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 can be compiled with Microsoft
373 <ulink url="&url.vstudio;">Visual Studio</ulink>.
374 Versions 2003 (C 7.1), 2005 (C 8.0), 2008 (C 9.0) and 2013 (C 12.0)
377 <section id="installation.windows.boost">
380 For Windows, it's easiest to get the precompiled Boost
381 package from <ulink url="&url.boost.windows.download;">here</ulink>.
382 Several versions of the Boost libraries may be selected when
383 installing Boost for windows. Please choose at least the
384 <emphasis>multithreaded</emphasis> (non-DLL) version because
385 the Metaproxy makefile uses that.
388 For more information about installing Boost refer to the
389 <ulink url="&url.boost.getting.started;">getting started</ulink>
394 <section id="installation.windows.libxslt">
395 <title>Libxslt</title>
397 <ulink url="&url.libxslt;">Libxslt</ulink> can be downloaded
399 <ulink url="&url.libxml2.download.windows;">here</ulink>.
402 Libxslt also requires libxml2 to operate.
406 <section id="installation.windows.yaz">
409 <ulink url="&url.yaz;">YAZ</ulink> can be downloaded
411 <ulink url="&url.yaz.download.win32;">here</ulink>.
415 <section id="installation.windows.yazplusplus">
418 Get <ulink url="&url.yazplusplus;">YAZ++</ulink> as well.
419 Version 1.5.2 or later is required.
422 YAZ++ includes NMAKE makefiles, similar to those found in the
427 <section id="installation.windows.metaproxy">
428 <title>Metaproxy</title>
430 Metaproxy is shipped with NMAKE makefiles as well - similar
431 to those found in the YAZ++/YAZ packages. Adjust this Makefile
432 to point to the proper locations of Boost, Libxslt, Libxml2,
433 zlib, iconv, yaz and yazpp.
437 <varlistentry><term><literal>DEBUG</literal></term>
439 If set to 1, the software is
440 compiled with debugging libraries (code generation is
441 multi-threaded debug DLL).
442 If set to 0, the software is compiled with release libraries
443 (code generation is multi-threaded DLL).
448 <term><literal>BOOST</literal></term>
451 Boost install location
457 <term><literal>BOOST_VERSION</literal></term>
460 Boost version (replace . with _).
466 <term><literal>BOOST_TOOLSET</literal></term>
475 <term><literal>LIBXSLT_DIR</literal>,
476 <literal>LIBXML2_DIR</literal> ..</term>
479 Specify the locations of Libxslt, libiconv, libxml2 and
488 After successful compilation you'll find
489 <literal>metaproxy.exe</literal> in the
490 <literal>bin</literal> directory.
498 <chapter id="yazproxy-comparison">
499 <title>YAZ Proxy Comparison</title>
501 The table below lists facilities either supported by either
502 <ulink url="&url.yazproxy;">YAZ Proxy</ulink> or Metaproxy.
504 <table id="yazproxy-comparison-table">
505 <title>Metaproxy / YAZ Proxy comparison</title>
509 <entry>Facility</entry>
510 <entry>Metaproxy</entry>
511 <entry>YAZ Proxy</entry>
516 <entry>Z39.50 server</entry>
517 <entry>Using filter <xref linkend="ref-frontend_net"/></entry>
518 <entry>Supported</entry>
521 <entry>SRU server</entry>
522 <entry>Supported with filter <xref linkend="ref-sru_z3950"/></entry>
523 <entry>Supported</entry>
526 <entry>Z39.50 client</entry>
527 <entry>Supported with filter <xref linkend="ref-z3950_client"/></entry>
528 <entry>Supported</entry>
531 <entry>SRU client</entry>
532 <entry>Supported with filter <xref linkend="ref-zoom"/></entry>
533 <entry>Unsupported</entry>
536 <entry>Connection reuse</entry>
537 <entry>Supported with filter <literal>session_shared</literal></entry>
538 <entry>Supported</entry>
541 <entry>Connection share</entry>
542 <entry>Supported with filter <literal>session_shared</literal></entry>
543 <entry>Unsupported</entry>
546 <entry>Result set reuse</entry>
547 <entry>Supported with filter <literal>session_shared</literal></entry>
548 <entry>Within one Z39.50 session / HTTP keep-alive</entry>
551 <entry>Record cache</entry>
552 <entry>Supported by filter <literal>session_shared</literal></entry>
553 <entry>Supported for last result set within one Z39.50/HTTP-keep alive session</entry>
556 <entry>Z39.50 Virtual database, i.e. select any Z39.50 target for database</entry>
557 <entry>Supported with filter <literal>virt_db</literal></entry>
558 <entry>Unsupported</entry>
561 <entry>SRU Virtual database, i.e. select any Z39.50 target for path</entry>
562 <entry>Supported with filter <literal>virt_db</literal>,
563 <literal>sru_z3950</literal></entry>
564 <entry>Supported</entry>
567 <entry>Multi target search</entry>
568 <entry>Supported with filter <literal>multi</literal> (round-robin)</entry>
569 <entry>Unsupported</entry>
572 <entry>Retrieval and search limits</entry>
573 <entry>Supported using filter <literal>limit</literal></entry>
574 <entry>Supported</entry>
577 <entry>Bandwidth limits</entry>
578 <entry>Supported using filter <literal>limit</literal></entry>
579 <entry>Supported</entry>
582 <entry>Connect limits</entry>
583 <entry>Supported by filter <literal>frontend_net</literal> (connect-max)</entry>
584 <entry>Supported</entry>
587 <entry>Retrieval sanity check and conversions</entry>
588 <entry>Supported using filter <literal>record_transform</literal></entry>
589 <entry>Supported</entry>
592 <entry>Query check</entry>
594 Supported by <literal>query_rewrite</literal> which may be check
595 a query and throw diagnostics (errors)
597 <entry>Supported</entry>
600 <entry>Query rewrite</entry>
601 <entry>Supported with <literal>query_rewrite</literal></entry>
602 <entry>Unsupported</entry>
605 <entry>Session invalidate for -1 hits</entry>
606 <entry>Unsupported</entry>
607 <entry>Supported</entry>
610 <entry>Architecture</entry>
611 <entry>Multi-threaded + select for networked modules such as
612 <literal>frontend_net</literal>)</entry>
613 <entry>Single-threaded using select</entry>
617 <entry>Extensability</entry>
618 <entry>Most functionality implemented as loadable modules</entry>
619 <entry>Unsupported and experimental</entry>
623 <entry><ulink url="&url.usemarcon;">USEMARCON</ulink></entry>
624 <entry>Supported with <literal>record_transform</literal></entry>
625 <entry>Supported</entry>
629 <entry>Portability</entry>
631 Requires YAZ, YAZ++ and modern C++ compiler supporting
632 <ulink url="&url.boost;">Boost</ulink>.
