BaseX-Talk

basex-talk@mailman.uni-konstanz.de

1 participants
5252 discussions

BaseX 8.0.1: Minor Patches
by Christian Grün 25 Feb '15

25 Feb '15

Hi there, We have just released a first post-Prague version of BaseX, which includes some minor optimizations and fixes (nothing critical): XQUERY - Faster execution of single, index-based results - Iterative evaluation of steps with multiple predicates FIXES - WebDAV locking - Archive Module - Adaptive serialization of arrays and maps - Digest Authentication - Save command-line history In future, we will try our best to release bug fixes more regularly. This way, you can stay up-to-date without the need to check for new versions five times a day. Have fun, Christian [1] http://basex.org/downloads

2 1

Re: [basex-talk] Modelling Checkin-Checkout using locks
by Christian Grün 24 Feb '15

24 Feb '15

I have added a GitHub issue on your locking request [1]. In the meanwhile, it's probably best to use RESTXQ (XQuery) and check if a resource is locked before returning it to a user. [1] https://github.com/BaseXdb/basex/issues/1083 On Mon, Feb 23, 2015 at 6:28 PM, <zopyxfilter(a)gmail.com> wrote: > Andreas Jung wrote: >> Christian Grün wrote: >>>> oops, this was intended for the exist-db guys...but the question >>>> are similar...anyone used BaseX with a checkout-checkin scenario? >>> First of all: I'm also interested to hear more about this from our >>> users.. >> >>> Talking about locking functions, we plan to generalize our WebDAV >>> locking feature to a general database locking mechanism. The WebDAV >>> locks already contain information on the user who created the lock, >>> a timeout, the scope (shared vs. exclusive), the lock id, etc., and >>> this info could be extended by the lock time and other (maybe even >>> application specific) info. Do you have some examples for data that >>> could be specific to an application? >> >> In our SGML CMS you were able to checkout parts of a bigger document >> for example a chapter. In this case we needed additional information >> inside the checked-out document - perhaps not on the lock itself - >> we used processing instructions for that as far as I recall. >> >> Being able to store arbitrary key-value pairs on a lock would be >> sufficient. Otherweise I would store such application specific >> data as an additional XML file containing these informations. > > At least needs to be some support for lock token in order detect > lock/unlock conflicts. > > Andreas

2 1

Xquery issue
by Rob Stapper 24 Feb '15

24 Feb '15

Hi guys, While debugging a xquery-program I saw something strange behavior( to me anyway). I reproduced it in the attached xquery-snipped. Looking at the evaluation data in the query-info screen of the basexGUI after running the program, one can see that function $f($a) is evaluated more than ones. The result is ok but I wondered if this would reduce the performance. I'm running version 8.01 beta d13ddc7 in local GUI-mode. Regards, Rob Stapper --- Dit e-mailbericht is gecontroleerd op virussen met Avast antivirussoftware. http://www.avast.com

3 3

Comparing Two Iter Objects.
by ankit kumar 24 Feb '15

24 Feb '15

Hi, I want to compare equality for two Iter objects. Each contains a set of DBNodes in unordered manner. I have created a wrapper for Iter object. Can you please help me in defining equals and hashcode method for the wrapper. Thanks Ankit

2 3

DBNode Not Overriding equals() and hashcode() method.
by ankit kumar 24 Feb '15

24 Feb '15

Hi, I am trying to maintain a Set of DBNode Object But, it was not working properly because DBNode is not overriding the equals() and hashcode() methods. Due to this reason i am facing difficulty to maintain similar DBNode. Looking forward for help. Thanks & Regards Ankit Kumar

2 1

CLI issue with 8.0.1
by Marco Lettere 24 Feb '15

24 Feb '15

Hello all, I'm unable to execute the example from documentation: bin/basex -b$v=example "declare variable $v external; $v" IT gives this exception: Stopped at ., 1/19: [XPST0003] Expecting '$', found 'e'. Any clues? Regards, MArco.

