3 Use YAZ backtrace facility PAZ-983. In case of crashes YAZ log may
4 include more details about the fault. Better install gdb and the
5 debug version of pazpar2 (pazpar2-dbg on Debian/Ubuntu).
7 Per target timeout PAZ-981
8 New setting pz:timeout which overrides z3950_operation timeout
13 Re-introduce journal-subpart PAZ-980
14 Both citation and journal-subpart have same value.
18 Add Ubuntu utopic 14.04 and Ubuntu saucy 13.10 to the list of packages
21 Remove from js/pz2.js the code that translates just two diagnostic
22 codes (1 and 2) into human-readable messages. This is no longer
23 necessary at all, as PazPar2 has been returning the message as well as
24 the code in its responses since December 2012. Fixes PAZ-978.
26 primo xsl produces citation rather than journal-subpart PAZ-972
27 (fix in 1.8.1 was incorrect)
29 Fix missing .url files PAZ-977
30 Add missing .url+.cfg and add proper check for it.
34 Turbomarc needs number check for date 903 map PAZ-973
36 primo-pz2.xsl creates citation rather than journal-subpart PAZ-972
38 pz2.js: on successful search, issue a "show" command only if a show
39 callback function has been installed. Fixes part of PAZ-975
43 New command=stop which stops further search&retrieval for active
46 CF/primo XSL generates {volume,issue,pages}-number that are defined
47 in default Pazpar2 service definition. PAZ-968 PAZ-970
51 Various fixes for threaded operation PAZ-960 thru PAZ-966
53 Dump main configuration file during start of Pazpar2 PAZ-957
57 Log ingest failures and record failures:
59 Clean up logging PAZ-942
60 Use session lead in more places. Log hits/records/filtered for each
61 target when fetching is complete. Log hit count when a search
62 (response) is received. Remove some less useful log messages.
64 Fix warning about about missing (internal) stylesheet PAZ-956
66 Report number of filtered records in bytarget PAZ-954
68 Cache XML documents as they appear after normalization. The key
69 for the cache is database+record offset. Cache is removed if query
70 is modified for search. This speeds up re-establishing a result set
71 for databases where new query is same previous query.
75 poll: add assert for number of FDs; abort on poll failure
79 Fix hang for 2nd use of target with unresolved DNS PAZ-949
83 Use poll rather than select for main event loop to overcome limit of
84 1024 file descriptors PAZ-947
86 Override setting with same name (do not append) PAZ-945
90 Separate package: pazpar2-xsl PAZ-932
94 Update OPAC stylesheets for availabilityDate PAZ-943
96 Fix too funny startrecs description
98 Fix broken ~ match if filter contains = PAZ-941
100 --- 1.6.43 2014/06/12
102 Add pz:redis option for ZOOM caching using redis. This is an alternative
103 to memcached. Requires YAZ 5.2.0 to work.
105 --- 1.6.42 2014/05/06
107 Re-ingest native facets PAZ-934
109 Fix medium override for solr-pz2.xsl.
111 --- 1.6.41 2014/04/23
113 Fix URL with leading http:// not working PAZ-933
115 Build packages for Ubuntu Trusty Tahr 14.04 LTS
117 cf.xsl: handle medium override AND md field
119 --- 1.6.40 2014/04/10
121 sortmap relevance always uses increasing=1. PAZ-930
123 solr-pz2.xsl: ensure medium is also applied if Solr record do not
126 Fix leak in getaddrinfo usage for HTTP listener. Not a big deal because
127 this code is only executed once.
129 Fix typos is sortmap documentation.
131 Doc: fix misplaced variablelist tag.
133 --- 1.6.39 2014/02/19
135 Doc: update WRT ICU which is configured for YAZ; not Pazpar2.
136 It's been like that since version 1.1.1 from August 2009.
138 Add support for result-set caching via memcached. See description of
139 pz:memcached. Require that Pazpar2 is using YAZ 5.0.12 or later.
