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

XQuery update: multiple document nodes with the same path
by Francis Avila 10 Aug '12

10 Aug '12

I'm having trouble understanding document/collection semantics with xquery update. If a create a document and add another document to the same path, that path becomes a collection with two document nodes. However it seems to be impossible to manipulate these document nodes individually. See the following console output: > create db test Database 'test' created in 137.25 ms. > open test Database 'test' was opened in 2.15 ms. > add to test.xml <root1/> Path "test.xml" added in 19.44 ms. > add to test.xml <root2/> Path "test.xml" added in 2.57 ms. > list test Input Path Type Content-Type Size --------------------------------------- test.xml xml application/xml 2 test.xml xml application/xml 2 2 Resources. > xquery collection('test/test.xml') <root1/><root2/> Query executed in 30.84 ms. > xquery doc('test/test.xml') Stopped at line 1, column 20: [BXDB0006] Database path 'test/test.xml' must point to a single document. > xquery import module namespace functx = "http://www.functx.com"; collection('test/test.xml') ! (document-uri(.), functx:node-kind(.), .) test/test.xml document-node<root1/> test/test.xml document-node<root2/> Query executed in 29.54 ms. > xquery delete node collection('test/test.xml')[2] Query executed in 1.18 ms. > xquery import module namespace functx = "http://www.functx.com"; collection('test/test.xml') ! (document-uri(.), functx:node-kind(.), .) test/test.xml document-node<root1/> test/test.xml document-node<root2/> Query executed in 43.83 ms. > xquery delete node db:open('test','test.xml')[2] Query executed in 2.15 ms. > xquery import module namespace functx = "http://www.functx.com"; collection('test/test.xml') ! (document-uri(.), functx:node-kind(.), .) test/test.xml document-node<root1/> test/test.xml document-node<root2/> Query executed in 36.77 ms. Note that it doesn't seem possible to delete one of the document nodes, even using db:replace() > xquery db:replace('test','test.xml',collection('test/test.xml')[1]) Stopped at line 1, column 60: [BXDB0006] Database path '%' must point to a single document. I don't understand this error message, unless db:replace is internally using the path of the document-node rather than the document-node itself. > delete test.xml 2 resource(s) deleted in 36.35 ms. But I can delete both items at the same path. The way I discovered this is that I accidentally added a document to an existing path using db:add(). However, I was unable to correct the mistake and remove the documents without completely deleting the path and re-adding its members. I don't see any way to do this with a single xquery update PUL. It seems to me that this is a bug. Either xquery update expressions like 'delete node' should work on document nodes, or adding document-nodes to a path which has a document-node should be illegal and raise an error when attempted. -- Francis Avila

2 5

freies Geschenk
by hi 09 Aug '12

09 Aug '12

1 0

Hadoop with QueryProcessor
by Tiago Freitas 07 Aug '12

07 Aug '12

Hi. I'm trying to use the BaseX QueryProcessor with Hadoop to process very large XML files. I use an XmlInputFormat to split the XML files, and process each node with BaseX. I prefer to use QueryProcessor for this and not the server (each individual element is not very large and I don't want to store the input files on a BaseX server). The problem is when I use modules in the query, BaseX can't open the module files in the distributed filesystem. I get "[XQST0059] Could not retrieve module …". I'm passing the query using "new QueryProcessor(query_string, basex_context);", and the bind method for the input. Is there a way to include the module programmatically so the QueryProcessor doesn't try to open the xquery file? Maybe with org.basex.query.util.pkg.ModuleLoader or QueryContext.module but I din't figure how to do it yet. Regards

2 1

Clustering/replicating/high availability
by Philipp Kamps 04 Aug '12

04 Aug '12

Hi, I'd like to pick up that topic about clustering and high availability. I found a message from October 2011, you mentioned that there are a couple students working on that topic. Any updates? Maybe they could describe the approaches? That would be awesome. Have a good weekend. Philipp

2 1

Hallo
by hi 04 Aug '12

04 Aug '12

1 0

Redo log
by Henrique Oliveira 03 Aug '12

03 Aug '12

Hi there, Does Basex use redo log to reduce the risk of a system failure occurring while updating the metadata store? regards -- Henrique Oliveira

