Again this night hudson failed to build VerveineJ :-(
---
Error Message
Forked Java VM exited abnormally. Please note the time in the report does not reflect the time until the VM exit.
Stacktrace
junit.framework.AssertionFailedError: Forked Java VM exited abnormally. Please note the time in the report does not reflect the time until the VM exit.
---
nicolas
Hi Doru,
I just tried to load Moose (default version) in a PharoCore 1.2.2a image and I got an error when loading Glamour.
The package Glamour-EyeSee-Presentations is not found in the repository.
Veronica
Status: New
Owner: ----
Labels: Type-Defect Priority-Medium
New issue 580 by alexandr...(a)gmail.com: DNU in EyeSee
http://code.google.com/p/moose-technology/issues/detail?id=580
I just tried it in a Pharo 1.2 and got 11 errors and 1 failure among the 13
tests.
Apparently #displayString is used a bit everywhere. This method does not
exist in Pharo.
After having added #displayString on Object, I got other problem with the
examples: #any is not understood by a collection, it is invoked in
Collection>>minValue.
After changing #any by #anyOne, I was able to get some nice graphics. There
are other bugs from here, but I will report them after....
Hello
What's the status of Java annotation import with infusion/VerveineJ? I have seen some mails in the past weeks but didn't take a close look.
Now I would like to use such information to draw dependencies linked to Spring annotations (for dependency injection).
I tried both infusion and verveinej on the sample petclinic app for Spring but both returned a minimalistic set of annotation types (Override, SuppressWarnings, XmlElement, and XmlRootElement)
Any idea what is going wrong?
--
Simon Denier
Status: New
Owner: tudor.gi...(a)gmail.com
Labels: Type-Defect Priority-Medium Component-Glamour Milestone-4.4
New issue 583 by tudor.gi...(a)gmail.com: Add EyeSeePresentation to Glamour
http://code.google.com/p/moose-technology/issues/detail?id=583
We need an EyeSeePresentation to embed an EyeSee diagram in Glamour.
Updates:
Status: Fixed
Labels: Milestone-4.4 Component-EyeSee
Comment #5 on issue 155 by tudor.gi...(a)gmail.com: Port EyeSee to Pharo
http://code.google.com/p/moose-technology/issues/detail?id=155
It is working quite well now. We can close this issue and open more
specific ones if needed.
Status: New
Owner: ----
Labels: Type-Defect Priority-Medium
New issue 540 by vuqui...(a)vub.ac.be: Glamour: removing dependency in
configuration - core group
http://code.google.com/p/moose-technology/issues/detail?id=540
In the configuration of Glamour, the core group was loading only Glamour
packages, and the extras group: mondrian and magritte presentations. This
allowed to load those presentations if needed.
Currently, the core group is also depending on Mondrian. The method
MOViewRenderer>>openWithStatusbar is causing this dependency in the package
Glamour-Scripting.
Can this dependency be removed from the group core?
* Type-Enhancement
* Component-Glamour
Hi,
The GLMUITheme changed:
- no more color fading, except for selected buttons and tabs
- more contrast between grays
- fixed the issue with lighter background color for inactive windows
This resulted in a simpler and slightly faster look.
Cheers,
Doru
--
www.tudorgirba.com
"Being happy is a matter of choice."
Updates:
Status: Accepted
Labels: -Priority-Medium Priority-High
Comment #4 on issue 446 by tudor.gi...(a)gmail.com: uml class shape aligned
the rectangles
http://code.google.com/p/moose-technology/issues/detail?id=446
A form of the problem is still around. This is an important issue that
prevents us from actually using the FormsBuilder.
Here is an example with a ChildrenShape inside. The green shape on top
should have a constant minimum size. Instead its height increases together
with the height of the children shape below:
| builder |
builder := MOFormsBuilder new.
builder column; fill.
builder row; fill; row; fill.
builder x: 1 y: 1 add:
(MORectangleShape new width: 30; fillColor: Color green).
builder x: 1 y: 2 add: (MORectangleShape new).
builder x: 1 y: 2 add: (MOChildrenShape new).
view shape: builder shape.
view nodes: (1 to: 5) forEach: [:each |
view nodes: (1 to: each).
view verticalLineLayout ]
Status: New
Owner: tudor.gi...(a)gmail.com
Labels: Type-Defect Priority-Medium Component-Kumpel Component-MooseCore
Milestone-4.4
New issue 578 by tudor.gi...(a)gmail.com: Add Kumpel to ConfigurationOfMoose
http://code.google.com/p/moose-technology/issues/detail?id=578
Just that.
