Call for action for Roassal

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
38 messages Options
12
Reply | Threaded
Open this post in threaded view
|

Call for action for Roassal

abergel
Dear community,

As you may have seen, Roassal has entered a stabilization phase. The book AgileVisualization.com will soon be released. After its release, Roassal will go over a new development phase. In order to prepare it, I am asking this question:

        What are the 3 aspects you would like to see improved in Roassal?

You can answer publicly or by sending private messages.

Kind regards,
Alexandre
--
_,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
Alexandre Bergel  http://www.bergel.eu
^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.



_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.list.inf.unibe.ch/listinfo/moose-dev
Reply | Threaded
Open this post in threaded view
|

Re: Call for action for Roassal

Anne Etien
Hi Alexandre,

In a wonderful world, I would like:
- a real graph layout (like in graphviz) that can take into account around hundred nodes and several hundred of edges and place the nodes in order to see something. Currently, I have to use the circle layout and it looks strange.
- the possibility to place hangout of the edges on nodes (as box) more flexible. Indeed, I am reverse engineering a database containing around 80 tables and lot of foreign keys and the result is not readable since the edge are orthogonal starting from the top or the bottom of the box corresponding to the column name. I would like a real entity-relation diagram.

For the remainder, I use only Telescope since I have just to specify my data and not to script the visualization taking into account the order of the instructions… So don’t change the interface of the existing API in order we can continue to use Telescope on top on Roassal.

 If you want, I can show you the current stage of the visu (once again done in Telescope on top of Roassal) in order that you see on real data how uggly it is.

Cheers,
Anne

> Le 24 févr. 2016 à 09:51, Alexandre Bergel <[hidden email]> a écrit :
>
> Dear community,
>
> As you may have seen, Roassal has entered a stabilization phase. The book AgileVisualization.com will soon be released. After its release, Roassal will go over a new development phase. In order to prepare it, I am asking this question:
>
> What are the 3 aspects you would like to see improved in Roassal?
>
> You can answer publicly or by sending private messages.
>
> Kind regards,
> Alexandre
> --
> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
> Alexandre Bergel  http://www.bergel.eu
> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>
>
>
> _______________________________________________
> Moose-dev mailing list
> [hidden email]
> https://www.list.inf.unibe.ch/listinfo/moose-dev

_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.list.inf.unibe.ch/listinfo/moose-dev
Reply | Threaded
Open this post in threaded view
|

Re: [Pharo-dev] Call for action for Roassal

abergel
> - the possibility to place hangout of the edges on nodes (as box) more flexible. Indeed, I am reverse engineering a database containing around 80 tables and lot of foreign keys and the result is not readable since the edge are orthogonal starting from the top or the bottom of the box corresponding to the column name. I would like a real entity-relation diagram.

Do you have a screenshot showing the issue?

> If you want, I can show you the current stage of the visu (once again done in Telescope on top of Roassal) in order that you see on real data how uggly it is.

Yes please, some screenshots would help

Alexandre

>
> Cheers,
> Anne
>
>> Le 24 févr. 2016 à 09:51, Alexandre Bergel <[hidden email]> a écrit :
>>
>> Dear community,
>>
>> As you may have seen, Roassal has entered a stabilization phase. The book AgileVisualization.com will soon be released. After its release, Roassal will go over a new development phase. In order to prepare it, I am asking this question:
>>
>> What are the 3 aspects you would like to see improved in Roassal?
>>
>> You can answer publicly or by sending private messages.
>>
>> Kind regards,
>> Alexandre
>> --
>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
>> Alexandre Bergel  http://www.bergel.eu
>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>>
>>
>>
>> _______________________________________________
>> Moose-dev mailing list
>> [hidden email]
>> https://www.list.inf.unibe.ch/listinfo/moose-dev
>
>

--
_,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
Alexandre Bergel  http://www.bergel.eu
^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.



