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>2005-2012</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 1.46.
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> version 4.0 and
169 <ulink url="&url.vstudio;">Microsoft Visual Studio</ulink> 2003/2005/2008.
172 <section id="installation.unix">
173 <title>Installation on Unix (from Source)</title>
175 Here is a quick step-by-step guide on how to compile all the
176 tools that Metaproxy uses. Only few systems have none of the required
177 tools binary packages. If, for example, Libxml2/libxslt are already
178 installed as development packages use those (and omit compilation).
181 <section id="libxml2.fromsource">
182 <title>Libxml2/libxslt</title>
187 gunzip -c libxml2-version.tar.gz|tar xf -
195 gunzip -c libxslt-version.tar.gz|tar xf -
203 <section id="yaz.fromsource">
204 <title>YAZ/YAZ++</title>
206 gunzip -c yaz-version.tar.gz|tar xf -
214 gunzip -c yazpp-version.tar.gz|tar xf -
223 <title id="boost.fromsource">Boost</title>
225 Metaproxy needs components thread and test from
229 gunzip -c boost-version.tar.gz|tar xf -
231 ./configure --with-libraries=thread,test,regex --with-toolset=gcc
237 However, under the hood bjam is used. You can invoke that with
240 ./bjam --toolset=gcc --with-thread --with-test --with-regex stage
243 Replace <literal>stage</literal> with <literal>clean</literal> /
244 <literal>install</literal> to perform clean and install respectively.
247 Add <literal>--prefix=DIR</literal> to install Boost in other
248 prefix than <literal>/usr/local</literal>.
251 <section id="metaproxy.fromsource">
252 <title>Metaproxy</title>
254 gunzip -c metaproxy-version.tar.gz|tar xf -
262 You may have to tell configure where Boost is installed by supplying
263 options <literal>--with-boost</literal> and <literal>--with-boost-toolset</literal>.
264 The former sets the PREFIX for Boost (same as --prefix for Boost above).
265 The latter the compiler toolset (eg. gcc34).
268 Pass <literal>--help</literal> to configure to get a list of
274 <section id="installation.debian">
275 <title>Installation on Debian GNU/Linux</title>
277 All dependencies for Metaproxy are available as
278 <ulink url="&url.debian;">Debian</ulink>
279 packages for the sarge (stable in 2005) and etch (testing in 2005)
283 The procedures for Debian based systems, such as
284 <ulink url="&url.ubuntu;">Ubuntu</ulink> is probably similar
287 There is currently no official Debian package for YAZ++.
288 And the official Debian package for YAZ is probably too old.
289 But Index Data builds "new" versions of those for Debian (i386, amd64 only).
292 Update the <filename>/etc/apt/sources.list</filename>
293 to include the Index Data repository.
294 See YAZ' <ulink url="&url.yaz.download.debian;">Download Debian</ulink>
295 for more information.
298 apt-get install libxslt1-dev
299 apt-get install libyazpp2-dev
300 apt-get install libboost-dev
301 apt-get install libboost-thread-dev
302 apt-get install libboost-test-dev
303 apt-get install libboost-regex-dev
306 With these packages installed, the usual configure + make
307 procedure can be used for Metaproxy as outlined in
308 <xref linkend="installation.unix"/>.
312 <section id="installation.rpm">
313 <title>Installation on RPM based Linux Systems</title>
315 All external dependencies for Metaproxy are available as
316 RPM packages, either from your distribution site, or from the
317 <ulink url="http://fr.rpmfind.net/">RPMfind</ulink> site.
320 For example, an installation of the requires Boost C++ development
321 libraries on RedHat Fedora C4 and C5 can be done like this:
323 wget ftp://fr.rpmfind.net/wlinux/fedora/core/updates/testing/4/SRPMS/boost-1.33.0-3.fc4.src.rpm
324 sudo rpmbuild --buildroot src/ --rebuild -p fc4/boost-1.33.0-3.fc4.src.rpm
325 sudo rpm -U /usr/src/redhat/RPMS/i386/boost-*rpm
329 The <ulink url="&url.yaz;">YAZ</ulink> library is needed to
330 compile &metaproxy;, see there
331 for more information on available RPM packages.
334 There is currently no official RPM package for YAZ++.
335 See the <ulink url="&url.yazplusplus;">YAZ++</ulink> pages
336 for more information on a Unix tarball install.
339 With these packages installed, the usual configure + make
340 procedure can be used for Metaproxy as outlined in
341 <xref linkend="installation.unix"/>.
345 <section id="installation.windows">
346 <title>Installation on Windows</title>
348 Metaproxy can be compiled with Microsoft
349 <ulink url="&url.vstudio;">Visual Studio</ulink>.
350 Versions 2003 (C 7.1), 2005 (C 8.0) and 2008 (C 9.0) are known to work.
352 <section id="installation.windows.boost">
355 For Windows, it's easiest to get the precompiled Boost
356 package from <ulink url="&url.boost.windows.download;">here</ulink>.
357 Several versions of the Boost libraries may be selected when
358 installing Boost for windows. Please choose at least the
359 <emphasis>multithreaded</emphasis> (non-DLL) version because
360 the Metaproxy makefile uses that.
363 For more information about installing Boost refer to the
364 <ulink url="&url.boost.getting.started;">getting started</ulink>
369 <section id="installation.windows.libxslt">
370 <title>Libxslt</title>
372 <ulink url="&url.libxslt;">Libxslt</ulink> can be downloaded
374 <ulink url="&url.libxml2.download.win32;">here</ulink>.
377 Libxslt has other dependencies, but these can all be downloaded
378 from the same site. Get the following package:
379 iconv, zlib, libxml2, libxslt.
383 <section id="installation.windows.yaz">
386 <ulink url="&url.yaz;">YAZ</ulink> can be downloaded
388 <ulink url="&url.yaz.download.win32;">here</ulink>.
392 <section id="installation.windows.yazplusplus">
395 Get <ulink url="&url.yazplusplus;">YAZ++</ulink> as well.
396 Version 1.2.7 or later is required.
399 YAZ++ includes NMAKE makefiles, similar to those found in the
404 <section id="installation.windows.metaproxy">
405 <title>Metaproxy</title>
407 Metaproxy is shipped with NMAKE makefiles as well - similar
408 to those found in the YAZ++/YAZ packages. Adjust this Makefile
409 to point to the proper locations of Boost, Libxslt, Libxml2,
410 zlib, iconv, yaz and yazpp.
414 <varlistentry><term><literal>DEBUG</literal></term>
416 If set to 1, the software is
417 compiled with debugging libraries (code generation is
418 multi-threaded debug DLL).
419 If set to 0, the software is compiled with release libraries
420 (code generation is multi-threaded DLL).
425 <term><literal>BOOST</literal></term>
428 Boost install location
434 <term><literal>BOOST_VERSION</literal></term>
437 Boost version (replace . with _).
443 <term><literal>BOOST_TOOLSET</literal></term>
452 <term><literal>LIBXSLT_DIR</literal>,
453 <literal>LIBXML2_DIR</literal> ..</term>
456 Specify the locations of Libxslt, libiconv, libxml2 and
465 After successful compilation you'll find
466 <literal>metaproxy.exe</literal> in the
467 <literal>bin</literal> directory.