3 3

Modelling Checkin-Checkout using locks
by Andreas Jung 23 Feb '15

23 Feb '15

Hi there, I am currently thinking about ways how to model a reliable checkout-checkin scenario with XML documents. In an older SGML CMS system that we build, we implemented the procedure roughly like that: - user chooses a document in the CMS and clicks on CHECKOUT - CHECKOUT copy the document to a private area of the user - CHECKOUT will lock the original file (reads allowed, write forbidden) - user edits the the copied content - user performs either a CHECKIN which will copy the private copy of the file into its original location and release the lock - or user will cancel the checkout (removing the copy and releasing the lock) I see that exist-db has support for read and exclusive locks (inside the util: module). Can they be used to implement such a scenario? Is it possible to associate (arbitrary metadata) with locks like lock time, user who created the lock, app specific data? Is it possible to introspect all current locks on the database? Is there something like a lock token (in WebDAV) does one needs to release a lock? Andreas

2 3

http:send-request / Digest - problem
by Hans-Juergen Rennau 23 Feb '15

23 Feb '15

Dear BaseX team, I have problems to get the http:send-request going when using Digest authentication. I get a response as if the password were wrong, although I think it isn't. Request: <http:request method='post' send-authorization='true' username='foofoo' password='secret' auth-method="Digest"> <http:body media-type='text/xml' method='xml'>{$useMsg}</http:body> </http:request> Response: <http:response xmlns:http="http://expath.org/ns/http-client" status="401" message="Unauthorized"> <http:header name="Server" value="Apache/2.2.22 (Ubuntu)"/> <http:header name="WWW-Authenticate" value="Digest realm="traveltainment", qop="auth", nonce="1424448664077:8d932e72fb644113f4fb1a76f942cf1d", opaque="pcyalPBREdBtyQHIHRxOQPgGANZiOjeW""/> <http:header name="Connection" value="close"/> <http:header name="Vary" value="Accept-Encoding"/> <http:header name="Content-Length" value="954"/> <http:header name="Date" value="Fri, 20 Feb 2015 16:11:04 GMT"/> <http:header name="Content-Type" value="text/html;charset=utf-8"/> </http:response> Interestingly, a JMeter test using the same URL, user name and password works correctly. (I used copy & paste to transfer the credentials from the JMeter GUI into the XQuery code, and the credentials are just what some documentation tells me; tried of course also typing them in manually.) Using WireShark, I observed the messages sent by JMeter and BaseX, respectively. BaseX does send two requests, as it is supposed to do; but the authorization data sent by the second one are in two respects different from the data sent by JMeter: JMeter sends a field: algorithm="MD5", which BaseX does not, and the cnonce is much longer. (See below for the authorization strings.) I use Java 8, and I suspect the problem is connected with that. I heard rumours that Digest authentication with Java 8 might require a little adaptation of the code. Did you try the module using Java8? Cheers,Hans-Juergen PS: (1) Authentication sent by BaseX, failing: User-Agent:Java/1.8.0_31 Authorization:Digest username=foofoo, realm=traveltainment,nonce=1424448664069:b6ed9add48830631ae90ad27cfcb5c5e,uri=/TTXml-1.6/Dispatcher/Search/Package/RegionList,qop=auth,nc=00000001,cnonce=9b2ff1ce6900217dd6be667aa6f99e12,response=4e35b40dd4ba7d62cd6123d9adebc046,opaque=pcyalPBREdBtyQHIHRxOQPgGANZiOjeW Authorization:Digest username=foofoo,realm=traveltainment,nonce=1424449896135:5fa18cec34de1a15d8ce2a36df77bd6a,uri=/TTXml-1.6/Dispatcher/Search/Package/RegionList,qop=auth,nc=00000001,cnonce=5be8fce766d843e8ea29936b73ed94c7,response=3757d8fb6cfc4c997030541c58e72d61,opaque=pcyalPBREdBtyQHIHRxOQPgGANZiOjeW (2) Authentication sent by JMeter, successfully (please note the "algorithm" field; also note the much shorter cnonce): User-Agent:Apache-HttpClient/4.2.6 (java 1.5) Autorization:Digest username="foofoo", realm="traveltainment",nonce="1424448476861:032a9011541d271a429d737844ec860a",uri="/TTXml-1.6/Dispatcher/Search/Package/RegionList",response="df6416855e49a0f28cff8020c30ad3a7",qop=auth,nc=00000001,cnonce="c4b43ae817866fb5",algorithm="MD5",opaque="pcyalPBREdBtyQHIHRxOQPgGANZiOjeW" Authorization:Digest username="foofoo", realm="traveltainment", nonce="1424450034667:82c9d9977a208442a7926a948e163e45", uri="/TTXml-1.6/Dispatcher/Search/Package/RegionList",response="74712ad74209c954bfb6e545b5f8670b",qop=auth,nc=00000001,cnonce="788dbfb1ed7a77ef",algorithm="MD5",opaque="pcyalPBREdBtyQHIHRxOQPgGANZiOjeW"

