1 <?xml version="1.0" standalone="no"?>
2 <!DOCTYPE book PUBLIC "-//OASIS//DTD DocBook XML V4.1//EN"
3 "http://www.oasis-open.org/docbook/xml/4.1/docbookx.dtd"
5 <!ENTITY % local SYSTEM "local.ent">
7 <!ENTITY % entities SYSTEM "entities.ent">
9 <!ENTITY % idcommon SYSTEM "common/common.ent">
14 <title>Pazpar2 - User's Guide and Reference</title>
16 <firstname>Sebastian</firstname><surname>Hammer</surname>
19 <firstname>Adam</firstname><surname>Dickmeiss</surname>
22 <firstname>Marc</firstname><surname>Cromme</surname>
25 <firstname>Jakub</firstname><surname>Skoczen</surname>
27 <releaseinfo>&version;</releaseinfo>
29 <year>©right-year;</year>
30 <holder>Index Data</holder>
34 Pazpar2 is a high-performance metasearch engine featuring
35 merging, relevance ranking, record sorting,
37 It is middleware: it has no user interface of its own, but can be
38 configured and controlled by an XML-over-HTTP web-service to provide
39 metasearching functionality behind any user interface.
42 This document is a guide and reference to Pazpar2 version &version;.
47 <imagedata fileref="common/id.png" format="PNG"/>
50 <imagedata fileref="common/id.eps" format="EPS"/>
57 <chapter id="introduction">
58 <title>Introduction</title>
60 Pazpar2 is a stand-alone metasearch engine with a web-service API, designed
61 to be used either from a browser-based client (JavaScript, Flash,
63 etc.), from server-side code, or any combination of the two.
64 Pazpar2 is a highly optimized client designed to
65 search many resources in parallel. It implements record merging,
66 relevance-ranking and sorting by arbitrary data content, and facet
67 analysis for browsing purposes. It is designed to be data-model
68 independent, and is capable of working with MARC, DublinCore, or any
69 other <ulink url="&url.xml;">XML</ulink>-structured response format
70 -- <ulink url="&url.xslt;">XSLT</ulink> is used to normalize and extract
71 data from retrieval records for display and analysis. It can be used
72 against any server which supports the
73 <ulink url="&url.z39.50;">Z39.50</ulink> or <ulink url="&url.sru;">SRU/SRW</ulink>
75 backend modules can function as connectors between these standard
76 protocols and any non-standard API, including web-site scraping, to
77 support a large number of other protocols
78 (please contact Index Data for further information about this).
81 Additional functionality such as
82 user management and attractive displays are expected to be implemented by
83 applications that use Pazpar2. Pazpar2 itself is user-interface independent.
84 Its functionality is exposed through a simple XML RPC web-service API,
85 designed to be easy to use from an AJAX-enabled browser, Flash
86 animation, Java applet, etc., or from a higher-level server-side language
87 like PHP, Perl or Java. Because session information can be shared between
88 browser-based logic and server-side scripting, there is tremendous
89 flexibility in how you implement application-specific logic on top
93 Once you launch a search in Pazpar2, the operation continues behind the
94 scenes. Pazpar2 connects to servers, carries out searches, and
95 retrieves, deduplicates, and stores results internally. Your application
96 code may periodically inquire about the status of an ongoing operation,
97 and ask to see records or result set facets. Results become
98 available immediately, and it is easy to build end-user interfaces than
99 feel extremely responsive, even when searching more than 100 servers
103 Pazpar2 is designed to be highly configurable. Incoming records are
104 normalized to XML/UTF-8, and then further normalized using XSLT to a
105 simple internal representation that is suitable for analysis. By
106 providing XSLT stylesheets for different kinds of result records, you
107 can configure Pazpar2 to work against different kinds of information
108 retrieval servers. Finally, metadata is extracted in a configurable
109 way from this internal record, to support display, merging, ranking,
110 result set facets, and sorting. Pazpar2 is not bound to a specific model
111 of metadata, such as DublinCore or MARC: by providing the right
112 configuration, it can work with any combination of different kinds of data in
113 support of many different applications.
116 Pazpar2 is designed to be efficient and scalable. You can set it up to
117 search several hundred targets in parallel, or you can use it to support
118 hundreds of concurrent users. It is implemented with the same attention
119 to performance and economy that we use in our indexing engines, so that
120 you can focus on building your application without worrying about the
121 details of metasearch logic. You can devote all of your attention to
122 usability and let Pazpar2 do what it does best -- metasearch.