Hi!
Since some of you are working on remodularizing code, maybe the following may inspire you.
In Mondrian, the tests are badly structured. I actually wrote a large part of them as a way for me to understand the initial version of Mondrian, back in 2007-2008. Most of the tests are contained in the class MOViewRendererTest. They actually contain a lot of duplicated scenario (not really code): I open a canvas, put some nodes and edges, run the layout...
I would like to cut this class into smaller unit tests. Fixture will have to be found for each new unit tests. How to structure them? Along the class hierarchy? Along the features of Mondrian? Along the features commonly used in Mondrian scripts (which is how they are roughly structured)?
I can remodularize my tests and come up with a satisfactory design in a couple of hours. But maybe someone worked on a tool that would help me on this.
Cheers,
Alexandre
--
_,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
Alexandre Bergel http://www.bergel.eu
^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
Hi guys
now that we have announcements it should be easy to have a code spy:
a window that when I browse any things in moose (being it a mondrian node, or
a whatever stuff announcement event) I could see in the spy window the code
of the entity.
This way I could move hyperfast my mouse on a nice mondrian view and get immediately
fats code overview wihtout having to click, spawn or wahtever action.
Do you like the idea?
Because I always got frustrtaed with visualaition because I could not get that behavior
in addition to the visualization.
Stef
Hello,
Last year I ported some packages about 'Simulated Annealing' from
VisualWorks.
I am not sure what this is about and what does 'Simaulated Annealing' mean
exactly, But I recently remembered that I had it in my repository with all
tests green at least :)
So in case it can be usefull or it can interest somoene. I just put them in
a new repository:
www.squeaksource.com/SimulatedAnnealing
Status: New
Owner: ----
Labels: Type-Defect Priority-Medium Maintainability Milestone-4.4
New issue 575 by tudor.gi...(a)gmail.com: Filesystem does not load
http://code.google.com/p/moose-technology/issues/detail?id=575
The Filesystem package is now empty and its contents have been moved to
sub-packages. Also, the new repository is squeaksource.com/fs
The problem is that ConfigurationOfFilesystem is not updated, so we first
need to fix this and then add it to the ConfigurationOfMoose.
Hi,
I noticed that Filesystem changed significantly recently both internally and in package structure (I guess it was Colin - the user says cwp). This led to some red tests in Moose that I now fixed. I am writing here to document these changes in case other people are using this nice library.
>From a package point of view Filesystem has now a cool structure with basically each category being in a separate package. Furthermore, the original Filesystem is now empty which might create problems if people are loading it like we do in Moose.
To get it loading, I created a new baseline (2.0.1-baseline) in the ConfigurationOfFilesystem.
There were also some refactorings at the level of the model. Now, FSFilesystem does not have subclasses anymore (e.g., FSDiskFilesystem does not exist anymore). Instead there is a composition with a Store, so you will have FSDiskStore.
Thus, if before you used something like "FSDiskFilesystem current", now you will have to use "FSDiskStore currentFilesystem"
That was it.
Everything is green now in Moose land.
Cheers,
Doru
--
www.tudorgirba.com
"Next time you see your life passing by, say 'hi' and get to know her."
Hi,
There are some cool things happening around Moose these days:
- EyeSee is almost fully working in Pharo thanks to the port of Andre, and it is now in the ConfigurationOfMoose
- Glamour is on the way to receive a Wizard and dialog ability thanks to the work of Cyrille
- Nicolas is doing a nice job on VerveineJ
- Alex and Vanessa improved significantly the tests of Moose
- Alex improved Mondrian to better render complex shapes
- Igor and Ben are working on the basic Pharo widgets which will boost the Glamour capabilities
- Fabrizio is working on parsing SQL
- I worked on Metanool
- I extended the Finder was with several views like the DSM, or the dependencies, and I worked on improving the bookmarking behavior
I might have missed something, but I really like this rhythm. Let's keep it :)
Cheers,
Doru
--
www.tudorgirba.com
"Be rather willing to give than demanding to get."
