Alexandre Poitras wrote:
On 4/19/06, Jonathan Revusky <[EMAIL PROTECTED]> wrote:
Dave Newton wrote:
Jonathan Revusky wrote:
[...] rather than trying to modernize/refactor it forward
From a developer's standpoint I'm not even sure how I'd go about
refactoring the existing Struts 1.x codebase...
As an example, I've always been pissy about the Action being tightly
coupled to the servlet spec., returning something only useful in the
context of Struts 1.x, etc. (Who hasn't?! :)
Okay, how do I remove that dependency? Just thinking out loud, so parts
of this will be dumb.
- App functionality already moved into injectable objects
- Change request processor to put params etc. into an "action context"
(just a map?) rather than directly accessing HttpServletRequest (or its
wrapper)
- Change action processing to accept a different type of forward (just a
string, no-brainer?)
- Minor validation rework to accept POJOs rather than ActionForms
- Pluggable request processing/path handling
- ...etc.
I dunno... Sure looks a lot like stuff that already exists. I wouldn't
rewrite/refactor it either; I would start over.
Now, it might be feasible to write a wrapper around an already-existing
framework (or minorly tweaked one) to process existing struts
code/configs/Actions etc. and this might provide an incremental
migration path, but...
*shudder*
Well, there all these issues, and yeah, I guess they could make you and
other people shudder. I mean, among the existing Struts committers there
is no stomach apparently to try to do anything significant with the
Struts 1.x codebase and they just prefer to either start something new
(Shale) or bring in a previously competing codebase (Webwork) and work
on that.
But the real key point I am wondering about is this: if the existing
Struts developers have no plans for developing the Struts 1.x codebase,
what is the justification for not letting people who want to work on
that (independently of whether this reflects good taste on their part or
not) come in and work on it?
Well why don't you do it yourself and stop BUGGING the users of this
list so they can stop receiving your "How great I am" emails.
I have no idea what on earth you're talking about. I have no
recollection of singing my own praises in these emails and even if I
was, it would not make any arguments I am presenting any more or less valid.
Isn't it the case, Alexandre, that you are starting with this kind of
incoherent personal attack because your counter-arguments to what I am
saying are kind of thin? (At best...)
You
should really check out this Apache license which from what I heard
give you the right to evolve Struts yourself and this great site
called SourceForge.net which give you a free CVS repository access.
It is true that anybody could take the Struts 1.x codebase and fork it
and put it on sourceforge.net. However, the problem is that, once you're
talking about a non-canonical forked version, you are basically in the
position of shouting into the wilderness. The exact same work will only
get a fraction (a *very* small one) of the attention and usage that it
would if it were part of the canonical Struts project on apache.org.
If the Struts developers have a certain vision of how Struts 1.x should
evolve, and you disagree with them, then, okay, maybe you have to go
fork off your own version. But why should someone like Phil or anybody
else who maybe wants to do something with Struts 1.x have to fork off a
separate project on sourceforge when the Struts developers themselves
have no intention of doing *anything* with the Struts 1.x codebase anyway?
Surely you see my point, don't you? Have you thought about this?
Seriously, I have never seen a guy more in love with himself.
Well, if that's really the case, you should get out more.
But seriously, the above is irrelevant. It does not invalidate any
argument I have made. It does not reinforce any argument you have made
(if you have in fact made any argument, that is...). It is simply
ad-hominem drivel. Do you think it really makes sense to behave this way
in a professional forum? You know, there is a permanent electronic
archive of these discussions.
Jonathan Revusky
--
lead developer, FreeMarker project, http://freemarker.org/
Given the basic parameters of the situation, what would there possibly
be to lose?
Jonathan Revusky
--
lead developer, FreeMarker project, http://freemarker.org/
(But an interesting *shudder* I'll admit... *ponder*)
Dave
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
--
Alexandre Poitras
Québec, Canada
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