125 If you wish to connect to commercial or other databases which do not
126 support open standards, please contact Index Data. We have a
127 proprietary framework for building connectors that enable Pazpar2
129 thousands of online databases, in addition to the vast number of catalogs
130 and online services that support the Z39.50/SRU/SRW protocols.
133 Pazpar2 is our attempt to re-think the traditional paradigms for
134 implementing and deploying metasearch logic, with an uncompromising
135 approach to performance, and attempting to make maximum use of the
136 capabilities of modern browsers. The demo user interface that
137 accompanies the distribution is but one example. If you think of new
138 ways of using Pazpar2, we hope you'll share them with us, and if we
139 can provide assistance with regards to training, design, programming,
140 integration with different backends, hosting, or support, please don't
141 hesitate to contact us. If you'd like to see functionality in Pazpar2
142 that is not there today, please don't hesitate to contact us. It may
143 already be in our development pipeline, or there might be a
144 possibility for you to help out by sponsoring development time or
145 code. Either way, get in touch and we will give you straight answers.
151 Pazpar2 is covered by the GNU General Public License (GPL) version 2.
152 See <xref linkend="license"/> for further information.
156 <chapter id="installation">
157 <title>Installation</title>
159 The Pazpar2 package is very small. It includes documentation as well
160 as the Pazpar2 server. The package also includes a simple user
161 interface test1 which consists of a single HTML page and a single
162 JavaScript file to illustrate the use of Pazpar2.
165 Pazpar2 depends on the following tools/libraries:
167 <varlistentry><term><ulink url="&url.yaz;">YAZ</ulink></term>
170 The popular Z39.50 toolkit for the C language.
171 YAZ <emphasis>must</emphasis> be compiled with Libxml2/Libxslt support.
175 <varlistentry><term><ulink url="&url.icu;">International
176 Components for Unicode (ICU)</ulink></term>
179 ICU provides Unicode support for non-English languages with
180 character sets outside the range of 7bit ASCII, like
181 Greek, Russian, German and French. Pazpar2 uses the ICU
182 Unicode character conversions, Unicode normalization, case
183 folding and other fundamental operations needed in
184 tokenization, normalization and ranking of records.
187 Compiling, linking, and usage of the ICU libraries is optional,
188 but strongly recommended for usage in an international
196 In order to compile Pazpar2, a C compiler which supports C99 or later
200 <section id="installation.unix">
201 <title>Installation on Unix (from Source)</title>
203 The latest source code for Pazpar2 is available from
204 <ulink url="&url.pazpar2.download;"/>.
205 Only few systems have none of the required
206 tools binary packages.
207 If, for example, Libxml2/libXSLT libraries
208 are already installed as development packages use these.
212 Ensure that the development libraries + header files are
213 available on your system before compiling Pazpar2. For installation
214 of YAZ, refer to the YAZ installation chapter.
217 gunzip -c pazpar2-version.tar.gz|tar xf -
225 The <literal>make install</literal> will install manpages as well as the
226 Pazpar2 server, <literal>pazpar2</literal>,
227 in PREFIX<literal>/sbin</literal>.
228 By default, PREFIX is <literal>/usr/local/</literal> . This can be
229 changed with configure option <option>--prefix</option>.
233 <section id="installation.win32">
234 <title>Installation on Windows (from Source)</title>
236 Pazpar2 can be built for Windows using
237 <ulink url="&url.vstudio;">Microsoft Visual Studio</ulink>.
238 The support files for building YAZ on Windows are located in the
239 <filename>win</filename> directory. The compilation is performed
240 using the <filename>win/makefile</filename> which is to be
241 processed by the NMAKE utility part of Visual Studio.
244 Ensure that the development libraries + header files are
245 available on your system before compiling Pazpar2. For installation
246 of YAZ, refer to the YAZ installation chapter.
247 It is easiest if YAZ and Pazpar2 are unpacked in the same
248 directory (side-by-side).
251 The compilation is tuned by editing the makefile of Pazpar2.
252 The process is similar to YAZ. Adjust the various directories
253 <literal>YAZ_DIR</literal>, <literal>ZLIB_DIR</literal>, ..
