[Pharo-users] Roassal2 on Ubuntu 64

[Evan Donahue]

Hello,

I'm trying to learn Roassal, but I can't seem to get the first
visualization from
https://dl.dropboxusercontent.com/u/31543901/AgileVisualization/QuickStart/0101-QuickStart.html
working on any combination of image and vm. I've tried the vm+40 and vm+50
from get.pharo and the moose 6.0 image from the moose site.

The issue seems to be that CairoLibrary>>unixModuleName does not contain
the /usr/lib/x86_64-linux-gnu/libcairo.so.2 path where 64-bit Ubuntu
installs cairo. If I symlink that library, I get an Error: External module
not found from AthensCairoSurface. I don't really know enough about the
external module system to know what this is telling me. Any ideas would be
much appreciated.

Thanks,
Evan

Re: [Pharo-users] [ARTICLE] Tiny, yet so beautiful

[Sven Van Caekenberghe]

Thanks a lot, Ben, for your remarks; I applied them all.

I am glad you liked the article.

Sven

> On 10 Mar 2016, at 01:21, Ben Coman  wrote:
> 
> On Thu, Mar 10, 2016 at 6:05 AM, Sven Van Caekenberghe  wrote:
>> 
>> I wrote another article:
>> 
>>  Tiny, yet so beautiful
>>  The smallest piece of great Pharo code
>>  https://medium.com/concerning-pharo/tiny-yet-so-beautiful-1ef5149c910e
>> 
> 
> I love reading these articles of yours.  Very well written.
> Feedback for some minor wordsmithing...
> 
> -  2 out of 5 reserved words in the language.)
> + 2 out of _the_ 5 reserved words in the language.) _They are
> instances respectively of classes True and False, that have common
> superclass Boolean._
> 
> -  where an identical message _results in_ different implementations
> + where an identical message _invokes_ different implementations
> 
> -  Manipulating objects using their API
> + Manipulating objects using their _pre-existing_ API
> 
> -  Switserland
> + Switzerland
> 
> cheers -ben
> 

Re: [Pharo-users] GTSpotter fuzzy search

[Damien Pollet]

+1

I wanted to copy the algo from selecta but never got to it…
https://github.com/garybernhardt/selecta/blob/master/selecta

On 9 March 2016 at 20:51, Tudor Girba  wrote:

> Hi,
>
> Would you like to try to propose a implementation for this? This would fit
> as a subclass of GTFilter?
>
> Cheers,
> Doru
>
>
> > On Mar 9, 2016, at 4:04 PM, Peter Uhnák  wrote:
> >
> > Would be possible to have fuzzy-search?
> > Or rather Pascal-based search?
> >
> > RTAE -> RTAbstractExample
> > BCF -> BlConfigurableFormatter
> >
> > etc.
>
> --
> www.tudorgirba.com
> www.feenk.com
>
> "Every successful trip needs a suitable vehicle."
>
>
>
>
>
>
>

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

[Peter Uhnák]

On Thu, Mar 10, 2016 at 12:36 AM, Martin Bähr  wrote:

> hi,
>
> i just remembered two things:
>
> i'd like to see rounded corners on nodes.
>

We've added this quite some time ago, unless you have something else in
mind.

~~
|v e|
v := RTView new.
e := RTRoundedBox new
size: 100;
borderRadius: 20;
borderColor: Color black;
borderWidth: 1;
element.
v add: e.
v
~~


​




>
> Excerpts from stepharo's message of 2016-02-24 12:42:49 +0100:
> >  - having a better output (arrows, edge)
>
> and related to edges:
>
> i'd like edges to connect to a node on the border of the box, not to the
> center
> of the node. this means the connecting point would always be at the middle
> of
> the side of one edge:
>

We have about 10 different attach points already. See the 'attach point'
protocol of RTAbstractLine.
Some of them also automatically add extra space if there are multiple lines
between the same elements.

As for bezier and multiline edges, their curvature isn't currently handled
properly, but it's in my todo list (I'll work on this probably during the
weekend).

I'm sure it can be presented in more revealing way, but it's too late for
me today (or is it too early?).

You can play with it (drag the elements to see how it behaves).

~~
|v attachOptions es e offset|
v := RTView new.
attachOptions := RTAbstractLine selectors select: [ :each | each
 beginsWith: #with ].

offset := 0 @ 0.
attachOptions do: [ :each |
es := RTBox new borderColor: Color black; size: 50; elementsOn: #(a b).
each = #withContinuousCircleAttachPoint
ifTrue: [ es := RTEllipse new borderColor: Color black; size: 50;
elementsOn: #(a b). ].
es translateBy: offset.
es @ RTDraggable.
offset := offset + (0 @ 100).
v addAll: es.

es second translateBy: 300 @ 25.

e := RTArrowedLine new
color: Color black;
perform: each;
edgeFrom: es first to: es second.
e model: each.
v add: e.
e @ (RTLabelled new color: Color black).
].

