- BaseX-Talk - mailman.uni-konstanz.de

date field ordering
by Hariprasad 20 Sep '12

20 Sep '12

Dear Basex, Could you please help us in writing xquery for ordering xml based on date field(bDate).For example I have a xml with following nodes <Results> <GROUPBYVAL> <bDate>01-JUN-12</bDate> </GROUPBYVAL> <GROUPBYVAL> <bDate>10-AUG-12</bDate> </GROUPBYVAL> <GROUPBYVAL> <bDate>21-FEB-12</bDate> </GROUPBYVAL> </Results> query should take input as the above xml and should return output in the below format <Results> <GROUPBYVAL> <bDate>10-AUG-12</bDate> </GROUPBYVAL> <GROUPBYVAL> <bDate>21-JUL-12</bDate> </GROUPBYVAL> <GROUPBYVAL> <bDate>01-FEB-12</bDate> </GROUPBYVAL> </Results> Thanks & Regards Hariprasad.d

2 1

Indexes update
by basex＠phpexpert.net 20 Sep '12

20 Sep '12

Is it necessary to update indexes and fulltext on every insert for the complete Database, or is there a flag to update changes automaticly Regards Marcus

2 1

(no subject)
by John Best 20 Sep '12

20 Sep '12

Hi there, Below is my XML Structure – <Docs> <Doc> <DocTitle C=”1”></DocTitle> <MainDesc></MainDesc> <FullDesc></FullDesc> </Doc> <Doc> <DocTitle C=”2”></DocTitle> <MainDesc></MainDesc> <FullDesc></FullDesc> </Doc> …………………. …………………. …………………. </Docs> There are at least 6000 Doc (more to be added soon). I have created a FT Index of the same. To search a particular word, which might be either in <MainDesc> or in <FullDesc>, below are my queries – Query No -1 (Time taken is 21-22 Secs) data(doc('sample')/Docs/Doc[MainDesc[. contains text {'unfavorable'}] or FullDesc[. contains text {'unfavorable'}]]/DocTitle/@C) Query No -2 (Time taken is 25-26 Secs) for $x in doc('sample')/Docs/Doc where $x[MainDesc[. contains text {'unfavorable'}] or FullDesc[. contains text {'unfavorable'}]] return $x/DocTitle/@C The time taken is too much. How can I lower this…? I tried ft:search(), but it doesn’t fits in here as ft:search wont return Nodes. * * Have a nice day Regards, JBest

2 1

How to lower the Full-Text search time??
by John Best 20 Sep '12

20 Sep '12

Hi there, Below is my XML Structure – <Docs> <Doc> <DocTitle C=”1”></DocTitle> <MainDesc></MainDesc> <FullDesc></FullDesc> </Doc> <Doc> <DocTitle C=”2”></DocTitle> <MainDesc></MainDesc> <FullDesc></FullDesc> </Doc> …………………. …………………. …………………. </Docs> There are at least 6000 Doc (more to be added soon). I have created a FT Index of the same. To search a particular word, which might be either in <MainDesc> or in <FullDesc>, below are my queries – Query No -1 (Time taken is 21-22 Secs) data(doc('sample')/Docs/Doc[MainDesc[. contains text {'unfavorable'}] or FullDesc[. contains text {'unfavorable'}]]/DocTitle/@C) Query No -2 (Time taken is 25-26 Secs) for $x in doc('sample')/Docs/Doc where $x[MainDesc[. contains text {'unfavorable'}] or FullDesc[. contains text {'unfavorable'}]] return $x/DocTitle/@C The time taken is too much. How can I lower this…? I tried ft:search(), but it doesn’t fits in here as ft:search wont return Nodes. * * Have a nice day Regards, JBest

1 0

restxq..Does the extension .xqm is mandatory?
by Seenivasan Gunabalan 20 Sep '12

20 Sep '12

Hi, Was .xqm extension for the xquery module that handles RESTXQ calls are mandatory? We used to create a separate file("customrestxq.xq") to handle the restxq calls, which was imported by the main "restxq.xqm" module. It was working fine till the 7.4 beta version, but at the latest developer release, Basex7.5 (BaseX75-20120918.130850), the calls were not mapped and the status " No function found that matches the request." was returned. we found that by changing the extension from .xq to .xqm i.e. to "customrestxq.xqm" fixed the issue. was .xqm extension is enforced? Best regards, Seenivasan