256 Compile Pazpar2 by invoking <application>nmake</application> in
257 the <filename>win</filename> directory.
258 The resulting binaries of the build process are located in the
259 <filename>bin</filename> of the Pazpar2 source
260 tree - including the <filename>pazpar2.exe</filename> and necessary DLLs.
263 The Windows version of Pazpar2 is a console application. It may
264 be installed as a Windows Service by adding option
265 <literal>-install</literal> for the pazpar2 program. This will
266 register Pazpar2 as a service and use the other options provided
267 in the same invocation. For example:
270 ..\bin\pazpar2 -install -f pazpar2.cfg -l pazpar2.log
272 The Pazpar2 service may now be controlled via the Service Control
273 Panel. It may be unregistered by passing the <literal>-remove</literal>
277 ..\bin\pazpar2 -remove
282 <section id="installation.test1">
283 <title>Installation of test1 interface</title>
285 In this section we outline how to install a simple interface that
286 is part of the Pazpar2 source package. Note that Debian users can
287 save time by just installing package <literal>pazpar2-test1</literal>.
290 A web server must be installed and running on the system, such as Apache.
294 Start the Pazpar2 daemon using the 'in-source' binary of the Pazpar2
295 daemon. On Unix the process is:
298 cp pazpar2.cfg.dist pazpar2.cfg
299 ../src/pazpar2 -f pazpar2.cfg
304 copy pazpar2.cfg.dist pazpar2.cfg
305 ..\bin\pazpar2 -f pazpar2.cfg
307 This will start a Pazpar2 listener on port 9004. It will proxy
308 HTTP requests to localhost - port 80, which we assume will be the regular
309 HTTP server on the system. Inspect and modify pazpar2.cfg as needed
310 if this is to be changed. The pazpar2.cfg includes settings from the
311 file <filename>settings/edu.xml</filename>
315 Make a new console and move to the other stuff.
316 For more information about pazpar2 options refer to the manpage.
320 The test1 UI is located in <literal>www/test1</literal>. Ensure this
321 directory is available to the web server by either copying
322 <literal>test1</literal> to the document root, create a symlink or
323 use Apache's <literal>Alias</literal> directive.
327 The interface test1 interface should now be available on port 8004.
330 If you don't see the test1 interface. See if test1 is really available
331 on the same URL but on port 80. If it's not, the Apache configuration
332 (or other) is not correct.
335 In order to use Apache as frontend for the interface on port 80
336 for public access etc., refer to
337 <xref linkend="installation.apache2proxy"/>.
341 <section id="installation.debian">
342 <title>Installation on Debian GNU/Linux</title>
344 Index Data provides Debian packages for Pazpar2. These are prepared
345 for Debian versions Etch and Lenny (as of 2007).
346 These packages are available at
347 <ulink url="&url.pazpar2.download.debian;"/>.
351 <section id="installation.apache2proxy">
352 <title>Apache 2 Proxy</title>
355 <ulink url="http://httpd.apache.org/docs/2.2/mod/mod_proxy.html">
357 </ulink> which allows Pazpar2 to become a backend to an Apache 2
358 based web service. The Apache 2 proxy must operate in the
359 <emphasis>Reverse</emphasis> Proxy mode.
363 On a Debian based Apache 2 system, the relevant modules can
366 sudo a2enmod proxy_http
371 Traditionally Pazpar2 interprets URL paths with suffix
372 <literal>/search.pz2</literal>.
375 url="http://httpd.apache.org/docs/2.2/mod/mod_proxy.html#proxypass"
376 >ProxyPass</ulink> directive of Apache must be used to map a URL path
377 the the Pazpar2 server (listening port).
382 The ProxyPass directive takes a prefix rather than
383 a suffix as URL path. It is important that the Java Script code
384 uses the prefix given for it.
388 <example id="installation.apache2proxy.example">
389 <title>Apache 2 proxy configuration</title>
391 If Pazpar2 is running on port 8004 and the portal is using
392 <filename>search.pz2</filename> inside portal in directory
393 <filename>/myportal/</filename> we could use the following
394 Apache 2 configuration:
397 <IfModule mod_proxy.c>
401 AddDefaultCharset off
406 ProxyPass /myportal/search.pz2 http://localhost:8004/search.pz2
417 <title>Using Pazpar2</title>
419 This chapter provides a general introduction to the use and
420 deployment of Pazpar2.