Status: New
Owner: ----
CC: alexandr...(a)gmail.com
Labels: Type-Defect Priority-Medium Component-Mondrian Milestone-4.4
New issue 572 by tudor.gi...(a)gmail.com: Nodes with complex shapes do not
move with the parent node
http://code.google.com/p/moose-technology/issues/detail?id=572
Try the following code and then try to drag and drop the top node.
view node: 'hello' forIt: [
view shape umlClass.
view nodes: (1 to: 5).
view verticalLineLayout
]
You will see that the inner nodes look like they remain in the same place.
However, clicking works well, which means that the bounds of the children
nodes are computed correctly. Only the drawing is somewhat strange.
Hi Andre,
I saw that your last commits say that the labels should be rotatable now. However, in the latest Moose image based on 1.2.1, it does not seem to work. Am I doing something wrong?
Cheers,
Doru
--
www.tudorgirba.com
"What we can governs what we wish."
Status: New
Owner: ----
Labels: Type-Enhancement Priority-Medium Component-Finder Milestone-4.4
New issue 573 by tudor.gi...(a)gmail.com: DSM should be integrated into Moose
Finder
http://code.google.com/p/moose-technology/issues/detail?id=573
DSM should be shown as a separate tab in the MooseFinder for namespaces and
packages
Status: New
Owner: tudor.gi...(a)gmail.com
Labels: Type-Enhancement Priority-Medium Milestone-4.4
New issue 565 by tudor.gi...(a)gmail.com: Implement Metanool bridge between
Fame and Magritte
http://code.google.com/p/moose-technology/issues/detail?id=565
Metanool was a VW project to provide editing of Moose entities properties.
Moose entities provide Fame descriptions. To make them editable, we can use
Magritte.
Hi,
i got a strange behaviour from Mondrian trying to build a shape for data base tables. Basically defining a shape which contains a rectangle with a title and a children shape it happen that the rectangle containing the title have the same size of the children shape. Is it a bug or is a problem of my code? Below the code that i use, copy and paste in MOShapeSelector the tableShape method and than in a normal mondrian easel window you should be able to reproduce the same behaviour with the snippet below.
By using this shape definition:
tableShape
| builder |
builder := MOFormsBuilder new.
builder column; fill; pref; grow.
builder row; fill; pref; grow; row; fill; pref; grow.
builder x: 1 y: 1 add: (MORectangleShape new text: [:table | table name]; fontSize: 10).
builder x: 1 y: 2 add: MORectangleShape new.
builder x: 1 y: 2 add: MOChildrenShape new.
^view shape: builder shape
and using this visualization script:
view shape tableShape.
view node: (FAMIXNamedEntity new name: 'test') forIt: [
view shape rectangle.
view nodes: (1 to: 20).
view verticalLineLayout. ].
I got the result in the attachment.
Thanks,
Fabrizio
Status: New
Owner: ----
Labels: Type-Defect Priority-Medium
New issue 498 by fabrizio...(a)gmail.com: the sizes of the rectangles into
built shape are the same
http://code.google.com/p/moose-technology/issues/detail?id=498
Defining a shape which contains a rectangle with a title and a children
shape it happen that the rectangle containing the title have the same size
of the children shape while i would expect that the rectangle size was
coherent with its content.
By executing the following code in Mondrian you will see the problem
| builder |
builder := MOFormsBuilder new.
builder column; fill; pref; grow.
builder row; fill; row; fill; pref; grow.
builder x: 1 y: 1 add: (MORectangleShape new text: [:table | table name];
fontSize: 10).
builder x: 1 y: 2 add: MORectangleShape new.
builder x: 1 y: 2 add: MOChildrenShape new.
view shape: builder shape.
view node: 'hello' forIt: [
view nodes: (1 to: 5).
view verticalLineLayout
]