141 Add support for a <metadata> container tag. Patfh from Sven Porst.
143 Add support for XInclude in main configuarion file. Patch from Sven Porst.
145 Add client_{lock,unlock} before call to non_block_events in
146 connection_continue. Problem is that if non_block_events fire records,
147 then we will not unlock client at all. And as a result, a dead-lock
150 --- 1.6.38 2014/01/07
152 Avoid removing slash and other characters for generic metadata that
153 looks like a URL (contains ://). PAZ-915
155 pz2.js: Only use XDR when doing CORS. Favor XHR in all other circumstances.
157 Cluster merge fixes. PAZ-901
159 Set client state to "working" when waiting for DNS resolver. PAZ-907
161 Log when session terminates. PAZ-905
163 Fix LSB/HS conformance for /etc/init.d/pazpar2 PAZ-904
164 Debian: add support for /etc/init.d/pazpar2 status
166 --- 1.6.37 2013/11/18
168 Fix 'Record Missing' for command=show PAZ-903
170 Preserve order of records within cluster PAZ-902
172 testing: use yaz-ztest -V if possible and use YAZ in sibling directory
175 Allow clustered records to be ingested during retrieval. PAZ-900
177 Fix void function error which cannot return a value.
179 --- 1.6.36 2013/11/11
181 SRU pz:authentication="a/b" may end up as x-username=a%2Fb PAZ-899
183 Upgrade to debhelper version 7 PAZ-898
185 Package with libyaz5 PAZ-897
187 Better error message for CCL parse errors PAZ-887
189 --- 1.6.35 2013/09/23
191 New target setting: pz:authentication_mode. PAZ-896
193 pazpar2 PRM: pazpar2 runs as user nobody. PAZ-893
195 Strip #-suffix for ZOOM URLs. PAZ-892
197 --- 1.6.34 2013/08/30
199 Fix issue with hanging show/bytarget. Could happen if new search was
200 fired with filter before show from previous show completed.
202 Add 'date' handling for solr-pz2.xsl.
204 --- 1.6.33 2013/08/13
206 pz:authentication may carry user group password. If pz:authentication
207 has at least one blank in it, that forces Pazpar2 to use idPass Z39.50
208 authentication. Two forms are allowed
211 If no blanks are present, open authentication is in use (as before).
213 Fix {prev,next}recid elements that are returned by record
214 command. These were incorrect in versions 1.6.23 thru 1.6.32.
216 --- 1.6.32 2013/06/14
218 Experimental support for snippets. If argument snippets=1 is given
219 for command record or show, the text will include <match>-tags for
220 those area that matches.
222 Fix memory leak WRT client structures (incorrect inc ref count).
224 Log when a HTTP response is serialized and the time elapsed.
226 New command, service, which returns current service XML in use for
229 --- 1.6.31 2013/05/24
231 Local limit for fields of type "generic" (ie text fields) are
232 now normalized by mergekey rule before comparison takes place.
233 Before, no normalization took place and comparsion was, thus, exact.
235 New facility: dynamic rank , which allows ranking to be changed
236 for an existing result set. The rank can be set for search or
237 show. This overrides the rank for metadata elements (which fields
240 New facility:dynamic mergekey , which allows mergekey to be changed
241 for an existing result set. The mergekey can be set for search or
244 HTTP server: host=@ listens on IPV6 and IPV4 if possible.
245 Also, allow ports to be given as names (not just integers).
246 Fix @-notation on systems that don't support IPV6 sockets.
248 --- 1.6.30 2013/04/26
250 Add MasterKey Connect configuration. Refer to section "non-standard
251 databases" for more information.
253 Allow timeformat for log to be specified with option -m (as Metaproxy
256 New Debian package pazpar2-dbg.
258 Fix NULL ptr reference of se->relevance. If a new search is initiated
259 and a CCL query fails to parse, Pazpar2 could crash.
261 Fix problem with host/port mangling in Pazpar2 HTTP server. Problem
262 existed in 1.6.29 only.
264 --- 1.6.29 2013/04/05
266 IPV6 updates. Allow IPv6 addresses for database hosts and IPv6 address
267 for HTTP server. By default the HTTP server uses IPv4 only, but that can be
268 changed by setting "host" attribute for the "listen" element.