_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.list.inf.unibe.ch/listinfo/moose-dev
Reply | Threaded
Open this post in threaded view
|

Re: Call for action for Roassal

Peter Uhnak
In reply to this post by Anne Etien


On Wed, Feb 24, 2016 at 10:28 AM, Anne Etien <[hidden email]> wrote:
Hi Alexandre,

In a wonderful world, I would like:
- a real graph layout (like in graphviz) that can take into account around hundred nodes and several hundred of edges and place the nodes in order to see something. Currently, I have to use the circle layout and it looks strange.

Unfortunately making real graph layouts is really hard issue… I've wrote my bachelor thesis about this and didn't get very far; it's bit of a nightmare if you don't have good foundations. :'(

In any case, I wrote a simpler layout delegator in Roassal that delegates the layouting to graphviz and it works reasonably well, so I could probably add it to Roassal (you however need graphviz installed and it would add another dependency to Roassal, which I'm not so keen on).

Maybe also OGDF (http://ogdf.net/doku.php) could be used if you are ok with GPL… they have their own graph format so there could be some interaction.

Peter

_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.list.inf.unibe.ch/listinfo/moose-dev
Reply | Threaded
Open this post in threaded view
|

Re: Call for action for Roassal

Anne Etien

Le 24 févr. 2016 à 10:35, Peter Uhnák <[hidden email]> a écrit :



On Wed, Feb 24, 2016 at 10:28 AM, Anne Etien <[hidden email]> wrote:
Hi Alexandre,

In a wonderful world, I would like:
- a real graph layout (like in graphviz) that can take into account around hundred nodes and several hundred of edges and place the nodes in order to see something. Currently, I have to use the circle layout and it looks strange.

Unfortunately making real graph layouts is really hard issue… I've wrote my bachelor thesis about this and didn't get very far; it's bit of a nightmare if you don't have good foundations. :’(

I know. I tried but my mathematical background and my time were not enough. I got the research paper explaining the algorithm of Graphviz but even with that, it was very complex. However, since some tools (as graphviz) do it, we should have it.

Anne

In any case, I wrote a simpler layout delegator in Roassal that delegates the layouting to graphviz and it works reasonably well, so I could probably add it to Roassal (you however need graphviz installed and it would add another dependency to Roassal, which I'm not so keen on).

Maybe also OGDF (http://ogdf.net/doku.php) could be used if you are ok with GPL… they have their own graph format so there could be some interaction.

Peter
_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.list.inf.unibe.ch/listinfo/moose-dev


_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.list.inf.unibe.ch/listinfo/moose-dev
Reply | Threaded
Open this post in threaded view
|

Re: [Pharo-users] Re: Call for action for Roassal

Peter Uhnak
In reply to this post by Peter Uhnak


On Wed, Feb 24, 2016 at 11:03 AM, Martin Bähr <[hidden email]> wrote:
Excerpts from Peter Uhnák's message of 2016-02-24 10:35:05 +0100:
> In any case, I wrote a simpler layout delegator in Roassal that delegates
> the layouting to graphviz and it works reasonably well, so I could probably
> add it to Roassal (you however need graphviz installed and it would add
> another dependency to Roassal, which I'm not so keen on).

how does that work? does it ask graphviz for the position of the nodes and then
recreate that in roassal?

Yes, I generate .dot file that describes the nodes I want the layout for (that includes the dimensions of it), and then the .dot file is passed to GraphViz… from there I parse the generated SVG file (they can also output other formats like txt, or xml) and move the Roassal nodes.
 

> Maybe also OGDF (http://ogdf.net/doku.php) could be used if you are ok with
> GPL… they have their own graph format so there could be some interaction.

this would work the same as with graphviz as an additional dependency on ogdf, right?

GraphViz is good enough for generic graphs, but OGDF is well-founded and much more powerful layouting library. However C++ coding is required if one wants to use it.

Peter

_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.list.inf.unibe.ch/listinfo/moose-dev
Reply | Threaded
Open this post in threaded view
|

Re: [Pharo-users] Re: Call for action for Roassal

abergel
In reply to this post by Peter Uhnak
> In any case, I wrote a simpler layout delegator in Roassal that delegates the layouting to graphviz and it works reasonably well, so I could probably add it to Roassal (you however need graphviz installed and it would add another dependency to Roassal, which I'm not so keen on).

