You are viewing a plain text version of this content. The canonical link for it is here.

Posted to user@uima.apache.org by Renaud Richardet <re...@gmail.com> on 2015/04/28 16:36:55 UTC

Ruta syntax simplifications?

Hello,

I have been using Ruta a lot lately, and it has made me much more
productive. Thanks to all the people that contributed to it!

I have been thinking about two small syntax simplifications and was
wondering if they would make sense:

A) Types declarations (2.5.1)

// Types with features may add a parent type in their declarations.

// *If no parent type is specified, Annotation is used as the implied
default parent type. *


// before

DECLARE Annotation MyAnnotation(STRING myFeature);

// after

DECLARE MyAnnotation(STRING myFeature);



B) Matching condition

// The matching condition of the rule element refers to the complete
document,

// or more specific to the annotation of the type “DocumentAnnotation”,
which covers the whole document.

// *If no matching rule is specified, the whole document
(DocumentAnnotation) is implicitely used. *


// before

Document{-> MARKFAST(Animal, 'Animals.txt')};

// after

MARKFAST(Animal, 'Animals.txt');



What do you think? Do they make sense? Would that introduce some
inconsistencies / ambiguity in the Ruta language? Would the implementation
be hard?

All the best, Renaud

Re: Ruta syntax simplifications?

Posted by Peter Klügl <pe...@averbis.com>.

btw, I will commit something to the documentation in the next hours. I 
am metioning this just to avoid merge conflicts.

Best,

Peter

Am 29.04.2015 um 13:18 schrieb Peter Klügl:
> Hi,
>
> that would be great if you can create a patch for the documentation.
>
> Best,
>
> Peter
>
> Am 29.04.2015 um 08:04 schrieb Renaud Richardet:
>> Hallo Peter,
>>
>> Can you create a jira issue for it? I will take care of it then.
>> Excellent, here we go:
>> * https://issues.apache.org/jira/browse/UIMA-4368
>> * https://issues.apache.org/jira/browse/UIMA-4369
>>
>> PS: You can of course also submit a patch for it, if you want :-)
>>
>>
>> I don't feel up to it at this point, I don't know enough of the Ruta
>> codebase. However, here is how I can help:
>> * I refactored the tests with the new syntax, and attached patches to 
>> the
>> above jira issues
>> * I would commit to refactoring the documentation if you like
>>
>> Best, Renaud
>>
>

Re: Ruta syntax simplifications?

Posted by Peter Klügl <pe...@averbis.com>.

Hi,

that would be great if you can create a patch for the documentation.

Best,

Peter

Am 29.04.2015 um 08:04 schrieb Renaud Richardet:
> Hallo Peter,
>
> Can you create a jira issue for it? I will take care of it then.
> Excellent, here we go:
> * https://issues.apache.org/jira/browse/UIMA-4368
> * https://issues.apache.org/jira/browse/UIMA-4369
>
> PS: You can of course also submit a patch for it, if you want :-)
>
>
> I don't feel up to it at this point, I don't know enough of the Ruta
> codebase. However, here is how I can help:
> * I refactored the tests with the new syntax, and attached patches to the
> above jira issues
> * I would commit to refactoring the documentation if you like
>
> Best, Renaud
>

Re: Ruta syntax simplifications?

Posted by Renaud Richardet <re...@gmail.com>.

Hallo Peter,

Can you create a jira issue for it? I will take care of it then.
>

Excellent, here we go:
* https://issues.apache.org/jira/browse/UIMA-4368
* https://issues.apache.org/jira/browse/UIMA-4369

PS: You can of course also submit a patch for it, if you want :-)


I don't feel up to it at this point, I don't know enough of the Ruta
codebase. However, here is how I can help:
* I refactored the tests with the new syntax, and attached patches to the
above jira issues
* I would commit to refactoring the documentation if you like

Best, Renaud

Re: Ruta syntax simplifications?

Posted by Peter Klügl <pe...@averbis.com>.

Hi,

(the mailing lists obiously have some problems right now (at least 
commits and dev). Let's see when this mail arrives.)

can you take a look at the current trunk and resolve the issues in case 
everything works as expected?

Best,

Peter

Am 28.04.2015 um 16:49 schrieb Peter Klügl:
> Hi,
>
> short answer: It makes sense :-)
>
> There are maybe some problems in the grammars concerning types and 
> variables vs external actions, but I think it's worth the trouble.
>
> Can you create a jira issue for it? I will take care of it then.
>
> Best,
>
> Peter
>
> PS: You can of course also submit a patch for it, if you want :-)
>
>
> Am 28.04.2015 um 16:36 schrieb Renaud Richardet:
>> Hello,
>>
>> I have been using Ruta a lot lately, and it has made me much more
>> productive. Thanks to all the people that contributed to it!
>>
>> I have been thinking about two small syntax simplifications and was
>> wondering if they would make sense:
>>
>> A) Types declarations (2.5.1)
>>
>> // Types with features may add a parent type in their declarations.
>>
>> // *If no parent type is specified, Annotation is used as the implied
>> default parent type. *
>>
>>
>> // before
>>
>> DECLARE Annotation MyAnnotation(STRING myFeature);
>>
>> // after
>>
>> DECLARE MyAnnotation(STRING myFeature);
>>
>>
>>
>> B) Matching condition
>>
>> // The matching condition of the rule element refers to the complete
>> document,
>>
>> // or more specific to the annotation of the type “DocumentAnnotation”,
>> which covers the whole document.
>>
>> // *If no matching rule is specified, the whole document
>> (DocumentAnnotation) is implicitely used. *
>>
>>
>> // before
>>
>> Document{-> MARKFAST(Animal, 'Animals.txt')};
>>
>> // after
>>
>> MARKFAST(Animal, 'Animals.txt');
>>
>>
>>
>> What do you think? Do they make sense? Would that introduce some
>> inconsistencies / ambiguity in the Ruta language? Would the 
>> implementation
>> be hard?
>>
>> All the best, Renaud
>>
>

Re: Ruta syntax simplifications?

Posted by Peter Klügl <pe...@averbis.com>.

Hi,

short answer: It makes sense :-)

There are maybe some problems in the grammars concerning types and 
variables vs external actions, but I think it's worth the trouble.

Can you create a jira issue for it? I will take care of it then.

Best,

Peter

PS: You can of course also submit a patch for it, if you want :-)


Am 28.04.2015 um 16:36 schrieb Renaud Richardet:
> Hello,
>
> I have been using Ruta a lot lately, and it has made me much more
> productive. Thanks to all the people that contributed to it!
>
> I have been thinking about two small syntax simplifications and was
> wondering if they would make sense:
>
> A) Types declarations (2.5.1)
>
> // Types with features may add a parent type in their declarations.
>
> // *If no parent type is specified, Annotation is used as the implied
> default parent type. *
>
>
> // before
>
> DECLARE Annotation MyAnnotation(STRING myFeature);
>
> // after
>
> DECLARE MyAnnotation(STRING myFeature);
>
>
>
> B) Matching condition
>
> // The matching condition of the rule element refers to the complete
> document,
>
> // or more specific to the annotation of the type “DocumentAnnotation”,
> which covers the whole document.
>
> // *If no matching rule is specified, the whole document
> (DocumentAnnotation) is implicitely used. *
>
>
> // before
>
> Document{-> MARKFAST(Animal, 'Animals.txt')};
>
> // after
>
> MARKFAST(Animal, 'Animals.txt');
>
>
>
> What do you think? Do they make sense? Would that introduce some
> inconsistencies / ambiguity in the Ruta language? Would the implementation
> be hard?
>
> All the best, Renaud
>