BaseX-Talk September 2015

basex-talk@mailman.uni-konstanz.de

32 participants
51 discussions

BaseX, Zorba, e-mail module
by Tim Thompson 04 Feb '16

04 Feb '16

Hello, I would like to be able to send e-mail from a Web app I am developing in BaseX and was wondering whether anyone else had implemented this functionality (or whether a BaseX e-mail module might be in the works?). I see that Zorba has an e-mail module with support for SMTP and IMAP (not to mention many other useful modules), which leads me to wonder whether anyone has worked on a BaseX connector for Zorba. In the meantime, I decided to try using the BaseX process module to call Zorba, like so: proc:system( 'zorba', ( '-q', convert:binary-to-string( db:retrieve( 'out', 'email.xq' ) ) ) ) I stored the XQuery code for Zorba as a raw file in the BaseX database, since storing it in the webapp directory caused an error, due to the unrecognized Zorba module declarations. This approach seems to work adequately, but does anyone have any suggestions for improvement, or a different approach? Thanks in advance, Tim -- Tim A. Thompson Metadata Librarian (Spanish/Portuguese Specialty) Princeton University Library

3 4

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

Origami presentation XML Amsterdam
by Marc van Grootel 04 Oct '15

04 Oct '15

Hi, Anyone on the list coming to Amsterdam the 6th of November? We may have some beers together. I will (finally) be presenting on the Origami templating library. I'm still working on it over on Github (in the unstable branch). After a pause in development it was almost completely rewritten. I intend to cut a new release just before the conference. Here's the anouncement I made for the conference website. XQuery 3.1 is still a candidate for recommendation by the W3C. Nevertheless there are already various implementations ready for use today. The introduction of maps and arrays, and the availability of functional weaponry such as fold and apply may change the way you write XQuery applications. You can still wield the trusty weapons of XPath and the FLWOR expression while, at the same time, adopting a more functional style of programming. Origami is an ongoing exploration of templating in XQuery and it makes heavy use of this functional style. It started with a quest for a more XSLT-like transformation capability and recently took a nice turn with the adoption of a flavor of micro-XML, with a heavy reliance on maps and arrays. Origami's unfolding design tries to honour principles of simplicity and composability. With Origami I aim to decomplect, or disentangle, several aspects common to templating problems and offer them in a powerful library using a couple of simple abstractions. With some practical code examples I will demonstrate how to use Origami to implement different templating styles. --Marc

2 1

about xi:include and xml:base
by Marco Randazzo 01 Oct '15

01 Oct '15

In my application I have a main xml file (general.xml) which includes several xml files (file1.xml, file2.xml...) using the xi:include tag. I successfully imported general.xml in basex using the gui: the xi:include directives were successfully processed, merging all data in a single database. I noticed in the result window that the database keeps track of the origin of the data, adding an xml:base info for each included block. I then performed some find and replace operation on the database, modifying some values which were originally belonging to different files. Now I want to export data to file. I successfully obtained a single xml file using the export command of the gui. The generated file still preserves the xml:base infos. However, I'd like to recreate the original structure, i.e. obtain multiple files, the same which were merged by xi:include directives at the beginning of the import process. I cannot find such an option in the export tab. Is it possible? Thank you for the help, MR

2 1

JSON serializer - slight bug
by Hans-Juergen Rennau 30 Sep '15

30 Sep '15

Dear BaseX, I think there is a very slight bug in the JSON serializer: when an element with type="object" has a <_> child, I think the serializer should report an error (as underscores must be doubled, if not using the name "_", reserved for array items. Instead, the serializer creates a property with the empty string as name. Example: json:serialize( <json type="object"> <BaseX type="object"> <_>is phantastic</_> </BaseX> </json> )=>{ "BaseX":{ "":"is phantastic" } } Compare with: json:serialize( <json type="object"> <BaseX type="object"> <__>is phantastic</__> </BaseX> </json> )=>{ "BaseX":{ "_":"is phantastic" } } It may look like pedantry, but this misbehaviour may cause confusion, as it is of course an interesting practise to create JSON by creating the XML format and in the end serializing: if then inspecting the end result, it is not immediately clear that the "":... is caused by the serializer, rather than by own code. Kind regards,Hans-Jürgen

2 2

crypto module with binary files
by Michel Banguerski 29 Sep '15

29 Sep '15

Greetings all, I would like to perform cryptographic operations with arbitrary binary files but am unable to figure how to convert the output of file:read-binary() to something that could be processed by crypto:hmac() for example. Can You give me a hint please? TIA Best regards Michel

2 1

Java access to JSON-to-XML transformation
by Hans-Juergen Rennau 29 Sep '15

29 Sep '15

Dear BaseX team, I have a question concerning the availability of the JSON-to-XML transformation. Before asking, this remark: I am very glad and grateful that BaseX offers this functionality (json:parse), as it exposes JSON to XPath, which enables important evaluations that otherwise would not even be attempted. And now my question. Is it possible - does it appear reasonable - to access the functionality within a Java application by calling the BaseX code (org.basex.build.json.JsonParser, I suppose)) immediately, without launching XQuery? I expect that one would be treading dangerous ground, relying on internal APIs which may change any time, not being meant for external use. If this is so, a brief confirmation would suffice to make me drop the idea. Thank you, and kind regards -Hans-Jürgen PS: Background of my question - I evaluate alternative possibilities how to validate JSON within Java applications. One (and probably the best) option is validating against JSON schema (json-schema.org), an alternative would be a to-XML transformation followed by XSD validation. A first experiment lettig BaseX via XQJ execute the epic query "json:parse(.)" revealed a performance which would not be sufficient to be competitive. So I wondered if removing the overhead of launching BaseX might be an approach to consider.

2 1

inspect:functions performance
by Andy Bunce 28 Sep '15

28 Sep '15

Hi, The following code runs (using v8.3) in about 1sec for me. import module namespace p="xquery-30" at " https://raw.githubusercontent.com/expkg-zone58/ex-xparse/master/src/main/co… "; p:parse-XQuery("2+2") Using inspect:functions let $f:= inspect:functions( " https://raw.githubusercontent.com/expkg-zone58/ex-xparse/master/src/main/co… ") [ local-name-from-QName(function-name(.)) = "parse-XQuery" and function-arity(.)=1] return $f("2+2") takes a very much longer time. Is this a bug? /Andy

2 4

Getting Line Number of Node in XML Doc
by ankit kumar 25 Sep '15

25 Sep '15

Hi, I just wanted to know , can basex provide line and column number of a element of a xml document loaded in BaseX database. Thanks, Anky.

2 2

BaseX 8.3: The Autumn Edition
by Christian Grün 23 Sep '15

23 Sep '15

Dear all, Summer is over, and a new version of BaseX is here! We are more than happy to present Version 8.3 of our XML database and XQuery 3.1 processor. This is what you get: SELECTIVE INDEXING - restrict value indexing to given elements and attributes - support extended to updatable and main-memory databases VALIDATION - support for RelaxNG validation (XML and compact syntax) - optional XML Schema 1.1 support - new function for creating validation reports XQUERY MODULES - Strings Module: Levenshtein, Soundex, Cologne Phonetic - updates in the Archive, Database, Admin and Process modules DBA - File panel: upload and download files BUG FIXES - automatic database optimization if node id turns negative - XQuery optimizations, performance tweaks, WebDAV issues You find the new release at http://basex.org. We are looking forward to your feedback, Christian BaseX Team

1 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

BaseX-Talk September 2015