270 Change semantics of pz:extendrecs. Allow for repeated fetches .
271 pz:extendrecs is now the number of extra records to fetch (was total
274 --- 1.6.28 2013/03/21
276 Fix hang or memory violation if show occurred before a search. (not that
277 it makes much sense to perform a show before a search).
279 --- 1.6.27 2013/03/08
281 pz2.js: Element_parseChildNodes concatenates all Text/CDATA nodes, instead
284 --- 1.6.26 2013/03/08
286 New pz:metadata attribute, empty="empty-value" for Pazpar2's
287 internal representation. With this attribute, Pazpar2 treates an empty
288 pz:metadata type as having the value for "empty" - if empty.
290 New setting, pz:extendrecs, which triggers extended fetch of records
291 for a database beyond pz:maxrecs for a show command.
293 Fix warning that was falsely issued for "missing limitmap".
295 Log message for Pazpart start/stop changed. Now using same style as
296 Metaproxy, ie Pazpar2 start SHA1 / Pazpar2 stop .
298 --- 1.6.25 2013/02/07
300 Fix hang of 2nd command=show with esn/syntax given.
302 --- 1.6.24 2013/01/18
304 New merge attribute type: 'first', which takes all metadata fields
305 from first target that returns the particular field.
307 --- 1.6.23 2013/01/02
309 Extend info command with hostname and YAZ SHA1
310 Indent results for both command stat and info.
312 Allow limit on merged content. The new configuration metadata
313 element, limitcluster, configures that a metadata element (name) be used
314 as limit name for search. Applies to the whole service (ie all targets),
315 unlike pz:limitmap which is configured per-target (database).
317 New feature: limitmap local:* matches against all metadata fields.
319 Allow repeated list in limitmap spec . Separated by comma. For
320 example: value="local:title,rpn:@attr 1=4".
322 New element <message> in bytarget response. Holds diagnostic message
323 of code (say 'Unsupported Use Attribute' for Bib-1 114).
325 Improved logging for record ingestion failures.
327 Avoid using struct icu_chain in non-YAZ_HAVE_ICU mode. In the rare case,
328 when YAZ is compiled without ICU support.
330 --- 1.6.22 2012/10/11
332 Requires YAZ-4.2.40 to support native solr support.
334 Fix and improve logic handling whether or not to re-do search on sort
335 order changes. A sort order with change in ascending/descending only
336 whould not trigger a new search, which is required for targets with
337 native sorting capabilities. Each client is now checked if instructions
338 (sortmap) exist for native sorting and only client that does requires
339 it is researched. Other clients is just re-ingesting the records,
340 they already have. The resultset is now cleared if any researching
343 Connection sharing between session has broken since version 1.6.8 with
344 introduction of logic that would minimize searching if pazpar2 could
345 detect this based on same query and limits and partly sort order.
346 This could lead to segementations violations.
348 Added a chapter in the manaul about relevance ranking.
350 --- 1.6.21 2012/09/24
352 Rank tweak: follow=number will increase mult by number if two terms
353 occur next to each other; number-1 if they are one term apart , .. 0
354 if they are number a part (all in order). Default is 0 (following
355 terms has no effect).
357 Rank tweak: lead=k will divide mult by 1 + log2(1+k*l) where k is
358 value given by lead and l is length from beginning of field where
359 term occurs (l=0 for first term, l=1 for second term, ..). Default
362 Rank tweak: length=strategy. length="linear" if mult is to be divided
363 by length (existing, default behavior), length="log" if mult is to be
364 divided by log2(1+length), length="none" if mult is not to be affected
367 --- 1.6.20 2012/09/21
369 Rank algorithm details may be printed as part of show response in
370 element <relevance_info>.. This is only printed if <rank debug="yes"/>
373 Record as returned by show/record command have a minimal indentation
374 which makes things human-readable.
376 New configuration of default sorting criteria (sort-default) in
377 service definition. If no criteria is defined it will be as before
380 Search command now supports sort parameter just as the show command.
381 If no parameter is give, it will use the service sort-default value.
383 --- 1.6.19 2012/09/18
385 Rank algorithm skips strings that gets normalized to empty string.
386 For example, & and ! could map to the empty string. The weight for
387 those terms is now 0 (as if they were not part of the query).