2 1

What is the easiest approach to edit data in a browser and have that data stored as xml in basex
by Alex Muir 01 Aug '12

01 Aug '12

Hi, So I've got a client interested to create a dataset of music information such as artist, track name, album release year... and store that in XML. Non technical users would need to view and edit the music info in a web interface and the data would be stored in XML. I was looking at Typo3 which seems to be able to work with basex. I wonder what is the easiest approach to edit data in a browser and have that data stored as xml in basex? What are CMS solutions that work with basex? Is typo3 a good solution? Regards -- - Alex G. Muir Software Engineering Consultant Linkedin Profile : http://ca.linkedin.com/pub/alex-muir/36/ab7/125

4 4

How to update DB with new XML Files??
by John Best 01 Aug '12

01 Aug '12

Hi there, I am have stored one large document in BaseX DB. Now I want to add more to the same DB. Is there any simple way to do the same? Say, I had a file country.xml given below <Countries> <Country>A</Country> <Country>B</Country> <Country>C</Country> <Country>D</Country> </Countries> I stored this in BaseX DB. Now I have another file with same format <Countries> <Country>E</Country> <Country>F</Country> <Country>G</Country> <Country>H</Country> </Countries> I want this file to update the previous DB.... so that the DB will have in all 8 countries. Is there any simple way to do this... -- Have a nice day JBest

2 1

Issue with expath HTTP request and response
by Seenivasan Gunabalan 01 Aug '12

01 Aug '12

Hi all, I want to share my experience with HTTP module while posting to restxq. Consider this example, -------- import module namespace http='http://expath.org/ns/http-client'; let $req := <http:request method="GET" href="http://localhost:8984/restxq/example" status-only = "false" username = "admin" password = "admin" send-authorization="true"> </http:request> let $resp := http:send-request($req) return $resp ---------- run the same query with status-only="false" and status-only="true". the first one(status-only="false") gives exception while the second one(status-only="true") returns proper response header. This is making difficult to handle the response of some requests which MAY contain body data. suppose we want to return the content of the document, if exists, as response body or a standard 410 status. The request should obviously contain status-only = "false". we also tried handling response with the one below, but the issue does exists. <rest:response> <http:response status="410" reason="Gone"> </http:response> </rest:response> I have tried with other http client like Firefox Poster plug-in and it is returning expected response. Is there anything wrong in the approach? Thanks, Seenivasan

2 7

HTML got changes after uploading into BaseX
by Seenivasan Gunabalan 31 Jul '12

31 Jul '12

Hi All, Sorry for flooding the basex mailing list :( . We have this very simple HTML for some testing purposes. And we upload it through basexgui (HTML file type). <html> <body> <h1>Hello world!</h1> <ul> <li>service</li> <li>advancedassociationquery</li> <ul> <li>service2</li> <li>advancedassociationquery2</li> </ul> </li> </body> </html> but when serve it to the browser we get back the following html response where the list is flattened. <html> <body> <ul> <li>service</li> <li>advancedassociationquery</li> </ul> <ul> <li>service2</li> <li>advancedassociationquery2</li> </ul> </body> </html> After validating we've seen that, according to W3, the STRICT HTML compliant form is the following: <html xmlns="http://www.w3.org/1999/xhtml"> <head><title>A title</title></head> <body> <ul> <li>service</li> <li>advancedassociationquery</li> <li> <ul> <li>service2</li> <li>advancedassociationquery2</li> </ul> </li> </body> </html> But even after uploading this STRICT HTML through basexgui we got back the flattened version. Our need is to just store and render the HTML with out any further processing. Also the HTML is generated using a tool. The thing is that the STRICT HTML file is served correctly when uploaded through the REST interface of BaseX (Firefox Poster with PUT method) whereas the incorrect file is served the wrong way(flattened) even if it uploaded through the REST interface. This sums up to an inconsistent behavior. Thanks and Regards, Marco & Seenivasan

3 3