276 <literal>z3950_client</literal> filter, with the response data
1277 filled by the external Z39.50 server targeted.
1278 All non-Z39.50 packages are passed through to the
1279 <literal>bounce</literal> filter, which definitely bounces
1280 everything, including fish, bananas, cold pyjamas,
1281 mutton, beef and trout packages.
1282 When the response arrives, it is handed
1283 back to the <literal>log</literal> filter, which emits another
1284 message; and then to the <literal>frontend_net</literal> filter,
1285 which returns the response to the client.
1289 <section id="config-file-modularity">
1290 <title>Config file modularity</title>
1292 Metaproxy XML configuration snippets can be reused by other
1293 filters using the <literal>XInclude</literal> standard, as seen in
1294 the <literal>/etc/config-sru-to-z3950.xml</literal> example SRU
1297 <filter id="sru" type="sru_z3950">
1298 <database name="Default">
1299 <xi:include xmlns:xi="http://www.w3.org/2001/XInclude"
1300 href="explain.xml"/>
1307 <section id="config-file-syntax-check">
1308 <title>Config file syntax checking</title>
1310 The distribution contains RelaxNG Compact and XML syntax checking
1311 files, as well as XML Schema files. These are found in the
1314 xml/schema/metaproxy.rnc
1315 xml/schema/metaproxy.rng
1316 xml/schema/metaproxy.xsd
1318 and can be used to verify or debug the XML structure of
1319 configuration files. For example, using the utility
1320 <filename>xmllint</filename>, syntax checking is done like this:
1322 xmllint --noout --schema xml/schema/metaproxy.xsd etc/config-local.xml
1323 xmllint --noout --relaxng xml/schema/metaproxy.rng etc/config-local.xml
1325 (A recent version of <literal>libxml2</literal> is required, as
1326 support for XML Schemas is a relatively recent addition.)
1329 You can of course use any other RelaxNG or XML Schema compliant tool
1337 <chapter id="multidb">
1338 <title>Virtual databases and multi-database searching</title>
1341 <section id="multidb-introductory-notes">
1342 <title>Introductory notes</title>
1344 Two of Metaproxy's filters are concerned with multiple-database
1345 operations. Of these, <literal>virt_db</literal> can work alone
1346 to control the routing of searches to one of a number of servers,
1347 while <literal>multi</literal> can work together with
1348 <literal>virt_db</literal> to perform multi-database searching, merging
1349 the results into a unified result-set - ``metasearch in a box''.
1352 The interaction between
1353 these two filters is necessarily complex: it reflects the real,
1354 irreducible complexity of multi-database searching in a protocol such
1355 as Z39.50 that separates initialization from searching, and in
1356 which the database to be searched is not known at initialization
1360 It's possible to use these filters without understanding the
1361 details of their functioning and the interaction between them; the
1362 next two sections of this chapter are ``HOW-TO'' guides for doing
1363 just that. However, debugging complex configurations will require
1364 a deeper understanding, which the last two sections of this
1365 chapters attempt to provide.
1370 <section id="multidb.virt_db">
1371 <title>Virtual databases with the <literal>virt_db</literal> filter</title>
1373 Working alone, the purpose of the
1374 <literal>virt_db</literal>
1375 filter is to route search requests to one of a selection of
1376 back-end databases. In this way, a single Z39.50 endpoint
1377 (running Metaproxy) can provide access to several different
1378 underlying services, including those that would otherwise be
1379 inaccessible due to firewalls. In many useful configurations, the
1380 back-end databases are local to the Metaproxy installation, but
1381 the software does not enforce this, and any valid Z39.50 servers
1382 may be used as back-ends.
1385 For example, a <literal>virt_db</literal>
1386 filter could be set up so that searches in the virtual database
1387 ``lc'' are forwarded to the Library of Congress bibliographic
1388 catalogue server, and searches in the virtual database ``marc''
1389 are forwarded to the toy database of MARC records that Index Data
1390 hosts for testing purposes. A <literal>virt_db</literal>
1391 configuration to make this switch would look like this:
1393 <screen><![CDATA[<filter type="virt_db">
1395 <database>lc</database>
1396 <target>z3950.loc.gov:7090/voyager</target>
1399 <database>marc</database>
1400 <target>indexdata.com/marc</target>
1402 </filter>]]></screen>
1404 As well as being useful in it own right, this filter also provides
1405 the foundation for multi-database searching.
