Hi,
We are currently making some visualizations of application with Usman. They rely on method invocations. In that case, it is better to know which method invokes which other one. We put arrow on our graph, but the only available arrow we found are 'UML like inheritance' arrow with big triangle. Would it be possible to have other type of arrow that are more discrete, but nevertheless indicates the sense? (If possible put acute and not obtuse angle between the two branches of the arrow, it is more easier to see when they are lot of links).
More over, would it be possible to have hook on the lines to transform them as broken lines. Indeed, when you have m1 linked to m2 and m2 linked to m1, there is superposition of the two links so not easy to read. It would be great if we can manually break the lines to see the two of them.
I put some examples of arrow below.
Thanks a lot in advance.
Anne
Hi guys,
I just saw with Usman when we tried to add a CompositeShape using the
RTMondrianViewBuilder it is not working.
Here the example :
| view elements |
view := RTMondrianViewBuilder new.
elements := (1 to: 10).
view nodes: elements.
view shape shape: (RTLine new color: Color red) + RTArrow new.
view edges: elements from: #yourself toAll: elements.
view horizontalLineLayout.
view
If you comment "+ RTArrow new" everything is ok but with the CompositeShape
it is not.
Tanks.
--
Leo Perard
Hi everyone,
I am wandering if anyone is using Arki. If yes, please tell me if it’s missing something, or what you like about it. Of no, then again why?
I want to try to implement something cool in terms of bug prediction in software, and have a real use cases will be very useful.
Cheers.
Uko
Hi all,
Can we export FAST entities to a MSE file?
Because I have a model with FAST + FAMIX entities and I have problems
exporting them (Message Not Understood).
usman
Hi guys
With igor we added a registration mechanism for Freetype fonts for 13149.
We fixed the one of Moose and we should publish the moose packages (but if we publish it in Moose it will break moose).
Now Roassal should not cache the fonts in ROPlatform current because this means that you always get bitmap
if you get bitmap the first time.
So let us know how we proceed?
Stef
Hi everyone.
I wander if there is any place I can read about the philosophy about how Moose is linked with a source code. I know that there are source anchors, and sourced entities. But as far as I can see, my entities do not have any anchors.
Right now I’m thinking that it would be cool to be able to browse a source code of any sourced entity that you have, but I’m not sure how this can be implemented.
Cheers,
Uko
Hi Folks,
I have been working on a tool to transform Rascal M3 models to Moose models. Although the work is far from complete, I presented the progress so far in the Software Composition seminar. The slides (with notes) are here:
http://www.slideshare.net/onierstrasz/m3-for-moose
The Rascal guys are planning to have lots of importers for different languages, so a fast path for M3 models into Moose would be very useful …
Thanks to Doru and Fabrizio for suggesting this path. (Before I tried to generate MSE from M3 within Rascal, but that was very painful and fragile.)
Cheers,
Oscar
PS: Does anyone know how to display presenter notes with slideshare presentations? Their web site says it works, but I can’t figure it out. (I converted my Keynote slides to PPT precisely so this would work, but it doesn’t.)
Hello,
I am looking for the alternatives to the MSE format for model persistency.
The current problems I see with MSE:
- It is all or nothing approach (hence loading takes time)
- need to add entities to model to be able to store them on the disk and
this approach does not work well when you want to persist duplication
information, caches and annotations.
There was a Fuel-based solution and I am wondering if it can be helpful.
While looking for persistency solutions for Pharo, I came across several
solutions like Voyage/Mongo, GLORP, StandStoneBD, and Magma.
http://www.seaside.st/documentation/persistence
Did anyone try any of these tools in the context Moose? I would do a pilot
(saving n retrieving a test model, playing with caches and properties) with
Voyage/Mongo to understand the impact of their usage.
Usman
