BaseX-Talk

basex-talk@mailman.uni-konstanz.de

5251 discussions

Replace command long execution time
by Andrey Fedosenko 14 Apr '18

14 Apr '18

Hello, The size of my database is about 23GB and about 650,000 records. The execution of the REPLACE command can take about 10 minutes. What could be the reason?

2 5

FW: Java 9 and BaseX 9
by Bram Vanroy 13 Apr '18

13 Apr '18

From: Bram Vanroy Sent: vrijdag 13 april 2018 11:03 To: 'basex-talk(a)mailman.uni-konstanz.de' <basex-talk(a)mailman.uni-konstanz.de> Subject: Java 9 and BaseX 9 Hello list I just downloaded and installed the most recent version of Java and tried to install the latest version (from the download web page) of Basex. However, upon launching the installer I get the message "Must install Java before running the installer". When looking through the mailing list, I already found this issue but it was marked as solved. [1] The problem still occurs, though. Perhaps the issue is that the {url}/releases/latest isn't the same download as you get from using the {url}//download page. If they are different, than perhaps these should be put in sync so that you always get the latest version when downloading. Kind regards Bram Vanroy [1] https://www.mail-archive.com/basex-talk@mailman.uni-konstanz.de/msg10417.ht…

2 3

Problems compiling BaseX in Eclipse Oxygen
by Ben Engbers 13 Apr '18

13 Apr '18

Hi, I want to compile and run the basex-examples in Eclipse (Oxygen) but even after adding Maven-support, I can't manage to compile the sources (were downloaded with git). When I look at the properties for the basex-examples project, I would expect to find the Java-classes in the target/classes directory but this directory does not contain any java classes. How can I compile the sources? Thanks, Ben

2 8

Re: [basex-talk] XSLT Module: feature request - XSLT 3.0 support
by Yitzhak Khabinsky 11 Apr '18

11 Apr '18

Just some minor touches for the XSLT Module for improvement. 1. If Saxon is present in the classpath set CP=%MAIN%\BaseX.jar;%MAIN%\lib\*;%MAIN%\lib\custom\*;d:\Kit\Saxonica\9.8.0.11\JavaEE\* return xslt:processor(), xslt:version() shall return Saxon EE 3.0 instead of Saxon EE 2.0 1. Documentation improvement http://docs.basex.org/wiki/Xslt_Module It currently says "...XSLT 2.0 is used instead if Version 9.x of the Saxon XSLT Processor<http://www.saxonica.com/> (saxon9he.jar, saxon9pe.jar, saxon9ee.jar) is found in the classpath...." Regards, Yitzhak Khabinsky Technical Services Lead Millicom International Services LLC 396 Alhambra Circle, Suite 1100 Coral Gables, FL 33134 Skype4B: +1 (305) 445-4172 Tel: (954) 684-8673 yitzhak.khabinsky(a)millicom.com<mailto:.khabinsky@millicom.com> www.millicom.com<https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.millico…>

2 2

Multiparts ... this time a proposal ...
by Marco Lettere 11 Apr '18

11 Apr '18

Dear Christian and BaseX community, working with multipart requests and RestXQ lately, we have the feeling that there could be a way to intercept the structure of the request with annotations in order to perform a more fine grained access to the single parts. At the moment the docs state the following example where, due to the consumes annotation reporting multipart/mixed, the data is returned as a sequence of items: |declare| |||%rest:path||(||"/multipart"||)| |||%rest:POST||(||"{$data}"||)| |||%rest:consumes||(||"multipart/mixed"||) ||(: optional :)| |function||page:multipart(||$data||as||item||()*) {| |||"Number of items: "|||| ||count||(||$data||)| |};| Do you think that an approach where the content negotiation is refined to single parts could be implementable? What we have in mind is something like the following snippet. Besides the usual sequence based representation of the body, parts are extracted as variables (part named “partname1” with $part1 and type validated/coerced to the function signature and so on). > |declare| > |||%rest:path||(||"/multipart"||)| > |||%rest:POST||(||"{$data}”||) (: |{$data} Could be removed at all or > is out of standard? :) > |||%rest:consumes||(||"multipart/mixed"||) ||(: optional :)| > |%input:part||(|||"{$part1}", |"partname1"||) ||(: partname could be > inferred from Content-Disposition and C|ontent-Type for type coercion > applied as usually with single-part requests :) |%input:part||(|||"{$part2}", |"partname2"||)| || |function||p||:multipart(||$data||as||item||()*, $part1 ||*as*||xs:binary, $part2 ||*as*||node()) {| | p||:do-something-with($part1)||||||||||| | };| || |Maybe even the serialization method could be declared... | |declare| |||%rest:path||(||"/multipart"||)| |||%rest:POST||(||"{$data}"||)| |||%rest:consumes||(||"multipart/mixed"||) ||(: optional :) ||%input:part||(|||"{$part1}", |"partname"||, "csv;separator=','") ||(: Just as an example ...:)||| |function||p||:multipart(||$data||as||item||()*, $part1 as node()) {| |||p||:do-something-with($part1)||||||||||| |}; ||Finally a check on the part count could be enforced in order to disallow requests that do not match *declare* %rest:path("/multipart") %rest:POST("{$data}") %rest:consumes("multipart/mixed") (: optional :) %input:part("{$part1}", "partname", "csv,separator=','")| ||%input:part-count(2)| *function* p:multipart($data as item()*, $part1 as node()) { |||p||:do-something-with($part1)| }; | This makes for a nice, declarative way of extracting parts or ensuring that the structure of the requests respects a given REST service contract. We understand that RestXQ is a shared formalization effort in the XML community but maybe there could be some opportunity to experiment and propose an addition to the specification? As usual thanks a lot for your attention. Best regards Marco.