1410 <section id="multidb.multi">
1411 <title>Multi-database search with the <literal>multi</literal> filter</title>
1413 To arrange for Metaproxy to broadcast searches to multiple back-end
1414 servers, the configuration needs to include two components: a
1415 <literal>virt_db</literal>
1416 filter that specifies multiple
1417 <literal><target></literal>
1418 elements, and a subsequent
1419 <literal>multi</literal>
1420 filter. Here, for example, is a complete configuration that
1421 broadcasts searches to both the Library of Congress catalogue and
1422 Index Data's tiny testing database of MARC records:
1424 <screen><![CDATA[<?xml version="1.0"?>
1425 <metaproxy xmlns="http://indexdata.com/metaproxy" version="1.0">
1426 <start route="start"/>
1429 <filter type="frontend_net">
1430 <threads>10</threads>
1433 <filter type="virt_db">
1435 <database>lc</database>
1436 <target>z3950.loc.gov:7090/voyager</target>
1439 <database>marc</database>
1440 <target>indexdata.com/marc</target>
1443 <database>all</database>
1444 <target>z3950.loc.gov:7090/voyager</target>
1445 <target>indexdata.com/marc</target>
1448 <filter type="multi"/>
1449 <filter type="z3950_client">
1450 <timeout>30</timeout>
1452 <filter type="bounce"/>
1455 </metaproxy>]]></screen>
1458 <literal>virt_db</literal>
1459 filter that specifies multiple
1460 <literal><target></literal>
1461 elements but without a subsequent
1462 <literal>multi</literal>
1463 filter yields surprising and undesirable results, as will be
1464 described below. Don't do that.)
1467 Metaproxy can be invoked with this configuration as follows:
1469 <screen>../src/metaproxy --config config-simple-multi.xml</screen>
1471 And thereafter, Z39.50 clients can connect to the running server
1472 (on port 9000, as specified in the configuration) and search in
1473 any of the databases
1474 <literal>lc</literal> (the Library of Congress catalogue),
1475 <literal>marc</literal> (Index Data's test database of MARC records)
1477 <literal>all</literal> (both of these). As an example, a session
1478 using the YAZ command-line client <literal>yaz-client</literal> is
1479 here included (edited for brevity and clarity):
1481 <screen><![CDATA[$ yaz-client @:9000
1485 Search was a success.
1486 Number of hits: 10000, setno 1
1490 Search was a success.
1491 Number of hits: 10, setno 3
1495 Search was a success.
1496 Number of hits: 10010, setno 4
1499 [marc]Record type: USmarc
1502 005 00000000000000.0
1503 008 910710c19910701nju 00010 eng
1507 100 10 $a Jack Collins
1508 245 10 $a How to program a computer
1514 [VOYAGER]Record type: USmarc
1516 005 20041229102447.0
1517 008 030910s2004 caua 000 0 eng
1518 035 $a (DLC) 2003112666
1519 906 $a 7 $b cbc $c orignew $d 4 $e epcn $f 20 $g y-gencatlg
1520 925 0 $a acquire $b 1 shelf copy $x policy default
1521 955 $a pc10 2003-09-10 $a pv12 2004-06-23 to SSCD; $h sj05 2004-11-30 $e sj05 2004-11-30 to Shelf.
1524 040 $a DLC $c DLC $d DLC
1525 050 00 $a MLCM 2004/03312 (G)
1526 245 10 $a 007, everything or nothing : $b Prima's official strategy guide / $c created by Kaizen Media Group.
1527 246 3 $a Double-O-seven, everything or nothing
1528 246 30 $a Prima's official strategy guide
1529 260 $a Roseville, CA : $b Prima Games, $c c2004.
1530 300 $a 161 p. : $b col. ill. ; $c 28 cm.
1531 500 $a "Platforms: Nintendo GameCube, Macintosh, PC, PlayStation 2 computer entertainment system, Xbox"--P. [4] of cover.
1532 650 0 $a Video games.
1533 710 2 $a Kaizen Media Group.
1534 856 42 $3 Publisher description $u http://www.loc.gov/catdir/description/random052/2003112666.html
1539 As can be seen, the first record in the result set is from the
1540 Index Data test database, and the second from the Library of
1541 Congress database. The result-set continues alternating records
1542 round-robin style until the point where one of the databases'
1543 records are exhausted.
1546 This example uses only two back-end databases; more may be used.
1547 There is no limitation imposed on the number of databases that may
1548 be metasearched in this way: issues of resource usage and
1549 administrative complexity dictate the practical limits.
