June 2012 - BaseX-Talk - mailman.uni-konstanz.de

Getting node by index.
by Alex Markin 08 Jun '12

08 Jun '12

Hi all. I've got another question about attribute indexing. The function index:attributes() returns strings like ... <entry count="1">id152429</entry> ... but how can I get objects in my data base that contain these indexes? Markin Alex.

2 11

CHECK Cmd
by Henrique Oliveira 08 Jun '12

08 Jun '12

Hi guys, I am trying to execute the command Check, I am getting the follow exception: java.io.IOException: Database 'TRADE' was not found. Shouldn't it create a new one DB if it does not exist? Regards -- Henrique Oliveira

2 2

query execution
by Radim Bača 08 Jun '12

08 Jun '12

Hello, I'm interested about BaseX technical details and its query exectution techniques. BaseX uses structural joins such as XPath accelerator as a main tree-join for a location step execution? In other words, do you use some other type joins a as well (i.e. holistic joins such as TwigStack)? I investigated the execution plans however I was not able to find any information about indexes used by each operation. Can I observe it from the plan like this: <QueryPlan> <GFLWOR> <For var="$b"> <IterPath> <DBNode name="xmark" pre="0"/> <IterStep axis="child" test="*:site"/> <IterStep axis="child" test="*:regions"/> <IterStep axis="child" test="*"/> <IterStep axis="child" test="*:item"/> </IterPath> </For> <Let var="$k"> <AxisPath> <VarRef> <Var name="$b" id="0"/> </VarRef> <IterStep axis="child" test="name"/> <IterStep axis="child" test="text()"/> </AxisPath> </Let> <Order> <OrderByExpr dir="ascending" empty="greatest"> <FNSimple name="zero-or-one(item)"> <AxisPath> <VarRef> <Var name="$b" id="0"/> </VarRef> <IterStep axis="child" test="location"/> </AxisPath> </FNSimple> </OrderByExpr> </Order> <Return> <CElem> <QNm value="item" type="xs:QName"/> <CAttr> <QNm value="name" type="xs:QName"/> <VarRef> <Var name="$k" id="1"/> </VarRef> </CAttr> <AxisPath> <VarRef> <Var name="$b" id="0"/> </VarRef> <IterStep axis="child" test="location"/> <IterStep axis="child" test="text()"/> </AxisPath> </CElem> </Return> </GFLWOR> </QueryPlan> XQuery: for $b in /site/regions//item let $k := $b/name/text() order by zero-or-one($b/location) ascending empty greatest return <item name="{$k}">{$b/location/text()}</item> Thank you for any response, Radim

2 1

Help for parsing http:send-request response
by Philippe Rathé 07 Jun '12

07 Jun '12

Hello folks, I'm using the HTTP Module[1] for sending request. The response has he http:response element which is XML. Then at the end of that element is the body of the response. If your ask for text/plain then you will have the text just after the closing </http:response>. <http:response status="200" message="OK"> ... </http:response>Hello World! In my case the media-type of the response is XML. I have something like the following: <http:response status="200" message="OK"> ... </http:response><service> <hello/> </service> How am I suppose to parse this output? That is the EXPpath spec, I have to deal with that. But parsing it with BaseX give the following error: No elements allowed after closed root element. True, that is not valid XML document. What is the best solution to parse separately the http:response and the body? Thanks, -- Philippe [1]: http://docs.basex.org/wiki/HTTP_Module

2 7

Using BaseX to query proprietary DB
by Brock, David 07 Jun '12

07 Jun '12

Wanting to develop and XQuery interface to proprietary DB and was hoping to use BaseX core for implementation. Is there a well-defined API for implementing such a thing, or would it be more a brute-force source-code sort of effort? Licensing implications? Thanks!

