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

can't make Geo Module work
by Eric Boisvert 03 Jan '16

03 Jan '16

Greetings I just downloaded BaseX 8.3.1 .war and installed it on Tomcat 7 (jre1.8.0_66) running Windows 7. enterprise (SP1). Did nothing (just loaded a single XML file) throught the web based admin interface and tried this XQuery import module namespace geo = "http://expath.org/ns/geo"; declare namespace gml='http://www.opengis.net/gml'; let $point := <gml:Point><gml:coordinates>1,2</gml:coordinates></gml:Point> return geo:geometry-type($point) (it's from Geo Module documentation at http://docs.basex.org/wiki/Geo_Module#geo:geometry-type) all I get is *[XPST0017] Unknown function 'geo:geometry-type'.* in the logs, I have [POST] http://localhost:8080/BaseX831/dba/eval-query?name=&resource=&sort=&loglist=; Stopped at C:/Program Files/Apache Software Foundation/Tomcat 7.0/webapps/BaseX831/dba/modules/util.xqm, 94/17: [XPST0017] Unknown function 'geo:geometry-type'. The java Geo.class are present (webapps\BaseX831\WEB-INF\classes\org\expath\ns) Other trivial XQuery do work Any idea what's wrong ? Thanks and Happy New Year

3 6

GUI feature request "Add as import"
by Andy Bunce 03 Jan '16

03 Jan '16

Hi, I often find myself adding module imports in the GUI and I wonder if the following suggestion might speed up that process. Add a new menu item to the filelist right click context menu "Add as import" when the file selection is a XQuery module. The action would be to add a line to the current edit window source with import module namespace {$prefix} = '{$namespace}' at {$path}; Where $prefix and $namespace} are read from the source of the selected file, and $path is calculated from the relative path from the edit file source to the selected filelist source. The line would be inserted at a suitable place in the prologue or at the current location if that is too tricky. /Andy

2 4

archive:extract:-binary can cause OutOfMemoryException
by E. Wray Johnson 02 Jan '16

02 Jan '16

I have written an XQuery that properly extracts binary(raw) archive resource contents to binary(raw) resources. However since archive:extract:-binary returns the contents as xs:base64Binary and not streamable items, large archive contents are loaded completely into memory and will often cause an OutOfMemoryException. I have increased the BaseX JVM heap size and the problem goes away, but this is not acceptable in a system where many users may be doing this concurrently. archive:extract-to seems to only extract to an external file-system location and not to binary(raw) resources. Example code is provided below that extracts to the same database folder as the archive, however the target location of the contents is not pertinent to the problem. let $database := 'db' let $archivePath := 'myfolder/myarchive.zip' let $archive := db:retrieve($database, $archivePath) let $basePath := replace($archivePath, '[^/]+$', '') let $entries := archive:entries($archive) let $contents := archive:extract-binary($archive) for $entry at $pos in $entries let $content := $contents[$pos] let $target := $basePath || '/' || $entry/text() return db:store($database, $target, $content)

2 1

Java module to add documents to database
by Kevin S. Clarke 01 Jan '16

01 Jan '16

Hi, I'm writing a Java module that will do a couple of things, in addition to adding documents to a BaseX database. I've gotten the other things done, but I'm stuck a bit about how to add docs to the database from a module (I've seen the example that returns xml fragments that can be used from the xquery side). I see on http://docs.basex.org/wiki/Java_Examples that there are local/embedded and client-api ways to interact, but a module would want to be able to work in either situation. Are there any module examples that write to the database that someone would like to share (or just point me in the right direction)? Thanks, Kevin

2 4

Versioning
by Laurent Pellegrino 31 Dec '15

31 Dec '15

Hello all, I discovered BaseX recently and I find it really powerfull. Unfortunately, I read that versioning was not supported out of the box in 2014 [1]. Is it still the case ? Is there no simple workaround ? My usecase is the following. I have thousands of XML files which are indexed using BaseX in a specific database, lets say "mydb". At some point of time, a user may want to add a new XML file which is a slightly modified version of one already indexed by BaseX. I would like to be able to: - store this new XML resource in "mydb" database without loosing the XML from which it is issued - to retrieve the latest version or a previous one. - filter resources based on the latest version A simple way to achieve such a behaviour would be to be able to index a XML resource in BaseX with some additional meta information such as an ID and a timestamp, and then offer the possibility to use these meta information in an XQuery to filter resources. Is it possible with BaseX 8.3 ? I am looking for a solution that does not a require another database. [1] https://mailman.uni-konstanz.de/pipermail/basex-talk/2014-March/006535.html Kind Regards, Laurent Pellegrino