635 Requires YAZ and YAZ++.
636 STL is not required so pretty much any C++ compiler out there should work.
645 <chapter id="architecture">
646 <title>The Metaproxy Architecture</title>
648 The Metaproxy architecture is based on three concepts:
649 the <emphasis>package</emphasis>,
650 the <emphasis>route</emphasis>
651 and the <emphasis>filter</emphasis>.
655 <term>Packages</term>
658 A package is request or response, encoded in some protocol,
659 issued by a client, making its way through Metaproxy, send to or
660 received from a server, or sent back to the client.
663 The core of a package is the protocol unit - for example, a
664 Z39.50 Init Request or Search Response, or an SRU searchRetrieve
665 URL or Explain Response. In addition to this core, a package
666 also carries some extra information added and used by Metaproxy
670 In general, packages are doctored as they pass through
671 Metaproxy. For example, when the proxy performs authentication
672 and authorization on a Z39.50 Init request, it removes the
673 authentication credentials from the package so that they are not
674 passed onto the back-end server; and when search-response
675 packages are obtained from multiple servers, they are merged
676 into a single unified package that makes its way back to the
685 Packages make their way through routes, which can be thought of
686 as programs that operate on the package data-type. Each
687 incoming package initially makes its way through a default
688 route, but may be switched to a different route based on various
689 considerations. Routes are made up of sequences of filters (see
698 Filters provide the individual instructions within a route, and
699 effect the necessary transformations on packages. A particular
700 configuration of Metaproxy is essentially a set of filters,
701 described by configuration details and arranged in order in one
702 or more routes. There are many kinds of filter - about a dozen
703 at the time of writing with more appearing all the time - each
704 performing a specific function and configured by different
708 The word ``filter'' is sometimes used rather loosely, in two
709 different ways: it may be used to mean a particular
710 <emphasis>type</emphasis> of filter, as when we speak of ``the
711 auth_simple filter'' or ``the multi filter''; or it may be used
712 to be a specific <emphasis>instance</emphasis> of a filter
713 within a Metaproxy configuration. For example, a single
714 configuration will often contain multiple instances of the
715 <literal>z3950_client</literal> filter. In
716 operational terms, of these is a separate filter. In practice,
717 context always make it clear which sense of the word ``filter''
721 Extensibility of Metaproxy is primarily through the creation of
722 plugins that provide new filters. The filter API is small and
723 conceptually simple, but there are many details to master. See
725 <link linkend="filters">Filters</link>.
731 Since packages are created and handled by the system itself, and
732 routes are conceptually simple, most of the remainder of this
733 document concentrates on filters. After a brief overview of the
734 filter types follows, along with some thoughts on possible future
741 <chapter id="filters">
742 <title>Filters</title>
745 <section id="filters-introductory-notes">
746 <title>Introductory notes</title>
748 It's useful to think of Metaproxy as an interpreter providing a small
749 number of primitives and operations, but operating on a very
750 complex data type, namely the ``package''.
753 A package represents a Z39.50 or SRU/W request (whether for Init,
754 Search, Scan, etc.) together with information about where it came
755 from. Packages are created by front-end filters such as
756 <literal>frontend_net</literal> (see below), which reads them from
757 the network; other front-end filters are possible. They then pass
758 along a route consisting of a sequence of filters, each of which
759 transforms the package and may also have side-effects such as
760 generating logging. Eventually, the route will yield a response,
761 which is sent back to the origin.
764 There are many kinds of filter: some that are defined statically
765 as part of Metaproxy, and others may be provided by third parties
766 and dynamically loaded. They all conform to the same simple API
767 of essentially two methods: <function>configure()</function> is
768 called at startup time, and is passed an XML DOM tree representing that
769 part of the configuration file that pertains to this filter
770 instance: it is expected to walk that tree extracting relevant
771 information; and <function>process()</function> is called every
772 time the filter has to processes a package.
775 While all filters provide the same API, there are different modes
776 of functionality. Some filters are sources: they create
778 (<literal>frontend_net</literal>);
779 others are sinks: they consume packages and return a result
780 (<literal>backend_test</literal>,
781 <literal>bounce</literal>,
782 <literal>http_file</literal>,
783 <literal>z3950_client</literal>);
784 the others are true filters, that read, process and pass on the
785 packages they are fed
786 (<literal>auth_simple</literal>,
787 <literal>log</literal>,
788 <literal>multi</literal>,
789 <literal>query_rewrite</literal>,
790 <literal>record_transform</literal>,
791 <literal>session_shared</literal>,
792 <literal>sru_z3950</literal>,
793 <literal>template</literal>,
794 <literal>virt_db</literal>).
799 <section id="overview.filter.types">
800 <title>Overview of filter types</title>
802 We now briefly consider each of the types of filter supported by
803 the core Metaproxy binary. This overview is intended to give a
804 flavor of the available functionality; more detailed information
805 about each type of filter is included below in
806 <xref linkend="reference"/>.
809 The filters are here named by the string that is used as the
810 <literal>type</literal> attribute of a
811 <literal><filter></literal> element in the configuration
812 file to request them, with the name of the class that implements
813 them in parentheses. (The classname is not needed for normal
814 configuration and use of Metaproxy; it is useful only to
818 The filters are here listed in alphabetical order:
825 New virt_db-alike that does inteligent peer choice, explain merging,
826 adds FD&N to explain. Keeps init responses (like "virt_db Classic"),
827 makes routing choices based on local explain knowledge. Ref IDDI
830 Filter to convert Explain Classic to ZeeRex.
832 CQL2PQF (which needs augmented ZeeRex) - MARC for Talis.
836 Figure out what additional information we need in:
837 ZeeRex (check against D3.1)
838 Init request (e.g. loop detection)
839 Query package (e.g. number of hops)
840 Query response (e.g. record source)
844 <section id="auth_simple">
845 <title><literal>auth_simple</literal>
846 (mp::filter::AuthSimple)</title>
848 Simple authentication and authorization. The configuration
849 specifies the name of a file that is the user register, which
850 lists <varname>username</varname>:<varname>password</varname>
851 pairs, one per line, colon separated. When a session begins, it
852 is rejected unless username and passsword are supplied, and match
853 a pair in the register. The configuration file may also specific
854 the name of another file that is the target register: this lists
855 lists <varname>username</varname>:<varname>dbname</varname>,<varname>dbname</varname>...