Where can we have it please?

Alexandre

--
_,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
Alexandre Bergel  http://www.bergel.eu
^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.



_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.list.inf.unibe.ch/listinfo/moose-dev
Reply | Threaded
Open this post in threaded view
|

Re: Call for action for Roassal

Stephan Eggermont-3
In reply to this post by abergel
On 24-02-16 09:51, Alexandre Bergel wrote:
> Dear community,
> What are the 3 aspects you would like to see improved in Roassal?

My favorite improvement would not really be a Roassal feature
but would save me a lot of time:
I would like the configuration versions to stop depending on numbered
versions of outside dependencies (Glamour, Neo-JSON-Core)

Stephan

_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.list.inf.unibe.ch/listinfo/moose-dev
Reply | Threaded
Open this post in threaded view
|

Re: [Pharo-users] Re: Call for action for Roassal

Peter Uhnak
In reply to this post by abergel
On Wed, Feb 24, 2016 at 11:51 AM, Alexandre Bergel <[hidden email]> wrote:
> In any case, I wrote a simpler layout delegator in Roassal that delegates the layouting to graphviz and it works reasonably well, so I could probably add it to Roassal (you however need graphviz installed and it would add another dependency to Roassal, which I'm not so keen on).

Where can we have it please?

I'll add it to Roassal in the evening (GMT+1).

_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.list.inf.unibe.ch/listinfo/moose-dev
Reply | Threaded
Open this post in threaded view
|

Re: [Pharo-users] Re: Call for action for Roassal

Peter Uhnak
Also personally I would like to have some more load options so I could load just the core of Roassal without all the chart and map builders and everything.

On Wed, Feb 24, 2016 at 12:37 PM, Peter Uhnák <[hidden email]> wrote:
On Wed, Feb 24, 2016 at 11:51 AM, Alexandre Bergel <[hidden email]> wrote:
> In any case, I wrote a simpler layout delegator in Roassal that delegates the layouting to graphviz and it works reasonably well, so I could probably add it to Roassal (you however need graphviz installed and it would add another dependency to Roassal, which I'm not so keen on).

Where can we have it please?

I'll add it to Roassal in the evening (GMT+1).


_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.list.inf.unibe.ch/listinfo/moose-dev
Reply | Threaded
Open this post in threaded view
|

Re: [Pharo-users] Re: Call for action for Roassal

abergel
In reply to this post by Peter Uhnak
Ok, great!

Alexandre


> On Feb 24, 2016, at 12:37 PM, Peter Uhnák <[hidden email]> wrote:
>
> On Wed, Feb 24, 2016 at 11:51 AM, Alexandre Bergel <[hidden email]> wrote:
> > In any case, I wrote a simpler layout delegator in Roassal that delegates the layouting to graphviz and it works reasonably well, so I could probably add it to Roassal (you however need graphviz installed and it would add another dependency to Roassal, which I'm not so keen on).
>
> Where can we have it please?
>
> I'll add it to Roassal in the evening (GMT+1).
> _______________________________________________
> Moose-dev mailing list
> [hidden email]
> https://www.list.inf.unibe.ch/listinfo/moose-dev

--
_,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
Alexandre Bergel  http://www.bergel.eu
^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.



_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.list.inf.unibe.ch/listinfo/moose-dev
Reply | Threaded
Open this post in threaded view
|

Re: Call for action for Roassal

abergel
In reply to this post by Stephan Eggermont-3
>> Dear community,
>> What are the 3 aspects you would like to see improved in Roassal?
>
> My favorite improvement would not really be a Roassal feature
> but would save me a lot of time:
> I would like the configuration versions to stop depending on numbered versions of outside dependencies (Glamour, Neo-JSON-Core)

Hi!

I do not get this. Relying on the baseline is not enough?