423 <section id="architecture">
424 <title>Pazpar2 and your systems architecture</title>
426 Pazpar2 is designed to provide asynchronous, behind-the-scenes
427 metasearching functionality to your application, exposing this
428 functionality using a simple webservice API that can be accessed
429 from any number of development environments. In particular, it is
430 possible to combine Pazpar2 either with your server-side dynamic
431 website scripting, with scripting or code running in the browser, or
432 with any combination of the two. Pazpar2 is an excellent tool for
433 building advanced, AJAX-based user interfaces for metasearch
434 functionality, but it isn't a requirement -- you can choose to use
435 Pazpar2 entirely as a backend to your regular server-side scripting.
436 When you do use Pazpar2 in conjunction
437 with browser scripting (JavaScript/AJAX, Flash, applets,
438 etc.), there are special considerations.
442 Pazpar2 implements a simple but efficient HTTP server, and it is
443 designed to interact directly with scripting running in the browser
444 for the best possible performance, and to limit overhead when
445 several browser clients generate numerous webservice requests.
446 However, it is still desirable to use a conventional webserver,
447 such as Apache, to serve up graphics, HTML documents, and
448 server-side scripting. Because the security sandbox environment of
449 most browser-side programming environments only allows communication
450 with the server from which the enclosing HTML page or object
451 originated, Pazpar2 is designed so that it can act as a transparent
452 proxy in front of an existing webserver (see <xref
453 linkend="pazpar2_conf"/> for details).
454 In this mode, all regular
455 HTTP requests are transparently passed through to your webserver,
456 while Pazpar2 only intercepts search-related webservice requests.
460 If you want to expose your combined service on port 80, you can
461 either run your regular webserver on a different port, a different
462 server, or a different IP address associated with the same server.
466 Pazpar2 can also work behind
467 a reverse Proxy. Refer to <xref linkend="installation.apache2proxy"/>)
468 for more information.
469 This allows your existing HTTP server to operate on port 80 as usual.
470 Pazpar2 can be started on another (internal) port.
474 Sometimes, it may be necessary to implement functionality on your
475 regular webserver that makes use of search results, for example to
476 implement data import functionality, emailing results, history
477 lists, personal citation lists, interlibrary loan functionality,
478 etc. Fortunately, it is simple to exchange information between
479 Pazpar2, your browser scripting, and backend server-side scripting.
480 You can send a session ID and possibly a record ID from your browser
481 code to your server code, and from there use Pazpar2s webservice API
482 to access result sets or individual records. You could even 'hide'
483 all of Pazpar2s functionality between your own API implemented on
484 the server-side, and access that from the browser or elsewhere. The
485 possibilities are just about endless.
489 <section id="data_model">
490 <title>Your data model</title>
492 Pazpar2 does not have a preconceived model of what makes up a data
493 model. There are no assumptions that records have specific fields or
494 that they are organized in any particular way. The only assumption
495 is that data comes packaged in a form that the software can work
496 with (presently, that means XML or MARC), and that you can provide
497 the necessary information to massage it into Pazpar2's internal
502 Handling retrieval records in Pazpar2 is a two-step process. First,
503 you decide which data elements of the source record you are
504 interested in, and you specify any desired massaging or combining of
505 elements using an XSLT stylesheet (MARC records are automatically
506 normalized to <ulink url="&url.marcxml;">MARCXML</ulink> before this step).
507 If desired, you can run multiple XSLT stylesheets in series to accomplish
508 this, but the output of the last one should be a representation of the
509 record in a schema that Pazpar2 understands.
513 The intermediate, internal representation of the record looks like
516 <record xmlns="http://www.indexdata.com/pazpar2/1.0"
517 mergekey="title The Shining author King, Stephen">
519 <metadata type="title">The Shining</metadata>
521 <metadata type="author">King, Stephen</metadata>
523 <metadata type="kind">ebook</metadata>
525 <!-- ... and so on -->
529 As you can see, there isn't much to it. There are really only a few
530 important elements to this file.
534 Elements should belong to the namespace
535 <literal>http://www.indexdata.com/pazpar2/1.0</literal>.
