Status: New
Owner: tu...(a)tudorgirba.com
Labels: Type-Defect Priority-Medium Component-Glamour Milestone-4.8
New issue 980 by tu...(a)tudorgirba.com: Integrate GraphET in Glamour
http://code.google.com/p/moose-technology/issues/detail?id=980
Just what it says
--
You received this message because this project is configured to send all
issue notifications to this address.
You may adjust your notification preferences at:
https://code.google.com/hosting/settings
Hi,
I'd like to announce the first version of the Moldable Debugger.
It is a framework for creating and working with domain specific debuggers.
It's integrated into Moose and it currently comes with three debuggers: one
for announcements, one for PetitParser and one for Glamour.
More information can be found here: *
http://scg.unibe.ch/research/moldabledebugger*
A video introducing the framework is here: *https://vimeo.com/73980431*
(more are on the way)
Cheers,
Andrei
Status: New
Owner: ----
Labels: Type-Defect Priority-Medium
New issue 979 by benjamin...(a)gmail.com: Default window sizes for large
screens
http://code.google.com/p/moose-technology/issues/detail?id=979
The following should probably be pushed upstream for Pharo 3.0, but I
thought it best to report here in case it is feasible to include in Moose
4.8. Let me know what you think and if needed I can submit a changeset or
slice (what is the preferred method for submissions for Moose)
I've got a 24 inch screen that I have been using with Moose-4.8-dev for a
couple of weeks. I am finding it annoying that the Glamorous Inspector and
Glamorous Debugger open to maximum full screen size.
1. For the Standard Debugger, I don't agree with making the default size
bigger for bigger screens. I think the extra screen space is useful for
inspectors and observing the application. Indeed, the debugger width should
match the width of the System Browser, since that is the basis for the
formatting of code. Matching these helps give a consistent UI feel for
users.
For Moose4.8/Pharo2.0 we have...
NautilusWindow>>initialExtent
^ 850@600
Actually, perhaps that shouldn't be hard coded and instead be...
NautilusWindow>>initialExtent
^ RealEstateAgent standardWindowExtent
then for Debugger we could have something similar to this...
Debugger>>initialExtent
dependents size < 9 ifTrue: [^ super initialExtent]. "Pre debug window"
RealEstateAgent standardWindowExtent y < 400 "a tiny screen"
ifTrue: [^ super initialExtent].
^ [ RealEstateAgent standardWindowExtent.
] on: Error do: [800@600]
2. For Glamourous Inspector I understand defaulting to maximum width is
useful, but its overkill to default to the full height of the screen. So
you could have...
GLMMorphicRender>>open: aRenderable
window := GLMMorphicWindowRenderer render: aRenderable from: self.
^ window openInWorld "removed 'maximize' from here"
GTInspector>>initialExtent
^ RealEstateAgent fullscreenWindowExtent x @ RealEstateAgent
standardWindowExtent y
which requires also...
RealEstateAgent class >> fullscreenWindowExtent
^ self maximumUsableArea insetBy: self fullscreenMargin.
and...
RealEstateAgent class >> fullscreenMargin
^ FullscreenMargin ifNil: [FullscreenMargin := 25]
The last was copied from SystemWindow class >> fullScreenMargin which as an
option can then be...
SystemWindow class >> fullScreenMagin
^ RealEstateAgent fullscreenMargin
3. Glamorous Debbugger needs a bit more vertical height, so perhaps...
GTDebugger >> initialExtent
^ RealEstateAgent fullscreenWindowExtent x @ RealEstateAgent
fullscreenWindowExtent y * 2 // 3
--
You received this message because this project is configured to send all
issue notifications to this address.
You may adjust your notification preferences at:
https://code.google.com/hosting/settings
Hi Fabrizio,
We made quite some FAMIX changes recently. All test were green,
but It can be a test coverage problem.
In the latest moose I only have
FAMIXParameterizableClass>>parameterizedTypes
<MSEProperty: #parameterizedTypes type: #FAMIXParameterizedType opposite: #parameterizableClass>
<multivalued> <derived>
^ parameterizedTypes
Do you have a sample MSE file for me exhibiting the problem?
Stephan
Hi,
Glamour now offers an integration of Graph-ET.
