[Uml-devel] Thoughts on Umbrello 1.3

[Jonathan Riddell]

Dearest Umbrello and KDE Developers,

KDE has been branched for 3.2 and Umbrello UML Modeller 1.2 will soon
be released with it.  My impression of Umbrello 1.2 is that it is a
good release with features that were defiantly lacking in the previous
version.  Developing as part of KDE has been a good thing, our release
cycle has matched well and we get quite a bit of development,
translation and packaging support as well as publicity with it
although I hope it doesn't put off non-KDE users from using Umbrello.
Two commercial UML tool have bought Google Adwords for 'Umbrello'
which affirms in my mind that we compete with the commercial
competition (and in many ways surpass it).  We have gained a couple of
top class developers in the last year and a few equally important
beastie reporters.  My most proud achievement is getting Americans to
use the word 'beastie' :)

Unfortunately 1.2 has far too many bugs in it, many of them crashes.
Cleaning up the unconfirmed beasties today it crashed several times.
Worse, they're hard to repeat and harder to track down type crashes
(which is why I didn't report them).  Changes to associations and the
code generators seem to be the main culprit.  I gather that the rules
for KDE 3.2 branch are the same as the recent deep freeze: fixes to
major beasties only and no i18n() changes.  So any fixes to major
problems should be done in HEAD and then merged back into 3.2
carefully (cvs -j is rumoured to be good for this, ask if you have any
problems).

Onto Umbrello 1.3.  Bartko pointed out the choices to me:

a) improvement of Umbrello 1.2
b) rewrite of Umbrello 1.2
c) merge with Umbrello2
d) something else?

I strongly suggest the first choice.  Umbrello 1.2 is good and
maintainable so we should maintain it and improve it.  There's plenty
to be improved and added.

Firstly if you have not tried Umbrello before please do so briefly
when you install KDE 3.2 and comment on what you think is unintuitive.

One of the most significant non-crash problems is that the direction
of new compositions changes.  This is probably a candidate for
backporting into 3.2 http://bugs.kde.org/show_bug.cgi?id=72615

I hesitate to mention associations because I'm not honestly sure what
the current status is with them (same with packages).  I suspect more
work may be needed to clean them up, probably showing the associations
of a class/whatever in the list view.  You should be able to see a
list of all the associations when you look at the class properties,
currently you can only see the associations on the current diagram.

We may also have to turn off fancy code generation by default, I
haven't tried Umbrello with large diagrams recently but
http://bugs.kde.org/show_bug.cgi?id=72644 [Large (Huge?) memory leak]
indicates that it may be taking up more memory than is fair.

When you create an item on the canvas the dialogue should give you the
option of using an existing class/interface/whatever and list the
packages it could go in.

Code import for more languages features in bugs.kde.org but it may be
best to tidy up the code generators we do have.  For example why does
the C++ code generator not add a variable for a unidirectional
association?

The refactoring agent probably needs some work to make it more user
friendly.

Adding What's This help would be good.

I like the icons we have in the toolbar, they're clean, simple and
useable.  The ones with English text in them should be changed
though.  Icons for diagrams can't be too hard (significant widget for
that diagram type, make them coloured, add a border to indicate
canvas).

I lost track of the printing which never got the work we hoped for.
If we could get SVG export of course that would be excellent.

Class diagrams should allow Objects in them (apparently, says the UML
spec).  This allows for object diagrams (which are just class diagrams
without the classes).

Automatic tests as suggested in
http://bugs.kde.org/show_bug.cgi?id=70256 would be useful.

Better error messages on a failed paste and file open are something
I've wanted for a while.

The oft-requested KDevelop integration is probably a long way off.  It
would be nice if done properly but is probably too much work.

QCanvas works well for us as a canvas but anti-aliasing and SVG export
would be nice.  I suspect it's still the best going at the moment.

Standards compliance with XMI is happening now which is great.
Eventually we may need to look at UML 2 but probably not for this
release.

Those are the issues on my mind today but there's plenty of other
entries marked 'Umbrello' in bugs.kde.org and Seb reported that
valgrind gave 30,000 errors when Umbrello was run over it so if anyone
is looking for a way to begin a KDE programming career this is your
opportunity!

Finally if anyone feels they would do a better job than me with the
title of project admin do say so.  I've not been the most active
developer in the last few months (having left university) but I do
make an effort to ensure beastie reporters are thanked, all e-mail is
replied to and the website is up to date.  Others do that as well
though so maybe I should drop the title.

Thanks and guid nicht.