4 11

Full-text search with ft:mark (linking back)
by kleist 31 Dec '15

31 Dec '15

Dear members, given for example: ft:mark(//p[.//text() contains text 'real']). This will yield some results. Is there any way, to link back to the documents, where p was found? I know, it would work without ft:mark, but I need the <mark>-tag inside of the results, especially for fuzzy-search-results. Any help would be appreciated. - Günter

2 1

Store command with InputStream
by buddyonweb-software＠yahoo.com 29 Dec '15

29 Dec '15

I am writing a Java program that creates some text data that I want to store as a 'raw' document in a database.In the JavaDocs for the Store command () it states for the constructor public Store(java.lang.String path): "Constructor, specifying a target path. The input needs to be set via Command.setInput(InputStream)." Since the input is not a file, but rather an InputStream of data in memory, I do indeed call setInput. However, I cannotget this to work. Am I doing something wrong? Must the input only be from a file (and not from data in memory)? Here is snippet of code and thanks in advance. Note: Using the client/server mode. ClientSession session = new ClientSession(host, port, user, pwd); session.open("MyDb"); String theData = someFunctionThatReturnsALargeStringOfData(); ByteArrayInputStream bais = new ByteArrayInputStream(theData.getBytes()); Store cmd = new Store("newFile.txt"); cmd.setInput(bais); session.execute(cmd); session.close(); Note sure if this helps, but this is the stack trace I get. org.basex.core.BaseXException: Stopped at , 1/63: Syntax: STORE (TO [path]) [input] Store raw data. Stores raw data to the specified [path]. at org.basex.api.client.ClientSession.receive(ClientSession.java:193) at org.basex.api.client.ClientSession.execute(ClientSession.java:162) at org.basex.api.client.ClientSession.execute(ClientSession.java:167) at org.basex.api.client.Session.execute(Session.java:36) ......