v
~~



​




>
>  ___\|/___
> | |
>\| |/
>   --| |--
>/| |\
> |_|
> /|\
>
> bezier edges should always leave at a 90deg angle:
>  |
> | |
> | |
>   --| |--
> | |
> |_|
>  |
>
> optionally, starting at the corners could also be nice:
>
> \_/
> | |
> | |
> | |
> | |
> |_|
> / \
>
> and in such a case bezier edges should leave at 135deg to each side.
>
> by default the edge should find the nearest connection point, but i'd like
> to
> configure a list of allowed points.
>
> finally, multiple edges in the same general direction should either leave
> at
> the same point, pick the nearest side that doesn't have an edge already.
> or spaced evenly along the same side if all available sides have edges
> already.
>
> the list of allowed connection points would have 12 options:
> the middle of 4 sides, 4 corners, and 4 sides as a whole when multiple
> edges
> should be spaced evenly.
>
> greetings, martin.
>
> --
> eKita   -   the online platform for your entire academic
> life
> --
> chief engineer
>  eKita.co
> pike programmer  pike.lysator.liu.secaudium.net
> societyserver.org
> secretary
> beijinglug.org
> mentor
> fossasia.org
> foresight developer  foresightlinux.org
> realss.com
> unix sysadmin
> Martin Bähr  working in china
> http://societyserver.org/mbaehr/
>
>

Re: [Pharo-users] [ARTICLE] Tiny, yet so beautiful

[Ben Coman]

On Thu, Mar 10, 2016 at 6:05 AM, Sven Van Caekenberghe  wrote:
>
> I wrote another article:
>
>   Tiny, yet so beautiful
>   The smallest piece of great Pharo code
>   https://medium.com/concerning-pharo/tiny-yet-so-beautiful-1ef5149c910e
>

I love reading these articles of yours.  Very well written.
Feedback for some minor wordsmithing...

-  2 out of 5 reserved words in the language.)
+ 2 out of _the_ 5 reserved words in the language.) _They are
instances respectively of classes True and False, that have common
superclass Boolean._

-  where an identical message _results in_ different implementations
+ where an identical message _invokes_ different implementations

-  Manipulating objects using their API
+ Manipulating objects using their _pre-existing_ API

-  Switserland
+ Switzerland

cheers -ben

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

[Martin Bähr]

hi,

i just remembered two things:

i'd like to see rounded corners on nodes.

Excerpts from stepharo's message of 2016-02-24 12:42:49 +0100:
>  - having a better output (arrows, edge)

and related to edges:

i'd like edges to connect to a node on the border of the box, not to the center
of the node. this means the connecting point would always be at the middle of
the side of one edge:

 ___\|/___ 
| |
   \| |/
  --| |--
   /| |\
|_|
/|\

bezier edges should always leave at a 90deg angle:
 | 
| |
| | 
  --| |--
| | 
|_|
 | 

optionally, starting at the corners could also be nice:

\_/
| |
| | 
| |  
| | 
|_|
/ \

and in such a case bezier edges should leave at 135deg to each side.

by default the edge should find the nearest connection point, but i'd like to
configure a list of allowed points.

finally, multiple edges in the same general direction should either leave at
the same point, pick the nearest side that doesn't have an edge already.
or spaced evenly along the same side if all available sides have edges already.

the list of allowed connection points would have 12 options: 
the middle of 4 sides, 4 corners, and 4 sides as a whole when multiple edges
should be spaced evenly.

greetings, martin.

-- 
eKita   -   the online platform for your entire academic life
-- 
chief engineer   eKita.co
pike programmer  pike.lysator.liu.secaudium.net societyserver.org
secretary  beijinglug.org
mentor   fossasia.org
foresight developer  foresightlinux.orgrealss.com
unix sysadmin
Martin Bähr  working in chinahttp://societyserver.org/mbaehr/

[Pharo-users] [VIDEO TUTORIAL] Automating Image Building

[Dimitris Chloupis]

This one is a lengthy one but for a good reason, building images can be a
tedious task but can be easily automated with the help of startup scripts
and make files plus the mandatory dosage of Pharo technomagic.

https://www.youtube.com/watch?v=rR4UmDoYAyQ&feature=youtu.be

[Pharo-users] [ARTICLE] Tiny, yet so beautiful

[Sven Van Caekenberghe]

I wrote another article:

  Tiny, yet so beautiful
  The smallest piece of great Pharo code
  https://medium.com/concerning-pharo/tiny-yet-so-beautiful-1ef5149c910e

Sven

Please consider joining us @ http://pharodays2016.pharo.org

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

[Peter Uhnák]

On Wed, Mar 9, 2016 at 10:02 PM, Johan Fabry  wrote:

> Heh, also a funny feature of your anchor constraints: they do not play
> well with layouts and animations. Try the code below (adapted from the
> example to use the layout and animation of LRP ) and you will see that the
> circles keep spinning around, they never stop. I think it would be better
> for them to stop ;-)
>

can't help myself :)

https://youtu.be/PGNiXGX2nLU?t=1m


> stepping := RTSpringLayoutStepping new
> view: v; layout: layout;
> afterBlock: [ v canvas camera focusOnCenter].
>

The problem is that RTSpringLayoutStepping>>view sets in effect the layout
on all elements in the view, which obviously won't work since they'll start
competing.

I'm not sure right now how to address it, but try this

stepping := RTSpringLayoutStepping new
view: RTView new; layout: layout;
nodes: es;
afterBlock: [ v canvas camera focusOnCenter].