856 sets, one per line, with multiple database names separated by
857 commas. When a search is processed, it is rejected unless the
858 database to be searched is one of those listed as available to
863 <section id="backend_test">
864 <title><literal>backend_test</literal>
865 (mp::filter::Backend_test)</title>
867 A partial sink that provides dummy responses in the manner of the
868 <literal>yaz-ztest</literal> Z39.50 server. This is useful only
869 for testing. Seriously, you don't need this. Pretend you didn't
870 even read this section.
874 <section id="bounce">
875 <title><literal>bounce</literal>
876 (mp::filter::Bounce)</title>
878 A sink that swallows <emphasis>all packages</emphasis>,
879 and returns them almost unprocessed.
880 It never sends any package of any type further down the row, but
881 sets Z39.50 packages to Z_Close, and HTTP_Request packages to
882 HTTP_Response err code 400 packages, and adds a suitable bounce
884 The bounce filter is usually added at end of each filter chain route
885 to prevent infinite hanging of for example HTTP
886 requests packages when only the Z39.50 client partial sink
887 filter is found in the
892 <section id="cql_rpn">
893 <title><literal>cql_rpn</literal>
894 (mp::filter::CQLtoRPN)</title>
896 A query language transforming filter which catches Z39.50
897 <literal>searchRequest</literal>
898 packages containing <literal>CQL</literal> queries, transforms
899 those to <literal>RPN</literal> queries,
900 and sends the <literal>searchRequests</literal> on to the next
901 filters. It is among other things useful in a SRU context.
905 <section id="frontend_net">
906 <title><literal>frontend_net</literal>
907 (mp::filter::FrontendNet)</title>
909 A source that accepts Z39.50 connections from a port
910 specified in the configuration, reads protocol units, and
911 feeds them into the next filter in the route. When the result is
912 received, it is returned to the original origin.
916 <section id="http_file">
917 <title><literal>http_file</literal>
918 (mp::filter::HttpFile)</title>
920 A partial sink which swallows only
921 <literal>HTTP_Request</literal> packages, and
922 returns the contents of files from the local
923 filesystem in response to HTTP requests.
924 It lets Z39.50 packages and all other forthcoming package types
927 does mean that Metaproxy is also a Web-server in its spare time. So
928 far it does not contain either an email-reader or a Lisp
929 interpreter, but that day is surely coming.)
933 <section id="load_balance">
934 <title><literal>load_balance</literal>
935 (mp::filter::LoadBalance)</title>
937 Performs load balancing for incoming Z39.50 init requests.
938 It is used together with the <literal>virt_db</literal> filter,
939 but unlike the <literal>multi</literal> filter it does send an
940 entire session to only one of the virtual backends. The
941 <literal>load_balance</literal> filter is assuming that
942 all backend targets have equal content, and chooses the backend
943 with least load cost for a new session.
946 This filter is experimental and yet not mature for heavy load
954 <title><literal>log</literal>
955 (mp::filter::Log)</title>
957 Writes logging information to standard output, and passes on
958 the package unchanged. A log file name can be specified, as well
959 as multiple different logging formats.
964 <title><literal>multi</literal>
965 (mp::filter::Multi)</title>
967 Performs multi-database searching.
969 <link linkend="multidb">the extended discussion</link>
970 of virtual databases and multi-database searching below.
974 <section id="query_rewrite">
975 <title><literal>query_rewrite</literal>
976 (mp::filter::QueryRewrite)</title>
978 Rewrites Z39.50 <literal>Type-1</literal>
979 and <literal>Type-101</literal> (``<literal>RPN</literal>'')
981 three-step process: the query is transliterated from Z39.50
982 packet structures into an XML representation; that XML
983 representation is transformed by an XSLT stylesheet; and the
984 resulting XML is transliterated back into the Z39.50 packet
990 <section id="record_transform">
991 <title><literal>record_transform</literal>
992 (mp::filter::RecordTransform)</title>
994 This filter acts only on Z3950 present requests, and let all
995 other types of packages and requests pass untouched. It's use is
996 twofold: blocking Z3950 present requests, which the backend
997 server does not understand and can not honor, and transforming
998 the present syntax and elementset name according to the rules
999 specified, to fetch only existing record formats, and transform
1000 them on the fly to requested record syntaxes.
1004 <section id="session_shared">
1005 <title><literal>session_shared</literal>
1006 (mp::filter::SessionShared)</title>
1008 This filter implements global sharing of
1009 result sets (i.e. between threads and therefore between
1010 clients), yielding performance improvements by clever resource
1015 <section id="sru_z3950">
1016 <title><literal>sru_z3950</literal>
1017 (mp::filter::SRUtoZ3950)</title>
1019 This filter transforms valid
1020 SRU GET/POST/SOAP searchRetrieve requests to Z3950 init, search,
1021 and present requests, and wraps the
1022 received hit counts and XML records into suitable SRU response
1024 The <literal>sru_z3950</literal> filter processes also SRU
1025 GET/POST/SOAP explain requests, returning
1026 either the absolute minimum required by the standard, or a full
1027 pre-defined ZeeReX explain record.
1029 <ulink url="&url.zeerex.explain;">ZeeReX Explain</ulink>
1030 standard pages and the
1031 <ulink url="&url.sru.explain;">SRU Explain</ulink> pages
1032 for more information on the correct explain syntax.
1033 SRU scan requests are not supported yet.
1037 <section id="template">
1038 <title><literal>template</literal>
1039 (mp::filter::Template)</title>
1041 Does nothing at all, merely passing the packet on. (Maybe it
1042 should be called <literal>nop</literal> or
1043 <literal>passthrough</literal>?) This exists not to be used, but
1044 to be copied - to become the skeleton of new filters as they are
1045 written. As with <literal>backend_test</literal>, this is not
1046 intended for civilians.
1050 <section id="virt_db">
1051 <title><literal>virt_db</literal>
1052 (mp::filter::VirtualDB)</title>
1054 Performs virtual database selection: based on the name of the
1055 database in the search request, a server is selected, and its
1056 address added to the request in a <literal>VAL_PROXY</literal>
1057 otherInfo packet. It will subsequently be used by a
1058 <literal>z3950_client</literal> filter.
1060 <link linkend="multidb">the extended discussion</link>
1061 of virtual databases and multi-database searching below.