475 <chapter id="yazproxy-comparison">
476 <title>YAZ Proxy Comparison</title>
478 The table below lists facilities either supported by either
479 <ulink url="&url.yazproxy;">YAZ Proxy</ulink> or Metaproxy.
481 <table id="yazproxy-comparison-table">
482 <title>Metaproxy / YAZ Proxy comparison</title>
486 <entry>Facility</entry>
487 <entry>Metaproxy</entry>
488 <entry>YAZ Proxy</entry>
493 <entry>Z39.50 server</entry>
494 <entry>Using filter <literal>frontend_net</literal></entry>
495 <entry>Supported</entry>
498 <entry>SRU server</entry>
499 <entry>Supported with filter <literal>sru_z3950</literal></entry>
500 <entry>Supported</entry>
503 <entry>Z39.50 client</entry>
504 <entry>Supported with filter <literal>z3950_client</literal></entry>
505 <entry>Supported</entry>
508 <entry>SRU client</entry>
509 <entry>Unsupported</entry>
510 <entry>Unsupported</entry>
513 <entry>Connection reuse</entry>
514 <entry>Supported with filter <literal>session_shared</literal></entry>
515 <entry>Supported</entry>
518 <entry>Connection share</entry>
519 <entry>Supported with filter <literal>session_shared</literal></entry>
520 <entry>Unsupported</entry>
523 <entry>Result set reuse</entry>
524 <entry>Supported with filter <literal>session_shared</literal></entry>
525 <entry>Within one Z39.50 session / HTTP keep-alive</entry>
528 <entry>Record cache</entry>
529 <entry>Supported by filter <literal>session_shared</literal></entry>
530 <entry>Supported for last result set within one Z39.50/HTTP-keep alive session</entry>
533 <entry>Z39.50 Virtual database, i.e. select any Z39.50 target for database</entry>
534 <entry>Supported with filter <literal>virt_db</literal></entry>
535 <entry>Unsupported</entry>
538 <entry>SRU Virtual database, i.e. select any Z39.50 target for path</entry>
539 <entry>Supported with filter <literal>virt_db</literal>,
540 <literal>sru_z3950</literal></entry>
541 <entry>Supported</entry>
544 <entry>Multi target search</entry>
545 <entry>Supported with filter <literal>multi</literal> (round-robin)</entry>
546 <entry>Unsupported</entry>
549 <entry>Retrieval and search limits</entry>
550 <entry>Supported using filter <literal>limit</literal></entry>
551 <entry>Supported</entry>
554 <entry>Bandwidth limits</entry>
555 <entry>Supported using filter <literal>limit</literal></entry>
556 <entry>Supported</entry>
559 <entry>Connect limits</entry>
560 <entry>Supported by filter <literal>frontend_net</literal> (connect-max)</entry>
561 <entry>Supported</entry>
564 <entry>Retrieval sanity check and conversions</entry>
565 <entry>Supported using filter <literal>record_transform</literal></entry>
566 <entry>Supported</entry>
569 <entry>Query check</entry>
571 Supported by <literal>query_rewrite</literal> which may be check
572 a query and throw diagnostics (errors)
574 <entry>Supported</entry>
577 <entry>Query rewrite</entry>
578 <entry>Supported with <literal>query_rewrite</literal></entry>
579 <entry>Unsupported</entry>
582 <entry>Session invalidate for -1 hits</entry>
583 <entry>Unsupported</entry>
584 <entry>Supported</entry>
587 <entry>Architecture</entry>
588 <entry>Multi-threaded + select for networked modules such as
589 <literal>frontend_net</literal>)</entry>
590 <entry>Single-threaded using select</entry>
594 <entry>Extensability</entry>
595 <entry>Most functionality implemented as loadable modules</entry>
596 <entry>Unsupported and experimental</entry>
600 <entry><ulink url="&url.usemarcon;">USEMARCON</ulink></entry>
601 <entry>Unsupported</entry>
602 <entry>Supported</entry>
606 <entry>Portability</entry>
608 Requires YAZ, YAZ++ and modern C++ compiler supporting
609 <ulink url="&url.boost;">Boost</ulink>.
612 Requires YAZ and YAZ++.
613 STL is not required so pretty much any C++ compiler out there should work.
622 <chapter id="architecture">
623 <title>The Metaproxy Architecture</title>
625 The Metaproxy architecture is based on three concepts:
626 the <emphasis>package</emphasis>,
627 the <emphasis>route</emphasis>
628 and the <emphasis>filter</emphasis>.
632 <term>Packages</term>
635 A package is request or response, encoded in some protocol,
636 issued by a client, making its way through Metaproxy, send to or
637 received from a server, or sent back to the client.
640 The core of a package is the protocol unit - for example, a
641 Z39.50 Init Request or Search Response, or an SRU searchRetrieve
642 URL or Explain Response. In addition to this core, a package
643 also carries some extra information added and used by Metaproxy
647 In general, packages are doctored as they pass through
648 Metaproxy. For example, when the proxy performs authentication
649 and authorization on a Z39.50 Init request, it removes the
650 authentication credentials from the package so that they are not
651 passed onto the back-end server; and when search-response
652 packages are obtained from multiple servers, they are merged
653 into a single unified package that makes its way back to the
662 Packages make their way through routes, which can be thought of
663 as programs that operate on the package data-type. Each
664 incoming package initially makes its way through a default
665 route, but may be switched to a different route based on various
666 considerations. Routes are made up of sequences of filters (see
675 Filters provide the individual instructions within a route, and
676 effect the necessary transformations on packages. A particular
677 configuration of Metaproxy is essentially a set of filters,
678 described by configuration details and arranged in order in one
679 or more routes. There are many kinds of filter - about a dozen
680 at the time of writing with more appearing all the time - each
681 performing a specific function and configured by different
685 The word ``filter'' is sometimes used rather loosely, in two
686 different ways: it may be used to mean a particular
687 <emphasis>type</emphasis> of filter, as when we speak of ``the
688 auth_simple filter'' or ``the multi filter''; or it may be used
689 to be a specific <emphasis>instance</emphasis> of a filter
690 within a Metaproxy configuration. For example, a single
691 configuration will often contain multiple instances of the
692 <literal>z3950_client</literal> filter. In
693 operational terms, of these is a separate filter. In practice,
694 context always make it clear which sense of the word ``filter''
698 Extensibility of Metaproxy is primarily through the creation of
699 plugins that provide new filters. The filter API is small and
700 conceptually simple, but there are many details to master. See
702 <link linkend="filters">Filters</link>.
708 Since packages are created and handled by the system itself, and
709 routes are conceptually simple, most of the remainder of this
710 document concentrates on filters. After a brief overview of the
711 filter types follows, along with some thoughts on possible future
718 <chapter id="filters">
719 <title>Filters</title>
722 <section id="filters-introductory-notes">
723 <title>Introductory notes</title>
725 It's useful to think of Metaproxy as an interpreter providing a small
726 number of primitives and operations, but operating on a very
727 complex data type, namely the ``package''.