>
> I even have a bug report for you … before, when I removed the edge, the
> label was also removed automatically. Now this is no longer the case. :-(
>
>
Ah right, thanks. I've extracted the class from my code and I have
different removal mechanism...
Please try the latest version, it should be fixed there.

Peter

Re: [Pharo-users] [ANN] SelectEntity, a new widget for Spec

[stepharo]

We should check and collect the changes made by hernan too.

Stef

Le 8/3/16 16:45, Stephan Eggermont a écrit :

On 08-03-16 13:52, Johan Fabry wrote:

That looks very cool, how about adding it to the standard distribution?

That is definitely the goal. Now is the time to point out stupid names,
bad design decisions, missed use cases and ugly layout.
I would welcome feedback.

While building it, I found out I don't like the way an EntryCompletion
is connected to its PluggableTextFieldMorph. It makes it difficult
to deal with objects that are not strings.

PluggableTextFieldMorph>>entryCompletion: anEntryCompletion
entryCompletion := anEntryCompletion.
entryCompletion
ifNotNil: [ entryCompletion chooseBlock isNil
ifTrue: [entryCompletion chooseBlock: [:v | self setText: 
v. self acceptTextInModel ]]

ifFalse: [ | blk |
blk := entryCompletion chooseBlock.
entryCompletion chooseBlock: [:v | self setText: v. 
self acceptTextInModel. blk value: v]]].



In the demo code it also is clear that the ListModel is incomplete,
I couldn't find ho wto  open a selected item with the keyboard
(unless I implement click on the item, but that then does not
have enough context to actually know what to do).

Stephan

Re: [Pharo-users] NBOpenGL on Pharo5

[Alexandre Bergel]

Yes we are. The binding uses native boost, the code has to be ported. 

Alexandre 

> Le 9 mars 2016 à 06:42, Richard Wettel  a écrit :
> 
> Hi
> 
> I was wondering what the status of NBOpenGL on Pharo5 was. As far as I know, 
> it does not work. I would like to have CodeCity work on Pharo5.
> 
> Is anybody else (for instance the Roassal3d team) affected by this?
> 
> Cheers
> Ricky

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

[Johan Fabry]

Heh, also a funny feature of your anchor constraints: they do not play well 
with layouts and animations. Try the code below (adapted from the example to 
use the layout and animation of LRP ) and you will see that the circles keep 
spinning around, they never stop. I think it would be better for them to stop 
;-)

| v lbls es e1 e2 a1 a2 layout stepping|
v := RTView new.
lbls := RTLabel new elementsOn: #(#First #Second).
es := RTEllipse new
size: 30;
borderColor: Color black;
elementsOn: #(#source #dest).
v
addAll: lbls;
addAll: es.
es @ RTDraggable.
es @ RTLabeled.
e1 := RTArrowedLine new
withContinuousCircleAttachPoint;
color: Color black;
edgeFrom: es first to: es second.
v add: e1.
e2 := RTArrowedLine new
withContinuousCircleAttachPoint;
color: Color black;
edgeFrom: es second to: es first.
v add: e2.
a1 := RTAnchorConstraint new.
a1 anchorShape size: 10.
a1 guideLine color: Color red.
a1
element: lbls first;
edge: e1;
balance: 0.2;
minDistance: 10;
build.
(a2 := RTAnchorConstraint new)
element: lbls second;
edge: e2;
balance: 0.2;
minDistance: 10;
build.
layout := RTForceBasedLayout new
charge: -450; length: 100; 
doNotUseProgressBar; applyOn: es; 
yourself.
layout initialLayout: RTSugiyamaLayout new. 
stepping := RTSpringLayoutStepping new 
view: v; layout: layout;
afterBlock: [ v canvas camera 
focusOnCenter].
v addAnimation: stepping.   
^ v

> On Mar 9, 2016, at 17:56, Johan Fabry  wrote:
> 
> 
>> On Mar 7, 2016, at 11:34, Peter Uhnák > > wrote:
>> 
>> I've just commited RTAnchorConstraint to latest Roassal, see class-side 
>> example there.
>> 
> I have been trying it, and there is a first version that works now, thanks!
> 
> I even have a bug report for you … before, when I removed the edge, the label 
> was also removed automatically. Now this is no longer the case. :-(
> 
> ---> Save our in-boxes! http://emailcharter.org  
> <---
> 
> Johan Fabry   -   http://pleiad.cl/~jfabry 
> PLEIAD and RyCh labs  -  Computer Science Department (DCC)  -  University of 
> Chile
> 
> ___
> Moose-dev mailing list
> moose-...@list.inf.unibe.ch
> https://www.list.inf.unibe.ch/listinfo/moose-dev



---> Save our in-boxes! http://emailcharter.org <---

Johan Fabry   -   http://pleiad.cl/~jfabry
PLEIAD and RyCh labs  -  Computer Science Department (DCC)  -  University of 
Chile

Re: [Pharo-users] Fwd: Fwd: Phratch + Debian

[stepharo]

Cool initiative!
Tx Jannik
I blogged it and twitted it :)


Le 9/3/16 08:47, jannik laval a écrit :

Finally, I made it.

