Sorry. I feel bad for the original poster, I'm the one that submitted
the patch thinking that FormSupport was an internal class only used by
Form, so I guess I'm the root cause for David's bad day at work.. I know
how I hate it when the rug gets pulled out from under me without
expecting it.
So I probably should have left the old signature there and marked it
deprecated.. shrug, relatively easy thing to do..
But the old signature was buggy.. so should we be supporting a buggy
interface.. or just pull the bandaid really fast and deal with the pain
as soon as possible.. open ended philosophical discusson.. but yes this
was a tiny minor release, it shouldn't have given anyone any source of
headaches.. sorry.. I'll see if I can build a patch to expose the old
signature deprecated.
Brian K. Wallace wrote:
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Personal opinion on this type of issue - hit us over the head! For a
bug-fix release, contractual interfaces shouldn't break - nor should
implementation details. For feature releases (.X and X.) there should at
least be notable mention and migration path, but those are where these
types of changes should appear - not bug fix releases.
I do feel that these two releases were vitally important due to the
length of time since the last releases as well as the number of fixes.
No excuse tho' - I'm personally +1 to a 4.0.2 to fix this (and, possibly
non-intrusive fixes such as those pesky validation strings ;-))
That said - I'd also bow to Jesse on the final decision as it's in the
4.X realm (and I don't have access to actually perform the release yet).
Brian
Leonardo Quijano Vincenzi wrote:
Jesse, but this is not a Tacos issue. I mean, even if Tacos didn't have
a duplicated Form component, Tapestry should take care on its own of not
breaking the API on small bugfix releases.
Rather, the issue is about Tapestry not having a deprecated or
overloaded component that fits the old interface.
But I know nobody wants to hit you in the head for doing this. You and
Brian have been doing a great job, actually. Rather, what I suggest is
to fix the Tapestry interface and do a small release. The question is:
is it worth it?
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.5 (MingW32)
iD8DBQFENeGeaCoPKRow/gARAnC5AJ9PlO9Q7rjkZGccDBa2upDFHAVSjQCg+Lfs
ngDk2QlpHb8LRR7a2QmFecA=
=ANKm
-----END PGP SIGNATURE-----
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