2 1

Change context of QueryProcessor to query item
by Chris Speare 23 Dec '15

23 Dec '15

Good afternoon, I have a QueryProcessor to grab the top level items which works, but then I want to iterate over these items to query sub-items. My XML is currently (sub-section, excuse any typos however it is well formed) <MappingRequest> <Mappings destination="invoice_sftp" map="pdf_invoice"> <Mapping path="/var/lib/michrosatbilling/billingrunoutput_testing/dummy/invoice/2/3/1/231749.pdf" xpath="//CommercialDocument[@id='231749']"/> <Mapping path="/var/lib/michrosatbilling/billingrunoutput_testing/dummy/invoice/2/3/1/231750.pdf" xpath="//CommercialDocument[@id='231750']"/> <Mapping path="/var/lib/michrosatbilling/billingrunoutput_testing/dummy/invoice/2/3/1/231751.pdf" xpath="//CommercialDocument[@id='231751']"/> <Mapping path="/var/lib/michrosatbilling/billingrunoutput_testing/dummy/invoice/2/3/1/231752.pdf" xpath="//CommercialDocument[@id='231752']"/> <Mapping path="/var/lib/michrosatbilling/billingrunoutput_testing/dummy/invoice/2/3/1/231753.pdf" xpath="//CommercialDocument[@id='231753']"/> </Mappings> <Mappings destination="invoice_sftp" map="pdf_device_report"> <Mapping path="/var/lib/michrosatbilling/billingrunoutput_testing/dummy/report/8/3/7/8375680.pdf" xpath="//Device[@report_id='8375680']"/> <Mapping path="/var/lib/michrosatbilling/billingrunoutput_testing/dummy/report/8/3/7/8375681.pdf" xpath="//Device[@report_id='8375681']"/> <Mapping path="/var/lib/michrosatbilling/billingrunoutput_testing/dummy/report/8/3/7/8375679.pdf" xpath="//Device[@report_id='8375679']"/> <Mapping path="/var/lib/michrosatbilling/billingrunoutput_testing/dummy/report/8/3/7/8375682.pdf" xpath="//Device[@report_id='8375682']"/> <Mapping path="/var/lib/michrosatbilling/billingrunoutput_testing/dummy/report/8/3/7/8375686.pdf" xpath="//Device[@report_id='8375686']"/> </Mappings> </MappingRequest> My code in brief is: final Context context = new Context(); new CreateDB("LocalDB", "/test.xml").execute(context); QueryProcessor proc = new QueryProcessor("/MappingRequest/Mappings", context); // Loop over the mappings, get the map name then iterate over the underlying mapping nodes for(Item item; (item = iter.next()) != null;) { Object elem = BXNode.get((ANode)item); BXElem node1 = (BXElem)elem; System.out.println(node1.getAttribute("map")); QueryProcessor proc2 = proc.context(item); proc2.??("/Mapping"); Iter iter2 = proc.iter(); for(Item item2; (item2 = iter2.next()) != null;) { elem = BXNode.get((ANode)item2); node1 = (BXElem)elem; System.out.println(node1.getNodeName()); } } Having worked with Saxon previously I was able to pass the XdmItem from one query to the context of another query - allowing further queries to be carried out, therefore I assumed I would be able to something similar with BaseX? Feel I'm almost there with various ideas but nothing seems to change the context to the an item, rather than sticking with the originally loaded document. Secondly to this I need to workout how to perform the queries via a ClientSession - however that only seems to return a ClientQuery where .next() returns a String, rather than a workable Item or other object. Kind regards Chris Speare The information contained in the email and any files transmitted with it is confidential and intended for the addressee only. If you have received this email in error please accept our apologies and notify the originator. Any disclosure, copying, distribution or any other use of this communication is strictly prohibited and may be unlawful. We reserve the right to monitor and intercept communications for lawful business purposes. Wireless Innovation Ltd , Unit D2, Churcham Business Park, Churcham, Gloucestershire, GL2 8AX, Great Britain. Company Registration Number 5240202, Place of Registration Companies House Cardiff.

2 2

Problem with 3rd party jars in repository
by E. Wray Johnson 22 Dec '15

22 Dec '15

I have a 3rd party jar which is open source, which has no startup eligible classes. The classes I need to use have class names (not package name) that are all lower case. This seems to be a problem for BaseX. So, I tried to create a wrapper class, and the wrapper class loads but BaseX cannot load the dependent 3rd party jar no matter what I try. I have even tried putting the jar in the basex\lib folder. I have been able to install my wrapper class and run it successfully, but when I include the 3rd party jar, I get "Module not initialized" for the classes in the 3rd party jar because they are in another package. I have tried packaging everything into one jar, but that did not help. I have added a main class to the manifest of the 3rd party jar and it loaded fine but I cannot include the namespace in my xquery. My example is similar to the FOP.jar example on the file server, however I do not see query ussage example of this example. I have not yet tried the expath-pkg.xml route.

2 2

index:attribute-names empty
by Benjamin Gathmann 18 Dec '15

18 Dec '15

Hi Christian & Co., I have created a DB with the following characteristics: - Size: 37 GB -Nodes: 1'532'608'670 - Documents: 5205 - Up-to-date: true - TEXTINDEX: false - ATTRINDEX: false If I run "index:attribute-names" in the GUI command line (as described here: http://docs.basex.org/wiki/Indexes) I get an empty result and the following Query Info: ¨Query: index:attribute-names Result: - Hit(s): 0 Items - Updated: 0 Items - Printed: 0 Bytes - Read Locking: local [database] - Write Locking: none Timing: - Parsing: 1.08 ms - Compiling: 0.2 ms - Evaluating: 52.83 ms - Printing: 0.02 ms - Total Time: 54.13 ms Query plan: <QueryPlan compiled="true"> <CachedPath> <IterStep axis="child" test="index:attribute-names"/> </CachedPath> </QueryPlan> Why is the result empty? Yours, Benjamin

2 1