730 A package represents a Z39.50 or SRU/W request (whether for Init,
731 Search, Scan, etc.) together with information about where it came
732 from. Packages are created by front-end filters such as
733 <literal>frontend_net</literal> (see below), which reads them from
734 the network; other front-end filters are possible. They then pass
735 along a route consisting of a sequence of filters, each of which
736 transforms the package and may also have side-effects such as
737 generating logging. Eventually, the route will yield a response,
738 which is sent back to the origin.
741 There are many kinds of filter: some that are defined statically
742 as part of Metaproxy, and others may be provided by third parties
743 and dynamically loaded. They all conform to the same simple API
744 of essentially two methods: <function>configure()</function> is
745 called at startup time, and is passed an XML DOM tree representing that
746 part of the configuration file that pertains to this filter
747 instance: it is expected to walk that tree extracting relevant
748 information; and <function>process()</function> is called every
749 time the filter has to processes a package.
752 While all filters provide the same API, there are different modes
753 of functionality. Some filters are sources: they create
755 (<literal>frontend_net</literal>);
756 others are sinks: they consume packages and return a result
757 (<literal>backend_test</literal>,
758 <literal>bounce</literal>,
759 <literal>http_file</literal>,
760 <literal>z3950_client</literal>);
761 the others are true filters, that read, process and pass on the
762 packages they are fed
763 (<literal>auth_simple</literal>,
764 <literal>log</literal>,
765 <literal>multi</literal>,
766 <literal>query_rewrite</literal>,
767 <literal>record_transform</literal>,
768 <literal>session_shared</literal>,
769 <literal>sru_z3950</literal>,
770 <literal>template</literal>,
771 <literal>virt_db</literal>).
776 <section id="overview.filter.types">
777 <title>Overview of filter types</title>
779 We now briefly consider each of the types of filter supported by
780 the core Metaproxy binary. This overview is intended to give a
781 flavor of the available functionality; more detailed information
782 about each type of filter is included below in
783 <xref linkend="reference"/>.
786 The filters are here named by the string that is used as the
787 <literal>type</literal> attribute of a
788 <literal><filter></literal> element in the configuration
789 file to request them, with the name of the class that implements
790 them in parentheses. (The classname is not needed for normal
791 configuration and use of Metaproxy; it is useful only to
795 The filters are here listed in alphabetical order:
802 New virt_db-alike that does inteligent peer choice, explain merging,
803 adds FD&N to explain. Keeps init responses (like "virt_db Classic"),
804 makes routing choices based on local explain knowledge. Ref IDDI
807 Filter to convert Explain Classic to ZeeRex.
809 CQL2PQF (which needs augmented ZeeRex) - MARC for Talis.
813 Figure out what additional information we need in:
814 ZeeRex (check against D3.1)
815 Init request (e.g. loop detection)
816 Query package (e.g. number of hops)
817 Query response (e.g. record source)
821 <section id="auth_simple">
822 <title><literal>auth_simple</literal>
823 (mp::filter::AuthSimple)</title>
825 Simple authentication and authorization. The configuration
826 specifies the name of a file that is the user register, which
827 lists <varname>username</varname>:<varname>password</varname>
828 pairs, one per line, colon separated. When a session begins, it
829 is rejected unless username and passsword are supplied, and match
830 a pair in the register. The configuration file may also specific
831 the name of another file that is the target register: this lists
832 lists <varname>username</varname>:<varname>dbname</varname>,<varname>dbname</varname>...
833 sets, one per line, with multiple database names separated by
834 commas. When a search is processed, it is rejected unless the
835 database to be searched is one of those listed as available to
840 <section id="backend_test">
841 <title><literal>backend_test</literal>
842 (mp::filter::Backend_test)</title>
844 A partial sink that provides dummy responses in the manner of the
845 <literal>yaz-ztest</literal> Z39.50 server. This is useful only
846 for testing. Seriously, you don't need this. Pretend you didn't
847 even read this section.
851 <section id="bounce">
852 <title><literal>bounce</literal>
853 (mp::filter::Bounce)</title>
855 A sink that swallows <emphasis>all packages</emphasis>,
856 and returns them almost unprocessed.
857 It never sends any package of any type further down the row, but
858 sets Z39.50 packages to Z_Close, and HTTP_Request packages to
859 HTTP_Response err code 400 packages, and adds a suitable bounce
861 The bounce filter is usually added at end of each filter chain route
862 to prevent infinite hanging of for example HTTP
863 requests packages when only the Z39.50 client partial sink
864 filter is found in the
869 <section id="cql_rpn">
870 <title><literal>cql_rpn</literal>
871 (mp::filter::CQLtoRPN)</title>
873 A query language transforming filter which catches Z39.50
874 <literal>searchRequest</literal>
875 packages containing <literal>CQL</literal> queries, transforms
876 those to <literal>RPN</literal> queries,
877 and sends the <literal>searchRequests</literal> on to the next
878 filters. It is among other things useful in a SRU context.
882 <section id="frontend_net">
883 <title><literal>frontend_net</literal>
884 (mp::filter::FrontendNet)</title>
886 A source that accepts Z39.50 connections from a port
887 specified in the configuration, reads protocol units, and
888 feeds them into the next filter in the route. When the result is
889 received, it is returned to the original origin.
893 <section id="http_file">
894 <title><literal>http_file</literal>
895 (mp::filter::HttpFile)</title>
897 A partial sink which swallows only
898 <literal>HTTP_Request</literal> packages, and
899 returns the contents of files from the local
900 filesystem in response to HTTP requests.
901 It lets Z39.50 packages and all other forthcoming package types
904 does mean that Metaproxy is also a Web-server in its spare time. So
905 far it does not contain either an email-reader or a Lisp
906 interpreter, but that day is surely coming.)
910 <section id="load_balance">
911 <title><literal>load_balance</literal>
912 (mp::filter::LoadBalance)</title>
914 Performs load balancing for incoming Z39.50 init requests.
915 It is used together with the <literal>virt_db</literal> filter,
916 but unlike the <literal>multi</literal> filter it does send an
917 entire session to only one of the virtual backends. The
918 <literal>load_balance</literal> filter is assuming that
919 all backend targets have equal content, and chooses the backend
920 with least load cost for a new session.
923 This filter is experimental and yet not mature for heavy load
931 <title><literal>log</literal>
932 (mp::filter::Log)</title>
934 Writes logging information to standard output, and passes on
935 the package unchanged. A log file name can be specified, as well
936 as multiple different logging formats.
941 <title><literal>multi</literal>
942 (mp::filter::Multi)</title>
944 Performs multi-database searching.
946 <link linkend="multidb">the extended discussion</link>
947 of virtual databases and multi-database searching below.