Roassal is very simply packaged. Only 4 packages: Roassal, Trachel, Roassal2Spec, Roassal2GT. It is very unlikely that this list of package will grow (not until we have a Git integration at least). You can make your configuration directly point to these packages no?

Cheers,
Alexandre

--
_,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
Alexandre Bergel  http://www.bergel.eu
^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.



_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.list.inf.unibe.ch/listinfo/moose-dev
Reply | Threaded
Open this post in threaded view
|

Re: Call for action for Roassal

Stephan Eggermont-3
On 24-02-16 13:11, Alexandre Bergel wrote:

>>> Dear community,
>>> What are the 3 aspects you would like to see improved in Roassal?
>>
>> My favorite improvement would not really be a Roassal feature
>> but would save me a lot of time:
>> I would like the configuration versions to stop depending on numbered versions of outside dependencies (Glamour, Neo-JSON-Core)
>
> Hi!
>
> I do not get this. Relying on the baseline is not enough?

#stable points to a numbered version, and that refers to numbered
versions of Glamour and Neo-JSON-Core. They always get out of sync.
the baseline is not specific enough

Stephan


_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.list.inf.unibe.ch/listinfo/moose-dev
Reply | Threaded
Open this post in threaded view
|

Re: Call for action for Roassal

abergel
I am not sure how to do this. Can you help?

Alexandre


> On Feb 24, 2016, at 2:02 PM, Stephan Eggermont <[hidden email]> wrote:
>
> On 24-02-16 13:11, Alexandre Bergel wrote:
>>>> Dear community,
>>>> What are the 3 aspects you would like to see improved in Roassal?
>>>
>>> My favorite improvement would not really be a Roassal feature
>>> but would save me a lot of time:
>>> I would like the configuration versions to stop depending on numbered versions of outside dependencies (Glamour, Neo-JSON-Core)
>>
>> Hi!
>>
>> I do not get this. Relying on the baseline is not enough?
>
> #stable points to a numbered version, and that refers to numbered versions of Glamour and Neo-JSON-Core. They always get out of sync.
> the baseline is not specific enough
>
> Stephan
>
>
> _______________________________________________
> Moose-dev mailing list
> [hidden email]
> https://www.list.inf.unibe.ch/listinfo/moose-dev

--
_,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
Alexandre Bergel  http://www.bergel.eu
^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.



_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.list.inf.unibe.ch/listinfo/moose-dev
Reply | Threaded
Open this post in threaded view
|

Re: Call for action for Roassal

Tudor Girba-2
In reply to this post by abergel
This is very cool. We will all benefit from this documentation.

One thing that I would still like to see fixed before the release is the edge building (this problem of not being able to properly specify random objects to be taken into account when building edges). Could we work on that?

Cheers,
Doru


> On Feb 24, 2016, at 9:51 AM, Alexandre Bergel <[hidden email]> wrote:
>
> Dear community,
>
> As you may have seen, Roassal has entered a stabilization phase. The book AgileVisualization.com will soon be released. After its release, Roassal will go over a new development phase. In order to prepare it, I am asking this question:
>
> What are the 3 aspects you would like to see improved in Roassal?
>
> You can answer publicly or by sending private messages.
>
> Kind regards,
> Alexandre
> --
> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
> Alexandre Bergel  http://www.bergel.eu
> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>
>
>
> _______________________________________________
> Moose-dev mailing list
> [hidden email]
> https://www.list.inf.unibe.ch/listinfo/moose-dev

--
www.tudorgirba.com
www.feenk.com

“Live like you mean it."

_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.list.inf.unibe.ch/listinfo/moose-dev
Reply | Threaded
Open this post in threaded view
|

Re: [Pharo-dev] Call for action for Roassal

abergel
Do you have an example? I am not sure to understand.