389 Rank algorithm does not use CCL from limitmap; only from the query
390 parameter (user query).
394 --- 1.6.18 2012/09/17
396 Rank algorithm configurable by 'rank' element inside service. So far
397 only, attribute 'cluster' is recognized. If cluster="yes", multiple
398 records inside a cluster boosts higher than single records. This
399 is default behavior and existing behavior. cluster="no" takes the
400 average score of each record in a cluster.
402 --- 1.6.17 2012/09/05
404 Fix bad re-use of connections (connections with changing proxy should not
407 --- 1.6.16 2012/08/22
409 Fix a bug introduce in 1.6.15 around the position sorting. It resetted the
410 resultset and sorting when the sort order is position. However this will be done on
411 every client poll, which will make pazpar2 continuing reset and fetching.
412 It should only be done on FIRST request where the sort order change.
414 Fix an issue on suggestion option: Also disable suggestions on empty string.
416 Clean up in turbo marc stylesheet.
418 Remove the hardcoded size of termlists.
420 --- 1.6.15 2012/06/27
422 New facility: ccldirective may be given in service definition. Allows
423 CCL parsing to be customized a bit, such as defining names of operators
426 New facility: raw record by checksum, rather than offset. The record
427 command optionally takes checksum which identifies certain record from
430 New facility: per field ranking. Rank may be given as M [F N] where
431 M is default rank and N is rank for CCL terms from field F.
433 --- 1.6.14 2012/06/04
435 Fix for IE7/8 in pz.js
437 Applied patch from Giannis Kosmas on keepAlive, which also adds keepAlive to init response.
439 Lower log level some places.
441 Remove some invalid test results.
443 --- 1.6.13 2012/05/23
445 Introducing a version=2 parameter for show, termlist and bytarget commands.
446 This enables pazpar2 to return approximation on hit and count count when
447 doing record filtering using the limit parameter on search and a
448 limitmap with a value of "local:"
450 Setting pz:xslt may embed local XSLT as an alternative to referring
452 Value is not CDATA but XML nodes embedded, so escaping is not necessary
453 but a root element *must* be present. For example:
454 <settings target="target="z3950.indexdata.com/marc">
463 Metadata field rank may given by XML internal document (pz:xslt
464 result). If rank is not given, the rank from service description is
467 Metadata field can now configured a default limitmap and facetmap.
468 Setting limitmap to "local:" would work for all kind of targets, but would
469 prob. not be the optimal solution. But at least better than the default behavior
470 of pazpar2 where no filtering is done.
472 A service definition can now also contains <set/> that defines service-wide
473 settings. These will override server-wide sets and will be overridded by
476 New setting, pz:present_chunk, that specifies number of records to fetch
477 at a time. Zero will disable chunkation; will fetch max_records at once.
479 --- 1.6.12 2012/03/14
481 Revert the format change in termlist response, that could break
482 some clients / UIs since they were expecting an (empty) element
483 if no facet values was found.
485 --- 1.6.11 2012/03/07
487 Revert the behavior of returning errors when unable to block
488 on termlist, bytarget and search, when unable to block due to
489 other block. The client will now receive a regular response,
490 but it will be logged in the server. A parameter (report) is
491 added to change behavior to return error response or WARNING
492 status message. Consider this "API" as private, as it is mostly
493 untested and could be changed in future releases.
495 Fix spell error in pz2.js fix in 1.6.10.
497 New Marc2TurboMarc.xsl (contribution from Sven Porst).
498 Can solve the missing marc21.xsl updates in some cases.
500 tmarc.xsl: Simplify the 6xx to subject-long and fix 1-based
501 substring (contribtion from Sven Porst)
503 marc21.xsl: fix 1-based substring call
505 tmarc.xsl and marc21.xsl: use 856$a as last option for electronic-text.
507 Add test_termlist_block to test suite
509 --- 1.6.10 2012/02/23
511 Fix SEGV for invalid PQFs and SRU/SOLR targets
512 Also refactor a bit the code that converts from PQF to SRU/SOLR queries.
514 Fix pz2.js: "null object" due to change in in bytarget result XML.
516 Fixes in tmarc.xsl: Subject-long shorten for extra commas only.
517 Added this normalization to the other subject-long fields (d6xx),
518 where it was missing.