951 <section id="query_rewrite">
952 <title><literal>query_rewrite</literal>
953 (mp::filter::QueryRewrite)</title>
955 Rewrites Z39.50 <literal>Type-1</literal>
956 and <literal>Type-101</literal> (``<literal>RPN</literal>'')
958 three-step process: the query is transliterated from Z39.50
959 packet structures into an XML representation; that XML
960 representation is transformed by an XSLT stylesheet; and the
961 resulting XML is transliterated back into the Z39.50 packet
967 <section id="record_transform">
968 <title><literal>record_transform</literal>
969 (mp::filter::RecordTransform)</title>
971 This filter acts only on Z3950 present requests, and let all
972 other types of packages and requests pass untouched. It's use is
973 twofold: blocking Z3950 present requests, which the backend
974 server does not understand and can not honor, and transforming
975 the present syntax and elementset name according to the rules
976 specified, to fetch only existing record formats, and transform
977 them on the fly to requested record syntaxes.
981 <section id="session_shared">
982 <title><literal>session_shared</literal>
983 (mp::filter::SessionShared)</title>
985 This filter implements global sharing of
986 result sets (i.e. between threads and therefore between
987 clients), yielding performance improvements by clever resource
992 <section id="sru_z3950">
993 <title><literal>sru_z3950</literal>
994 (mp::filter::SRUtoZ3950)</title>
996 This filter transforms valid
997 SRU GET/POST/SOAP searchRetrieve requests to Z3950 init, search,
998 and present requests, and wraps the
999 received hit counts and XML records into suitable SRU response
1001 The <literal>sru_z3950</literal> filter processes also SRU
1002 GET/POST/SOAP explain requests, returning
1003 either the absolute minimum required by the standard, or a full
1004 pre-defined ZeeReX explain record.
1006 <ulink url="&url.zeerex.explain;">ZeeReX Explain</ulink>
1007 standard pages and the
1008 <ulink url="&url.sru.explain;">SRU Explain</ulink> pages
1009 for more information on the correct explain syntax.
1010 SRU scan requests are not supported yet.
1014 <section id="template">
1015 <title><literal>template</literal>
1016 (mp::filter::Template)</title>
1018 Does nothing at all, merely passing the packet on. (Maybe it
1019 should be called <literal>nop</literal> or
1020 <literal>passthrough</literal>?) This exists not to be used, but
1021 to be copied - to become the skeleton of new filters as they are
1022 written. As with <literal>backend_test</literal>, this is not
1023 intended for civilians.
1027 <section id="virt_db">
1028 <title><literal>virt_db</literal>
1029 (mp::filter::VirtualDB)</title>
1031 Performs virtual database selection: based on the name of the
1032 database in the search request, a server is selected, and its
1033 address added to the request in a <literal>VAL_PROXY</literal>
1034 otherInfo packet. It will subsequently be used by a
1035 <literal>z3950_client</literal> filter.
1037 <link linkend="multidb">the extended discussion</link>
1038 of virtual databases and multi-database searching below.
1042 <section id="z3950_client">
1043 <title><literal>z3950_client</literal>
1044 (mp::filter::Z3950Client)</title>
1046 A partial sink which swallows only Z39.50 packages.
1047 It performs Z39.50 searching and retrieval by proxying the
1048 packages that are passed to it. Init requests are sent to the
1049 address specified in the <literal>VAL_PROXY</literal> otherInfo
1050 attached to the request: this may have been specified by client,
1051 or generated by a <literal>virt_db</literal> filter earlier in
1052 the route. Subsequent requests are sent to the same address,
1053 which is remembered at Init time in a Session object.
1054 HTTP_Request packages and all other forthcoming package types
1055 are passed untouched.
1060 <section id="zeerex_explain">
1061 <title><literal>zeerex_explain</literal>
1062 (mp::filter::ZeerexExplain)</title>
1064 This filter acts as a sink for
1065 Z39.50 explain requests, returning a static ZeeReX
1066 Explain XML record from the config section. All other packages
1069 <ulink url="&url.zeerex.explain;">ZeeReX Explain</ulink>
1071 for more information on the correct explain syntax.
1075 This filter is not yet completed.
1084 <section id="future.directions">
1085 <title>Future directions</title>
1087 Some other filters that do not yet exist, but which would be
1088 useful, are briefly described. These may be added in future
1089 releases (or may be created by third parties, as loadable
1095 <term><literal>frontend_cli</literal> (source)</term>
1098 Command-line interface for generating requests.
1103 <term><literal>sru_client</literal> (sink)</term>
1106 SRU/GET and SRU/SOAP searching and retrieval.
1111 <term><literal>opensearch_client</literal> (sink)</term>
1114 A9 OpenSearch searching and retrieval.
1124 <chapter id="configuration">
1125 <title>Configuration: the Metaproxy configuration file format</title>
1128 <section id="configuration-introductory-notes">
1129 <title>Introductory notes</title>
1131 If Metaproxy is an interpreter providing operations on packages, then
1132 its configuration file can be thought of as a program for that
1133 interpreter. Configuration is by means of a single XML file, the name
1134 of which is supplied as the sole command-line argument to the
1135 <command>metaproxy</command> program. (See
1136 <xref linkend="reference"/> below for more information on invoking
1141 <section id="overview.xml.structure">
1142 <title>Overview of the config file XML structure</title>
1144 All elements and attributes are in the namespace
1145 <ulink url="http://indexdata.com/metaproxy"/>.
1146 This is most easily achieved by setting the default namespace on
1147 the top-level element, as here:
1150 <metaproxy xmlns="http://indexdata.com/metaproxy" version="1.0">
1153 The top-level element is <metaproxy>. This contains
1154 a <dlpath> element,
1155 a <start> element,
1156 a <filters> element and
1157 a <routes> element, in that order. <dlpath> and
1158 <filters> are optional; the other two are mandatory.
1159 All four are non-repeatable.
1162 The <dlpath;> element contains a text element which
1163 specifies the location of filter modules. This is only needed
1164 if Metaproxy must load 3rd party filters (most filters with Metaproxy
1165 are built into the Metaproxy application).
1168 The <start> element is empty, but carries a
1169 <literal>route</literal> attribute, whose value is the name of
1170 route at which to start running - analogous to the name of the
1171 start production in a formal grammar.
1174 If present, <filters> contains zero or more <filter>
1175 elements. Each filter carries a <literal>type</literal> attribute
1176 which specifies what kind of filter is being defined
1177 (<literal>frontend_net</literal>, <literal>log</literal>, etc.)
1178 and contain various elements that provide suitable configuration
1179 for a filter of its type. The filter-specific elements are
1181 <xref linkend="reference"/>.
1182 Filters defined in this part of the file must carry an
1183 <literal>id</literal> attribute so that they can be referenced
1187 <routes> contains one or more <route> elements, each
1188 of which must carry an <literal>id</literal> element. One of the
1189 routes must have the ID value that was specified as the start
1190 route in the <start> element's <literal>route</literal>
1191 attribute. Each route contains zero or more <filter>
1192 elements. These are of two types. They may be empty, but carry a
1193 <literal>refid</literal> attribute whose value is the same as the
1194 <literal>id</literal> of a filter previously defined in the
1195 <filters> section. Alternatively, a route within a filter
1196 may omit the <literal>refid</literal> attribute, but contain
1197 configuration elements similar to those used for filters defined
1198 in the <filters> section. (In other words, each filter in a
1199 route may be included either by reference or by physical
1205 <section id="example.configuration">
1206 <title>An example configuration</title>
1208 The following is a small, but complete, Metaproxy configuration
1209 file (included in the distribution as
1210 <literal>metaproxy/etc/config1.xml</literal>).