Alexandre


> On Feb 25, 2016, at 12:40 PM, Tudor Girba <[hidden email]> wrote:
>
> This is very cool. We will all benefit from this documentation.
>
> One thing that I would still like to see fixed before the release is the edge building (this problem of not being able to properly specify random objects to be taken into account when building edges). Could we work on that?
>
> Cheers,
> Doru
>
>
>> On Feb 24, 2016, at 9:51 AM, Alexandre Bergel <[hidden email]> wrote:
>>
>> Dear community,
>>
>> As you may have seen, Roassal has entered a stabilization phase. The book AgileVisualization.com will soon be released. After its release, Roassal will go over a new development phase. In order to prepare it, I am asking this question:
>>
>> What are the 3 aspects you would like to see improved in Roassal?
>>
>> You can answer publicly or by sending private messages.
>>
>> Kind regards,
>> Alexandre
>> --
>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
>> Alexandre Bergel  http://www.bergel.eu
>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>>
>>
>>
>> _______________________________________________
>> Moose-dev mailing list
>> [hidden email]
>> https://www.list.inf.unibe.ch/listinfo/moose-dev
>
> --
> www.tudorgirba.com
> www.feenk.com
>
> “Live like you mean it."
>
>

--
_,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
Alexandre Bergel  http://www.bergel.eu
^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.



_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.list.inf.unibe.ch/listinfo/moose-dev
Reply | Threaded
Open this post in threaded view
|

Re: [Pharo-dev] Call for action for Roassal

Tudor Girba-2
Hi,

We discussed this before. Try to draw a hierarchy using classes and inheritance objects (so, without using the class to superclass navigation). Something like this:

view := RTMondrian new.
view nodes: classes.
view edges objects: inheritances; connectFrom: #superclass to: #subclass.
view layout tree.
view

It won’t work.

Cheers,
Doru


> On Feb 25, 2016, at 1:29 PM, Alexandre Bergel <[hidden email]> wrote:
>
> Do you have an example? I am not sure to understand.
>
> Alexandre
>
>
>> On Feb 25, 2016, at 12:40 PM, Tudor Girba <[hidden email]> wrote:
>>
>> This is very cool. We will all benefit from this documentation.
>>
>> One thing that I would still like to see fixed before the release is the edge building (this problem of not being able to properly specify random objects to be taken into account when building edges). Could we work on that?
>>
>> Cheers,
>> Doru
>>
>>
>>> On Feb 24, 2016, at 9:51 AM, Alexandre Bergel <[hidden email]> wrote:
>>>
>>> Dear community,
>>>
>>> As you may have seen, Roassal has entered a stabilization phase. The book AgileVisualization.com will soon be released. After its release, Roassal will go over a new development phase. In order to prepare it, I am asking this question:
>>>
>>> What are the 3 aspects you would like to see improved in Roassal?
>>>
>>> You can answer publicly or by sending private messages.
>>>
>>> Kind regards,
>>> Alexandre
>>> --
>>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
>>> Alexandre Bergel  http://www.bergel.eu
>>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>>>
>>>
>>>
>>> _______________________________________________
>>> Moose-dev mailing list
>>> [hidden email]
>>> https://www.list.inf.unibe.ch/listinfo/moose-dev
>>
>> --
>> www.tudorgirba.com
>> www.feenk.com
>>
>> “Live like you mean it."
>>
>>
>
> --
> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
> Alexandre Bergel  http://www.bergel.eu
> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>
>
>
> _______________________________________________
> Moose-dev mailing list
> [hidden email]
> https://www.list.inf.unibe.ch/listinfo/moose-dev

--
www.tudorgirba.com
www.feenk.com

"Value is always contextual."




_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.list.inf.unibe.ch/listinfo/moose-dev
Reply | Threaded
Open this post in threaded view
|

Re: [Pharo-dev] Re: Call for action for Roassal

abergel
Oh, but this has been in Roassal for many months (I have implemented since right after ESUG). You need to use #source:connectFrom:toAll:

Here is an example that randomly generate a graph and render it using this facility:
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
nbOfNodes := 40.
nbOfRandomEdges := 40.

nodes := 1 to: nbOfNodes.
edges := (1 to: nbOfRandomEdges) 
collect: [ :notUsed | nodes atRandom -> {nodes atRandom . nodes atRandom} ].

