Hi Rafael,

that's interesting but still needs some manual writing ;). That said
the organisation could be the file system one and
http://tomee.apache.org/examples/ could be generated easily, good
point.

My main concern and why I thought starting manually was good was to
avoid to just "cat" all sources in a page and expose it. This doesn't
promote why the sample has been written so it is like not having it.

Finally with javaeeX-samples (think the 8 is on his way) wonder if we
shouldn't just merge our portable examples and only keep proprietary
ones moving tests of other examples to our main tests.

Any opinion on that points (rewriting them to make it obvious):

- removing portable examples and ensuring they are in javaee-sample
initiative without loosing in test coverage (if that's the case)
- avoid to generate examples without documentation
- reorganize the structure to match the category (we should surely
rework it to have spec + proprietary features + tests + tools as
subparts)


Romain Manni-Bucau
@rmannibucau |  Blog | Github | LinkedIn | Tomitriber


2016-03-21 21:38 GMT+01:00 Rafael Pestano <rmpest...@gmail.com>:
> Hi Romain,
>
> Nice work, just two ideas about how the examples could be generated:
>
> 1 - read from the tests as hibernate user guide is doing¹. Here is the guide
> <https://github.com/hibernate/hibernate-orm/tree/master/documentation/src/main/asciidoc/userguide>
> and the tests
> <https://github.com/hibernate/hibernate-orm/tree/master/documentation/src/test/java/org/hibernate/userguide>
> .
> 2 - read from a github repo and parse sources as javaee-support² is doing.
> Here are the sources <https://github.com/javaee-samples/javaee7-samples>
> and here
> <https://github.com/javaee-samples/javaee-samples.github.io/blob/develop/_ext/asciidocify.rb>
> is how they are parsed.
>
>
> [1] http://hibernate.org/validator/documentation/getting-started/
> [2] http://javaee.support/
>
>
> 2016-03-21 13:54 GMT-03:00 Jean-Louis Monteiro <jlmonte...@tomitribe.com>:
>
>> I'll have a deeper look tonight Romain.
>> Thanks for putting more content in there. Might be useful to see how it
>> renders.
>>
>> --
>> Jean-Louis Monteiro
>> http://twitter.com/jlouismonteiro
>> http://www.tomitribe.com
>>
>> On Mon, Mar 21, 2016 at 1:38 PM, Romain Manni-Bucau <rmannibu...@gmail.com
>> >
>> wrote:
>>
>> > Hi guys,
>> >
>> > pushed some more content and GUI fixes. What about deploying it live
>> > on tomee.apache.org/site-ng/?
>> >
>> > Main missing part ATM is the example page but not sure how to tackle
>> > it. Think it should be a manual task cause anything generated either
>> > doesn't render well or doesn't serve the end users very well in term
>> > of content. Can try to start hacking few of them or if anyone wants to
>> > join the website hacking he is very welcomed.
>> >
>> >
>> > Romain Manni-Bucau
>> > @rmannibucau |  Blog | Github | LinkedIn | Tomitriber
>> >
>> >
>> > 2016-03-17 19:57 GMT+01:00 Romain Manni-Bucau <rmannibu...@gmail.com>:
>> > > Tried to push current state/idea to avoid you to have to build it
>> > > locally: http://home.apache.org/~rmannibucau/tomeeng/#
>> > >
>> > > Romain Manni-Bucau
>> > > @rmannibucau |  Blog | Github | LinkedIn | Tomitriber
>> > >
>> > >
>> > > 2016-03-17 11:57 GMT+01:00 Robert Panzer <rpan...@tomitribe.com>:
>> > >> Hi,
>> > >>
>> > >> even though not a committer I’d like to give my 2 cents on this:
>> > >>
>> > >> Regarding a GH* based workflow:
>> > >>
>> > >> Having provided some updates to the documentation recently I think the
>> > current process does not really promote contributions and collaboration.
>> > >> For example I did not get any notification from the CMS that my
>> > proposal was received and in fact it wasn’t received and I had to attach
>> > svn patches to Jira tickets.
>> > >> Nor is there any possibility for review and discussion afterwards.
>> > >>
>> > >> So I am strongly for a Github Pull Request-alike workflow, where
>> > everyone can actively search and discuss contributions.
>> > >> Ideally this workflow should be lightweight enough that you could
>> > propose an update to the documentation after as a user you discovered
>> > something that is not yet documented.
>> > >> As a supporter it would make sense to update the documentation when
>> you
>> > answered a question that was not obvious just by pasting the interesting
>> > parts out of your email response.
>> > >> Most often this is the best documentation: To the point and it solves
>> a
>> > concrete problem.
>> > >>
>> > >>
>> > >> Regarding a JBake based solution:
>> > >>
>> > >> The current documentation is completely based on Markdown, which makes
>> > it kind of a lottery how the final output will look like.
>> > >> The update I proposed looked completely different on my machine than
>> > finally on the website, spaces were added to code snippets where they
>> don’t
>> > belong, links get sometimes rendered propery, sometimes not.
>> > >> Being a member of the AsciidoctorJ developers I certainly appreciate
>> > having support for Asciidoctor via JBake as well.
>> > >> JBake still supports Markdown and plain HTML as well.
>> > >>
>> > >> Cheers
>> > >> Robert
>> > >>
>> > >>
>> > >> * Don’t nail it down to Github, could be something else that provides
>> a
>> > similar workflow.
>> > >>
>> > >>
>> > >> Am 16.03.2016 um 19:56 schrieb Romain Manni-Bucau <
>> > rmannibu...@gmail.com>:
>> > >>>
>> > >>> Hi guys,
>> > >>>
>> > >>> trying to work on the website ATM, created a placeholder project on
>> my
>> > >>> github to share the idea:
>> https://github.com/rmannibucau/site-tomee-ng
>> > >>> (mvn jbake:inline then go on http://localhost:8080).
>> > >>>
>> > >>> Idea is:
>> > >>>
>> > >>> - get a more modern website
>> > >>> - restructure the doc to be more hierarchic and browsable
>> > >>> - get rid of the outdated doc
>> > >>> - make it easier to PR on github
>> > >>>
>> > >>> If encouraged I would like to still use the CMS as storing/publishing
>> > >>> solution but not generation (the edit feature is broken and not that
>> > >>> user friendly when you are not a committer and when you are you don't
>> > >>> really need).
>> > >>>
>> > >>> wdyt?
>> > >>>
>> > >>> Romain Manni-Bucau
>> > >>> @rmannibucau |  Blog | Github | LinkedIn | Tomitriber
>> > >>
>> >
>>
>
>
>
> --
> <http://www.advancedit.com.br/>Att,
>
> Rafael M. Pestano
>
> Desenvolvedor Java Cia. de Processamento de Dados do Rio Grande do Sul
> http://rpestano.wordpress.com/
> @realpestano <https://twitter.com/realpestano>

Reply via email to