520 Fixes in marc21.xsl: Updated with most of the new tmarc.xsl.
521 Still differences around medium and holdings. marc21.xsl is not
522 longer active used by Index Data, and should be considered unsupported.
523 Use tmarc.xsl instead.
527 Fix SEGV that could occur for failed connections.
531 Fix bug for command sort that could return no results for active clients
532 (from previous search). This bug was present in 1.6.6-1.6.7.
534 Fix bug in results that could include results that should have been
535 filtered out. This bug was present in 1.6.6-1.6.7.
539 Fix bug introduced in 1.6.6 where a connection re-use could stall
542 Local filtering may now specify a local metadata field, eg.
543 pz:limitmap:somefield[t]=local:otherfield
547 For search, when limit and or filtering is in place and search
548 is identical to previous search, the result set is re-used and the
549 target is not searched.
551 Limits may work perform local filtering as well, by using "local:"
556 Updated bytarget command to contain a suggestions element with misspelled
557 words and suggestions to these. pz2.js has been updated to deliver this
558 onwards as well. Only target that currently delivers this is the solr
559 client in YAZ 4.2.18.
563 New service definition element, xslt, that allows an embedded stylesheet
564 to be defined. This can be referred to from pz:xslt as an alternative to
567 New pz:sortmap:field setting for specifying hints on how to make
568 a target natively sort on a field. This is used for command=show in
569 conjunction with sort.
571 New pz:url setting for specifying the actual URL for a target. When
572 this is used the target ID is not used as URL anymore and the target ID
573 may be almost any string (not including []).
575 command=termlist without name parameter returns all termlists/facets.
576 Previously if name parameter was omitted, only "subject" was returned.
580 Make termlist sorting stable. Terms with same frequency are now sorted by
581 their display name. This makes a pretty display and improves our
582 regression test because qsort is not a stable sort.
584 New sort parameter value 'position'. The 'position' sorts merged records
585 by their original position from the remote target. This is primarily useful
586 for debugging and may be used for targets that already perform some kind
587 of relevance ranking. Note that sort by default is decreasing; so to get
588 records in their original order sort=position:1 must be used.
592 tmarc.xsl: yet another 773$g fix. Was broken in 1.6.1 as well.
594 Facility to change working directory for pazpar2 daemon. Option -wdir
595 sets working directory to dir. This facility is useful if core dumps
596 must be saved. In this case, the current working directory must be
597 writable by the running user, such as "nobody".
601 New configuration element <icu_chain> for <server>/<service> which
602 allows a named ICU rule (chain) to be defined. The names relevance,
603 sort, mergekey and facet are used for those operations. The definition
604 <icu_chain id="sort" locale="en"> .. </icu_chain>
606 <sort> <icu_chain locale="en> ... </icu_chain> </sort>
607 And so on.. for relevance, mergekey and facet as well. The latter
608 style is deprecated. The facet terms are normalized by the facet
609 rule by default. This may be changed on a metadata field basis by
610 defining the new attribute 'facetrule' for the metadata element.
612 <icu_chain id="myrule" locale="en"> ... </icu_chain>
613 <metadata name="author" termlist="yes" facetrule="myrule"/>
615 Preserve rorder for merged metadata. Fixes issue as reported by Sven
616 Porst: http://lists.indexdata.dk/pipermail/yazlist/2011-July/003230.html
618 tmarc.xsl: set journal-subpart to 773$ only.
622 Modify the behavior for the limit parameter (first defined in 1.5.7).
623 Mapping of limit searches are now defined by the new configuration item
624 pz:limitmap. Fix a dead-lock problem with the limit parameter.
626 Extend tmarc.xsl to extract 773$g data (OpenURL).
630 ICU default maps remove backquote (`).
632 Command 'search' takes limit parameter (optional). The limit parameter
633 allows a search to be limited one or more facets and the corresponding
634 values. This is for server side filtering.
636 Configure tweak: Use -lm for log(3) if needed
640 Fix a problem with skiparticle sortkey that could be completely
641 ignored (and reduced to "").