b := RTMondrian new.

b shape circle color: (Color black alpha: 0.5).
b nodes: nodes.

b shape line color: (Color gray alpha: 0.3).
b edges
source: edges connectFrom: #key toAll: #value.

b layout force.
b
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=


Cheers,
Alexandre

On Feb 25, 2016, at 1:37 PM, Tudor Girba <[hidden email]> wrote:

Hi,

We discussed this before. Try to draw a hierarchy using classes and inheritance objects (so, without using the class to superclass navigation). Something like this:

view := RTMondrian new.
view nodes: classes.
view edges objects: inheritances; connectFrom: #superclass to: #subclass.
view layout tree.
view 

It won’t work.

Cheers,
Doru


On Feb 25, 2016, at 1:29 PM, Alexandre Bergel <[hidden email]> wrote:

Do you have an example? I am not sure to understand. 

Alexandre


On Feb 25, 2016, at 12:40 PM, Tudor Girba <[hidden email]> wrote:

This is very cool. We will all benefit from this documentation.

One thing that I would still like to see fixed before the release is the edge building (this problem of not being able to properly specify random objects to be taken into account when building edges). Could we work on that?

Cheers,
Doru


On Feb 24, 2016, at 9:51 AM, Alexandre Bergel <[hidden email]> wrote:

Dear community,

As you may have seen, Roassal has entered a stabilization phase. The book AgileVisualization.com will soon be released. After its release, Roassal will go over a new development phase. In order to prepare it, I am asking this question:

What are the 3 aspects you would like to see improved in Roassal?

You can answer publicly or by sending private messages.

Kind regards,
Alexandre
-- 
_,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
Alexandre Bergel  http://www.bergel.eu
^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.



_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.list.inf.unibe.ch/listinfo/moose-dev

--
www.tudorgirba.com
www.feenk.com 

“Live like you mean it."



-- 
_,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
Alexandre Bergel  http://www.bergel.eu
^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.



_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.list.inf.unibe.ch/listinfo/moose-dev

--
www.tudorgirba.com
www.feenk.com

"Value is always contextual."






-- 
_,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
Alexandre Bergel  http://www.bergel.eu
^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.




_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.list.inf.unibe.ch/listinfo/moose-dev
Reply | Threaded
Open this post in threaded view
|

Re: [Pharo-dev] Re: Call for action for Roassal

Tudor Girba-2
Oh, I missed this one :). So, now I can refactor the code that is still using the RTEdge class side methods :).

Great!