1065 <section id="z3950_client">
1066 <title><literal>z3950_client</literal>
1067 (mp::filter::Z3950Client)</title>
1069 A partial sink which swallows only Z39.50 packages.
1070 It performs Z39.50 searching and retrieval by proxying the
1071 packages that are passed to it. Init requests are sent to the
1072 address specified in the <literal>VAL_PROXY</literal> otherInfo
1073 attached to the request: this may have been specified by client,
1074 or generated by a <literal>virt_db</literal> filter earlier in
1075 the route. Subsequent requests are sent to the same address,
1076 which is remembered at Init time in a Session object.
1077 HTTP_Request packages and all other forthcoming package types
1078 are passed untouched.
1083 <section id="zeerex_explain">
1084 <title><literal>zeerex_explain</literal>
1085 (mp::filter::ZeerexExplain)</title>
1087 This filter acts as a sink for
1088 Z39.50 explain requests, returning a static ZeeReX
1089 Explain XML record from the config section. All other packages
1092 <ulink url="&url.zeerex.explain;">ZeeReX Explain</ulink>
1094 for more information on the correct explain syntax.
1098 This filter is not yet completed.
1107 <section id="future.directions">
1108 <title>Future directions</title>
1110 Some other filters that do not yet exist, but which would be
1111 useful, are briefly described. These may be added in future
1112 releases (or may be created by third parties, as loadable
1118 <term><literal>frontend_cli</literal> (source)</term>
1121 Command-line interface for generating requests.
1126 <term><literal>sru_client</literal> (sink)</term>
1129 SRU/GET and SRU/SOAP searching and retrieval.
1134 <term><literal>opensearch_client</literal> (sink)</term>
1137 A9 OpenSearch searching and retrieval.
1147 <chapter id="configuration">
1148 <title>Configuration: the Metaproxy configuration file format</title>
1151 <section id="configuration-introductory-notes">
1152 <title>Introductory notes</title>
1154 If Metaproxy is an interpreter providing operations on packages, then
1155 its configuration file can be thought of as a program for that
1156 interpreter. Configuration is by means of a single XML file, the name
1157 of which is supplied as the sole command-line argument to the
1158 <command>metaproxy</command> program. (See
1159 <xref linkend="reference"/> below for more information on invoking
1164 <section id="overview.xml.structure">
1165 <title>Overview of the config file XML structure</title>
1167 All elements and attributes are in the namespace
1168 <ulink url="http://indexdata.com/metaproxy"/>.
1169 This is most easily achieved by setting the default namespace on
1170 the top-level element, as here:
1173 <metaproxy xmlns="http://indexdata.com/metaproxy" version="1.0">
1176 The top-level element is <metaproxy>. This contains
1177 a <dlpath> element,
1178 a <start> element,
1179 a <filters> element and
1180 a <routes> element, in that order. <dlpath> and
1181 <filters> are optional; the other two are mandatory.
1182 All four are non-repeatable.
1185 The <dlpath;> element contains a text element which
1186 specifies the location of filter modules. This is only needed
1187 if Metaproxy must load 3rd party filters (most filters with Metaproxy
1188 are built into the Metaproxy application).
1191 The <start> element is empty, but carries a
1192 <literal>route</literal> attribute, whose value is the name of
1193 route at which to start running - analogous to the name of the
1194 start production in a formal grammar.
1197 If present, <filters> contains zero or more <filter>
1198 elements. Each filter carries a <literal>type</literal> attribute
1199 which specifies what kind of filter is being defined
1200 (<literal>frontend_net</literal>, <literal>log</literal>, etc.)
1201 and contain various elements that provide suitable configuration
1202 for a filter of its type. The filter-specific elements are
1204 <xref linkend="reference"/>.
1205 Filters defined in this part of the file must carry an
1206 <literal>id</literal> attribute so that they can be referenced
1210 <routes> contains one or more <route> elements, each
1211 of which must carry an <literal>id</literal> element. One of the
1212 routes must have the ID value that was specified as the start
1213 route in the <start> element's <literal>route</literal>
1214 attribute. Each route contains zero or more <filter>
1215 elements. These are of two types. They may be empty, but carry a
1216 <literal>refid</literal> attribute whose value is the same as the
1217 <literal>id</literal> of a filter previously defined in the
1218 <filters> section. Alternatively, a route within a filter
1219 may omit the <literal>refid</literal> attribute, but contain
1220 configuration elements similar to those used for filters defined
1221 in the <filters> section. (In other words, each filter in a
1222 route may be included either by reference or by physical
1228 <section id="example.configuration">
1229 <title>An example configuration</title>
1231 The following is a small, but complete, Metaproxy configuration
1232 file (included in the distribution as
1233 <literal>metaproxy/etc/config1.xml</literal>).
1234 This file defines a very simple configuration that simply proxies
1235 to whatever back-end server the client requests, but logs each
1236 request and response. This can be useful for debugging complex
1237 client-server dialogues.
1239 <screen><![CDATA[<?xml version="1.0"?>
1240 <metaproxy xmlns="http://indexdata.com/metaproxy" version="1.0">
1241 <dlpath>/usr/lib/metaproxy/modules</dlpath>
1242 <start route="start"/>
1244 <filter id="frontend" type="frontend_net">
1247 <filter id="backend" type="z3950_client">
1252 <filter refid="frontend"/>
1253 <filter type="log"/>
1254 <filter refid="backend"/>
1255 <filter type="bounce"/>
1261 It works by defining a single route, called
1262 <literal>start</literal>, which consists of a sequence of four
1263 filters. The first and last of these are included by reference:
1264 their <literal><filter></literal> elements have
1265 <literal>refid</literal> attributes that refer to filters defined
1266 within the prior <literal><filters></literal> section. The
1267 middle filter is included inline in the route.
1270 The four filters in the route are as follows: first, a
1271 <literal>frontend_net</literal> filter accepts Z39.50 requests
1272 from any host on port 9000; then these requests are passed through
1273 a <literal>log</literal> filter that emits a message for each
1274 request; they are then fed into a <literal>z3950_client</literal>
1275 filter, which forwards all Z39.50 requests to the client-specified
1276 back-end Z39.509 server. Those Z39.50 packages are returned by the
1277 <literal>z3950_client</literal> filter, with the response data
1278 filled by the external Z39.50 server targeted.
1279 All non-Z39.50 packages are passed through to the
1280 <literal>bounce</literal> filter, which definitely bounces
1281 everything, including fish, bananas, cold pyjamas,
1282 mutton, beef and trout packages.