536 If the root node contains the
537 attribute 'mergekey', then every record that generates the same
538 merge key (normalized for case differences, white space, and
539 truncation) will be joined into a cluster. In other words, you
540 decide how records are merged. If you don't include a merge key,
541 records are never merged. The 'metadata' elements provide the meat
542 of the elements -- the content. the 'type' attribute is used to
543 match each element against processing rules that determine what
544 happens to the data element next.
548 The next processing step is the extraction of metadata from the
549 intermediate representation of the record. This is governed by the
550 'metadata' elements in the 'service' section of the configuration
551 file. See <xref linkend="config-server"/> for details. The metadata
552 in the retrieval record ultimately drives merging, sorting, ranking,
553 the extraction of browse facets, and display, all configurable.
557 <section id="client">
558 <title>Client development overview</title>
560 You can use Pazpar2 from any environment that allows you to use
561 webservices. The initial goal of the software was to support
562 AJAX-based applications, but there literally are no limits to what
563 you can do. You can use Pazpar2 from Javascript, Flash, Java, etc.,
564 on the browser side, and from any development environment on the
565 server side, and you can pass session tokens and record IDs freely
566 around between these environments to build sophisticated applications.
567 Use your imagination.
571 The webservice API of Pazpar2 is described in detail in <xref
572 linkend="pazpar2_protocol"/>.
576 In brief, you use the 'init' command to create a session, a
577 temporary workspace which carries information about the current
578 search. You start a new search using the 'search' command. Once the
579 search has been started, you can follow its progress using the
580 'stat', 'bytarget', 'termlist', or 'show' commands. Detailed records
581 can be fetched using the 'record' command.
587 <section id="nonstandard">
588 <title>Connecting to non-standard resources</title>
590 Pazpar2 uses Z39.50 as its switchboard language -- i.e. as far as it
591 is concerned, all resources speak Z39.50, or its webservices derivatives,
592 SRU/SRW. It is, however, equipped
593 to handle a broad range of different server behavior, through
594 configurable query mapping and record normalization. If you develop
595 configuration, stylesheets, etc., for a new type of resources, we
596 encourage you to share your work. But you can also use Pazpar2 to
597 connect to hundreds of resources that do not support standard
602 For a growing number of resources, Z39.50 is all you need. Over the
603 last few years, a number of commercial, full-text resources have
604 implemented Z39.50. These can be used through Pazpar2 with little or
605 no effort. Resources that use non-standard record formats will
606 require a bit of XSLT work, but that's all.
610 But what about resources that don't support Z39.50 at all? Some resources might
611 support OpenSearch, private, XML/HTTP-based protocols, or something
612 else entirely. Some databases exist only as web user interfaces and
613 will require screen-scraping. Still others exist only as static
614 files, or perhaps as databases supporting the OAI-PMH protocol.
615 There is hope! Read on.
619 Index Data continues to advocate the support of open standards. We
620 work with database vendors to support standards, so you don't have
621 to worry about programming against non-standard services. We also
622 provide tools (see <ulink
623 url="http://www.indexdata.com/simpleserver">SimpleServer</ulink>)
624 which make it comparatively easy to build gateways against servers
625 with non-standard behavior. Again, we encourage you to share any
626 work you do in this direction.
630 But the bottom line is that working with non-standard resources in
631 metasearching is really, really hard. If you want to build a
632 project with Pazpar2, and you need access to resources with
633 non-standard interfaces, we can help. We run gateways to more than
634 2,000 popular, commercial databases and other resources,
636 to plug them directly into Pazpar2. For a small annual fee per
637 database, we can help you establish connections to your licensed
638 resources. Meanwhile, you can help! If you build your own
639 standards-compliant gateways, host them for others, or share the
640 code! And tell your vendors that they can save everybody money and
641 increase the appeal of their resources by supporting standards.
645 There are those who will ask us why we are using Z39.50 as our
646 switchboard language rather than a different protocol. Basically,
647 we believe that Z39.50 is presently the most widely implemented
648 information retrieval protocol that has the level of functionality
649 required to support a good metasearching experience (structured
650 searching, structured, well-defined results). It is also compact and
651 efficient, and there is a very broad range of tools available to
656 <section id="unicode">
657 <title>Unicode Compliance</title>
659 Pazpar2 is Unicode compliant and language and locale aware but relies
660 on character encoding for the targets to be specified correctly if
661 the targets themselves are not UTF-8 based (most aren't).