2 1

Error-code when reading from a socket (RbaseClient)
by Ben Engbers 11 Apr '18

11 Apr '18

Hi, Since a few weeks I am working on a client for the R language. I use BaseXClient.java as example. And in R I use the R6 package for object orientation. And at this moment I can login and issue COMMANDs. I guess that at this moment and for this question, these parts of the code are the most essential: - socket-definition: private$sock <- socketConnection(host = "localhost", port = 1984L, open = "w+b", server = FALSE, blocking = TRUE, encoding = "utf-8") - execute a command: execute = function(command) { flush(private$sock) private$send(command) private$info <- private$receive() private$testSock() result <- private$show_receive() return(result) } - get the server-response: receive = function(input, output) { private$info <- NULL if (missing(input) || missing(output)) { output <- raw(0) output <- private$receive(private$sock, output) return(rawToChar(output)) } else { while((rd <- readBin(input, what = "raw", n =1)) >0) { if (rd == 0xff) next output <- append(output, rd) } return(output) } - test the result (Exceptionhandling): testSock = function() { tryCatch({ test <- readBin(private$sock, what = "raw", n =1) }, error = function(e) { print("ERROR") print(e) return(FALSE) }) return(test == 0x00) } I create a session and issue some commands from in a script: Session <- BasexClient$new("localhost", 1984, "admin", "admin") print(Session$execute("xquery 1 to 5")) print(Session$execute("xquery 6 to 10")) print(Session$execute("list")) When testsock is used in the first call, the readBin in the third line returns 00! Perfect! Result is [[1]] [1] "1" "2" "3" "4" "5" When testsock is used for the second call returns 0a ;-( and output is [[1]] character(0) When I comment out the first two calls, result is: [[1]] [1] "Name Resources Size Input Path " [2] "---------------------------------------------------------------------------------------------------------------" [3] "EUR 164 12288704 /home/bengbers/Programs/basex/ " [4] "EUR_Debat 1492 44857696 /home/bengbers/DataScience/TextMining-Web-Analytics/Text-mining case/xml_files/ " [5] "test1 0 4568 " [6] "" [7] "3 database(s)." My question for this moment is, if the response '0a' in testSock is truly the response from the basexserver or is this an error that is returned by the socket? Is the a list that explains the errorcodes returnde by the basex-server? Ben

2 1

OPTIMIZE command error
by Andrey Fedosenko 10 Apr '18

10 Apr '18

Hello, I have the following error while executing the "OPTIMIZE ALL" command: mproper use? Potential bug? Your feedback is welcome: Contact: basex-talk(a)mailman.uni-konstanz.de Version: BaseX 8.6.7 Java: Oracle Corporation, 1.7.0_151 OS: Linux, amd64 Stack Trace: java.lang.ArrayIndexOutOfBoundsException: 3 at org.basex.util.Compress.unpack(Compress.java:120) at org.basex.data.DiskData.txt(DiskData.java:284) at org.basex.data.DiskData.text(DiskData.java:251) at org.basex.io.serial.Serializer.node(Serializer.java:415) at org.basex.io.serial.Serializer.node(Serializer.java:158) at org.basex.io.serial.Serializer.node(Serializer.java:345) at org.basex.io.serial.Serializer.node(Serializer.java:158) at org.basex.io.serial.Serializer.serialize(Serializer.java:109) at org.basex.core.cmd.OptimizeAll$DBParser.parse(OptimizeAll.java:200) at org.basex.build.Builder.parse(Builder.java:77) at org.basex.build.DiskBuilder.build(DiskBuilder.java:77) at org.basex.core.cmd.OptimizeAll.optimizeAll(OptimizeAll.java:122) at org.basex.core.cmd.OptimizeAll$1.run(OptimizeAll.java:42) at org.basex.core.cmd.ACreate.update(ACreate.java:90) at org.basex.core.cmd.OptimizeAll.run(OptimizeAll.java:38) at org.basex.core.Command.run(Command.java:257) at org.basex.core.Command.execute(Command.java:93) at org.basex.server.ClientListener.run(ClientListener.java:139)

2 5

Re: [basex-talk] XSLT Module: feature request - XSLT 3.0 support
by Yitzhak Khabinsky 09 Apr '18