1283 When the response arrives, it is handed
1284 back to the <literal>log</literal> filter, which emits another
1285 message; and then to the <literal>frontend_net</literal> filter,
1286 which returns the response to the client.
1290 <section id="config-file-modularity">
1291 <title>Config file modularity</title>
1293 Metaproxy XML configuration snippets can be reused by other
1294 filters using the <literal>XInclude</literal> standard, as seen in
1295 the <literal>/etc/config-sru-to-z3950.xml</literal> example SRU
1298 <filter id="sru" type="sru_z3950">
1299 <database name="Default">
1300 <xi:include xmlns:xi="http://www.w3.org/2001/XInclude"
1301 href="explain.xml"/>
1308 <section id="config-file-syntax-check">
1309 <title>Config file syntax checking</title>
1311 The distribution contains RelaxNG Compact and XML syntax checking
1312 files, as well as XML Schema files. These are found in the
1315 xml/schema/metaproxy.rnc
1316 xml/schema/metaproxy.rng
1317 xml/schema/metaproxy.xsd
1319 and can be used to verify or debug the XML structure of
1320 configuration files. For example, using the utility
1321 <filename>xmllint</filename>, syntax checking is done like this:
1323 xmllint --noout --schema xml/schema/metaproxy.xsd etc/config-local.xml
1324 xmllint --noout --relaxng xml/schema/metaproxy.rng etc/config-local.xml
1326 (A recent version of <literal>libxml2</literal> is required, as
1327 support for XML Schemas is a relatively recent addition.)
1330 You can of course use any other RelaxNG or XML Schema compliant tool
1338 <chapter id="multidb">
1339 <title>Virtual databases and multi-database searching</title>
1342 <section id="multidb-introductory-notes">
1343 <title>Introductory notes</title>
1345 Two of Metaproxy's filters are concerned with multiple-database
1346 operations. Of these, <literal>virt_db</literal> can work alone
1347 to control the routing of searches to one of a number of servers,
1348 while <literal>multi</literal> can work together with
1349 <literal>virt_db</literal> to perform multi-database searching, merging
1350 the results into a unified result-set - ``metasearch in a box''.
1353 The interaction between
1354 these two filters is necessarily complex: it reflects the real,
1355 irreducible complexity of multi-database searching in a protocol such
1356 as Z39.50 that separates initialization from searching, and in
1357 which the database to be searched is not known at initialization
1361 It's possible to use these filters without understanding the
1362 details of their functioning and the interaction between them; the
1363 next two sections of this chapter are ``HOW-TO'' guides for doing
1364 just that. However, debugging complex configurations will require
1365 a deeper understanding, which the last two sections of this
1366 chapters attempt to provide.
1371 <section id="multidb.virt_db">
1372 <title>Virtual databases with the <literal>virt_db</literal> filter</title>
1374 Working alone, the purpose of the
1375 <literal>virt_db</literal>
1376 filter is to route search requests to one of a selection of
1377 back-end databases. In this way, a single Z39.50 endpoint
1378 (running Metaproxy) can provide access to several different
1379 underlying services, including those that would otherwise be
1380 inaccessible due to firewalls. In many useful configurations, the
1381 back-end databases are local to the Metaproxy installation, but
1382 the software does not enforce this, and any valid Z39.50 servers
1383 may be used as back-ends.
1386 For example, a <literal>virt_db</literal>
1387 filter could be set up so that searches in the virtual database
1388 ``lc'' are forwarded to the Library of Congress bibliographic
1389 catalogue server, and searches in the virtual database ``marc''
1390 are forwarded to the toy database of MARC records that Index Data
1391 hosts for testing purposes. A <literal>virt_db</literal>
1392 configuration to make this switch would look like this:
1394 <screen><![CDATA[<filter type="virt_db">
1396 <database>lc</database>
1397 <target>z3950.loc.gov:7090/voyager</target>
1400 <database>marc</database>
1401 <target>indexdata.com/marc</target>
1403 </filter>]]></screen>
1405 As well as being useful in it own right, this filter also provides
1406 the foundation for multi-database searching.
1411 <section id="multidb.multi">
1412 <title>Multi-database search with the <literal>multi</literal> filter</title>
1414 To arrange for Metaproxy to broadcast searches to multiple back-end
1415 servers, the configuration needs to include two components: a
1416 <literal>virt_db</literal>
1417 filter that specifies multiple
1418 <literal><target></literal>
1419 elements, and a subsequent
1420 <literal>multi</literal>
1421 filter. Here, for example, is a complete configuration that
1422 broadcasts searches to both the Library of Congress catalogue and
1423 Index Data's tiny testing database of MARC records:
1425 <screen><![CDATA[<?xml version="1.0"?>
1426 <metaproxy xmlns="http://indexdata.com/metaproxy" version="1.0">
1427 <start route="start"/>
1430 <filter type="frontend_net">
1431 <threads>10</threads>
1434 <filter type="virt_db">
1436 <database>lc</database>
1437 <target>z3950.loc.gov:7090/voyager</target>
1440 <database>marc</database>
1441 <target>indexdata.com/marc</target>
1444 <database>all</database>
1445 <target>z3950.loc.gov:7090/voyager</target>
1446 <target>indexdata.com/marc</target>
1449 <filter type="multi"/>
1450 <filter type="z3950_client">
1451 <timeout>30</timeout>
1453 <filter type="bounce"/>
1456 </metaproxy>]]></screen>
1459 <literal>virt_db</literal>
1460 filter that specifies multiple
1461 <literal><target></literal>
1462 elements but without a subsequent
1463 <literal>multi</literal>
1464 filter yields surprising and undesirable results, as will be
1465 described below. Don't do that.)
1468 Metaproxy can be invoked with this configuration as follows:
1470 <screen>../src/metaproxy --config config-simple-multi.xml</screen>
1472 And thereafter, Z39.50 clients can connect to the running server
1473 (on port 9000, as specified in the configuration) and search in
1474 any of the databases
1475 <literal>lc</literal> (the Library of Congress catalogue),
1476 <literal>marc</literal> (Index Data's test database of MARC records)
1478 <literal>all</literal> (both of these). As an example, a session
1479 using the YAZ command-line client <literal>yaz-client</literal> is
1480 here included (edited for brevity and clarity):
1482 <screen><![CDATA[$ yaz-client @:9000
1486 Search was a success.
1487 Number of hits: 10000, setno 1
1491 Search was a success.
1492 Number of hits: 10, setno 3
1496 Search was a success.
1497 Number of hits: 10010, setno 4
1500 [marc]Record type: USmarc
1503 005 00000000000000.0
1504 008 910710c19910701nju 00010 eng
1508 100 10 $a Jack Collins
1509 245 10 $a How to program a computer
1515 [VOYAGER]Record type: USmarc
1517 005 20041229102447.0
1518 008 030910s2004 caua 000 0 eng
1519 035 $a (DLC) 2003112666
1520 906 $a 7 $b cbc $c orignew $d 4 $e epcn $f 20 $g y-gencatlg
1521 925 0 $a acquire $b 1 shelf copy $x policy default
1522 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.