1552 What happens when one of the databases doesn't respond? By default,
1553 the entire multi-database search fails, and the appropriate
1554 diagnostic is returned to the client. This is usually appropriate
1555 during development, when technicians need maximum information, but
1556 can be inconvenient in deployment, when users typically don't want
1557 to be bothered with problems of this kind and prefer just to get
1558 the records from the databases that are available. To obtain this
1559 latter behavior add an empty
1560 <literal><hideunavailable></literal>
1562 <literal>multi</literal> filter:
1564 <screen><![CDATA[ <filter type="multi">
1566 </filter>]]></screen>
1568 Under this regime, an error is reported to the client only if
1569 <emphasis>all</emphasis> the databases in a multi-database search
1575 <section id="multidb.what">
1576 <title>What's going on?</title>
1578 <title>Lark's vomit</title>
1580 This section goes into a level of technical detail that is
1581 probably not necessary in order to configure and use Metaproxy.
1582 It is provided only for those who like to know how things work.
1583 You should feel free to skip on to the next section if this one
1584 doesn't seem like fun.
1588 Hold on tight - this may get a little hairy.
1591 In the general course of things, a Z39.50 Init request may carry
1592 with it an otherInfo packet of type <literal>VAL_PROXY</literal>,
1593 whose value indicates the address of a Z39.50 server to which the
1594 ultimate connection is to be made. (This otherInfo packet is
1595 supported by YAZ-based Z39.50 clients and servers, but has not yet
1596 been ratified by the Maintenance Agency and so is not widely used
1597 in non-Index Data software. We're working on it.)
1598 The <literal>VAL_PROXY</literal> packet functions
1599 analogously to the absoluteURI-style Request-URI used with the GET
1600 method when a web browser asks a proxy to forward its request: see
1602 <ulink url="http://www.w3.org/Protocols/rfc2616/rfc2616-sec5.html#sec5.1.2"
1603 >Request-URI</ulink>
1605 <ulink url="http://www.w3.org/Protocols/rfc2616/rfc2616.html"
1606 >the HTTP 1.1 specification</ulink>.
1609 Within Metaproxy, Search requests that are part of the same
1610 session as an Init request that carries a
1611 <literal>VAL_PROXY</literal> otherInfo are also annotated with the
1612 same information. The role of the <literal>virt_db</literal>
1613 filter is to rewrite this otherInfo packet dependent on the
1614 virtual database that the client wants to search.
1617 When Metaproxy receives a Z39.50 Init request from a client, it
1618 doesn't immediately forward that request to the back-end server.
1619 Why not? Because it doesn't know <emphasis>which</emphasis>
1620 back-end server to forward it to until the client sends a Search
1621 request that specifies the database that it wants to search in.
1622 Instead, it just treasures the Init request up in its heart; and,
1623 later, the first time the client does a search on one of the
1624 specified virtual databases, a connection is forged to the
1625 appropriate server and the Init request is forwarded to it. If,
1626 later in the session, the same client searches in a different
1627 virtual database, then a connection is forged to the server that
1628 hosts it, and the same cached Init request is forwarded there,
1632 All of this clever Init-delaying is done by the
1633 <literal>frontend_net</literal> filter. The
1634 <literal>virt_db</literal> filter knows nothing about it; in
1635 fact, because the Init request that is received from the client
1636 doesn't get forwarded until a Search request is received, the
1637 <literal>virt_db</literal> filter (and the
1638 <literal>z3950_client</literal> filter behind it) doesn't even get
1639 invoked at Init time. The <emphasis>only</emphasis> thing that a
1640 <literal>virt_db</literal> filter ever does is rewrite the
1641 <literal>VAL_PROXY</literal> otherInfo in the requests that pass
1645 It is possible for a <literal>virt_db</literal> filter to contain
1647 <literal><target></literal>
1648 elements. What does this mean? Only that the filter will add
1649 multiple <literal>VAL_PROXY</literal> otherInfo packets to the
1650 Search requests that pass through it. That's because the virtual
1651 DB filter is dumb, and does exactly what it's told - no more, no
1653 If a Search request with multiple <literal>VAL_PROXY</literal>
1654 otherInfo packets reaches a <literal>z3950_client</literal>
1655 filter, this is an error. That filter doesn't know how to deal
1656 with multiple targets, so it will either just pick one and search
1657 in it, or (better) fail with an error message.