09 Apr '18

Hello, Good news on my end. I was running the following 3 lines in the BaseX Editor: let $doc := 'd:\Temp\CDW\IFRS\InforSunSystems\id0001_TZ20180200_IFRS15_SunSystems_3TZ002.xml' let $stylesheet := 'd:\Temp\CDW\IFRS\InforSunSystems\IFRS15_SunSystemsOut.xslt' return xslt:transform($doc, $stylesheet) And getting the following error message: "" (Line 1): Content is not allowed in prolog. My XSLT stylesheet starts as follows: <?xml version="1.0"?> <xsl:stylesheet version="3.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform" xpath-default-namespace="http://www.millicom.com" xmlns:functx="http://www.functx.com" xmlns:xs="http://www.w3.org/2001/XMLSchema"> <xsl:output method="text" encoding="UTF-8"/> <xsl:strip-space elements="*"/> Based on Martin's input, I changed the last line to: return xslt:transform-text($doc, $stylesheet) And everything is working. Thanks a lot! Regards, Yitzhak Khabinsky Technical Services Lead Millicom International Services LLC 396 Alhambra Circle, Suite 1100 Coral Gables, FL 33134 Skype4B: +1 (305) 445-4172 Tel: (954) 684-8673 yitzhak.khabinsky(a)millicom.com<mailto:.khabinsky@millicom.com> www.millicom.com<https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.millico…>

1 0

XSLT Module: feature request - XSLT 3.0 support
by Yitzhak Khabinsky 09 Apr '18

09 Apr '18

Hello, We have Saxon EE edition v.9.8.* licenses and are using XSLT stylesheets compliant with the XSLT 3.0 standard. Unfortunately, BaseX XSLT module currently supports up to XSLT 2.0 standard only. return xslt:processor(), xslt:version() yields the following: Saxon EE 2.0 Though Saxon XSLT processor supports XSLT 3.0 standard. XSLT 3.0 features are available in all Saxon editions. http://saxonica.com/documentation/index.html#!using-xsl/xslt30 And Michael Kay chairs the w3c XSLT group. Are there any plans to add the XSLT 3.0 standard support to the BaseX XSLT module? [cid:image001.png@01D3D004.BB4AE290] Regards, Yitzhak Khabinsky Technical Services Lead Millicom International Services LLC 396 Alhambra Circle, Suite 1100 Coral Gables, FL 33134 Skype4B: +1 (305) 445-4172 Tel: (954) 684-8673 yitzhak.khabinsky(a)millicom.com<mailto:.khabinsky@millicom.com> www.millicom.com<https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.millico…>

2 1

query is not using attribute index with function
by andre weber 09 Apr '18

09 Apr '18

Hi, with the new Version 9 I have encountered a strange performance issue: - Running a simple xpath-query is fast, as it is using the attribute-index. Execution time: 59.19 ms - Putting the same simple xpath-query into a function results in a very poor performance. No use of the attribute-index here. Execution time: 12578.55 ms There where no issues with that in the previous Versions 8.x Here’s the setup… xquery version "3.1"; declare function local:slow_query_here() as node()* { prof:time( db:open("mars","data“)/sec[ @isin = "FR0000125338" ] ) (: 12578.55 ms :) }; let $fast_query_here := prof:time( db:open("mars","data“)/sec[ @isin = "FR0000125338" ] ) (: 59.19 ms :) return local:slow_query_here() Here’s the query plan… <QueryPlan compiled="true" updating="false"> <GFLWOR type="element()*"> <Let type="element()*"> <Var name="$fast_query_here" id="0" type="element()*"/> <ProfTime name="time(value[,label])" type="element()*"> <CachedPath type="element()*"> <ValueAccess index="attribute" type="attribute()*"> <IndexStaticDb name="mars" database="mars"/> <Str type="xs:string">FR0000125338</Str> </ValueAccess> <IterStep axis="self" test="*:base_isin" type="attribute()*"/> <CachedStep axis="parent" test="*:option" type="element()*"> <CachedPath> <IterStep axis="parent" test="*:option_list" type="element()*"/> <IterStep axis="parent" test="*:finance" type="element()*"/> <IterStep axis="parent" test="document-node()" type="document-node()*"/> </CachedPath> </CachedStep> </CachedPath> </ProfTime> </Let> <ProfTime name="time(value[,label])" type="element()*"> <CachedPath type="element()*" database="mars"> <DBNode database="mars" pre="0" type="document-node()"/> <IterStep axis="descendant" test="*:option_list" type="element()*"/> <IterStep axis="child" test="*:option" type="element()*"> <CmpSimpleG op="=" type="xs:boolean" size="1"> <CachedPath type="attribute()?"> <IterStep axis="attribute" test="*:base_isin" type="attribute()?"/> </CachedPath> <Str type="xs:string">FR0000125338</Str> </CmpSimpleG> </IterStep> </CachedPath> </ProfTime> </GFLWOR> </QueryPlan> Any idea what’s causing the performance drop? Thank you for your thoughts! Andre

2 2

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

BaseX-Talk