Cheló̱na, a Parboiled2 based Scala parser for the W3C RDF 1.1 Turtle Language

[Jürgen Pfundt]

Hi,

a first draft of a Parboiled2 based Scala parser for the W3C RDF 1.1 Turtle Language has been published.

Cheló̱na, which is the Greek name for turtle, can be found here: https://github.com/JuPfu/chelona/blob/master/src/main/scala/org/chelona/ChelonaParser.scala 

Any participation in the project or suggestions for improvement are kindly welcome!

Kind regards

Jürgen Pfundt

[Mathias Doenitz]

Nice!

Thanks, Jürgen, for the pointer!

Cheers,
Mathias

---
mat...@parboiled.org
http://www.parboiled.org

> --
> You received this message because you are subscribed to the Google Groups "parboiled2.org User List" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to parboiled-use...@googlegroups.com.
> Visit this group at http://groups.google.com/group/parboiled-user.
> To view this discussion on the web visit https://groups.google.com/d/msgid/parboiled-user/f632f620-274e-4214-973a-4f9851747a26%40googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.

[Andrew Bate]

Hi Jürgen,

I just saw that you posted about your Cheló̱na parser for turtle.

It looks very interesting. Do you have any idea how the performance compares against RDF4J (Sesame) or Jena?

Best wishes,

Andrew

[Juergen Pfundt]

Hi Andrew,

sorry, I did not do any performance comparisons against Sesame or Jena.

Cheló̱na handles RDF-formats in two phases. In the first phase the RDF-format is parsed using Parboiled2 and an AST is built. In the second phase the AST is interpreted and the selected output format is generated.

Parboiled2 itself is fast, but not as fast as a hand crafted parser. The architectural decision for a two phase processing has some impact on performance, too, even though parsing and AST interpretation are done asynchronously.

So my assumption is that Sesame and Jena might be faster than Cheló̱na.

May be, Cheló̱nas support of JavaScript is some compensation for the assumed performance deviation.

The current yet unpublished version has some modifications with respect to asynchronous parsing and AST evaluation. This results in some performance improvements compared to the published version.

In the upcoming version block size for parsing NTripels and NQuads can be adjusted from 1 to n lines. I can give you more details on this if you are interested.

Kind regards

Jürgen

Anfang der weitergeleiteten Nachricht:

-- 
You received this message because you are subscribed to a topic in the Google Groups "parboiled2.org User List" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/parboiled-user/evoGCzltJ3A/unsubscribe.
To unsubscribe from this group and all its topics, send an email to parboiled-use...@googlegroups.com.
Visit this group at https://groups.google.com/group/parboiled-user.
To view this discussion on the web visit https://groups.google.com/d/msgid/parboiled-user/80fc1175-1c05-4944-93ac-5eac71f19048%40googlegroups.com.