2 1

Cycle detected: a cycle exists in the type hierarchy between Prop and AProp
by Brock, David 06 Jun '12

06 Jun '12

When I load the baseX project in Eclipse 3.6 I get the following issues in org.basex.core.Prop.java which creates errors in all derived classes Cycle detected: a cycle exists in the type hierarchy between Prop and AProp The hierarchy of the type AProp is inconsistent

3 6

how to cache a pre-compiled query?
by Mauricio S. Castro 06 Jun '12

06 Jun '12

I am trying to use QueryProcessor to compile and cache for later, in case no used modules have changed, is this ok? Can I concurrently use QueryProcessor within the same JVM in different threads? Thanks. -- Maurício Santiago de Castro mauricioscastro(a)hotmail.com <mailto:mauricioscastro@hotmail.com><http://br.linkedin.com/pub/mauricio-castro/17/275/a5b><http://www.delicious.com/mscastro><http://twitter.com/#%21/mauricioscastro>mauriciosantiagodecastro@facebook <http://facebook.com/mauriciosantiagodecastro><https://mscastro.startssl.com/>

2 1

BaseX Standalone -i flag may need clarification
by Philippe Rathé 06 Jun '12

06 Jun '12

Hello there, I'm using the CLI to load a xml file from my file system: java -cp basex.jar org.basex.BaseX -i myfile.xml myscript.xqy The first time I run the command, everything is fine, my file is loaded and my xquery script is executed. But the second time I run it, whenever I change the content of myfile.xml, it is not taken in consideration. I realized that it stores my file in the database the first time and after that it loads it from the database. I was confused because it worked the first time. Could it refresh the content of the database whenever a file exists? And fallback to open the file from database when the specified file does not exist? Whatever, maybe just a clarification about the behavior in the doc could prevent bad usage. Thanks.

2 6

Re: [basex-talk] Nothing gets printed after insert node
by Anupam Bakshi 06 Jun '12