1660 The <literal>multi</literal> filter comes to the rescue! This is
1661 the only filter that knows how to deal with multiple
1662 <literal>VAL_PROXY</literal> otherInfo packets, and it does so by
1663 making multiple copies of the entire Search request: one for each
1664 <literal>VAL_PROXY</literal>. Each of these new copies is then
1665 passed down through the remaining filters in the route. (The
1666 copies are handled in parallel though the
1667 spawning of new threads.) Since the copies each have only one
1668 <literal>VAL_PROXY</literal> otherInfo, they can be handled by the
1669 <literal>z3950_client</literal> filter, which happily deals with
1670 each one individually. When the results of the individual
1671 searches come back up to the <literal>multi</literal> filter, it
1672 merges them into a single Search response, which is what
1673 eventually makes it back to the client.
1678 <imagedata fileref="multi.pdf" format="PDF" scale="50"/>
1681 <imagedata fileref="multi.png" format="PNG"/>
1684 <!-- Fall back if none of the images can be used -->
1686 [Here there should be a diagram showing the progress of
1687 packages through the filters during a simple virtual-database
1688 search and a multi-database search, but is seems that your
1689 tool chain has not been able to include the diagram in this
1694 <para>A picture is worth a thousand words (but only five hundred on 64-bit architectures)</para>
1701 <chapter id="sru-server">
1702 <title>Combined SRU webservice and Z39.50 server configuration</title>
1704 Metaproxy can act as
1705 <ulink url="&url.sru;">SRU</ulink> and
1706 web service server, which translates web service requests to
1707 <ulink url="&url.z39.50;">ANSI/NISO Z39.50</ulink> packages and
1708 sends them off to common available targets.
1711 A typical setup for this operation needs a filter route including the
1715 <table id="sru-server-table-config" frame="top">
1716 <title>SRU/Z39.50 Server Filter Route Configuration</title>
1720 <entry>Filter</entry>
1721 <entry>Importance</entry>
1722 <entry>Purpose</entry>
1728 <entry><literal>frontend_net</literal></entry>
1729 <entry>required</entry>
1730 <entry>Accepting HTTP connections and passing them to following
1731 filters. Since this filter also accepts Z39.50 connections, the
1732 server works as SRU and Z39.50 server on the same port.</entry>
1735 <entry><literal>sru_z3950</literal></entry>
1736 <entry>required</entry>
1737 <entry>Accepting SRU GET/POST/SOAP explain and
1738 searchRetrieve requests for the the configured databases.
1739 Explain requests are directly served from the static XML configuration.
1740 SearchRetrieve requests are
1741 transformed to Z39.50 search and present packages.
1742 All other HTTP and Z39.50 packages are passed unaltered.</entry>
1745 <entry><literal>http_file</literal></entry>
1746 <entry>optional</entry>
1747 <entry>Serving HTTP requests from the filesystem. This is only
1748 needed if the server should serve XSLT stylesheets, static HTML
1749 files or Java Script for thin browser based clients.
1750 Z39.50 packages are passed unaltered.</entry>
1753 <entry><literal>cql_rpn</literal></entry>
1754 <entry>required</entry>
1755 <entry>Usually, Z39.50 servers do not talk CQL, hence the
1756 translation of the CQL query language to RPN is mandatory in
1757 most cases. Affects only Z39.50 search packages.</entry>
1760 <entry><literal>record_transform</literal></entry>
1761 <entry>optional</entry>
1762 <entry>Some Z39.50 backend targets can not present XML record
1763 syntaxes in common wanted element sets. using this filter, one
1764 can transform binary MARC records to MARCXML records, and
1765 further transform those to any needed XML schema/format by XSLT
1766 transformations. Changes only Z39.50 present packages.</entry>
1769 <entry><literal>session_shared</literal></entry>
1770 <entry>optional</entry>
1771 <entry>The stateless nature of web services requires frequent
1772 re-searching of the same targets for display of paged result set
1773 records. This might be an unacceptable burden for the accessed
1774 backend Z39.50 targets, and this mosule can be added for
1775 efficient backend target resource pooling.</entry>
1778 <entry><literal>z3950_client</literal></entry>
1779 <entry>required</entry>
1780 <entry>Finally, a Z39.50 package sink is needed in the filter
1781 chain to provide the response packages. The Z39.50 client module
1782 is used to access external targets over the network, but any
1783 coming local Z39.50 package sink could be used instead of.</entry>
1786 <entry><literal>bounce</literal></entry>
1787 <entry>required</entry>
1788 <entry>Any Metaproxy package arriving here did not do so by
1789 purpose, and is bounced back with connection closure. this
1790 prevents inifinite package hanging inside the SRU server.</entry>
1796 A typical minimal example <ulink url="&url.sru;">SRU</ulink>
1797 server configuration file is found in the tarball distribution at
1798 <literal>etc/config-sru-to-z3950.xml</literal>.