2 2

Re: [basex-talk] Fulltext search
by Jens Erat 19 Sep '12

19 Sep '12

Hi Marcus, this is because `ft:search` can scan the index directly, `contains text` needs to construct some objects before. `ft:search` doesn't return strings but text nodes, so you can use a parent axis step to get the containing node: ft:search($x, 'double')/parent::node() Kind regards from Lake Constance, Germany, Jens -- Jens Erat [phone]: tel:+49-151-56961126 [mail]: mailto:email@jenserat.de [jabber]: xmpp:jabber@jenserat.de [web]: http://www.jenserat.de PGP: 350E D9B6 9ADC 2DED F5F2 8549 CBC2 613C D745 722B Am 19.09.2012 um 10:22 schrieb <basex(a)phpexpert.net>: > My searches are > > <no>{let $quasi := (for $x in /record where ft:search($x, 'double') and ($x/item[@name="STATUS"]/value/string() = "edited") return <quasi>{data($x/@id)}<content>{$x//value[text() contains text 'double']}</content></quasi>) return $quasi }</no> > > > <no>{let $quasi := (for $x in /record where ($x//value[text() contains text 'double']) and ($x/item[@name="STATUS"]/value/string() = "edited") return <quasi>{data($x/@id)}<content>{$x//value[text() contains text 'double']}</content></quasi>) return $quasi }</no> > > > These searches take a lot of time (about 300 seconds) cause of nearly 600.000 objects (about 4 GB data) to search through > > A Fulltext search wich only delivers the textnodes only takes 2 seconds to deliver text results > > > > On Wed, 19 Sep 2012 09:55:50 +0200, Jens Erat <jens.erat(a)uni-konstanz.de> wrote: >> Hi Marcus, >> >> `ft:search(…)` is built for returning text nodes, have a look at the >> [documentation]. If you want to get the nodes, you should use the more >> general [full text expressions]. >> >> If you need help transforming your query, would you please paste your >> query (or better: the full text search relevant part of it)? >> >> Regards, >> Jens >> >> [documentation]: http://docs.basex.org/wiki/Full-Text_Module#ft:search >> [full text expressions]: http://docs.basex.org/wiki/Full-Text >> >> -- >> Jens Erat >> >> [phone]: tel:+49-151-56961126 >> [mail]: mailto:email@jenserat.de >> [jabber]: xmpp:jabber@jenserat.de >> [web]: http://www.jenserat.de >> >> PGP: 350E D9B6 9ADC 2DED F5F2 8549 CBC2 613C D745 722B >> >> >> >> Am 19.09.2012 um 09:33 schrieb <basex(a)phpexpert.net>: >> >>> Hi @all, >>> >>> is there a way if i use ft:search on the database to get the nodes of results back instead of just the results? >>> >>> >>> Regards >>> >>> >>> Marcus >>> >>> _______________________________________________ >>> BaseX-Talk mailing list >>> BaseX-Talk(a)mailman.uni-konstanz.de >>> https://mailman.uni-konstanz.de/mailman/listinfo/basex-talk >

1 0

Fulltext search
by basex＠phpexpert.net 19 Sep '12

19 Sep '12

Hi @all, is there a way if i use ft:search on the database to get the nodes of results back instead of just the results? Regards Marcus

2 1

BaseX 7.3: unexpected Stack Overflow
by Torsten Grust 18 Sep '12

18 Sep '12

Dear all, I have been playing with the following toy implementation of a super-simple associative map (it's mainly an exercise in demonstrating the use of h-o functions in XQuery 3.0): > declare namespace map = "http://db.inf.uni-tuebingen.de/map"; > > declare function map:entry($map, $k, $v) { > function($x) { if ($x eq $k) then $v else $map($x) } > }; > > declare function map:empty() > { > function($x) { () } > }; > > > let $map := map:empty(), > $map1 := map:entry($map, 1, ("one", "eins", "uno")), > $map2 := map:entry($map1, 2, ("two", "zwei", "due")) > return $map2(1) The execution of this script with BaseX 7.3 unexpectedly yields Stack Overflow: circular variable declaration? Note that replacing $map2(1) by $map1(1) in the last line returns the expected result ("one", "eins", "uno"). I'd be grateful for any comment/insight. Cheers and keep up the good work, --Torsten -- | Prof. Dr. Torsten Grust | Database Systems — Universität Tübingen (Germany) | torsten.grust(a)uni-tuebingen.de | db.inf.uni-tuebingen.de

2 2

REST "directory" listing
by basex＠phpexpert.net 18 Sep '12

18 Sep '12

Hi @all, since BaseX 7.3 it seems there ist no more resource listing of sub paths of a database http://localhost:8984/rest/factbook/tmp [1] as example won't work There will be an error like Name 'factbook/tmp' is invalid. Listing in http://localhost:8984/rest/factbook/ [2] shows the path and it's contents factbook.xml tmp/factbook.xml in BaseX7.2 the listing of paths was possible Regards Marcus Links: ------ [1] http://localhost:8984/rest/factbook/tmp [2] http://localhost:8984/rest/factbook/tmp

2 2

Is BaseX suitable for storing XML files that follow different schemas ?
by Tom Bailly-Salins 14 Sep '12

14 Sep '12

Hello BaseX list, After looking at the BaseX website and XML databases in general I would like to know if BaseX would be suitable for the following application. I would like to store a relatively large amount of XML files (couple of hundreds) that contain metadata information that is compliant with various standard metadata schemas (Dublincore, ISO, INSPIRE, etc...). Once stored in one (or several - 1 per schema type?) BaseX database(s) I would like to be able to: - search these XML files accross schemas and namespaces to perform general keyword searches - search for specific information within a known schema and namespace - return query results in an HTML format to display directly on a web page I would like to know if BaseX would be suitable for such application and how flexible BaseX is in storing XML files compliant with different schemas and containing several namespaces. If suitable, what would be the best practice for storing the different XML files? Should all the XML files that follow a same schema be stored in the same database (for faster indexing, easier searching?) or I can I use a single database to store all XML files regardless of the fact that they follow different XSD schemas ? Thanks a lot for any input Tom

2 1