Hi Torsten,
Sorry for not following up but I was away for a JSR-301 F2F.
You proposed earlier to provide a new patch to restore the unresolved TODO and
FIXMEs, but rolling back and only applying the real changes should be fine as
well.
And we definitely need a proper list of outstanding issues, especially those related to the JSR-286 RI, so if you can provide that based on your own information
*as well* as the remaining TODO and FIXMEs that would be nice.
Regards,
Ate
Torsten Dettborn wrote:
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hi Ate,
because the discussion about this has stopped, here my plan, (I need
half a day and than I'm ready)
I roll the patch back, send a new patch without deleting undone TODO and
Fixme and send a list for this items.
For this list we have to compare it with the jira issues.
Hopefully this is ok for everybody
Torsten
[EMAIL PROTECTED] schrieb:
| I'm currently not able to check on these changes as I'm at a JSR-301 F2F
| and can't get my laptop to connect to the internet (I'm typing this
| response at a public booth) which it looks like will remain so the next 2
| days :(
|
| Anyway, if this is all true, I fully agree these changes need to be rolled
| back. Its clear there are areas in Pluto 2.0 trunk which needs fixing and
| implementing before it can be considered usable and more or less complete.
| Simply ripping out FIXME and TODO comment won't bring us any closer to
| that, on the contrary...
|
| My suggestion is th following:
| - rollback these changes
| - make a list of all *must* have issues/FIXMEs
| - make a list of all *nice* to have TODOs
| - if need be, discuss these on the dev list first
| - create JIRA issues for all the above, making (at least) the first list
| items required for the 2.0 container release
|
| Regards,
|
| Ate
|
|> Torsten,
|>
|> I very much appreciated the work you and your group has done to create
|> Pluto 2.0, but I was very disappointed in many of the commits you did
|> today. Rather than fixing most of the TODOs and FIXMEs, you just erased
|> them. Some of the TODOs and FIXMEs are inconsequential and not necessary
|> and should be erased and some FIXMEs are not that important and should be
|> changed to TODOs. But many of these should have been fixed before the
|> FIXME
|> or TODO was removed.
|>
|> Here are a few specific comments:
|> 1. Many of the TODOs are those automatically added by Eclipse and are
|> labelled "Auto-generated catch block" and include a printStackTrace()
call
|> to stdout. This should have been changed to log to Pluto's log file
|> including a relevant message and the exception so the stack trace will
|> appear in the Pluto log. Consideration should also be made to rethrow the
|> exception so it is propagated up the stack trace to the user.
|> 2. The FIXME in SupportedModesServiceImpl.isPortletManagedMode()
refers to
|> the fact that this method is not properly implemented and had notes
on how
|> to implement the method. You just erased the FIXME and notes rather than
|> fixing the method as detailed in the FIXME comment.
|> 3. In PortletURLTag286.doStartTag(), rather than implementing the FIXME
|> suggestion, you just erased the FIXME and the suggested code that was
|> commented out.
|>
|> I think the commits where FIXMEs and TODOs are just erased should be
|> rolled
|> back except in those cases where the FIXME or TODO does not add value or
|> the issue pointed out by the FIXME or TODO was fixed. Does anybody else
|> have this opinion or am I making too much out of nothing here?
|>
|
|
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFHkKZl0Ji0BqEIlIURArsPAJ4x4lT1Jc7B9yyr+omCO7LS3qsmNQCdE7s4
tm5Zi/SZOcc0/42f9OG0Nd4=
=60sh
-----END PGP SIGNATURE-----