1801 Off course, any other metaproxy modules can be integrated into a
1802 SRU server solution, including, but not limited to, load balancing,
1803 multiple target querying
1804 (see <xref linkend="multidb"/>), and complex RPN query rewrites.
1811 <chapter id="extensions">
1812 <title>Writing extensions for Metaproxy</title>
1813 <para>### To be written</para>
1819 <chapter id="classes">
1820 <title>Classes in the Metaproxy source code</title>
1823 <section id="classes-introductory-notes">
1824 <title>Introductory notes</title>
1826 <emphasis>Stop! Do not read this!</emphasis>
1827 You won't enjoy it at all. You should just skip ahead to
1828 <xref linkend="reference"/>,
1830 <!-- The remainder of this paragraph is lifted verbatim from
1831 Douglas Adams' _Hitch Hiker's Guide to the Galaxy_, chapter 8 -->
1832 you things you really need to know, like the fact that the
1833 fabulously beautiful planet Bethselamin is now so worried about
1834 the cumulative erosion by ten billion visiting tourists a year
1835 that any net imbalance between the amount you eat and the amount
1836 you excrete whilst on the planet is surgically removed from your
1837 bodyweight when you leave: so every time you go to the lavatory it
1838 is vitally important to get a receipt.
1841 This chapter contains documentation of the Metaproxy source code, and is
1842 of interest only to maintainers and developers. If you need to
1843 change Metaproxy's behavior or write a new filter, then you will most
1844 likely find this chapter helpful. Otherwise it's a waste of your
1845 good time. Seriously: go and watch a film or something.
1846 <citetitle>This is Spinal Tap</citetitle> is particularly good.
1849 Still here? OK, let's continue.
1852 In general, classes seem to be named big-endianly, so that
1853 <literal>FactoryFilter</literal> is not a filter that filters
1854 factories, but a factory that produces filters; and
1855 <literal>FactoryStatic</literal> is a factory for the statically
1856 registered filters (as opposed to those that are dynamically
1861 <section id="individual.classes">
1862 <title>Individual classes</title>
1864 The classes making up the Metaproxy application are here listed by
1865 class-name, with the names of the source files that define them in
1869 <section id="class-FactoryFilter">
1870 <title><literal>mp::FactoryFilter</literal>
1871 (<filename>factory_filter.cpp</filename>)</title>
1873 A factory class that exists primarily to provide the
1874 <literal>create()</literal> method, which takes the name of a
1875 filter class as its argument and returns a new filter of that
1876 type. To enable this, the factory must first be populated by
1877 calling <literal>add_creator()</literal> for static filters (this
1878 is done by the <literal>FactoryStatic</literal> class, see below)
1879 and <literal>add_creator_dyn()</literal> for filters loaded
1884 <section id="class-FactoryStatic">
1885 <title><literal>mp::FactoryStatic</literal>
1886 (<filename>factory_static.cpp</filename>)</title>
1888 A subclass of <literal>FactoryFilter</literal> which is
1889 responsible for registering all the statically defined filter
1890 types. It does this by knowing about all those filters'
1891 structures, which are listed in its constructor. Merely
1892 instantiating this class registers all the static classes. It is
1893 for the benefit of this class that <literal>struct
1894 metaproxy_1_filter_struct</literal> exists, and that all the filter
1895 classes provide a static object of that type.
1899 <section id="class-filter-Base">
1900 <title><literal>mp::filter::Base</literal>
1901 (<filename>filter.cpp</filename>)</title>
1903 The virtual base class of all filters. The filter API is, on the
1904 surface at least, extremely simple: two methods.
1905 <literal>configure()</literal> is passed an XML DOM tree representing
1906 that part of the configuration file that pertains to this filter
1907 instance, and is expected to walk that tree extracting relevant
1908 information. And <literal>process()</literal> processes a
1909 package (see below). That surface simplicity is a bit
1910 misleading, as <literal>process()</literal> needs to know a lot
1911 about the <literal>Package</literal> class in order to do
1916 <section id="class-AuthSimple">
1917 <title><literal>mp::filter::AuthSimple</literal>,
1918 <literal>Backend_test</literal>, etc.
