Hello Pavel,

I'll take a look at those. Are these being ported to java for 
performance-reasons, or bootstrappability?

-Jesper

> On 07/06/2015, at 18.23, Дионисьев Павел <[email protected]> wrote:
> 
> Hello Jesper,
> 
> BTW, there is an ongoing (nearly-slow) work on Java-fication of that GSoC 
> code, as it was written in Groovy at the GSoC time. Part of that code is now 
> decomposed, and translated to Java, and maybe can be reused. 
> https://github.com/xSeagullx/antlrv4_groovy_grammar/tree/feature/migrate_to_java
> Basically, Statements and Expressions part has been translated.
> 
> Regards,
> Pavel.
> 
> 2015-06-07 17:25 GMT+03:00 Pascal Schumacher <[email protected]>:
>> Hi Jesper Steen,
>> 
>> great news. :)
>> 
>> Your approach sounds very good to me.
>> 
>> There is also a pull request to the pre-Apache git repo 
>> (https://github.com/groovy/groovy-core/pull/513) by the Google Summer of 
>> Code student who started the Antlr4 work. Maybe Jochen or somebody else can 
>> comment on how it relates to the Antlr4 work?
>> 
>> Kind regards,
>> Pascal
>> 
>> 
>>> Am 06.06.2015 um 01:07 schrieb Jesper Steen Møller:
>>> Hi list
>>> 
>>> I’m new on the list, and I’m volunteering to do some work on Groovy, 
>>> starting with the Antlr4 parser. I’m not completely new to language and 
>>> compiler implementation, I did a number of things on the Eclipse JDT for 
>>> Java 8, and also the Eclipse XPath 2.0 engine, and have worked with Antlr4 
>>> on work projects.
>>> 
>>> As stated on the issue GROOVY-5299, I’ve taken steps to make it actually 
>>> work with Groovy, in the proper package, but things are slow in the 
>>> beginning (I’m not really proficient in complex Gradle builds, for 
>>> instance), and I’m not really well aquatinted with the Groovy codebase.
>>> 
>>> I’d really appreciate any feedback I could get — I’ve set up a fork on 
>>> Github: https://github.com/jespersm/incubator-groovy/commits/antlr4
>>> I’m figuring that the next steps would be to fix the tests I had to comment 
>>> out, and the increase the test coverage and tackle the known weaknesses. 
>>> After that, I’d rebase the antlr4 branch with updates from ‘master’ and 
>>> work on tracking the changes (traits and other stuff),
>>> 
>>> Does that sound like a reasonable approach?
>>> 
>>> Kind regards,
>>> Jesper Steen Møller
> 
> 
> 
> -- 
> С уважением, Дионисьев П.А.

Reply via email to