It is available here: http://www.phratch.com/download/phratch.deb

Best regards,
Jannik

2016-03-07 12:18 GMT+01:00 Hilaire >:


In last resort, it is still possible for a LiveCD to get Pratch
installed even if not a Deb package.

But yes, Pratch will greatly gain exposure being a Deb package,
propagated to the debian family distribution. I write that, but
DrGeo is
not deb packaged as well...

Good luck.

Hilaire

Le 07/03/2016 08:32, jannik laval a écrit :
>
> I send the email to the pharo community.
> Can someone help us to create a phratch .deb package ?
>
> The need is for a french professor who needs to create a debian live
> distrib for education. He want to include phratch instead of
scratch /
> snap ...
> So, it is a great opportunity for  us.
>
> Thank you for your help.
> Jannik

--
Dr. Geo
http://drgeo.eu





--
~~Jannik Laval~~
Enseignant-chercheur
Responsable Pédagogique Licence Coordonnateur de Projet en Système 
d'Information

IUT Lumière, Université Lyon Lumière
laboratoire DISP
http://www.jannik-laval.eu
http://www.phratch.com
http://www.approchealpes.info

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

[Johan Fabry]

> On Mar 7, 2016, at 11:34, Peter Uhnák  wrote:
> 
> I've just commited RTAnchorConstraint to latest Roassal, see class-side 
> example there.
> 
I have been trying it, and there is a first version that works now, thanks!

I even have a bug report for you … before, when I removed the edge, the label 
was also removed automatically. Now this is no longer the case. :-(

---> Save our in-boxes! http://emailcharter.org <---

Johan Fabry   -   http://pleiad.cl/~jfabry
PLEIAD and RyCh labs  -  Computer Science Department (DCC)  -  University of 
Chile

[Pharo-users] Upcoming deadlines SPLASH'16 in Amsterdam: OOPSLA & Onward!

[Tijs van der Storm]

//
ACM Conference on Systems, Programming, Languages, and Applications:
Software for Humanity (SPLASH'16)

Amsterdam, The Netherlands
Sun 30th October - Fri 4th November , 2016

http://2016.splashcon.org

Sponsored by ACM SIGPLAN
//


# Important dates

OOPSLA Papers:
 * Submissions due: March 23, 2016
 * Author Response: May 19 - May 21, 2016
 * Author Notification: June 1, 2016
 * Camera Ready: August 26, 2016

Onward! Papers & Essays:
 * Submissions due: April 1, 2016
 * First notification: May 20, 2016
 * Revised papers: July 15, 2016
 * Final notification: July 31, 2016
 * Camera ready: August 26, 2016


# SPLASH 2016

The ACM SIGPLAN conference on Systems, Programming, Languages and
Applications: Software for Humanity (SPLASH) embraces all aspects of
software construction, to make it the premier conference at the
intersection of programming, languages, systems, and software engineering.
We invite high quality submissions describing original and unpublished work.


## OOPSLA Research Papers

Papers that address any aspect of software development are welcome,
including requirements, modeling, prototyping, design, implementation,
generation, analysis, verification, testing, evaluation, maintenance,
reuse, replacement, and retirement of software systems. Papers may address
these topics in a variety of ways, including new tools (such as languages,
program analyses, and runtime systems), new techniques (such as
methodologies, design processes, code organization approaches, and
management techniques), and new evaluations (such as formalisms and proofs,
corpora analyses, user studies, and surveys).

Submissions due: Wed 23 March, 2016
http://2016.splashcon.org/track/splash-2016-oopsla


## Onward! Research Papers

Onward! is a premier multidisciplinary conference focused on everything to
do with programming and software: including processes, methods, languages,
communities, and applications. Onward! is more radical, more visionary, and
more open than other conferences to ideas that are well-argued but not yet
proven. We welcome different ways of thinking about, approaching, and
reporting on programming language and software engineering research.

Submissions due: Fri 1 April, 2016
http://2016.splashcon.org/track/onward2016/onward-2016-papers


## Onward! Essays

Onward! Essays is looking for clear and compelling pieces of writing about
topics important to the software community. An essay can be an exploration
of a topic, its impact, or the circumstances of its creation; it can
present a personal view of what is, explore a terrain, or lead the reader
in an act of discovery; it can be a philosophical digression or a deep
analysis. It can describe a personal journey, perhaps by which the author
reached an understanding of such a topic. The subject area should be
interpreted broadly and can include the relationship of software to human
endeavors, or its philosophical, sociological, psychological, historical,
or anthropological underpinnings.