1919 (<filename>filter_auth_simple.cpp</filename>,
1920 <filename>filter_backend_test.cpp</filename>, etc.)</title>
1922 Individual filters. Each of these is implemented by a header and
1923 a source file, named <filename>filter_*.hpp</filename> and
1924 <filename>filter_*.cpp</filename> respectively. All the header
1925 files should be pretty much identical, in that they declare the
1926 class, including a private <literal>Rep</literal> class and a
1927 member pointer to it, and the two public methods.
1930 The source file for each filter needs to supply:
1935 A definition of the private <literal>Rep</literal> class.
1940 Some boilerplate constructors and destructors.
1945 A <literal>configure()</literal> method that uses the
1946 appropriate XML fragment.
1951 Most important, the <literal>process()</literal> method that
1952 does all the actual work.
1958 <section id="class-Package">
1959 <title><literal>mp::Package</literal>
1960 (<filename>package.cpp</filename>)</title>
1962 Represents a package on its way through the series of filters
1963 that make up a route. This is essentially a Z39.50 or SRU APDU
1964 together with information about where it came from, which is
1965 modified as it passes through the various filters.
1969 <section id="class-Pipe">
1970 <title><literal>mp::Pipe</literal>
1971 (<filename>pipe.cpp</filename>)</title>
1973 This class provides a compatibility layer so that we have an IPC
1974 mechanism that works the same under Unix and Windows. It's not
1975 particularly exciting.
1979 <section id="class-RouterChain">
1980 <title><literal>mp::RouterChain</literal>
1981 (<filename>router_chain.cpp</filename>)</title>
1987 <section id="class-RouterFleXML">
1988 <title><literal>mp::RouterFleXML</literal>
1989 (<filename>router_flexml.cpp</filename>)</title>
1995 <section id="class-Session">
1996 <title><literal>mp::Session</literal>
1997 (<filename>session.cpp</filename>)</title>
2003 <section id="class-ThreadPoolSocketObserver">
2004 <title><literal>mp::ThreadPoolSocketObserver</literal>
2005 (<filename>thread_pool_observer.cpp</filename>)</title>
2011 <section id="class-util">
2012 <title><literal>mp::util</literal>
2013 (<filename>util.cpp</filename>)</title>
2015 A namespace of various small utility functions and classes,
2016 collected together for convenience. Most importantly, includes
2017 the <literal>mp::util::odr</literal> class, a wrapper for YAZ's
2022 <section id="class-xml">
2023 <title><literal>mp::xml</literal>
2024 (<filename>xmlutil.cpp</filename>)</title>
2026 A namespace of various XML utility functions and classes,
2027 collected together for convenience.
2033 <section id="other.source.files">
2034 <title>Other Source Files</title>
2036 In addition to the Metaproxy source files that define the classes
2037 described above, there are a few additional files which are
2038 briefly described here:
2042 <term><literal>metaproxy_prog.cpp</literal></term>
2045 The main function of the <command>metaproxy</command> program.
2050 <term><literal>ex_router_flexml.cpp</literal></term>
2053 Identical to <literal>metaproxy_prog.cpp</literal>: it's not clear why.
2058 <term><literal>test_*.cpp</literal></term>
2061 Unit-tests for various modules.
2067 ### Still to be described:
2068 <literal>ex_filter_frontend_net.cpp</literal>,
2069 <literal>filter_dl.cpp</literal>,
2070 <literal>plainfile.cpp</literal>,
2071 <literal>tstdl.cpp</literal>.
2077 <reference id="reference">
2078 <title>Reference</title>
2079 <partintro id="reference-introduction">
2081 The material in this chapter is drawn directly from the individual
2082 manual entries. In particular, the Metaproxy invocation section is
2083 available using <command>man metaproxy</command>, and the section
2084 on each individual filter is available using the name of the filter
2085 as the argument to the <command>man</command> command.
2091 <appendix id="license">
2092 <title>License</title>
2097 Metaproxy is free software; you can redistribute it and/or modify it under
2098 the terms of the GNU General Public License as published by the Free
2099 Software Foundation; either version 2, or (at your option) any later
2104 Metaproxy is distributed in the hope that it will be useful, but WITHOUT ANY
2105 WARRANTY; without even the implied warranty of MERCHANTABILITY or
2106 FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
2111 You should have received a copy of the GNU General Public License
2112 along with Metaproxy; see the file LICENSE. If not, write to the
2113 Free Software Foundation,
2114 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
2122 <!-- Keep this comment at the end of the file
2125 nxml-child-indent: 1