Thanks,
Doru


> On Feb 25, 2016, at 1:59 PM, Alexandre Bergel <[hidden email]> wrote:
>
> Oh, but this has been in Roassal for many months (I have implemented since right after ESUG). You need to use #source:connectFrom:toAll:
>
> Here is an example that randomly generate a graph and render it using this facility:
> -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
> nbOfNodes := 40.
> nbOfRandomEdges := 40.
>
> nodes := 1 to: nbOfNodes.
> edges := (1 to: nbOfRandomEdges)
> collect: [ :notUsed | nodes atRandom -> {nodes atRandom . nodes atRandom} ].
>
> b := RTMondrian new.
>
> b shape circle color: (Color black alpha: 0.5).
> b nodes: nodes.
>
> b shape line color: (Color gray alpha: 0.3).
> b edges
> source: edges connectFrom: #key toAll: #value.
>
> b layout force.
> b
> -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
>
> <Screen Shot 2016-02-25 at 1.57.24 PM.png>
>
> Cheers,
> Alexandre
>
>> On Feb 25, 2016, at 1:37 PM, Tudor Girba <[hidden email]> wrote:
>>
>> Hi,
>>
>> We discussed this before. Try to draw a hierarchy using classes and inheritance objects (so, without using the class to superclass navigation). Something like this:
>>
>> view := RTMondrian new.
>> view nodes: classes.
>> view edges objects: inheritances; connectFrom: #superclass to: #subclass.
>> view layout tree.
>> view
>>
>> It won’t work.
>>
>> Cheers,
>> Doru
>>
>>
>>> On Feb 25, 2016, at 1:29 PM, Alexandre Bergel <[hidden email]> wrote:
>>>
>>> Do you have an example? I am not sure to understand.
>>>
>>> Alexandre
>>>
>>>
>>>> On Feb 25, 2016, at 12:40 PM, Tudor Girba <[hidden email]> wrote:
>>>>
>>>> This is very cool. We will all benefit from this documentation.
>>>>
>>>> One thing that I would still like to see fixed before the release is the edge building (this problem of not being able to properly specify random objects to be taken into account when building edges). Could we work on that?
>>>>
>>>> Cheers,
>>>> Doru
>>>>
>>>>
>>>>> On Feb 24, 2016, at 9:51 AM, Alexandre Bergel <[hidden email]> wrote:
>>>>>
>>>>> Dear community,
>>>>>
>>>>> As you may have seen, Roassal has entered a stabilization phase. The book AgileVisualization.com will soon be released. After its release, Roassal will go over a new development phase. In order to prepare it, I am asking this question:
>>>>>
>>>>> What are the 3 aspects you would like to see improved in Roassal?
>>>>>
>>>>> You can answer publicly or by sending private messages.
>>>>>
>>>>> Kind regards,
>>>>> Alexandre
>>>>> --
>>>>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
>>>>> Alexandre Bergel  http://www.bergel.eu
>>>>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>>>>>
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Moose-dev mailing list
>>>>> [hidden email]
>>>>> https://www.list.inf.unibe.ch/listinfo/moose-dev
>>>>
>>>> --
>>>> www.tudorgirba.com
>>>> www.feenk.com
>>>>
>>>> “Live like you mean it."
>>>>
>>>>
>>>
>>> --
>>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
>>> Alexandre Bergel  http://www.bergel.eu
>>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>>>
>>>
>>>
>>> _______________________________________________
>>> Moose-dev mailing list
>>> [hidden email]
>>> https://www.list.inf.unibe.ch/listinfo/moose-dev
>>
>> --
>> www.tudorgirba.com
>> www.feenk.com
>>
>> "Value is always contextual."
>>
>>
>>
>>
>>
>
> --
> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
> Alexandre Bergel  http://www.bergel.eu
> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>
>
>
> _______________________________________________
> Moose-dev mailing list
> [hidden email]
> https://www.list.inf.unibe.ch/listinfo/moose-dev

--
www.tudorgirba.com
www.feenk.com

"Next time you see your life passing by, say 'hi' and get to know her."




_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.list.inf.unibe.ch/listinfo/moose-dev
Reply | Threaded
Open this post in threaded view
|

Re: [Pharo-dev] Re: Call for action for Roassal

Peter Uhnak
Hi Anne,

I've just added a Roassal plugin with the GraphViz-based layout.

Note that I had to hack it to work with RTMultiLine (it was originally designed for different lines which are currently not part of Roassal),
so it's possible there will be some issues.

You can load it (if you have latest Roassal installed)  via the Roassal plugin menu,
or by executing the following

Metacello new
baseline: 'GraphVizLayout';
repository: 'github://peteruhnak/graphviz-layout:master/repository';
load

Then look at the class-side examples in RTGraphVizLayout.

Finally if you use this, you shouldn't move elements within the layout (because graphviz layout hard-codes the pathing)… we could probably change this in the future, but I am not sure how (ideally the layout would have to be recomputed, but that's expensive to do live),

So to summarize:
* you must have GraphViz installed in your system
* you must use RTMultiLine for lines (because only they support bending)

Let me know how it goes,





Peter

On Thu, Feb 25, 2016 at 2:02 PM, Tudor Girba <[hidden email]> wrote:
Oh, I missed this one :). So, now I can refactor the code that is still using the RTEdge class side methods :).

Great!