643 Fix dependency problem in pazpar2 RPM package (did not require
644 libyaz4 as it should).
648 Fix memory leak that occurred for command=termlist&name=xtargets .
650 Pazpar2 may save HTTP requests. Enabled by option -R.
654 Experimental support for DTIC DADS target. New dads-pz2.xsl.
656 Support for query_syntax (overrides the default for SRU | Z39.50)
658 Support for extraArgs (ZOOM "extraArgs" option) for targets
660 New commands: status-server and status-session
664 Fix for threaded runs: Client now have a copy of the database URL,
665 which can used after the database has been release from the client.
666 This makes the logging in the connection idle timeout of the client nicer (no NOURL) and should be thread-safe.
668 tmarc.xsl: Add journal-title-abbrev and full text.
670 cf.xsl: new fields: isbn, issn, journaltitle, volume, issue
672 Fix for cmd=record before search.
674 Session Logging clean up.
676 Fix wrong termlist factor when maxrecs is different from 100.
680 Fix missing pz:termlist_term_factor in settings.c messed up pz:preferred.
681 Term factor is default enabled but can be diseabled by
682 pz:termlist_term_factor=0
686 Add scaling of facet count. Currently always enabled, needs fixing.
687 Allow user-defined info for target suffix. This has no meaning in
688 Pazpar2 except to distinguish targets from each other. The suffix
689 data begins with #. For example z3950.indexdata.com/gils#Mydata
691 Added exact-match recordfilter; format name=value
695 SOLR support. Pazpar2 may operate as web service client for SOLR.
699 Fix for show command and block=1 (dead lock). Bug was introduced in
704 New RPM packages: pazpar2, pazpar2-js, pazpar2-doc. These have been
705 tested on CentOS 5.5 only.
709 Fix problem with result sets being removed from a client session
710 if the connection for it was resused by another session. Bug #3489.
712 New iphone UI for Pazpar2 (www/iphone).
716 Fixes for threaded operation.
718 New stylesheets for TurboMARC: tmarc.xsl and opac_turbomarc.xsl.
720 New example services in etc/services in source. In the Debian packages
721 these are located in /etc/pazpar2/services-available
723 Threaded mode operational on Windows. Requires Windows 7 or Windows
726 Default value of setting pz:max_connections is 0 which means that there
727 is no limit on number of connections.
731 Pazpar2 may operate in threaded mode. Enabled by element threads in
732 the configuration. See pazpar2_conf for details.
734 New setting setting: pz:max_connections. Setting pz:max_connections is
735 a limit of number of sockets to a host. When this limit is reached,
736 Pazpar2 will wait up to 5 seconds for a connection to becomes available.
737 The client will be marked Client_Error when it can not be searched
738 (other clients in a session may work). If pz:max_connections is not set
739 for a target, a value of 30 will be used. Note: the pz:max_connections
740 will only work in threaded mode.
742 pz2.js: JSON support for show.
744 Debian package: Enable default service, default.xml, before starting
745 Pazpar2 only if there is no service already in /etc/pazpar2/services-enabled.
749 Debian version depends on on libyaz4. Note that Pazpar2 will still
750 compile from source with YAZ 3.
752 Split services into separate files. The example configuration file
753 pazpar2.cfg.dist now includes a default service default.xml (part of
754 etc). And default.xml includes settings/edu.xml. The default.xml file,
755 not to be confused with settings/defaults.xml, is a template for jsdemo
756 and other services. The Debian package installs /etc/pazpar2/server.xml
757 which is now the main pazpar2 configuration (used to be called pazpar2.cfg).
758 server.xml includes services from /etc/pazpar2/services-enabled/*.xml .
759 The default.xml (from etc) is installed in /etc/pazpar2/services-available
760 and a symlink to it is created from services-enabled. The default.xml
761 service is unnamed and, thus, will be used by jsdemo and test1.
763 New setting pz:negotiation_charset. Patch from Andrei V. Toutoukine. The
764 new setting pz:negotiation_charset specifies character set for Z39.50 Init.
768 Support for additional fields in cf.xsl and pazpar2.conf.dist:
769 publisher, available, due, location (=locallocation), callno
770 (=callnumber), thumburl and score.