1211 This file defines a very simple configuration that simply proxies
1212 to whatever back-end server the client requests, but logs each
1213 request and response. This can be useful for debugging complex
1214 client-server dialogues.
1216 <screen><![CDATA[<?xml version="1.0"?>
1217 <metaproxy xmlns="http://indexdata.com/metaproxy" version="1.0">
1218 <dlpath>/usr/lib/metaproxy/modules</dlpath>
1219 <start route="start"/>
1221 <filter id="frontend" type="frontend_net">
1224 <filter id="backend" type="z3950_client">
1229 <filter refid="frontend"/>
1230 <filter type="log"/>
1231 <filter refid="backend"/>
1232 <filter type="bounce"/>
1238 It works by defining a single route, called
1239 <literal>start</literal>, which consists of a sequence of four
1240 filters. The first and last of these are included by reference:
1241 their <literal><filter></literal> elements have
1242 <literal>refid</literal> attributes that refer to filters defined
1243 within the prior <literal><filters></literal> section. The
1244 middle filter is included inline in the route.
1247 The four filters in the route are as follows: first, a
1248 <literal>frontend_net</literal> filter accepts Z39.50 requests
1249 from any host on port 9000; then these requests are passed through
1250 a <literal>log</literal> filter that emits a message for each
1251 request; they are then fed into a <literal>z3950_client</literal>
1252 filter, which forwards all Z39.50 requests to the client-specified
1253 back-end Z39.509 server. Those Z39.50 packages are returned by the
1254 <literal>z3950_client</literal> filter, with the response data
1255 filled by the external Z39.50 server targeted.
1256 All non-Z39.50 packages are passed through to the
1257 <literal>bounce</literal> filter, which definitely bounces
1258 everything, including fish, bananas, cold pyjamas,
1259 mutton, beef and trout packages.
1260 When the response arrives, it is handed
1261 back to the <literal>log</literal> filter, which emits another
1262 message; and then to the <literal>frontend_net</literal> filter,
1263 which returns the response to the client.
1267 <section id="config-file-modularity">
1268 <title>Config file modularity</title>
1270 Metaproxy XML configuration snippets can be reused by other
1271 filters using the <literal>XInclude</literal> standard, as seen in
1272 the <literal>/etc/config-sru-to-z3950.xml</literal> example SRU
1275 <filter id="sru" type="sru_z3950">
1276 <database name="Default">
1277 <xi:include xmlns:xi="http://www.w3.org/2001/XInclude"
1278 href="explain.xml"/>
1285 <section id="config-file-syntax-check">
1286 <title>Config file syntax checking</title>
1288 The distribution contains RelaxNG Compact and XML syntax checking
1289 files, as well as XML Schema files. These are found in the
1292 xml/schema/metaproxy.rnc
1293 xml/schema/metaproxy.rng
1294 xml/schema/metaproxy.xsd
1296 and can be used to verify or debug the XML structure of
1297 configuration files. For example, using the utility
1298 <filename>xmllint</filename>, syntax checking is done like this:
1300 xmllint --noout --schema xml/schema/metaproxy.xsd etc/config-local.xml
1301 xmllint --noout --relaxng xml/schema/metaproxy.rng etc/config-local.xml
1303 (A recent version of <literal>libxml2</literal> is required, as
1304 support for XML Schemas is a relatively recent addition.)
1307 You can of course use any other RelaxNG or XML Schema compliant tool
1315 <chapter id="multidb">
1316 <title>Virtual databases and multi-database searching</title>
1319 <section id="multidb-introductory-notes">
1320 <title>Introductory notes</title>
1322 Two of Metaproxy's filters are concerned with multiple-database
1323 operations. Of these, <literal>virt_db</literal> can work alone
1324 to control the routing of searches to one of a number of servers,
1325 while <literal>multi</literal> can work together with
1326 <literal>virt_db</literal> to perform multi-database searching, merging
1327 the results into a unified result-set - ``metasearch in a box''.
1330 The interaction between
1331 these two filters is necessarily complex: it reflects the real,
1332 irreducible complexity of multi-database searching in a protocol such
1333 as Z39.50 that separates initialization from searching, and in
1334 which the database to be searched is not known at initialization
1338 It's possible to use these filters without understanding the
1339 details of their functioning and the interaction between them; the
1340 next two sections of this chapter are ``HOW-TO'' guides for doing
1341 just that. However, debugging complex configurations will require
1342 a deeper understanding, which the last two sections of this
1343 chapters attempt to provide.
1348 <section id="multidb.virt_db">
1349 <title>Virtual databases with the <literal>virt_db</literal> filter</title>
1351 Working alone, the purpose of the
1352 <literal>virt_db</literal>
1353 filter is to route search requests to one of a selection of
1354 back-end databases. In this way, a single Z39.50 endpoint
1355 (running Metaproxy) can provide access to several different
1356 underlying services, including those that would otherwise be
1357 inaccessible due to firewalls. In many useful configurations, the
1358 back-end databases are local to the Metaproxy installation, but
1359 the software does not enforce this, and any valid Z39.50 servers
1360 may be used as back-ends.
1363 For example, a <literal>virt_db</literal>
1364 filter could be set up so that searches in the virtual database
1365 ``lc'' are forwarded to the Library of Congress bibliographic
1366 catalogue server, and searches in the virtual database ``marc''
1367 are forwarded to the toy database of MARC records that Index Data
1368 hosts for testing purposes. A <literal>virt_db</literal>
1369 configuration to make this switch would look like this:
1371 <screen><![CDATA[<filter type="virt_db">
1373 <database>lc</database>
1374 <target>z3950.loc.gov:7090/voyager</target>
1377 <database>marc</database>
1378 <target>indexdata.com/marc</target>
1380 </filter>]]></screen>
1382 As well as being useful in it own right, this filter also provides
1383 the foundation for multi-database searching.
1388 <section id="multidb.multi">
1389 <title>Multi-database search with the <literal>multi</literal> filter</title>
1391 To arrange for Metaproxy to broadcast searches to multiple back-end
1392 servers, the configuration needs to include two components: a
1393 <literal>virt_db</literal>
1394 filter that specifies multiple
1395 <literal><target></literal>
1396 elements, and a subsequent
1397 <literal>multi</literal>
1398 filter. Here, for example, is a complete configuration that
1399 broadcasts searches to both the Library of Congress catalogue and
1400 Index Data's tiny testing database of MARC records:
1402 <screen><![CDATA[<?xml version="1.0"?>
1403 <metaproxy xmlns="http://indexdata.com/metaproxy" version="1.0">
1404 <start route="start"/>
1407 <filter type="frontend_net">
1408 <threads>10</threads>
1411 <filter type="virt_db">
1413 <database>lc</database>
1414 <target>z3950.loc.gov:7090/voyager</target>
1417 <database>marc</database>
1418 <target>indexdata.com/marc</target>
1421 <database>all</database>
1422 <target>z3950.loc.gov:7090/voyager</target>
1423 <target>indexdata.com/marc</target>
1426 <filter type="multi"/>
1427 <filter type="z3950_client">
1428 <timeout>30</timeout>
1430 <filter type="bounce"/>
1433 </metaproxy>]]></screen>
1436 <literal>virt_db</literal>
1437 filter that specifies multiple
1438 <literal><target></literal>
1439 elements but without a subsequent
1440 <literal>multi</literal>
1441 filter yields surprising and undesirable results, as will be
1442 described below. Don't do that.)