Submissions due: Fri 1 April, 2016
http://2016.splashcon.org/track/onward2016/onward2016-essays

## More information

Contact: public...@splashcon.org
Website: http://2016.splashcon.org

Location: Mövenpick Hotel Amsterdam City Centre
Amsterdam, The Netherlands

## Organization:

SPLASH General Chair: Eelco Visser (TU Delft)
OOPSLA Papers Chair: Yannis Smaragdakis (University of Athens)
Onward! Papers Chair: Emerson Murphy-Hill (North Carolina State University)
Onward! Essays Chair: Crista Lopes (University of California, Irvine)
Artifacts Co-Chairs: Michael Hind (IBM Research) and Michael Bond (Ohio
State University)
Publications Chair: Alex Potanin (Victoria University of Wellington)
Sponsorship Chair: Jurgen Vinju (Purdue University)
Publicity and Web Co-Chairs: Tijs van der Storm (CWI) and Ron Garcia
(University of British Columbia)
Web Technology Chair: Eelco Visser (TU Delft)

//


-- 
Researcher Centrum Wiskunde & Informatica (CWI)
Master of Software Engineering Universiteit van Amsterdam (UvA)

Dr. Tijs van der Storm @ Centrum Wiskunde & Informatica (CWI)
Office: L225| Phone: +31 (0)20 5924164 | Address: Science Park 123
P.O. Box 94079  | Postal code: 1090 GB | Amsterdam, The Netherlands

Re: [Pharo-users] GTSpotter fuzzy search

[Tudor Girba]

Hi,

Would you like to try to propose a implementation for this? This would fit as a 
subclass of GTFilter?

Cheers,
Doru


> On Mar 9, 2016, at 4:04 PM, Peter Uhnák  wrote:
> 
> Would be possible to have fuzzy-search?
> Or rather Pascal-based search?
> 
> RTAE -> RTAbstractExample
> BCF -> BlConfigurableFormatter
> 
> etc.

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

"Every successful trip needs a suitable vehicle."

Re: [Pharo-users] [Pharo-dev] Call about Numerical Methods in Pharo :)

[Serge Stinckwich]

On Wed, Mar 9, 2016 at 7:33 PM, Blondeau Vincent
 wrote:
>
>
>> -Message d'origine-
>> De : Pharo-dev [mailto:pharo-dev-boun...@lists.pharo.org] De la part de
>> Serge Stinckwich
>> Envoyé : jeudi 3 mars 2016 11:04
>> À : Any question about pharo is welcome
>> Cc : Didier Besset; Souissi Sami; Pharo Development List
>> Objet : Re: [Pharo-dev] [Pharo-users] Call about Numerical Methods in Pharo
>> :)
>>
>> On Thu, Mar 3, 2016 at 10:43 AM, Blondeau Vincent
>>  wrote:
>> > Hi,
>> >
>> > That's a very good idea.
>> > For now, I use a Pharo-R binding to make the statistics, but it is better 
>> > if it is
>> directly in the image.
>> >
>> > For my part, it should be nice to have:
>> > - Principal Component Analysis
>> > - Time series
>>
>> I already a very simple time series class. What kind of operations you want 
>> to
>> do on time series ?
>
> For now, I am not using time series.
> But one year ago, I was using Holt Winters and Seasonal Decomposition 
> (HoltWinters and stl functions under R).
> And it is possible that I reuse them in a close future.

I have no idea what is it :-) Can you add some issues regarding your
needs in the SciSmalltalk issue tracker here:
https://github.com/SergeStinckwich/SciSmalltalk/issues

Alex already add some issues as enhancement.

-- 
Serge Stinckwich
UCBN & UMI UMMISCO 209 (IRD/UPMC)
Every DSL ends up being Smalltalk
http://www.doesnotunderstand.org/

Re: [Pharo-users] [Pharo-dev] Call about Numerical Methods in Pharo :)

[Blondeau Vincent]

> -Message d'origine-
> De : Pharo-dev [mailto:pharo-dev-boun...@lists.pharo.org] De la part de
> Serge Stinckwich
> Envoyé : jeudi 3 mars 2016 11:04
> À : Any question about pharo is welcome
> Cc : Didier Besset; Souissi Sami; Pharo Development List
> Objet : Re: [Pharo-dev] [Pharo-users] Call about Numerical Methods in Pharo
> :)
>
> On Thu, Mar 3, 2016 at 10:43 AM, Blondeau Vincent
>  wrote:
> > Hi,
> >
> > That's a very good idea.
> > For now, I use a Pharo-R binding to make the statistics, but it is better 
> > if it is
> directly in the image.
> >
> > For my part, it should be nice to have:
> > - Principal Component Analysis
> > - Time series
>
> I already a very simple time series class. What kind of operations you want to
> do on time series ?

For now, I am not using time series.
But one year ago, I was using Holt Winters and Seasonal Decomposition 
(HoltWinters and stl functions under R).
And it is possible that I reuse them in a close future.

>
> > - Correlation with several variables (Correlation matrix) +
> > significance tests (pearson, spearman...)
> > - Standard statistical tests: normality, variance, mean, distribution
> > - And the more important is to have nice tools to visualize the data (Box
> plot, histogram, scatterplots, PCA ones) with a simple API for standard things
> but that we can customize. The best is that we are able to draw regression
> lines on them. But I think that is more a Roassal part.
>
> You have to ask Alex to the viz :-)

;)
>
> Please join the SciSmalltalk ml in order to express your needs ;-)
> --
> Serge Stinckwich
> UCBN & UMI UMMISCO 209 (IRD/UPMC)
> Every DSL ends up being Smalltalk
> http://www.doesnotunderstand.org/


!!!*
"Ce message et les pièces jointes sont confidentiels et réservés à l'usage 
exclusif de ses destinataires. Il peut également être protégé par le secret 
professionnel. Si vous recevez ce message par erreur, merci d'en avertir 
immédiatement l'expéditeur et de le détruire. L'intégrité du message ne pouvant 
être assurée sur Internet, la responsabilité de Worldline ne pourra être 
recherchée quant au contenu de ce message. Bien que les meilleurs efforts 
soient faits pour maintenir cette transmission exempte de tout virus, 
l'expéditeur ne donne aucune garantie à cet égard et sa responsabilité ne 
saurait être recherchée pour tout dommage résultant d'un virus transmis.