772 Describe pz:xslt and the auto setting.
774 Move mergekey definition away from the normalization stylesheets and
775 define a mergekey common for all target types in pazpar2.cfg.
777 Code update: Use the Odr_int type for hit counts. This is part of
778 YAZ 3.0.47 and later and so configure checks for that.
782 Metadata attribute 'skiparticle' also works for ICU based
783 normalization. (was only working for the non-ICU/ASCII before).
785 Command bytarget with argument settings=1 will show settings per
786 target.. This is to be able to verify correct settings and be able to
787 test that they are correct. The database settings array size is now
788 also stored.. Problems with database settings array is that if not
789 careful it will be too small (smaller than dictionary per-service
792 Make record list sorting stable by comparing mergekey for records if
793 relevance/title or other sorting criteria all match. This is merely to
794 ensure that our regressions tests works (reproducible output).
796 Relevance calculation changes: use a different denominator (length) for
797 per-field relevance scoring.. Instead of length of all ranked fields we
798 now use length of individual fields (as if they were individual "free"
799 text fields). This will ensure that documents with a long field with no
800 match (say description) will not "hurt" a title match.
802 Diagnostic member was not set on connection error. Fixed
806 Command search takes two optional parameters, startecs and maxrecs,
807 that specifies the start offset (0, 1, ...) and maximum number of records
808 to fetch for each target.
810 XSLTs + MARC maps are cached within a session so we don't re-parse
811 them over and over again. Even for a session with a single search
812 there's much to be gained because many targets use the same
815 The metadata attribute 'mergekey' now takes one of three values 'no',
816 'required', 'optional' . And the resulting mergekey from metadata
817 is now ordered in the same way as metadata in the service definition.
818 Older Pazpar2 version use the order in which metadata appeared in a
821 The search argument 'filter' now offers a new operator ~ which does a
822 substring match. The = operator works as before: string match for
823 anything but pz:id, or target match for pz:id.
825 New setting pz:recordfilter. The value of this setting takes the
826 form name[~value]. This setting makes Pazpar2 ignore all retrieved
827 records that do not have the metadata element name with value substring
830 Pazpar2 allows YAZ log level to be set (option -v).
834 For WS responses Pazpar2 creates XML header. Exception: raw record.
836 Setting XML files are now stored in etc/settings instead of etc. This
837 reflects the layout with the Debian package layout.
839 Settings may be posted for command=settings. The POSTed settings must
840 have root element 'settings' like regular setting files. In order to be
841 recognized, the POST request must use Content-Type=text/xml.
843 A service may be posted for command=init. This service will be used
844 during the session. The service may have its own target settings,
845 ICU config, timeout, etc. In order to be recognized, the POST request
846 must use Content-Type=text/xml.
848 Timeout values may be given per-service. That's element 'timeout'
849 which takes three attribute values (a subset may be given): 'session',
850 'z3950_operation', 'z3950_session'. Option -T is no longer supported
851 - used to specify session timeout.
853 Option -t tests the Pazpar2 configuration and returns exit code
854 (0=success, non-zero=failure). In previous version of Pazpar2, -t
855 specified local settings.
857 In version 1.2.0 the configuration file - after include processing -
858 was dumped to stdout. Now, the configuration is only dumped to the
859 yaz log file if option -d is given.
863 Configuration may now have multiple server areas. This means that a
864 Pazpar2 instance may listen on multiple ports. Virtual hosting is not
865 yet supported - on a server basis. Configuration may also have multiple
866 services .. That is repeating service elements inside a server. Each
867 has an attribute 'id' which serves as service ID. This ID in turn may
868 be used in a Pazpar2 session, by specifying parameter service=ID for
869 command init. There can be at most one unnamed service inside a server
870 which can be referred to by not specifying an service ID for command
871 init (backwards compatible). In order to partition multiple servers and
872 services a new include directive has been added. This takes an attribute
873 'src' which specifies one or more sub-files. For example to include
874 service files, one might use:
875 <server >.. <include src=/"etc/pazpar2/conf.d/*.xml"/> .. </server>.
876 It is the intention that that completely makes the settings directive
879 Fix problem where the record command would wait forever if there were
880 no targets to wait for (activeclients == 0).