662 Just a few bad behaving targets can spoil the search experience
663 considerably if for example Greek, Russian or otherwise non 7-bit ASCII
664 search terms are entered. In these cases some targets return
665 records irrelevant to the query, and the result screens will be
666 cluttered with noise.
669 While noise from misbehaving targets can not be removed, it can
670 be reduced using truly Unicode based ranking. This is an
671 option which is available to the system administrator if ICU
672 support is compiled into Pazpar2, see
673 <xref linkend="installation"/> for details.
676 In addition, the ICU tokenization and normalization rules must
677 be defined in the master configuration file described in
678 <xref linkend="config-server"/>.
682 <section id="load_balancing">
683 <title>Load balancing</title>
685 Just like any web server, Pazpar2, can be load balanced by a standard hardware or software load balancer as long as the session stickiness is ensured. If you are already running the Apache2 web server in front of Pazpar2 and use the apache mod_proxy module to 'relay' client requests to Pazpar2, this set up can be easily extended to include load balancing capabilites. To do so you need to enable the <ulink url="http://httpd.apache.org/docs/2.2/mod/mod_proxy_balancer.html">
687 </ulink> module in your Apache2 installation.
691 On a Debian based Apache 2 system, the relevant modules can
694 sudo a2enmod proxy_http
699 The mod_proxy_balancer can pass all 'sessionsticky' requests to the same backend worker as long as the requests are marked with the originating worker's ID (called 'route'). If the Pazpar2 serverID is configured (by setting an 'id' attribute on the 'server' element in the Pazpar2 configuration file) Pazpar2 will append it to the 'session' element returned during the 'init' in a mod_proxy_balancer compatible manner. Since the 'session' is then re-sent by the client (for all pazpar2 request besides 'init'), the balancer can use the marker to pass the request to the right route. To do so the balancer needs to be configured to inspect the 'session' parameter.
702 <example id="load_balancing.example">
703 <title>Apache 2 load balancing configuration</title>
705 Having 4 Pazpar2 instances running on the same host, port range of 8004-8007 and serverIDs of: pz1, pz2, pz3 and pz4 respectively we could use the following Apache 2 configuration to expose a single pazpar2 'endpoint' on a standard (<filename>/pazpar2/search.pz2</filename>) location:
709 AddDefaultCharset off
715 # 'route' has to match the configured pazpar2 server ID
716 <Proxy balancer://pz2cluster>
717 BalancerMember http://localhost:8004 route=pz1
718 BalancerMember http://localhost:8005 route=pz2
719 BalancerMember http://localhost:8006 route=pz3
720 BalancerMember http://localhost:8007 route=pz4
723 # route is resent in the 'session' param which has the form:
724 # 'sessid.serverid', understandable by the mod_proxy_load_balancer
725 # this is not going to work if the client tampers with the 'session' param
726 ProxyPass /pazpar2/search.pz2 balancer://pz2cluster lbmethod=byrequests stickysession=session nofailover=On]]></screen>
728 The 'ProxyPass' line sets up a reverse proxy for request ‘/pazpar2/search.pz2’ and delegates all requests to the load balancer (virtual worker) with name ‘pz2cluster’. Sticky sessions are enabled and implemented using the ‘session’ parameter. The ‘Proxy’ section lists all the servers (real workers) which the load balancer can use.
736 </chapter> <!-- Using Pazpar2 -->
738 <reference id="reference">
739 <title>Reference</title>
740 <partintro id="reference-introduction">
742 The material in this chapter is drawn directly from the individual
749 <appendix id="license"><title>License</title>
753 Copyright © ©right-year; Index Data.
757 Pazpar2 is free software; you can redistribute it and/or modify it under
758 the terms of the GNU General Public License as published by the Free
759 Software Foundation; either version 2, or (at your option) any later
764 Pazpar2 is distributed in the hope that it will be useful, but WITHOUT ANY
765 WARRANTY; without even the implied warranty of MERCHANTABILITY or
766 FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
771 You should have received a copy of the GNU General Public License
772 along with Pazpar2; see the file LICENSE. If not, write to the
773 Free Software Foundation,
774 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
783 <!-- Keep this comment at the end of the file
788 sgml-minimize-attributes:nil
789 sgml-always-quote-attributes:t
792 sgml-parent-document: nil
793 sgml-local-catalogs: nil
794 sgml-namecase-general:t