1445 Metaproxy can be invoked with this configuration as follows:
1447 <screen>../src/metaproxy --config config-simple-multi.xml</screen>
1449 And thereafter, Z39.50 clients can connect to the running server
1450 (on port 9000, as specified in the configuration) and search in
1451 any of the databases
1452 <literal>lc</literal> (the Library of Congress catalogue),
1453 <literal>marc</literal> (Index Data's test database of MARC records)
1455 <literal>all</literal> (both of these). As an example, a session
1456 using the YAZ command-line client <literal>yaz-client</literal> is
1457 here included (edited for brevity and clarity):
1459 <screen><![CDATA[$ yaz-client @:9000
1463 Search was a success.
1464 Number of hits: 10000, setno 1
1468 Search was a success.
1469 Number of hits: 10, setno 3
1473 Search was a success.
1474 Number of hits: 10010, setno 4
1477 [marc]Record type: USmarc
1480 005 00000000000000.0
1481 008 910710c19910701nju 00010 eng
1485 100 10 $a Jack Collins
1486 245 10 $a How to program a computer
1492 [VOYAGER]Record type: USmarc
1494 005 20041229102447.0
1495 008 030910s2004 caua 000 0 eng
1496 035 $a (DLC) 2003112666
1497 906 $a 7 $b cbc $c orignew $d 4 $e epcn $f 20 $g y-gencatlg
1498 925 0 $a acquire $b 1 shelf copy $x policy default
1499 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.
1502 040 $a DLC $c DLC $d DLC
1503 050 00 $a MLCM 2004/03312 (G)
1504 245 10 $a 007, everything or nothing : $b Prima's official strategy guide / $c created by Kaizen Media Group.
1505 246 3 $a Double-O-seven, everything or nothing
1506 246 30 $a Prima's official strategy guide
1507 260 $a Roseville, CA : $b Prima Games, $c c2004.
1508 300 $a 161 p. : $b col. ill. ; $c 28 cm.
1509 500 $a "Platforms: Nintendo GameCube, Macintosh, PC, PlayStation 2 computer entertainment system, Xbox"--P. [4] of cover.
1510 650 0 $a Video games.
1511 710 2 $a Kaizen Media Group.
1512 856 42 $3 Publisher description $u http://www.loc.gov/catdir/description/random052/2003112666.html
1517 As can be seen, the first record in the result set is from the
1518 Index Data test database, and the second from the Library of
1519 Congress database. The result-set continues alternating records
1520 round-robin style until the point where one of the databases'
1521 records are exhausted.
1524 This example uses only two back-end databases; more may be used.
1525 There is no limitation imposed on the number of databases that may
1526 be metasearched in this way: issues of resource usage and
1527 administrative complexity dictate the practical limits.
1530 What happens when one of the databases doesn't respond? By default,
1531 the entire multi-database search fails, and the appropriate
1532 diagnostic is returned to the client. This is usually appropriate
1533 during development, when technicians need maximum information, but
1534 can be inconvenient in deployment, when users typically don't want
1535 to be bothered with problems of this kind and prefer just to get
1536 the records from the databases that are available. To obtain this
1537 latter behavior add an empty
1538 <literal><hideunavailable></literal>
1540 <literal>multi</literal> filter:
1542 <screen><![CDATA[ <filter type="multi">
1544 </filter>]]></screen>
1546 Under this regime, an error is reported to the client only if
1547 <emphasis>all</emphasis> the databases in a multi-database search
1553 <section id="multidb.what">
1554 <title>What's going on?</title>
1556 <title>Lark's vomit</title>
1558 This section goes into a level of technical detail that is
1559 probably not necessary in order to configure and use Metaproxy.
1560 It is provided only for those who like to know how things work.
1561 You should feel free to skip on to the next section if this one
1562 doesn't seem like fun.
1566 Hold on tight - this may get a little hairy.
1569 In the general course of things, a Z39.50 Init request may carry
1570 with it an otherInfo packet of type <literal>VAL_PROXY</literal>,
1571 whose value indicates the address of a Z39.50 server to which the
1572 ultimate connection is to be made. (This otherInfo packet is
1573 supported by YAZ-based Z39.50 clients and servers, but has not yet
1574 been ratified by the Maintenance Agency and so is not widely used
1575 in non-Index Data software. We're working on it.)
1576 The <literal>VAL_PROXY</literal> packet functions
1577 analogously to the absoluteURI-style Request-URI used with the GET
1578 method when a web browser asks a proxy to forward its request: see
1580 <ulink url="http://www.w3.org/Protocols/rfc2616/rfc2616-sec5.html#sec5.1.2"
1581 >Request-URI</ulink>
1583 <ulink url="http://www.w3.org/Protocols/rfc2616/rfc2616.html"
1584 >the HTTP 1.1 specification</ulink>.
1587 Within Metaproxy, Search requests that are part of the same
1588 session as an Init request that carries a
1589 <literal>VAL_PROXY</literal> otherInfo are also annotated with the
1590 same information. The role of the <literal>virt_db</literal>
1591 filter is to rewrite this otherInfo packet dependent on the
1592 virtual database that the client wants to search.
1595 When Metaproxy receives a Z39.50 Init request from a client, it
1596 doesn't immediately forward that request to the back-end server.
1597 Why not? Because it doesn't know <emphasis>which</emphasis>
1598 back-end server to forward it to until the client sends a Search
1599 request that specifies the database that it wants to search in.
1600 Instead, it just treasures the Init request up in its heart; and,
1601 later, the first time the client does a search on one of the
1602 specified virtual databases, a connection is forged to the
1603 appropriate server and the Init request is forwarded to it. If,
1604 later in the session, the same client searches in a different
1605 virtual database, then a connection is forged to the server that
1606 hosts it, and the same cached Init request is forwarded there,
1610 All of this clever Init-delaying is done by the
1611 <literal>frontend_net</literal> filter. The
1612 <literal>virt_db</literal> filter knows nothing about it; in
1613 fact, because the Init request that is received from the client
1614 doesn't get forwarded until a Search request is received, the
1615 <literal>virt_db</literal> filter (and the
1616 <literal>z3950_client</literal> filter behind it) doesn't even get
1617 invoked at Init time. The <emphasis>only</emphasis> thing that a
1618 <literal>virt_db</literal> filter ever does is rewrite the
1619 <literal>VAL_PROXY</literal> otherInfo in the requests that pass
1623 It is possible for a <literal>virt_db</literal> filter to contain
1625 <literal><target></literal>
1626 elements. What does this mean? Only that the filter will add
1627 multiple <literal>VAL_PROXY</literal> otherInfo packets to the
1628 Search requests that pass through it. That's because the virtual
1629 DB filter is dumb, and does exactly what it's told - no more, no
1631 If a Search request with multiple <literal>VAL_PROXY</literal>
1632 otherInfo packets reaches a <literal>z3950_client</literal>
1633 filter, this is an error. That filter doesn't know how to deal
1634 with multiple targets, so it will either just pick one and search
1635 in it, or (better) fail with an error message.