884 One result set is created per session (last search) rather than for
885 each connection which happen to be shared (bug #3009).
887 marc21 stylesheets changed for efficiency.
891 Session timeout may be specified on the command-line as option -T.
893 Pazpar2 may now be operated in a no-merged mode for records.. All records
894 will be considered unique. This mode is enabled if no mergekey is
895 generated by the normalization stylesheet (pz:xslt).
897 Pazpar2 caches original records from each target and the 'record' command
898 with offset returns the original record if 'syntax' and 'esn' are NOT
899 specified. This speeds up retrieval of original records but also means
900 that Pazpar2 uses more memory. The cached records will be freed when the
901 session terminates or a new search is executed.
903 Pazpar2 no longer uses its own ICU wrapper. Instead the ICU wrapper
904 library part of YAZ is used.
906 Added SRU client support.
908 Automatically computes pz:nativesyntax if not provided. Works for XML and
911 --- 1.0.13 2008/11/24
913 Command bytarget returns name of target (if defined).
915 --- 1.0.12 2008/11/04
917 Fixed bug #2021.. location now holds all brief elements.
919 --- 1.0.11 2008/10/15
921 Fixed check for application/x-www-form-urlencoded parameters.
923 --- 1.0.10 2008/10/14
925 Fixes for IE in pz2.js.
927 Fixed bug #2021: non-merged, brief meta data NOT included for command=show.
931 Changed the JS library pz2.js to use POST for long URL (+ params).
933 Added installation instructions for Windows. Note: NT services is
934 NOT available until we make a new release of YAZ.
936 Preserve order of repeated metadata fields (they were reversed before).
938 More MARC21 information extracted for metadata.
942 Fixed bug #1162: HTML entities are not escaped properly.
944 Native Windows port of Pazpar2. Makefile for Visual Studio provided.
948 Marc21 stylesheet updated to reflect multiple full text fields
952 Fixed bug in pz2.js WRT DOMElement attributes on IE.
954 Fixed bug 2100: Database wildcards not working
958 Added support for retrieval of records in binary.
960 Fixed bug 1794: Pazpar2 does not return valid XML.
962 Deal with ICU not returning sortkey (resulted in SEGV before).
966 JavaScript library pzw2.js throws error if WS response (from Pazpar2 or
967 other) is malformed (non-wellformed XML or missing Pazpar2 OK status).
969 Improved diagnostics when Pazpar2 HTTP decoding fails.
971 Pazpar2 requests may be POSTed as using Content-Type
972 application/x-www-form-urlencoded.
974 Pazpar2 honors LF in HTTP headers.
976 Handle targets that handle negative hit counts (should not happen, but it
981 ICU is used for tokenization and normalization of the following: mergekey,
982 sorting, relevance terms.
984 Debian package now enables ICU tokenization and normalization by default.
988 Exposed user setting values (i.e. non-pz: names) to the record systems in two
989 ways: Either as parameters to the normalization stylesheets (which would allow the
990 programmer to postprocess or use the values in any way) or after the normalization
991 step, in which case values are made part of the normalized record (and available for
992 sorting, termlists, display, or other interface-related use.
994 Implemented sorting by year.
996 Option -d dumps records to the current log file instead of stderr.
998 Fixes for compilation on cygwin.
1000 Z39.50 client code uses pz:elements. pz:elements was recognized in
1001 earlier Pazpar2 versions but it was not used for anything.
1003 icu_chain_test is using fgets instead of getline - fixes compilation
1006 Loosen the CCL query parsing so that Pazpar2 only returns error if _all_
1007 query conversions fail (rather than _any_). This means targets that do
1008 not support some fields are ignored in a search.
1010 --- 1.0.1 2007/07/03
1012 Improved handling of socket timeout for Z39.50 connections.
1014 Misc documentation updates and spell fixes.
1016 Debian package pazpar2 creates log rotate entry.
1018 Debian package pazpar2-apache2 reloads Apache2.
1020 jsdemo included in distribution. It illustrates the use of the js/pz2.js
1023 --- 1.0.0 2007/06/20
1025 First public release.