groovy-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Owen Rubel <oru...@gmail.com>
Subject Re: Groovydoc and java code
Date Mon, 09 Nov 2015 16:07:31 GMT
I've talked about this alot. The easiest way to do this is through OPTIONS.

Implement the request.method OPTIONS for all endpoints so each endpoint can
send back its information. Then at runtime, you can cache this information
and show it per AUTHORITY/ROLE in your apidocs view.

I've already implemented this in the old Grails API Toolkit; you can
download and look at the source code I believe.


Owen Rubel
415-971-0976
orubel@gmail.com

On Mon, Nov 9, 2015 at 7:54 AM, Dzmitry Kazimirchyk <dkazimirchyk@gmail.com>
wrote:

> Hi,
>
> We are trying to build API documentation for a mixed source java/groovy
> project. We see groovydoc tool from standard groovy distribution as a good
> fit for our needs since it supports doc generation from both java and
> groovy sources and also allows us to use groovy templates which is very
> useful for building our heavily customized docs.
>
> However we hit a few obstacles related to the ability of groovydoc to
> handle java sources:
>
> 1. It fails to parse files containing Java 8 specific syntax;
> 2. It also fails to parse files containing Java 7 "diamond" operator;
> 3. Javadocs for individual enum constants from java classes don't make it
> into final GroovyClassDoc metadata and hence into the docs (this works fine
> for groovy sources though).
>
> We can mostly live with not having Java 8 syntax support for now, and work
> around Java 7 diamond issue, but having docs for enum constants is
> absolutely vital.
>
> In general I wonder if we are on a right track here using groovydoc. Is it
> supposed to handle java (mixed java/groovy) sources? If so whether support
> for Java 7 and 8 syntax is coming soon?
>
> Also should I create issue in JIRA regarding the java enum docs problem?
>
> I have tried to wrap my head around the parser logic and found that it
> uses old ANTLR v2 grammar. So I wonder if it is even worth looking into
> trying to figure out and fix this ourselves or there are plans on moving to
> ANTLR v3 or some other parser tool in the near future?
>
>
> Thanks for your help,
> Dima
>

Mime
View raw message