1638 The <literal>multi</literal> filter comes to the rescue! This is
1639 the only filter that knows how to deal with multiple
1640 <literal>VAL_PROXY</literal> otherInfo packets, and it does so by
1641 making multiple copies of the entire Search request: one for each
1642 <literal>VAL_PROXY</literal>. Each of these new copies is then
1643 passed down through the remaining filters in the route. (The
1644 copies are handled in parallel though the
1645 spawning of new threads.) Since the copies each have only one
1646 <literal>VAL_PROXY</literal> otherInfo, they can be handled by the
1647 <literal>z3950_client</literal> filter, which happily deals with
1648 each one individually. When the results of the individual
1649 searches come back up to the <literal>multi</literal> filter, it
1650 merges them into a single Search response, which is what
1651 eventually makes it back to the client.
1656 <imagedata fileref="multi.pdf" format="PDF" scale="50"/>
1659 <imagedata fileref="multi.png" format="PNG"/>
1662 <!-- Fall back if none of the images can be used -->
1664 [Here there should be a diagram showing the progress of
1665 packages through the filters during a simple virtual-database
1666 search and a multi-database search, but is seems that your
1667 tool chain has not been able to include the diagram in this
1672 <para>A picture is worth a thousand words (but only five hundred on 64-bit architectures)</para>
1679 <chapter id="sru-server">
1680 <title>Combined SRU webservice and Z39.50 server configuration</title>
1682 Metaproxy can act as
1683 <ulink url="&url.sru;">SRU</ulink> and
1684 web service server, which translates web service requests to
1685 <ulink url="&url.z39.50;">ANSI/NISO Z39.50</ulink> packages and
1686 sends them off to common available targets.
1689 A typical setup for this operation needs a filter route including the
1693 <table id="sru-server-table-config" frame="top">
1694 <title>SRU/Z39.50 Server Filter Route Configuration</title>
1698 <entry>Filter</entry>
1699 <entry>Importance</entry>
1700 <entry>Purpose</entry>
1706 <entry><literal>frontend_net</literal></entry>
1707 <entry>required</entry>
1708 <entry>Accepting HTTP connections and passing them to following
1709 filters. Since this filter also accepts Z39.50 connections, the
1710 server works as SRU and Z39.50 server on the same port.</entry>
1713 <entry><literal>sru_z3950</literal></entry>
1714 <entry>required</entry>
1715 <entry>Accepting SRU GET/POST/SOAP explain and
1716 searchRetrieve requests for the the configured databases.
1717 Explain requests are directly served from the static XML configuration.
1718 SearchRetrieve requests are
1719 transformed to Z39.50 search and present packages.
1720 All other HTTP and Z39.50 packages are passed unaltered.</entry>
1723 <entry><literal>http_file</literal></entry>
1724 <entry>optional</entry>
1725 <entry>Serving HTTP requests from the filesystem. This is only
1726 needed if the server should serve XSLT stylesheets, static HTML
1727 files or Java Script for thin browser based clients.
1728 Z39.50 packages are passed unaltered.</entry>
1731 <entry><literal>cql_rpn</literal></entry>
1732 <entry>required</entry>
1733 <entry>Usually, Z39.50 servers do not talk CQL, hence the
1734 translation of the CQL query language to RPN is mandatory in
1735 most cases. Affects only Z39.50 search packages.</entry>
1738 <entry><literal>record_transform</literal></entry>
1739 <entry>optional</entry>
1740 <entry>Some Z39.50 backend targets can not present XML record
1741 syntaxes in common wanted element sets. using this filter, one
1742 can transform binary MARC records to MARCXML records, and
1743 further transform those to any needed XML schema/format by XSLT
1744 transformations. Changes only Z39.50 present packages.</entry>
1747 <entry><literal>session_shared</literal></entry>
1748 <entry>optional</entry>
1749 <entry>The stateless nature of web services requires frequent
1750 re-searching of the same targets for display of paged result set
1751 records. This might be an unacceptable burden for the accessed
1752 backend Z39.50 targets, and this mosule can be added for
1753 efficient backend target resource pooling.</entry>
1756 <entry><literal>z3950_client</literal></entry>
1757 <entry>required</entry>
1758 <entry>Finally, a Z39.50 package sink is needed in the filter
1759 chain to provide the response packages. The Z39.50 client module
1760 is used to access external targets over the network, but any
1761 coming local Z39.50 package sink could be used instead of.</entry>
1764 <entry><literal>bounce</literal></entry>
1765 <entry>required</entry>
1766 <entry>Any Metaproxy package arriving here did not do so by
1767 purpose, and is bounced back with connection closure. this
1768 prevents inifinite package hanging inside the SRU server.</entry>
1774 A typical minimal example <ulink url="&url.sru;">SRU</ulink>
1775 server configuration file is found in the tarball distribution at
1776 <literal>etc/config-sru-to-z3950.xml</literal>.
1779 Off course, any other metaproxy modules can be integrated into a
1780 SRU server solution, including, but not limited to, load balancing,
1781 multiple target querying
1782 (see <xref linkend="multidb"/>), and complex RPN query rewrites.
1789 <chapter id="extensions">
1790 <title>Writing extensions for Metaproxy</title>
1791 <para>### To be written</para>
1797 <chapter id="classes">
1798 <title>Classes in the Metaproxy source code</title>
1801 <section id="classes-introductory-notes">
1802 <title>Introductory notes</title>
1804 <emphasis>Stop! Do not read this!</emphasis>
1805 You won't enjoy it at all. You should just skip ahead to
1806 <xref linkend="reference"/>,
1808 <!-- The remainder of this paragraph is lifted verbatim from
1809 Douglas Adams' _Hitch Hiker's Guide to the Galaxy_, chapter 8 -->
1810 you things you really need to know, like the fact that the
1811 fabulously beautiful planet Bethselamin is now so worried about
1812 the cumulative erosion by ten billion visiting tourists a year
1813 that any net imbalance between the amount you eat and the amount
1814 you excrete whilst on the planet is surgically removed from your
1815 bodyweight when you leave: so every time you go to the lavatory it
1816 is vitally important to get a receipt.
1819 This chapter contains documentation of the Metaproxy source code, and is
1820 of interest only to maintainers and developers. If you need to
1821 change Metaproxy's behavior or write a new filter, then you will most
1822 likely find this chapter helpful. Otherwise it's a waste of your
1823 good time. Seriously: go and watch a film or something.