1525 040 $a DLC $c DLC $d DLC
1526 050 00 $a MLCM 2004/03312 (G)
1527 245 10 $a 007, everything or nothing : $b Prima's official strategy guide / $c created by Kaizen Media Group.
1528 246 3 $a Double-O-seven, everything or nothing
1529 246 30 $a Prima's official strategy guide
1530 260 $a Roseville, CA : $b Prima Games, $c c2004.
1531 300 $a 161 p. : $b col. ill. ; $c 28 cm.
1532 500 $a "Platforms: Nintendo GameCube, Macintosh, PC, PlayStation 2 computer entertainment system, Xbox"--P. [4] of cover.
1533 650 0 $a Video games.
1534 710 2 $a Kaizen Media Group.
1535 856 42 $3 Publisher description $u http://www.loc.gov/catdir/description/random052/2003112666.html
1540 As can be seen, the first record in the result set is from the
1541 Index Data test database, and the second from the Library of
1542 Congress database. The result-set continues alternating records
1543 round-robin style until the point where one of the databases'
1544 records are exhausted.
1547 This example uses only two back-end databases; more may be used.
1548 There is no limitation imposed on the number of databases that may
1549 be metasearched in this way: issues of resource usage and
1550 administrative complexity dictate the practical limits.
1553 What happens when one of the databases doesn't respond? By default,
1554 the entire multi-database search fails, and the appropriate
1555 diagnostic is returned to the client. This is usually appropriate
1556 during development, when technicians need maximum information, but
1557 can be inconvenient in deployment, when users typically don't want
1558 to be bothered with problems of this kind and prefer just to get
1559 the records from the databases that are available. To obtain this
1560 latter behavior add an empty
1561 <literal><hideunavailable></literal>
1563 <literal>multi</literal> filter:
1565 <screen><![CDATA[ <filter type="multi">
1567 </filter>]]></screen>
1569 Under this regime, an error is reported to the client only if
1570 <emphasis>all</emphasis> the databases in a multi-database search
1576 <section id="multidb.what">
1577 <title>What's going on?</title>
1579 <title>Lark's vomit</title>
1581 This section goes into a level of technical detail that is
1582 probably not necessary in order to configure and use Metaproxy.
1583 It is provided only for those who like to know how things work.
1584 You should feel free to skip on to the next section if this one
1585 doesn't seem like fun.
1589 Hold on tight - this may get a little hairy.
1592 In the general course of things, a Z39.50 Init request may carry
1593 with it an otherInfo packet of type <literal>VAL_PROXY</literal>,
1594 whose value indicates the address of a Z39.50 server to which the
1595 ultimate connection is to be made. (This otherInfo packet is
1596 supported by YAZ-based Z39.50 clients and servers, but has not yet
1597 been ratified by the Maintenance Agency and so is not widely used
1598 in non-Index Data software. We're working on it.)
1599 The <literal>VAL_PROXY</literal> packet functions
1600 analogously to the absoluteURI-style Request-URI used with the GET
1601 method when a web browser asks a proxy to forward its request: see
1603 <ulink url="http://www.w3.org/Protocols/rfc2616/rfc2616-sec5.html#sec5.1.2"
1604 >Request-URI</ulink>
1606 <ulink url="http://www.w3.org/Protocols/rfc2616/rfc2616.html"
1607 >the HTTP 1.1 specification</ulink>.
1610 Within Metaproxy, Search requests that are part of the same
1611 session as an Init request that carries a
1612 <literal>VAL_PROXY</literal> otherInfo are also annotated with the
1613 same information. The role of the <literal>virt_db</literal>
1614 filter is to rewrite this otherInfo packet dependent on the
1615 virtual database that the client wants to search.
1618 When Metaproxy receives a Z39.50 Init request from a client, it
1619 doesn't immediately forward that request to the back-end server.
1620 Why not? Because it doesn't know <emphasis>which</emphasis>
1621 back-end server to forward it to until the client sends a Search
1622 request that specifies the database that it wants to search in.
1623 Instead, it just treasures the Init request up in its heart; and,
1624 later, the first time the client does a search on one of the
1625 specified virtual databases, a connection is forged to the
1626 appropriate server and the Init request is forwarded to it. If,
1627 later in the session, the same client searches in a different
1628 virtual database, then a connection is forged to the server that
1629 hosts it, and the same cached Init request is forwarded there,
1633 All of this clever Init-delaying is done by the
1634 <literal>frontend_net</literal> filter. The
1635 <literal>virt_db</literal> filter knows nothing about it; in
1636 fact, because the Init request that is received from the client
1637 doesn't get forwarded until a Search request is received, the
1638 <literal>virt_db</literal> filter (and the
1639 <literal>z3950_client</literal> filter behind it) doesn't even get
1640 invoked at Init time. The <emphasis>only</emphasis> thing that a
1641 <literal>virt_db</literal> filter ever does is rewrite the
1642 <literal>VAL_PROXY</literal> otherInfo in the requests that pass
1646 It is possible for a <literal>virt_db</literal> filter to contain
1648 <literal><target></literal>
1649 elements. What does this mean? Only that the filter will add
1650 multiple <literal>VAL_PROXY</literal> otherInfo packets to the
1651 Search requests that pass through it. That's because the virtual
1652 DB filter is dumb, and does exactly what it's told - no more, no
1654 If a Search request with multiple <literal>VAL_PROXY</literal>
1655 otherInfo packets reaches a <literal>z3950_client</literal>
1656 filter, this is an error. That filter doesn't know how to deal
1657 with multiple targets, so it will either just pick one and search
1658 in it, or (better) fail with an error message.
1661 The <literal>multi</literal> filter comes to the rescue! This is
1662 the only filter that knows how to deal with multiple
1663 <literal>VAL_PROXY</literal> otherInfo packets, and it does so by
1664 making multiple copies of the entire Search request: one for each
1665 <literal>VAL_PROXY</literal>. Each of these new copies is then
1666 passed down through the remaining filters in the route. (The
1667 copies are handled in parallel though the
1668 spawning of new threads.) Since the copies each have only one
1669 <literal>VAL_PROXY</literal> otherInfo, they can be handled by the
1670 <literal>z3950_client</literal> filter, which happily deals with
1671 each one individually. When the results of the individual
1672 searches come back up to the <literal>multi</literal> filter, it
1673 merges them into a single Search response, which is what
1674 eventually makes it back to the client.