Thanks,
Doru


> On Feb 25, 2016, at 1:59 PM, Alexandre Bergel <[hidden email]> wrote:
>
> Oh, but this has been in Roassal for many months (I have implemented since right after ESUG). You need to use #source:connectFrom:toAll:
>
> Here is an example that randomly generate a graph and render it using this facility:
> -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
> nbOfNodes := 40.
> nbOfRandomEdges := 40.
>
> nodes := 1 to: nbOfNodes.
> edges := (1 to: nbOfRandomEdges)
>                               collect: [ :notUsed | nodes atRandom -> {nodes atRandom . nodes atRandom} ].
>
> b := RTMondrian new.
>
> b shape circle color: (Color black alpha: 0.5).
> b nodes: nodes.
>
> b shape line color: (Color gray alpha: 0.3).
> b edges
>       source: edges connectFrom: #key toAll: #value.
>
> b layout force.
> b
> -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
>
> <Screen Shot 2016-02-25 at 1.57.24 PM.png>
>
> Cheers,
> Alexandre
>
>> On Feb 25, 2016, at 1:37 PM, Tudor Girba <[hidden email]> wrote:
>>
>> Hi,
>>
>> We discussed this before. Try to draw a hierarchy using classes and inheritance objects (so, without using the class to superclass navigation). Something like this:
>>
>> view := RTMondrian new.
>> view nodes: classes.
>> view edges objects: inheritances; connectFrom: #superclass to: #subclass.
>> view layout tree.
>> view
>>
>> It won’t work.
>>
>> Cheers,
>> Doru
>>
>>
>>> On Feb 25, 2016, at 1:29 PM, Alexandre Bergel <[hidden email]> wrote:
>>>
>>> Do you have an example? I am not sure to understand.
>>>
>>> Alexandre
>>>
>>>
>>>> On Feb 25, 2016, at 12:40 PM, Tudor Girba <[hidden email]> wrote:
>>>>
>>>> This is very cool. We will all benefit from this documentation.
>>>>
>>>> One thing that I would still like to see fixed before the release is the edge building (this problem of not being able to properly specify random objects to be taken into account when building edges). Could we work on that?
>>>>
>>>> Cheers,
>>>> Doru
>>>>
>>>>
>>>>> On Feb 24, 2016, at 9:51 AM, Alexandre Bergel <[hidden email]> wrote:
>>>>>
>>>>> Dear community,
>>>>>
>>>>> As you may have seen, Roassal has entered a stabilization phase. The book AgileVisualization.com will soon be released. After its release, Roassal will go over a new development phase. In order to prepare it, I am asking this question:
>>>>>
>>>>>   What are the 3 aspects you would like to see improved in Roassal?
>>>>>
>>>>> You can answer publicly or by sending private messages.
>>>>>
>>>>> Kind regards,
>>>>> Alexandre
>>>>> --
>>>>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
>>>>> Alexandre Bergel  http://www.bergel.eu
>>>>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>>>>>
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Moose-dev mailing list
>>>>> [hidden email]
>>>>> https://www.list.inf.unibe.ch/listinfo/moose-dev
>>>>
>>>> --
>>>> www.tudorgirba.com
>>>> www.feenk.com
>>>>
>>>> “Live like you mean it."
>>>>
>>>>
>>>
>>> --
>>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
>>> Alexandre Bergel  http://www.bergel.eu
>>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>>>
>>>
>>>
>>> _______________________________________________
>>> Moose-dev mailing list
>>> [hidden email]
>>> https://www.list.inf.unibe.ch/listinfo/moose-dev
>>
>> --
>> www.tudorgirba.com
>> www.feenk.com
>>
>> "Value is always contextual."
>>
>>
>>
>>
>>
>
> --
> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
> Alexandre Bergel  http://www.bergel.eu
> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>
>
>
> _______________________________________________
> Moose-dev mailing list
> [hidden email]
> https://www.list.inf.unibe.ch/listinfo/moose-dev

--
www.tudorgirba.com
www.feenk.com

"Next time you see your life passing by, say 'hi' and get to know her."




_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.list.inf.unibe.ch/listinfo/moose-dev


_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.list.inf.unibe.ch/listinfo/moose-dev
12