This e-mail and the documents attached are confidential and intended solely for 
the addressee; it may also be privileged. If you receive this e-mail in error, 
please notify the sender immediately and destroy it. As its integrity cannot be 
secured on the Internet, the Worldline liability cannot be triggered for the 
message content. Although the sender endeavours to maintain a computer 
virus-free network, the sender does not warrant that this transmission is 
virus-free and will not be liable for any damages resulting from any virus 
transmitted.!!!"

Re: [Pharo-users] [Pharo-dev] Call about Numerical Methods in Pharo :)

[Blondeau Vincent]

Hi Alex,

This binding allows to execute (almost) any function provided by R via Native 
Calls.

You can find some documentation here: 
http://smalltalkhub.com/#!/~VincentBlondeau/RProjectConnector
I just released a new stable version,  you should be able to load it from the 
catalog (Pharo40 only).
However, you must install R and copy the libraries in the VM folder (see doc).
And it is working only with pre-spur images (the migration to FFI has to be 
done).

Try to run the tests (in RConnector package). Iff they are green, you succeeded 
to configure the binding.

And you can plot the famous Iris Data:
iris := 'eval' asREval: { RObject findVar: 'iris' }.
'plot'
asREval:
{(iris at: 'Petal.Length').
(iris at: 'Petal.Width').
(#pch -> 21).
(#xlab -> 'length').
(#ylab -> 'Width').
(#bg  ->((iris at: 'Species') collect: [ :value 
| {'red'. 'green3'. 'blue'} at: value ])).
(#main -> 'Edgar Anderson''s Iris Data')}.


plot(iris$Petal.Length, iris$Petal.Width, pch=21, 
bg=c("red","green3","blue")[unclass(iris$Species)], main="Edgar Anderson's Iris 
Data")

is the genuine R function.

And inspect them:
iris inspect

You can use any collection of pharo and put it as argument, it should do the 
boxing automatically.

Warning: while testing new functions, don’t forget to save because evaluation 
of a non-existing R function throws a error that is not caught and makes Pharo 
crashing.

Cheers,
Vincent


De : Pharo-dev [mailto:pharo-dev-boun...@lists.pharo.org] De la part de 
Alexandre Bergel
Envoyé : mardi 8 mars 2016 20:23
À : Any question about pharo is welcome
Cc : Didier Besset; Souissi Sami; Pharo Development List
Objet : Re: [Pharo-dev] [Pharo-users] Call about Numerical Methods in Pharo :)

Vincent, can you tell a bit more what this binding is about?
how can I try?

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




On Mar 3, 2016, at 6:43 AM, Blondeau Vincent 
mailto:vincent.blond...@worldline.com>> wrote:

Hi,

That's a very good idea.
For now, I use a Pharo-R binding to make the statistics, but it is better if it 
is directly in the image.

For my part, it should be nice to have:
- Principal Component Analysis
- Time series
- Correlation with several variables (Correlation matrix) + significance tests 
(pearson, spearman...)
- Standard statistical tests: normality, variance, mean, distribution
- And the more important is to have nice tools to visualize the data (Box plot, 
histogram, scatterplots, PCA ones) with a simple API for standard things but 
that we can customize. The best is that we are able to draw regression lines on 
them. But I think that is more a Roassal part.

Thanks for the initiative,

Cheers,
Vincent


-Message d'origine-
De : Pharo-users [mailto:pharo-users-boun...@lists.pharo.org] De la part de
stepharo
Envoyé : mercredi 2 mars 2016 14:17
À : Pharo Development List; Any question about pharo is welcome; Didier
Besset; Souissi Sami
Objet : [Pharo-users] Call about Numerical Methods in Pharo :)

Hi guys

I met Didier Besset and we had a great hacking session and discussions with
Serge Stinckwich.
Didier would like to help Pharo and the numerical part of it. ***Big
thanks*** Didier.

We would like to do several things:

- Work on "Hows to"
  The numerical methods in Pharo is good but the gap between us and the
math is too large :)
 so the idea is to have a series of "how to ..."
- histomgram (simple, based on distribution)

- Improve the SciPharo/NumPha (previously SciSmalltalk) library
This morning we started to implement a
ComponentPrincipalDecomposition by combining two of the
objects available in SciPharo.
Then we started to enhance the distributions to make sure that we can
plug other distribution for having
controlled random number.

- Do a public call to know what is missing for you: this is this mail :)
Didier would like to work on concrete cases. I love that attitude
So tell us :)

Hernan ??
Alex: ?? pvalue? better distribution?
Vincent: covariance? CPA?
Philippe: times series
Serge R frames?
Sami: Better random number and various distributions?

- Organise a two day lectures with practices on concrete case in September
with a recording session.
Either at IRD Bondy or Lille.

