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

STORE Command
by Maximilian Gärber 19 Oct '15

19 Oct '15

Hi, any way to add multiple binary files at once? <store path="source/_img/">../source/_img/*.*</store> would be nice. Regards, Max

2 3

restxq multipart/mixed application/octet-stream
by Dolejsi Jiri 19 Oct '15

19 Oct '15

When i send multipart/mixed document with 2.part application/octet-stream with value 0xFF0xFF Into the following function, it writes into file character ?. I am using BaseX 8.3 declare %rest:path("test") %rest:PUT("{$doc}") function nib:insert-doc($doc as item()*) as empty-sequence() { let $data := $doc[2] return file:write-text("d:\abc.bin", $data) }; ________________________________ Odes?latel tohoto e-mailu informuje adres?ta, ?e ?TK vy?aduje k platnosti pr?vn?ho jedn?n? p?semnou formu s v?jimkou jedn?n? realizovan?ho prost?ednictv?m elektronick?ch syst?m? ?TK (Fotobanka, Videobanka, Protext, NewsMarket, Infobanka). D?le informuje adres?ta, ?e smlouva s ?TK nem??e b?t uzav?ena, pokud nen? v?slovn? shoda o v?ech jej?ch n?le?itostech. ?TK si vyhrazuje pr?vo ukon?it jedn?n? o uzav?en? smlouvy kdykoli, a to z jak?hokoliv d?vodu nebo i bez uveden? d?vodu. Obsahuje-li tento e-mail nab?dku, ?TK vylu?uje p?ijet? nab?dky s dodatkem ?i odchylkou. Pokud je tento mail kvitanc? ve smyslu ? 1949, odst. 1 z?k. ?. 89/2012 Sb., ?TK v?slovn? sd?luje, ?e kvitance o zaplacen? jak?koliv spl?tky dluhu neznamen?, ?e byla sou?asn? zaplacena jak?koliv ??st t?ho? dluhu splatn? d??ve, a kvitance o spl?tce jistiny neznamen?, ?e bylo zaplaceno tak? p??slu?enstv? pohled?vky.

2 1

xml-to-json vs json:serialize
by Marco Lettere 16 Oct '15

16 Oct '15

Hi all, I have a bit of difficulty in understanding the rationale behind the XML to Json (and vice-versa) conversions. I must admit I'm a bit confused on when to use which format and how those map to serialization/parsing rules. In particular I'm wondering whether it's really necessary to use both the formalisms (and functions) in the following example. My feelings are that using only the standard (?) XQuery 3.1 representation would be much clearer and would enable a cleaner service chaining mechanism. Has anyone 2 minutes left to explain to me the model behind this sort of things? I really appreciate your support! Regards, Marco. module namespace jt = "urn:json:test"; declare %rest:path("jsonsrv") %output:method("text") %output:media-type("application/json") function jt:service() { xml-to-json( <map xmlns="http://www.w3.org/2005/xpath-functions"> <string key="a">12345678</string> </map>) }; declare %rest:path("testreq") %output:method("text") %output:media-type("application/json") function jt:test() { let $response := http:send-request( <http:request href="http://localhost:9984/jsonsrv" method="POST"> <http:body media-type="application/x-www-form-urlencoded" method="text"/> </http:request>) return json:serialize($response[2]) };

4 6

Location for import modules
by Paul Swennenhuis 15 Oct '15

15 Oct '15

Hi, I installed BaseX 8.3 as Tomcat webapp on a Ubuntu server. Everything works fine, including the DBA interface (very much appreciated, thanks for that!) Except one thing: I can't get the import of XQuery modules right. When I import a module in a .xq file, like this: import module namespace lib="My Library" at "library.xq"; BaseX gives me a "[XQST0059] Could not retrieve module '/library.xq' " error. Earlier I installed BaseX on a EC2 instance at Amazon, and there I found that import modules had to be placed in /usr/share/tomcat7 to be found. But that directory does not exist on this server, and even when I create it and place library.xq there it is still not found. I tried placing library.xq in the RESTPATH directory, in the REPOPATH directory, in the home directory of BaseX, but all in vain. I only get it to work when I place library.xq in the root of the server. And that is obviously not something I want to do. Or, alternatively, when I place it in [some directory] and then change the import line to import module namespace lib="My Library" at "[some directory]/library.xq"; In other words, when I change the relative path to an absolute one. But I'd have to do that for all of my XQuery files (a lot!) and, worse, I'd have to change it again if I deploy the .xq files on another server. So, to summarize: in which directory/directories does BaseX try to locate import modules? And what would I have to do to change the default location(s)? Thanks in advance. Paul