1679 <imagedata fileref="multi.pdf" format="PDF" scale="50"/>
1682 <imagedata fileref="multi.png" format="PNG"/>
1685 <!-- Fall back if none of the images can be used -->
1687 [Here there should be a diagram showing the progress of
1688 packages through the filters during a simple virtual-database
1689 search and a multi-database search, but is seems that your
1690 tool chain has not been able to include the diagram in this
1695 <para>A picture is worth a thousand words (but only five hundred on 64-bit architectures)</para>
1702 <chapter id="sru-server">
1703 <title>Combined SRU webservice and Z39.50 server configuration</title>
1705 Metaproxy can act as
1706 <ulink url="&url.sru;">SRU</ulink> and
1707 web service server, which translates web service requests to
1708 <ulink url="&url.z39.50;">ANSI/NISO Z39.50</ulink> packages and
1709 sends them off to common available targets.
1712 A typical setup for this operation needs a filter route including the
1716 <table id="sru-server-table-config" frame="top">
1717 <title>SRU/Z39.50 Server Filter Route Configuration</title>
1721 <entry>Filter</entry>
1722 <entry>Importance</entry>
1723 <entry>Purpose</entry>
1729 <entry><literal>frontend_net</literal></entry>
1730 <entry>required</entry>
1731 <entry>Accepting HTTP connections and passing them to following
1732 filters. Since this filter also accepts Z39.50 connections, the
1733 server works as SRU and Z39.50 server on the same port.</entry>
1736 <entry><literal>sru_z3950</literal></entry>
1737 <entry>required</entry>
1738 <entry>Accepting SRU GET/POST/SOAP explain and
1739 searchRetrieve requests for the the configured databases.
1740 Explain requests are directly served from the static XML configuration.
1741 SearchRetrieve requests are
1742 transformed to Z39.50 search and present packages.
1743 All other HTTP and Z39.50 packages are passed unaltered.</entry>
1746 <entry><literal>http_file</literal></entry>
1747 <entry>optional</entry>
1748 <entry>Serving HTTP requests from the filesystem. This is only
1749 needed if the server should serve XSLT stylesheets, static HTML
1750 files or Java Script for thin browser based clients.
1751 Z39.50 packages are passed unaltered.</entry>
1754 <entry><literal>cql_rpn</literal></entry>
1755 <entry>required</entry>
1756 <entry>Usually, Z39.50 servers do not talk CQL, hence the
1757 translation of the CQL query language to RPN is mandatory in
1758 most cases. Affects only Z39.50 search packages.</entry>
1761 <entry><literal>record_transform</literal></entry>
1762 <entry>optional</entry>
1763 <entry>Some Z39.50 backend targets can not present XML record
1764 syntaxes in common wanted element sets. using this filter, one
1765 can transform binary MARC records to MARCXML records, and
1766 further transform those to any needed XML schema/format by XSLT
1767 transformations. Changes only Z39.50 present packages.</entry>
1770 <entry><literal>session_shared</literal></entry>
1771 <entry>optional</entry>
1772 <entry>The stateless nature of web services requires frequent
1773 re-searching of the same targets for display of paged result set
1774 records. This might be an unacceptable burden for the accessed
1775 backend Z39.50 targets, and this mosule can be added for
1776 efficient backend target resource pooling.</entry>
1779 <entry><literal>z3950_client</literal></entry>
1780 <entry>required</entry>
1781 <entry>Finally, a Z39.50 package sink is needed in the filter
1782 chain to provide the response packages. The Z39.50 client module
1783 is used to access external targets over the network, but any
1784 coming local Z39.50 package sink could be used instead of.</entry>
1787 <entry><literal>bounce</literal></entry>
1788 <entry>required</entry>
1789 <entry>Any Metaproxy package arriving here did not do so by
1790 purpose, and is bounced back with connection closure. this
1791 prevents inifinite package hanging inside the SRU server.</entry>
1797 A typical minimal example <ulink url="&url.sru;">SRU</ulink>
1798 server configuration file is found in the tarball distribution at
1799 <literal>etc/config-sru-to-z3950.xml</literal>.
1802 Off course, any other metaproxy modules can be integrated into a
1803 SRU server solution, including, but not limited to, load balancing,
1804 multiple target querying
1805 (see <xref linkend="multidb"/>), and complex RPN query rewrites.
1812 <chapter id="extensions">
1813 <title>Writing extensions for Metaproxy</title>
1814 <para>### To be written</para>
1820 <chapter id="classes">
1821 <title>Classes in the Metaproxy source code</title>
1824 <section id="classes-introductory-notes">
1825 <title>Introductory notes</title>
1827 <emphasis>Stop! Do not read this!</emphasis>
1828 You won't enjoy it at all. You should just skip ahead to
1829 <xref linkend="reference"/>,
1831 <!-- The remainder of this paragraph is lifted verbatim from
1832 Douglas Adams' _Hitch Hiker's Guide to the Galaxy_, chapter 8 -->
1833 you things you really need to know, like the fact that the
1834 fabulously beautiful planet Bethselamin is now so worried about
1835 the cumulative erosion by ten billion visiting tourists a year
1836 that any net imbalance between the amount you eat and the amount
1837 you excrete whilst on the planet is surgically removed from your
1838 bodyweight when you leave: so every time you go to the lavatory it
1839 is vitally important to get a receipt.
1842 This chapter contains documentation of the Metaproxy source code, and is
1843 of interest only to maintainers and developers. If you need to
1844 change Metaproxy's behavior or write a new filter, then you will most
1845 likely find this chapter helpful. Otherwise it's a waste of your
1846 good time. Seriously: go and watch a film or something.
1847 <citetitle>This is Spinal Tap</citetitle> is particularly good.