- Make sure that the Numerical Method book will get on lulu :) with a
better cover and title :)


Stef


!!!*
"Ce message et les pièces jointes sont confidentiels et réservés à l'usage 
exclusif de ses destinataires. Il peut également ê

Re: [Pharo-users] GTSpotter fuzzy search

[Alexandre Bergel]

Would be fantastic!
-- 
_,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
Alexandre Bergel  http://www.bergel.eu
^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.



> On Mar 9, 2016, at 12:04 PM, Peter Uhnák  wrote:
> 
> Would be possible to have fuzzy-search?
> Or rather Pascal-based search?
> 
> RTAE -> RTAbstractExample
> BCF -> BlConfigurableFormatter
> 
> etc.

[Pharo-users] GTSpotter fuzzy search

[Peter Uhnák]

Would be possible to have fuzzy-search?
Or rather Pascal-based search?

RTAE -> *RTA*bstract*E*xample
BCF -> *B*l*C*onfigurable*F*ormatter

etc.

Re: [Pharo-users] Fwd: Fwd: Phratch + Debian

[jannik laval]

Thank you Hilaire,
I will register on this list.

Jannik

2016-03-09 15:12 GMT+01:00 Hilaire :

> Hey:)
> There are some discussions on educ@April, Phratch being mentionned a
> couple of times (some by myself ;)
>
> Keep the good work.
>
> Hilaire
>
> Le 09/03/2016 08:47, jannik laval a écrit :
> > Finally, I made it.
> >
> > It is available here: http://www.phratch.com/download/phratch.deb
> >
>
> --
> Dr. Geo
> http://drgeo.eu
>
>
>


-- 
~~Jannik Laval~~
Enseignant-chercheur
Responsable Pédagogique Licence Coordonnateur de Projet en Système
d'Information
IUT Lumière, Université Lyon Lumière
laboratoire DISP
http://www.jannik-laval.eu
http://www.phratch.com
http://www.approchealpes.info

Re: [Pharo-users] Fwd: Fwd: Phratch + Debian

[Hilaire]

Hey:)
There are some discussions on educ@April, Phratch being mentionned a
couple of times (some by myself ;)

Keep the good work.

Hilaire

Le 09/03/2016 08:47, jannik laval a écrit :
> Finally, I made it.
> 
> It is available here: http://www.phratch.com/download/phratch.deb
> 

-- 
Dr. Geo
http://drgeo.eu

Re: [Pharo-users] Voyage, Magritte descriptions, and nil exceptions

[Sabine Manaa]

Hi Norbert,
Aha! I unterstand.
This is an interesting aspect. Fortunately I dont have this case at the
moment. I will keep this in mind for the future.
Regards

Am Mittwoch, 9. März 2016 schrieb NorbertHartl [via Smalltalk] :

> Sabine,
>
> maybe my description wasn't clear enough. If you start to use voyage like
> that you have the constraint that all objects returned from the collection
> need to be the same or a subclass of the one you are using to search.
> Meaning
>
> If you query
>
> One selectAll
>
> then all of the objects in the Numbers collection need to be the same
> class or a subclass of One. So this use case is at first only feasible if
> you want to find all instances of Numbers. Your basic query needs to be
>
> Numbers selectAll
>
> If you then want to deal only with instances of One you would need to add
> this to the query
>
> Numbers selectMany: {
> '#instanceOf' -> #One } asDictionary
>
> This is not a nice approach but for this use case it should give you the
> desire result.
>
> Norbert
>
> Am 09.03.2016 um 10:52 schrieb Sabine Manaa <[hidden email]
> >:
>
> Norbert, thank you. Ok, the Numbers collection name is clear.
>
> If I use
>
> 
> ^ VOMongoContainer new
> collectionName: 'Numbers';
> yourself
>
> the data is stored correctly in the Numbers collection in Mongo. Also the
> relationship is fine.
> Everything fine till here.
>
> But after
>
> VOMongoRepository current reset.
>
> I have problems with fetching the data from mongo
>
> If I define
>
> mongoContainer
> 
> ^ VOMongoContainer new
> enableMissingContent;
> collectionName: 'Numbers';
> "kind: self;"
> yourself
>
> Then I get the error "Bad object requested (Three does not inherits from
> One)"
>
> If I define
>
> mongoContainer
> 
> ^ VOMongoContainer new
> enableMissingContent;
> collectionName: 'Numbers';
> kind: self;
> yourself
>
> Then I get an Inspector with 3 instances of "One" (I refer to the
> workspace of the image from tommasso)
> It should be only one instance of One
> Clicking on the one instances, I get an emergency screen with
>  ***System error handling failed
> VOMongoError: Lazy reference not found ...
>
> Tommasso, do you get the same results?
>
> I am curious about the solution.
>
>
> ***This is the workspace of tommasso
> repository dropDatabase.
> repository := VOMongoRepository
> host: 'localhost'
> database: 'VoyageMongo-Link-Test-5'.
> repository enableSingleton.
>
> one := One new.
> two := Two new.
> three := Three new.
>
> one link: two.
> one link: three.
>
> one save.
> two save.
> three save.
>
> VOMongoRepository current reset.
>
> One   selectAll
>
> --
> View this message in context: Re: Voyage, Magritte descriptions, and nil
> exceptions
> 
> Sent from the Pharo Smalltalk Users mailing list archive
>  at Nabble.com
> .
>
>
>
>
> --
> If you reply to this email, your message will be added to the discussion
> below:
>
> http://forum.world.st/Voyage-Magritte-descriptions-and-nil-exceptions-tp4882726p4883475.html
> To start a new topic under Pharo Smalltalk Users, email
> ml-node+s1294792n1310670...@n4.nabble.com
> 
> To unsubscribe from Pharo Smalltalk Users, click here
> 
> .
> NAML
> 
>