In the latest Moose image you can do:
GLMCompositePresentation new with: [ :c |
c graphET
chart: [:chart :interval |
(chart lineDiagram)
models: interval;
y: [ :x | x*x ] ] ];
openOn: (1 to: 10)
It requires some more polishing, but it works.
Doru
--
www.tudorgirba.com
"Every thing has its own flow"
I made a screencast of Graph-ET :-)
It is my GSoC project and will be presented on ESUG this year.
http://www.youtube.com/watch?v=PkKNALQHa88
Any comments or suggestions to this mail.Enjoy!
- Daniel
Hi
just a short question
does FAMIX-AST forces to have the AST node subclass of a Moose entity
what is the independence of the tree? can we reuse him outside of famix and moose?
Stef
Hi,
I just generated an mse with inFamix and I got a MNU while importing the
file into the latest Moose.
Is there a newer release of inFamix I should download? Is it a know problem?
Thanks,
Fabrizio
Hi,
If you have to compare two MooseModels, we got tools for you.
We created the ConfigurationOf FamixDiff and OrionMerge.
- FamixDiff enables to do a diff between two MooseModels (or one OrionModel and a MooseModel). The result is a set of changes between the two models.
- OrionMerge takes this set of changes and creates a versioned OrionModel by applying the Orion action associated to each change. The result is thus an OrionModel corresponding to the target model of the diff, but taking less space of memory since its benefits from the way of Orion to manage model (between two versions, only the modified entities are stored. A clever system enables to access each entity whatever, it has been modified or not).
More information about Orion can be found there http://www.moosetechnology.org/tools/orion
I hope it is understandable.
So if you need to use them, please don't hesitate.
Anne
Nicolas wrote:
>did you run the tests before?
>I believe I saw some of them fail in the past ... ?
Uhm, we broke the build from 981 to 984,
but I'm not aware of earlier problems.
Stephan
Hi doru
i'm trying to get the parsers written by moser to work.
I cannot get the red executed (VW mess Sunittoo vs Sunit argggggghhhhhh)
I cannot get the parsers regenerated in VW.
So I'm trying to find a fix point.
stef
Hi all,
We (Stephan and myself) decided to add a Delphi unit to the FAMIX hierarchy in the Famix-Delphi package, since this is both a package and a namespace. But here we ran into trouble: we can inherit from package, from namespace, or from a scoping entity, but in all cases the method allPackages and allNamespaces do not return the desired value. Also some derived functionality, like the code browser, does not function properly, since it uses namespaces to browse the code. So we propose the following changes:
1) allPackages and allNamespaces should return "allWithSubTypesOf: " instead of withAll:
2) The message "allWithSubTypesOf: aSmalltalkClass ofGroupClass: aGroupSmalltalkClass" should iterate over a new method call "withMooseSubclasses" instead of "withAllSubclasses"
3) Default return value of withMooseSubclasses is withAllSubclasses, except when a class registers itself as a Moose subclass, indicating it is of the "Moose type" of the class it registers to.
This allows us to create a "merged" type of package and namespace for Delphi and still allowing for the codeBrowser to function properly.
What do you think about this change. I implemented it in my image, and it seems to work fine.
Cheers,
Diego
Hello,
Mouse Enter and Leave in Roassal is not working so well. Dynamic edges are
removed sometimes but sometimes they are not. Can you please have a look?
tx,
Usman
Hi,
I committed a little enhancement to the load MSE menu action: it no longer
blocks the ui. Thus, even if you load a larger MSE file, you can still use
the user interface.
Cheers,
Doru
--
www.tudorgirba.com
"Every thing has its own flow"
Status: New
Owner: ----
Labels: Type-Defect Priority-Medium Component-Finder Milestone-4.8
New issue 978 by tu...(a)tudorgirba.com: Loading MSE should not block the UI
http://code.google.com/p/moose-technology/issues/detail?id=978
Right now, choosing to load from MSE blocks the UI. A better solution would
be to load the MSE in parallel.
The implementation should be in MPImportMSECommand.
--
You received this message because this project is configured to send all
issue notifications to this address.
You may adjust your notification preferences at:
https://code.google.com/hosting/settings