1850 Still here? OK, let's continue.
1853 In general, classes seem to be named big-endianly, so that
1854 <literal>FactoryFilter</literal> is not a filter that filters
1855 factories, but a factory that produces filters; and
1856 <literal>FactoryStatic</literal> is a factory for the statically
1857 registered filters (as opposed to those that are dynamically
1862 <section id="individual.classes">
1863 <title>Individual classes</title>
1865 The classes making up the Metaproxy application are here listed by
1866 class-name, with the names of the source files that define them in
1870 <section id="class-FactoryFilter">
1871 <title><literal>mp::FactoryFilter</literal>
1872 (<filename>factory_filter.cpp</filename>)</title>
1874 A factory class that exists primarily to provide the
1875 <literal>create()</literal> method, which takes the name of a
1876 filter class as its argument and returns a new filter of that
1877 type. To enable this, the factory must first be populated by
1878 calling <literal>add_creator()</literal> for static filters (this
1879 is done by the <literal>FactoryStatic</literal> class, see below)
1880 and <literal>add_creator_dyn()</literal> for filters loaded
1885 <section id="class-FactoryStatic">
1886 <title><literal>mp::FactoryStatic</literal>
1887 (<filename>factory_static.cpp</filename>)</title>
1889 A subclass of <literal>FactoryFilter</literal> which is
1890 responsible for registering all the statically defined filter
1891 types. It does this by knowing about all those filters'
1892 structures, which are listed in its constructor. Merely
1893 instantiating this class registers all the static classes. It is
1894 for the benefit of this class that <literal>struct
1895 metaproxy_1_filter_struct</literal> exists, and that all the filter
1896 classes provide a static object of that type.
1900 <section id="class-filter-Base">
1901 <title><literal>mp::filter::Base</literal>
1902 (<filename>filter.cpp</filename>)</title>
1904 The virtual base class of all filters. The filter API is, on the
1905 surface at least, extremely simple: two methods.
1906 <literal>configure()</literal> is passed an XML DOM tree representing
1907 that part of the configuration file that pertains to this filter
1908 instance, and is expected to walk that tree extracting relevant
1909 information. And <literal>process()</literal> processes a
1910 package (see below). That surface simplicity is a bit
1911 misleading, as <literal>process()</literal> needs to know a lot
1912 about the <literal>Package</literal> class in order to do
1917 <section id="class-AuthSimple">
1918 <title><literal>mp::filter::AuthSimple</literal>,
1919 <literal>Backend_test</literal>, etc.
1920 (<filename>filter_auth_simple.cpp</filename>,
1921 <filename>filter_backend_test.cpp</filename>, etc.)</title>
1923 Individual filters. Each of these is implemented by a header and
1924 a source file, named <filename>filter_*.hpp</filename> and
1925 <filename>filter_*.cpp</filename> respectively. All the header
1926 files should be pretty much identical, in that they declare the
1927 class, including a private <literal>Rep</literal> class and a
1928 member pointer to it, and the two public methods.
1931 The source file for each filter needs to supply:
1936 A definition of the private <literal>Rep</literal> class.
1941 Some boilerplate constructors and destructors.
1946 A <literal>configure()</literal> method that uses the
1947 appropriate XML fragment.
1952 Most important, the <literal>process()</literal> method that
1953 does all the actual work.
1959 <section id="class-Package">
1960 <title><literal>mp::Package</literal>
1961 (<filename>package.cpp</filename>)</title>
1963 Represents a package on its way through the series of filters
1964 that make up a route. This is essentially a Z39.50 or SRU APDU
1965 together with information about where it came from, which is
1966 modified as it passes through the various filters.
1970 <section id="class-Pipe">
1971 <title><literal>mp::Pipe</literal>
1972 (<filename>pipe.cpp</filename>)</title>
1974 This class provides a compatibility layer so that we have an IPC
1975 mechanism that works the same under Unix and Windows. It's not
1976 particularly exciting.
1980 <section id="class-RouterChain">
1981 <title><literal>mp::RouterChain</literal>
1982 (<filename>router_chain.cpp</filename>)</title>
1988 <section id="class-RouterFleXML">
1989 <title><literal>mp::RouterFleXML</literal>
1990 (<filename>router_flexml.cpp</filename>)</title>
1996 <section id="class-Session">
1997 <title><literal>mp::Session</literal>
1998 (<filename>session.cpp</filename>)</title>
2004 <section id="class-ThreadPoolSocketObserver">
2005 <title><literal>mp::ThreadPoolSocketObserver</literal>
2006 (<filename>thread_pool_observer.cpp</filename>)</title>
2012 <section id="class-util">
2013 <title><literal>mp::util</literal>
2014 (<filename>util.cpp</filename>)</title>
2016 A namespace of various small utility functions and classes,
2017 collected together for convenience. Most importantly, includes
2018 the <literal>mp::util::odr</literal> class, a wrapper for YAZ's
2023 <section id="class-xml">
2024 <title><literal>mp::xml</literal>
2025 (<filename>xmlutil.cpp</filename>)</title>
2027 A namespace of various XML utility functions and classes,
2028 collected together for convenience.
2034 <section id="other.source.files">
2035 <title>Other Source Files</title>
2037 In addition to the Metaproxy source files that define the classes
2038 described above, there are a few additional files which are
2039 briefly described here:
2043 <term><literal>metaproxy_prog.cpp</literal></term>
2046 The main function of the <command>metaproxy</command> program.
2051 <term><literal>ex_router_flexml.cpp</literal></term>
2054 Identical to <literal>metaproxy_prog.cpp</literal>: it's not clear why.
2059 <term><literal>test_*.cpp</literal></term>
2062 Unit-tests for various modules.
2068 ### Still to be described:
2069 <literal>ex_filter_frontend_net.cpp</literal>,
2070 <literal>filter_dl.cpp</literal>,
2071 <literal>plainfile.cpp</literal>,
2072 <literal>tstdl.cpp</literal>.
2078 <reference id="reference">
2079 <title>Reference</title>
2080 <partintro id="reference-introduction">
2082 The material in this chapter is drawn directly from the individual
2083 manual entries. In particular, the Metaproxy invocation section is
2084 available using <command>man metaproxy</command>, and the section
2085 on each individual filter is available using the name of the filter
2086 as the argument to the <command>man</command> command.
2092 <appendix id="license">
2093 <title>License</title>
2098 Metaproxy is free software; you can redistribute it and/or modify it under
2099 the terms of the GNU General Public License as published by the Free
2100 Software Foundation; either version 2, or (at your option) any later
2105 Metaproxy is distributed in the hope that it will be useful, but WITHOUT ANY
2106 WARRANTY; without even the implied warranty of MERCHANTABILITY or
2107 FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
2112 You should have received a copy of the GNU General Public License
2113 along with Metaproxy; see the file LICENSE. If not, write to the
2114 Free Software Foundation,
2115 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
2123 <!-- Keep this comment at the end of the file
2126 nxml-child-indent: 1