06 Jun '12

Hello Maurice, Christian, I'm grateful to you for providing me help. This is exactly what I needed. Thanks a lot. Now I can move forward with the application development. Best regards, Anupam ________________________________ From: M. van Keulen <m.vankeulen(a)utwente.nl> To: Anupam Bakshi <bakshia(a)yahoo.com> Sent: Wednesday, June 6, 2012 3:37 AM Subject: Re: [basex-talk] Nothing gets printed after insert node Dear Anupam, You seem to misunderstand the term "update". An update modifies the document in the database. So if you ask for it after that (with doc(...)) you'll get the modified document. In your case: let $s := db:open("BaseX","sample1.xml") >let $r := db:open("BaseX","result1.xml") >for $m in $s//*:M >return > insert nodes $r//*:R[@id=$m/@id] into $m > modifies the sample1.xml in the database. A subsequent query db:open("BaseX","sample1.xml") would produce > <M id="1"> > <R id="1"/> > </M> > <M id="2"> > <R id="2"/> > </M> > <M id="3"/> > > If you don't want to store the document in the database and modify it, but simply want to generate a *new* intermediary document that is similar to sample1.xml but which includes the desired R-elements, then you don't update anything (the original file stays the same). You can do this with { >let $s := db:open("BaseX","sample1.xml") >let $r := db:open("BaseX","result1.xml") >for $m in $s//*:M >let $o := $r//*:R[@id=$m/@id] >return <M>{$m/@*,$m/*,$o}</M> >} > or by using the transform-expression let $s := db:open("BaseX","sample1.xml") >let $r := db:open("BaseX","result1.xml") >return >copy $new := $s >modify ( > for $m in $new//*:M > return > insert nodes $r//*:R[@id=$m/@id] into $m >) >return $new > Maurice On 06-06-12 08:12, Anupam Bakshi wrote: Hello Maurice, > Thanks for your response. > How do I communicate the intermediate result (modified document) between the two queries? > > Do I store the intermediate file back in the database? I tried to store the result of an update in a variable but it said that not allowed - "Let expression: No updating expression allowed". > So the question is how can I run the first query to update the document and the second query to retrieve the modified document? > > > If someone can point to an example, that will help too. >Best regards, >Anupam > > > > > >________________________________ > From: M. van Keulen <m.vankeulen(a)utwente.nl> >To: Anupam Bakshi <bakshia(a)yahoo.com> >Cc: basex-talk(a)mailman.uni-konstanz.de >Sent: Wednesday, June 6, 2012 1:38 AM >Subject: Re: [basex-talk] Nothing gets printed after insert node > > >Dear Anupam, > >One workaround would be to do it in two queries: one to update your document, and one to retrieve the modified nodes. > >The problem is not that you can't pick nodes from another resource, but that XQuery doesn't allow you to update and query inside the same query. In other words, within one query doing an update means you can't produce a (query) result, and having a query result means you can't do any update. > >Regards, >Maurice. > >On 06-06-12 05:53, Anupam Bakshi wrote: >Then how about option 'b' ? >>At this point any way forward, any work around will do. >>Please let me know if anyone has any suggestion. >> >>What I don't understand is, if I can add nodes why can't I pick those nodes from another resource? Why is this not possible? >> >>But I'm looking for some workaround. >> >>Thanks for your help. >>Anupam >> >> >> >>________________________________ >> From: Christian Grün <christian.gruen(a)gmail.com>; >>To: Anupam Bakshi <bakshia(a)yahoo.com>; >>Cc: <basex-talk(a)mailman.uni-konstanz.de>; >>Subject: Re: [basex-talk] Nothing gets printed after insert node >>Sent: Wed, Jun 6, 2012 12:24:18 AM >> >> >>> I'm interested in 'a'. >>> The two files are in basex, I need to write a query that will return the >>> modified nodes at run time. >> >>Due to the nature of the XQuery Update Specification, that's generally >>not possible. Sorry for that. >> >> >>> From: Christian Grün <christian.gruen(a)gmail.com>; >>> To: Anupam Bakshi <bakshia(a)yahoo.com>; >>> Cc: basex-talk(a)mailman.uni-konstanz.de >>> <basex-talk(a)mailman.uni-konstanz.de>; >>> Subject: Re: [basex-talk] Nothing gets printed after insert node >>> Sent: Tue, Jun 5, 2012 9:09:06 PM >>> >>> Are you mostly interested in >>> >>> a) updating your data, and returning the modified node(s) as query result, >>> or >>> b) finding the correct updating query to modify your data? >>> ___________________________ >>> >>> > Here is what I actually want to do. Maybe someone can suggest a >>> > straightforward way. >>> > >>> > File : sample1.xml >>> > >>> > <M id="1"/> >>> > <M id="2"/> >>> > <M id="3"/> >>> > >>> > >>> > File: result1.xml >>> > <Q> >>> > <R id="1"/> >>> > <R id="2"/> >>> > </Q> >>> > >>> > Desired output : >>> > >>> > <M id="1"> >>> > <R id="1"/> >>> > </M> >>> > <M id="2"> >>> > <R id="2"/> >>> > </M> >>> > <M id="3"/> >>> > >>> > >>> > So insert the node from 2nd file into 1st where the ids match. >>> > >>> > Thanks, >>> > Amupam >>> > >>> > ________________________________ >>> > From: Christian Grün <christian.gruen(a)gmail.com> >>> > To: Anupam Bakshi <bakshia(a)yahoo.com> >>> > Cc: basex-talk(a)mailman.uni-konstanz.de >>> > Sent: Tuesday, June 5, 2012 12:02 PM >>> > Subject: Re: [basex-talk] Nothing gets printed after insert node >>> > >>> >> Unfortunately I did not see how I can return the value into a variable >>> >> for >>> >> further processing. >>> >> Is there some other way to do it? >>> > >>> > Did you already try db:output()? >>> > >>> > >> > >-- ---------------------------------------------------------------------- Dr.Ir. M. van Keulen - Associate Professor, Data Management Technology Univ. of Twente, Dept of EEMCS, POBox 217, 7500 AE Enschede, Netherlands Email: m.vankeulen(a)utwente.nl, Phone: +31 534893688, Fax: +31 534892927 Room: ZI 3039, WWW: http://www.cs.utwente.nl/~keulen > > -- ---------------------------------------------------------------------- Dr.Ir. M. van Keulen - Associate Professor, Data Management Technology Univ. of Twente, Dept of EEMCS, POBox 217, 7500 AE Enschede, Netherlands Email: m.vankeulen(a)utwente.nl, Phone: +31 534893688, Fax: +31 534892927 Room: ZI 3039, WWW: http://www.cs.utwente.nl/~keulen