2 6

Re: [basex-talk] BaseX-Talk Digest, Vol 62, Issue 31
by Jothivel S 23 Feb '15

23 Feb '15

Hi Christian, I just query the root element like "/root" from the xml(database). When I was getting the out of memory, the thread did hold more than ~200MB in size. but actual xml size is ~1.2 MB. And that 200MB file have some encrypted data like "òª©" and the same code got repeated many times(probably this would cause the 200 MB). Is there any issue with it? Also I am using the same context for read and write. That is single context is used for read and write operation parallely and read and write operation for the same database is synchronized. please let me know if you need more information in this regard. Thanks,Jothivel On Friday, 20 February 2015 6:58 PM, Christian Grün <christian.gruen(a)gmail.com> wrote: Hi Jothivel, > I am getting out of memory error when i do the execute query operation. Sorry, I need more information on your particular query. Unfortunately, we cannot provide free support for older versions. What happens with Version 8.0? > 2. without migrating the basex 7.6 to 8.0, is it possible to avoid the > increase of size. Yes, as I said in my previous mail, we have worked on that in the meanwhile. Hope this helps, Christian PS: Please always write to the basex-talk mailing list; this way, others can benefit from the feedback as well.

2 5

Re: [basex-talk] impressed! and already trying to run before I can walk
by Niels Grundtvig Nielsen 23 Feb '15

23 Feb '15

Some years ago … running anything interesting on a Linux platform was a geekish activity involving compilation, dependency-hunting and many other time-consuming activities that put quite a few people off. In the interim, non-technical Linux users (like me!) have become used to packages that do all this black magic for them. So I was a bit disappointed when unzipping the Ubuntu installation files gave me nothing I could run and no clear instructions (at least at my level) for what I had to do with the contents of the archive. Can someone, please, give me clear, simple instructions that cover: - what extra programs I may need to install before being able to compile/installe BaseX - what I have to do, step by step, to compile/installe BaseX Of course, an even better answer would be "run this install script" … :-} Thanks in advance for help and advice. On Wed, Feb 4, 2015 at 1:48 PM, Niels Grundtvig Nielsen < communicator.ngn(a)gmail.com> wrote: > The next big test for me is, will BaseX install as easily on my Ubuntu box > at home as it did on a Windows PC at work … hope so, because the interface > is a much better match for the sort of things I want to do with an XML > database than the/an other free option I tried. > > On Tue, Feb 3, 2015 at 1:16 PM, Menashè Eliezer <meliezer(a)ogs.trieste.it> > wrote: > >> Hi Niels, >> Just confirming BaseX is very good. I've confronted its performance and >> support to the biggest open source alternative, which I won't mention its >> name here. >> >> With kind regards, >> Menashè >> >> On 02/02/2015 02:48 PM, Bridger Dyson-Smith wrote: >> >> Hi Niels, >> >> BaseX has some very nice documentation in the wiki [1]. The WIkibooks >> XQuery document can be helpful[2] and Priscilla Walmsley's XQuery book is >> an absolutely fantastic reference. >> >> Hope that's helpful. >> Cheers, >> Bridger >> >> [1] http://docs.basex.org/wiki/XQuery >> [2] http://en.wikibooks.org/wiki/XQuery >> >> On Mon, Feb 2, 2015 at 8:42 AM, Niels Grundtvig Nielsen < >> communicator.ngn(a)gmail.com> wrote: >> >>> BaseX already looks/feels like a program I am going to enjoy using, >>> even with the general challenge that I don't know enough about xQuery. I >>> would welcome suggestions for anywhere to find a friendly introduction! >>> >>> My first specific challenge is trying to under the syntax for replace. >>> In the code, I've managed to retrieve a single xml document where <title> >>> contains "35. Andan" … I made a copy/paste error, and in this document the >>> value of title should be '35. Andantino' instead of '35. Andantino mesto'. >>> My experiment is obviously incorrect, so I'd welcome tips on what's wrong >>> with it. >>> >>> //score[contains(title, '35. Andan')]//title >>> replace value of (title) with '35. Andantino' >>> >>> Thanks in advance! >>> >> >> >> >

4 4

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

BaseX-Talk