1824 <citetitle>This is Spinal Tap</citetitle> is particularly good.
1827 Still here? OK, let's continue.
1830 In general, classes seem to be named big-endianly, so that
1831 <literal>FactoryFilter</literal> is not a filter that filters
1832 factories, but a factory that produces filters; and
1833 <literal>FactoryStatic</literal> is a factory for the statically
1834 registered filters (as opposed to those that are dynamically
1839 <section id="individual.classes">
1840 <title>Individual classes</title>
1842 The classes making up the Metaproxy application are here listed by
1843 class-name, with the names of the source files that define them in
1847 <section id="class-FactoryFilter">
1848 <title><literal>mp::FactoryFilter</literal>
1849 (<filename>factory_filter.cpp</filename>)</title>
1851 A factory class that exists primarily to provide the
1852 <literal>create()</literal> method, which takes the name of a
1853 filter class as its argument and returns a new filter of that
1854 type. To enable this, the factory must first be populated by
1855 calling <literal>add_creator()</literal> for static filters (this
1856 is done by the <literal>FactoryStatic</literal> class, see below)
1857 and <literal>add_creator_dyn()</literal> for filters loaded
1862 <section id="class-FactoryStatic">
1863 <title><literal>mp::FactoryStatic</literal>
1864 (<filename>factory_static.cpp</filename>)</title>
1866 A subclass of <literal>FactoryFilter</literal> which is
1867 responsible for registering all the statically defined filter
1868 types. It does this by knowing about all those filters'
1869 structures, which are listed in its constructor. Merely
1870 instantiating this class registers all the static classes. It is
1871 for the benefit of this class that <literal>struct
1872 metaproxy_1_filter_struct</literal> exists, and that all the filter
1873 classes provide a static object of that type.
1877 <section id="class-filter-Base">
1878 <title><literal>mp::filter::Base</literal>
1879 (<filename>filter.cpp</filename>)</title>
1881 The virtual base class of all filters. The filter API is, on the
1882 surface at least, extremely simple: two methods.
1883 <literal>configure()</literal> is passed an XML DOM tree representing
1884 that part of the configuration file that pertains to this filter
1885 instance, and is expected to walk that tree extracting relevant
1886 information. And <literal>process()</literal> processes a
1887 package (see below). That surface simplicity is a bit
1888 misleading, as <literal>process()</literal> needs to know a lot
1889 about the <literal>Package</literal> class in order to do
1894 <section id="class-AuthSimple">
1895 <title><literal>mp::filter::AuthSimple</literal>,
1896 <literal>Backend_test</literal>, etc.
1897 (<filename>filter_auth_simple.cpp</filename>,
1898 <filename>filter_backend_test.cpp</filename>, etc.)</title>
1900 Individual filters. Each of these is implemented by a header and
1901 a source file, named <filename>filter_*.hpp</filename> and
1902 <filename>filter_*.cpp</filename> respectively. All the header
1903 files should be pretty much identical, in that they declare the
1904 class, including a private <literal>Rep</literal> class and a
1905 member pointer to it, and the two public methods.
1908 The source file for each filter needs to supply:
1913 A definition of the private <literal>Rep</literal> class.
1918 Some boilerplate constructors and destructors.
1923 A <literal>configure()</literal> method that uses the
1924 appropriate XML fragment.
1929 Most important, the <literal>process()</literal> method that
1930 does all the actual work.
1936 <section id="class-Package">
1937 <title><literal>mp::Package</literal>
1938 (<filename>package.cpp</filename>)</title>
1940 Represents a package on its way through the series of filters
1941 that make up a route. This is essentially a Z39.50 or SRU APDU
1942 together with information about where it came from, which is
1943 modified as it passes through the various filters.
1947 <section id="class-Pipe">
1948 <title><literal>mp::Pipe</literal>
1949 (<filename>pipe.cpp</filename>)</title>
1951 This class provides a compatibility layer so that we have an IPC
1952 mechanism that works the same under Unix and Windows. It's not
1953 particularly exciting.
1957 <section id="class-RouterChain">
1958 <title><literal>mp::RouterChain</literal>
1959 (<filename>router_chain.cpp</filename>)</title>
1965 <section id="class-RouterFleXML">
1966 <title><literal>mp::RouterFleXML</literal>
1967 (<filename>router_flexml.cpp</filename>)</title>
1973 <section id="class-Session">
1974 <title><literal>mp::Session</literal>
1975 (<filename>session.cpp</filename>)</title>
1981 <section id="class-ThreadPoolSocketObserver">
1982 <title><literal>mp::ThreadPoolSocketObserver</literal>
1983 (<filename>thread_pool_observer.cpp</filename>)</title>
1989 <section id="class-util">
1990 <title><literal>mp::util</literal>
1991 (<filename>util.cpp</filename>)</title>
1993 A namespace of various small utility functions and classes,
1994 collected together for convenience. Most importantly, includes
1995 the <literal>mp::util::odr</literal> class, a wrapper for YAZ's
2000 <section id="class-xml">
2001 <title><literal>mp::xml</literal>
2002 (<filename>xmlutil.cpp</filename>)</title>
2004 A namespace of various XML utility functions and classes,
2005 collected together for convenience.
2011 <section id="other.source.files">
2012 <title>Other Source Files</title>
2014 In addition to the Metaproxy source files that define the classes
2015 described above, there are a few additional files which are
2016 briefly described here:
2020 <term><literal>metaproxy_prog.cpp</literal></term>
2023 The main function of the <command>metaproxy</command> program.
2028 <term><literal>ex_router_flexml.cpp</literal></term>
2031 Identical to <literal>metaproxy_prog.cpp</literal>: it's not clear why.
2036 <term><literal>test_*.cpp</literal></term>
2039 Unit-tests for various modules.
2045 ### Still to be described:
2046 <literal>ex_filter_frontend_net.cpp</literal>,
2047 <literal>filter_dl.cpp</literal>,
2048 <literal>plainfile.cpp</literal>,
2049 <literal>tstdl.cpp</literal>.
2055 <reference id="reference">
2056 <title>Reference</title>
2057 <partintro id="reference-introduction">
2059 The material in this chapter is drawn directly from the individual
2060 manual entries. In particular, the Metaproxy invocation section is
2061 available using <command>man metaproxy</command>, and the section
2062 on each individual filter is available using the name of the filter
2063 as the argument to the <command>man</command> command.
2069 <appendix id="license">
2070 <title>License</title>
2075 Metaproxy is free software; you can redistribute it and/or modify it under
2076 the terms of the GNU General Public License as published by the Free
2077 Software Foundation; either version 2, or (at your option) any later
2082 Metaproxy is distributed in the hope that it will be useful, but WITHOUT ANY
2083 WARRANTY; without even the implied warranty of MERCHANTABILITY or
2084 FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
2089 You should have received a copy of the GNU General Public License
2090 along with Metaproxy; see the file LICENSE. If not, write to the
2091 Free Software Foundation,
2092 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
2100 <!-- Keep this comment at the end of the file
2103 nxml-child-indent: 1