We have a 'beta' for some time now. There is no release yet, which has
to do with lack of resources.
Issues I feel that are show stoppers.
- DataType validation causes some problems with saving nodes which have
a 'confirm password' type field. E.g. the security editors of cloud
context security are not working properly (at least not of a clean
install).
To solve this I will have to finish the implementation of mm:form and
e.g. will need to make some additions in bridge.
When this is tested, and field type validation is working in generic
editors, the 'datatypes' project is tested sufficiently, and will be
proven to be useful. That's why I think it is important.
- The actual packaging of the distro is not yet checked for sanity. At
least not by me, but I'm certain that there are issues.
E.g.
- Which taglib jar must be shipped? (one provides all features, but
doesn't work in tomcat 4, the other one is more conservative). More
or less it was decided to ignore tomcat 4 compatibility. Still that
needs to be arranged in the build.
- Must rmmci be installed in the distro? I think the RMMCI package, does
contrary to other packages not contain a directory.
- I think that several packages ship 'INDEX' files not only in jars,
which should not be.
- Should the checksum field of images.xml be removed again.
- ... There must be more of these kind of issues, this is only what I
can come up with right now....
- Apps 1 import/export. Does it work properly? Also for new-style
builders and so on? This must be tested, and ensured that it works. I
don't think that it does. If it cannot be made to work perfectly, at
the very least this must be stated somewhere.
- Gui-values of enumerations. Seems simple to fix. DataTypes
related. Must work.
- Code-reviews. Could perhaps be considered as no show-stopper, because
I think the workings are tested. But there are many new classes in
e.g. org.mmbase.datatypes, org.mmbase.core, org.mmbase.cache,
org.mmbase.core.events
which have not been reviewed or judged by anyone besides the original
author(s). I feel that a situation arises in which we have to say 'speak
now, or remain silent for ever!'. Perhaps it is not wise to make major
changes in much code on these terrains, but perhaps it is not too late
to mark code on which people have doubts as 'experimental!' or as
'possibly subject to future changes!'
So, please, if you feel that you are able to have opinions on these
kind of matters, read (new) code, and complain.
- Open issues in the mmbase bugtracker. I don't say that all of them
should be considered show-stoppers, but I do think that all open
issues need judgement whether they are or not. There are about 200
open issues, show-stoppers must receive priority high,
non-show-stoppers must receive lower priority than high, if they are
as such now.
Of course issues may be fixed too..
- Other issues?
Some other things that I'm working on for the general benefit, and which
I personally like to have finished soon (and in 1.8.0).
- Fixes on HugeNodeListIterator
- CommandServer for image-magick. Using this would avoid the problems
arising from the fork done by the JVM for the Runtime.exec.
On behalf of, and btw more or less being, the release team,
Michiel
--
Michiel Meeuwissen mihxil'
Peperbus 111 MediaPark H'sum [] ()
+31 (0)35 6772979 nl_NL eo_XX en_US
_______________________________________________
Developers mailing list
[email protected]
http://lists.mmbase.org/mailman/listinfo/developers