2 4

Best way to upgrade from 8.2.3 to 8.3?
by Chuck Bearden 14 Oct '15

14 Oct '15

Hi all, Is there a recommended best way to upgrade the BaseX software while retaining the data? Perhaps to generate backups of the databases, then restore them to the new instance? I mostly run in standalone CL mode. Thanks, Chuck

2 3

Http: database creation
by meumapple 14 Oct '15

14 Oct '15

Hi, I have developed with RESTXQ a web application that compares texts (the user specifies the urls of two texts in a form and the computation is performed with their submission). It works but it is too slow if the texts are relatively long (before the actual comparison I pre-process the texts deleting and adding nodes via transform expressions). I am not sure what the cause is, but I was wondering whether everything would be faster if I created a database with the texts and then applied the queries. If so, how is it possible to a allow one click submission (via RESTXQ or REST) to create a DB and then open it and perform queries on it, since the database creation -if I am not wrong- cannot be associated with other operations within the same query? Thanks. J.

2 1

Strange behaviour running unit test suites
by Marc van Grootel 11 Oct '15

11 Oct '15

Hi, Just noticed this on 8.3 I have multiple unit test suites in a test folder. When I run `basex -t test` or `basex -t test/` all suites are run (no errors) But when I try to run hem with something like `basex -t test/*.xqm` the test runner stops after the first suite with the following message: Stopped at /Users/marcvangrootel/data/fold-webapp/webapp/origami/test/test-data.xqm, 6/8: [XPST0003] Library modules cannot be evaluated. Can anyone confirm this behaviour? --Marc

2 2

archive:extract-to out of memory
by Lizzi, Vincent 11 Oct '15

11 Oct '15

I've been using the new method archive:extract-to almost since it became available in BaseX, and it generally works very well. I've run into trouble with an out of memory error a few times trying to extract .zip files that are fairly large. For example, the .zip file I'm looking at now is 268 Mb. The out of memory error seems to actually be thrown on the file:read-binary method, which must be passed as input to archive:extract-to. Increasing the amount of memory available to BaseX by increasing -Xmx from 512 to 768 helps, but that doesn't seem like a good solution. For example, using the latest BaseX (BaseX831-20150929.165521.zip), a 268 Mb .zip file, a smaller 65 Mb .zip file, and the following code: declare namespace _ = "test"; declare function _:unzip1($zip as xs:anyURI) as xs:anyURI { let $tempDir := file:create-temp-dir('basex', 'unzip') return ( archive:extract-to($tempDir, file:read-binary($zip)), xs:anyURI($tempDir) ) }; declare function _:unzip2($zip as xs:anyURI) as xs:anyURI { let $tempDir := file:create-temp-dir('basex', 'unzip') return ( let $entries := zip:entries($zip) for $e in $entries//zip:entry let $path := string-join(($e/parent::zip:dir/@name, $e/@name), '/') return ( file:create-dir(file:parent($tempDir || $path)), file:write-binary($tempDir || $path, zip:binary-entry($zip, $path)) ), xs:anyURI($tempDir) ) }; declare function _:unzip3($zip as xs:anyURI) as xs:anyURI { let $tempDir := file:create-temp-dir('basex', 'unzip') return ( for $f in file:read-binary($zip) return archive:extract-to($tempDir, $f), xs:anyURI($tempDir) ) }; let $zip := xs:anyURI('C:\temp\bigfile.zip') let $unzipped := _:unzip2($zip) (: change this to test 1 2 3 :) return $unzipped When given a small .zip file _:unzip1, _:unzip2, and _:unzip3 extract all files from the .zip. _:unzip1 and _:unzip3 seem to complete in less time than _:unzip2. When given a large .zip file, _:unzip2 and _:unzip3 work but _:unzip1 produces an out of memory error. However, when I use the _:unzip1, _:unzip2 or _:unzip3 code in a larger program, which does some processing on the unzipped file, I still get an out of memory error and can see that not all files were extracted. Any help would be appreciated. Thanks, Vincent

2 1

Hitting index for multi-db query
by Marco Lettere 11 Oct '15

11 Oct '15

Hello all, I have to query all the databases on one server for docs matching a query. I do it with this query for example: db:list() ! db: open(.)/entry[@about='abc'] This query never hits the available attribute index. I could rewrite the query to something like db:list() ! attribute(., 'abc', 'about')/.. and in this case it takes advantage of the index of course. The problem is that I have to implement more queries which have much more generic predicate structures where I probably won't be able to use indexes explicitly. Is there any possibility to rewrite this query in order to use the indexes implicitly? Thanks, Marco.

2 3

Java from XQuery
by meumapple 09 Oct '15

09 Oct '15

Hi All, It is not clear to me if I can run from XQuery a .jar file (it is a simple linguistic parser). Any idea about that? Thanks. Joseph

3 7