Re: Survey for preparing FreeMarker development effort

Sunday, March 10, 2013, 5:00:06 PM, Johno Crawford wrote:

> Hi Daniel,
>
> I would like to see the following;
> 1.      FreeMarker source moved to GitHub
> °       This would simplify the work involved for developers and
>         maintainers to manage contributions / bug fixes and issues.

That I have already done a year ago or so. Back then I hoped that the
GIT branch will be released as 2.4.0 earlier, and then the 2.3.0
branch on sf.net SVN can be abandoned. Anyway, if everything goes as I
have planned, this will happen in the summer.

> 2.      Minimum JDK 5, or even 6 (which is now EOL)

Sure, the GIT branch already requires Java 5.

> °       This would greatly reduce the complexity of the code, some
> parts of the current code are riddled with JDK compat checks.
> °       Ability to use the concurrent package to improve performance!

If you are talking about the class introspection cache, this will
probably already happen in 2.3.20.

> 3.      Continue deploying artifacts to central Maven.

Of course. (Especially since, as I said, I plan to switch to Maven for
building. Plus I want to make this a multi-module Maven project, so
you can mark only the modules you really need as dependency. Not that
people complain about the size of the monolithic freemarker.jar... but
with a Maven build I have to do that, because FM has to support
multiple versions of Servlet, Jython, etc., and then why not do it
properly.)

-- 
Best regards,
 Daniel Dekany


> Cheers,
>
> Johno
>
> On 10/03/2013 2:06 PM, Daniel Dekany wrote:
> Dear All,
>
> While I'm not one of the original authors of the 2.x.x (non-classic)
> line, seems I have gradually become the main maintainer of the
> FreeMarker template engine over the last few years or so, as the
> original authors (I mostly mean Jonathan and Attila) are inactive
> around here. Only I couldn't allocate enough time for this either, and
> this starts to become a problem for FreeMarker.
>
> So, to catch up, I plan to allocate serious amount of time [*] for
> FreeMarker development stating from late spring or early summer. The
> goal of this effort is fixing long-standing issues (mostly design
> issues that can be addressed without breaking backward compatibility
> too much), adding frequently requested missing features (an those that
> people don't request but I know they need it...   ), increasing
> usability (better error messages, better documentation, terminology
> simplification) and to make FreeMarker more accessible for potential
> contributors (fixing the branching mess, switching to Maven, Git,
> modularization, addressing ownership/licensing and CLA issues).
>
> Thus, as the first step, I would like to hear about what the wishes of
> you as FreeMarker user are. What features you crave for the most, how
> could FreeMarker integrate better with your product, what are the
> things you or your fellow developers most often has problems with,
> etc. (Of course, I do have a long list of things to fix/improve, but I
> don't want to influence the discussion at this point.)
>
> *: How much time, you might ask. I'm thinking of a few hundred hours
>    here. Yeah, that's bold. What life will allow (work, family, etc.),
>    is another question. We will see.
>


------------------------------------------------------------------------------
Symantec Endpoint Protection 12 positioned as A LEADER in The Forrester  
Wave(TM): Endpoint Security, Q1 2013 and "remains a good choice" in the  
endpoint security space. For insight on selecting the right partner to 
tackle endpoint security challenges, access the full report. 
http://p.sf.net/sfu/symantec-dev2dev