1 0

Re: [basex-talk] Nothing gets printed after insert node
by M. van Keulen 06 Jun '12

06 Jun '12

Forgot to send my response to the list: -------- Original Message -------- Subject: Re: [basex-talk] Nothing gets printed after insert node Date: Wed, 06 Jun 2012 09:37:25 +0200 From: M. van Keulen <m.vankeulen(a)utwente.nl> Organization: University of Twente To: Anupam Bakshi <bakshia(a)yahoo.com> Dear Anupam, You seem to misunderstand the term "update". An update modifies the document in the database. So if you ask for it after that (with doc(...)) you'll get the modified document. In your case: let $s := db:open("BaseX","sample1.xml") let $r := db:open("BaseX","result1.xml") for $m in $s//*:M return insert nodes $r//*:R[@id=$m/@id] into $m modifies the sample1.xml in the database. A subsequent query db:open("BaseX","sample1.xml") would produce <M id="1"> <R id="1"/> </M> <M id="2"> <R id="2"/> </M> <M id="3"/> If you don't want to store the document in the database and modify it, but simply want to generate a *new* intermediary document that is similar to sample1.xml but which includes the desired R-elements, then you don't update anything (the original file stays the same). You can do this with { let $s := db:open("BaseX","sample1.xml") let $r := db:open("BaseX","result1.xml") for $m in $s//*:M let $o := $r//*:R[@id=$m/@id] return <M>{$m/@*,$m/*,$o}</M> } or by using the transform-expression let $s := db:open("BaseX","sample1.xml") let $r := db:open("BaseX","result1.xml") return copy $new := $s modify ( for $m in $new//*:M return insert nodes $r//*:R[@id=$m/@id] into $m ) return $new Maurice On 06-06-12 08:12, Anupam Bakshi wrote: > Hello Maurice, > Thanks for your response. > How do I communicate the intermediate result (modified document) > between the two queries? > Do I store the intermediate file back in the database? I tried to > store the result of an update in a variable but it said that not > allowed - "Let expression: No updating expression allowed". > So the question is how can I run the first query to update the > document and the second query to retrieve the modified document? > > If someone can point to an example, that will help too. > Best regards, > Anupam > > > ------------------------------------------------------------------------ > *From:* M. van Keulen <m.vankeulen(a)utwente.nl> > *To:* Anupam Bakshi <bakshia(a)yahoo.com> > *Cc:* basex-talk(a)mailman.uni-konstanz.de > *Sent:* Wednesday, June 6, 2012 1:38 AM > *Subject:* Re: [basex-talk] Nothing gets printed after insert node > > Dear Anupam, > > One workaround would be to do it in two queries: one to update your > document, and one to retrieve the modified nodes. > > The problem is not that you can't pick nodes from another resource, > but that XQuery doesn't allow you to update and query inside the same > query. In other words, within one query doing an update means you > can't produce a (query) result, and having a query result means you > can't do any update. > > Regards, > Maurice. > > On 06-06-12 05:53, Anupam Bakshi wrote: >> Then how about option 'b' ? >> At this point any way forward, any work around will do. >> Please let me know if anyone has any suggestion. >> >> What I don't understand is, if I can add nodes why can't I pick those >> nodes from another resource? Why is this not possible? >> >> But I'm looking for some workaround. >> >> Thanks for your help. >> Anupam >> >> >> ------------------------------------------------------------------------ >> *From: * Christian Grün <christian.gruen(a)gmail.com> >> <mailto:christian.gruen@gmail.com>; >> *To: * Anupam Bakshi <bakshia(a)yahoo.com> <mailto:bakshia@yahoo.com>; >> *Cc: * <basex-talk(a)mailman.uni-konstanz.de> >> <mailto:basex-talk@mailman.uni-konstanz.de>; >> *Subject: * Re: [basex-talk] Nothing gets printed after insert node >> *Sent: * Wed, Jun 6, 2012 12:24:18 AM >> >> > I'm interested in 'a'. >> > The two files are in basex, I need to write a query that will >> return the >> > modified nodes at run time. >> >> Due to the nature of the XQuery Update Specification, that's generally >> not possible. Sorry for that. >> >> >> > From: Christian Grün <christian.gruen(a)gmail.com>; >> > To: Anupam Bakshi <bakshia(a)yahoo.com>; >> > Cc: basex-talk(a)mailman.uni-konstanz.de >> > <basex-talk(a)mailman.uni-konstanz.de>; >> > Subject: Re: [basex-talk] Nothing gets printed after insert node >> > Sent: Tue, Jun 5, 2012 9:09:06 PM >> > >> > Are you mostly interested in >> > >> > a) updating your data, and returning the modified node(s) as query >> result, >> > or >> > b) finding the correct updating query to modify your data? >> > ___________________________ >> > >> > > Here is what I actually want to do. Maybe someone can suggest a >> > > straightforward way. >> > > >> > > File : sample1.xml >> > > >> > > <M id="1"/> >> > > <M id="2"/> >> > > <M id="3"/> >> > > >> > > >> > > File: result1.xml >> > > <Q> >> > > <R id="1"/> >> > > <R id="2"/> >> > > </Q> >> > > >> > > Desired output : >> > > >> > > <M id="1"> >> > > <R id="1"/> >> > > </M> >> > > <M id="2"> >> > > <R id="2"/> >> > > </M> >> > > <M id="3"/> >> > > >> > > >> > > So insert the node from 2nd file into 1st where the ids match. >> > > >> > > Thanks, >> > > Amupam >> > > >> > > ________________________________ >> > > From: Christian Grün <christian.gruen(a)gmail.com> >> > > To: Anupam Bakshi <bakshia(a)yahoo.com> >> > > Cc: basex-talk(a)mailman.uni-konstanz.de >> > > Sent: Tuesday, June 5, 2012 12:02 PM >> > > Subject: Re: [basex-talk] Nothing gets printed after insert node >> > > >> > >> Unfortunately I did not see how I can return the value into a >> variable >> > >> for >> > >> further processing. >> > >> Is there some other way to do it? >> > > >> > > Did you already try db:output()? >> > > >> > > >> > > -- > ---------------------------------------------------------------------- > Dr.Ir. M. van Keulen - Associate Professor, Data Management Technology > Univ. of Twente, Dept of EEMCS, POBox 217, 7500 AE Enschede, Netherlands > Email:m.vankeulen@utwente.nl <mailto:m.vankeulen@utwente.nl>, Phone: +31 534893688, Fax: +31 534892927 > Room: ZI 3039, WWW:http://www.cs.utwente.nl/~keulen > > -- ---------------------------------------------------------------------- Dr.Ir. M. van Keulen - Associate Professor, Data Management Technology Univ. of Twente, Dept of EEMCS, POBox 217, 7500 AE Enschede, Netherlands Email:m.vankeulen@utwente.nl, Phone: +31 534893688, Fax: +31 534892927 Room: ZI 3039, WWW:http://www.cs.utwente.nl/~keulen -- ---------------------------------------------------------------------- Dr.Ir. M. van Keulen - Associate Professor, Data Management Technology Univ. of Twente, Dept of EEMCS, POBox 217, 7500 AE Enschede, Netherlands Email: m.vankeulen(a)utwente.nl, Phone: +31 534893688, Fax: +31 534892927 Room: ZI 3039, WWW: http://www.cs.utwente.nl/~keulen

1 0