--
View this message in context: 
http://forum.world.st/Voyage-Magritte-descriptions-and-nil-exceptions-tp4882726p4883498.html
Sent from the Pharo Smalltalk Users mailing list archive at Nabble.com.

Re: [Pharo-users] Voyage, Magritte descriptions, and nil exceptions

[Norbert Hartl]

Sabine,

maybe my description wasn't clear enough. If you start to use voyage like that 
you have the constraint that all objects returned from the collection need to 
be the same or a subclass of the one you are using to search. Meaning

If you query

One selectAll

then all of the objects in the Numbers collection need to be the same class or 
a subclass of One. So this use case is at first only feasible if you want to 
find all instances of Numbers. Your basic query needs to be

Numbers selectAll

If you then want to deal only with instances of One you would need to add this 
to the query

Numbers selectMany: {
'#instanceOf' -> #One } asDictionary

This is not a nice approach but for this use case it should give you the desire 
result.

Norbert

> Am 09.03.2016 um 10:52 schrieb Sabine Manaa :
> 
> Norbert, thank you. Ok, the Numbers collection name is clear.
> 
> If I use
> 
> 
>   ^ VOMongoContainer new
>   collectionName: 'Numbers';
>   yourself
> 
> the data is stored correctly in the Numbers collection in Mongo. Also the 
> relationship is fine.
> Everything fine till here.
> 
> But after 
> 
> VOMongoRepository current reset.
> 
> I have problems with fetching the data from mongo
> 
> If I define
> 
> mongoContainer
>   
>   ^ VOMongoContainer new
>   enableMissingContent;
>   collectionName: 'Numbers';
>   "kind: self;"
>   yourself
> 
> Then I get the error "Bad object requested (Three does not inherits from One)"
> 
> If I define 
> 
> mongoContainer
>   
>   ^ VOMongoContainer new
>   enableMissingContent;
>   collectionName: 'Numbers';
>   kind: self;
>   yourself
> 
> Then I get an Inspector with 3 instances of "One" (I refer to the workspace 
> of the image from tommasso)
> It should be only one instance of One
> Clicking on the one instances, I get an emergency screen with
>  ***System error handling failed
> VOMongoError: Lazy reference not found ...
>  
> Tommasso, do you get the same results?
> 
> I am curious about the solution.
> 
> 
> ***This is the workspace of tommasso
> repository dropDatabase.
> repository := VOMongoRepository
>   host: 'localhost'
>   database: 'VoyageMongo-Link-Test-5'.
>   
> repository enableSingleton.
> 
> one := One new.
> two := Two new.
> three := Three new.
> 
> one link: two.
> one link: three.
> 
> one save.
> two save.
> three save.
> 
> VOMongoRepository current reset.
> 
> One   selectAll  
> 
> View this message in context: Re: Voyage, Magritte descriptions, and nil 
> exceptions 
> 
> Sent from the Pharo Smalltalk Users mailing list archive 
>  at Nabble.com.

Re: [Pharo-users] Voyage, Magritte descriptions, and nil exceptions

[Sabine Manaa]

Norbert, thank you. Ok, the Numbers collection name is clear.

If I use


^ VOMongoContainer new
collectionName: 'Numbers';
yourself

the data is stored correctly in the Numbers collection in Mongo. Also the
relationship is fine.
Everything fine till here.

But after

VOMongoRepository current reset.

I have problems with fetching the data from mongo

If I define

mongoContainer

^ VOMongoContainer new
enableMissingContent;
collectionName: 'Numbers';
"kind: self;"
yourself

Then I get the error "Bad object requested (Three does not inherits from
One)"

If I define

mongoContainer

^ VOMongoContainer new
enableMissingContent;
collectionName: 'Numbers';
kind: self;
yourself

Then I get an Inspector with 3 instances of "One" (I refer to the workspace
of the image from tommasso)
It should be only one instance of One
Clicking on the one instances, I get an emergency screen with
 ***System error handling failed
VOMongoError: Lazy reference not found ...

Tommasso, do you get the same results?

I am curious about the solution.


***This is the workspace of tommasso
repository dropDatabase.
repository := VOMongoRepository
host: 'localhost'
database: 'VoyageMongo-Link-Test-5'.
repository enableSingleton.

one := One new.
two := Two new.
three := Three new.

one link: two.
one link: three.

one save.
two save.
three save.

VOMongoRepository current reset.

One   selectAll




--
View this message in context: 
http://forum.world.st/Voyage-Magritte-descriptions-and-nil-exceptions-tp4882726p4883467.html
Sent from the Pharo Smalltalk Users mailing list archive at Nabble.com.

[Pharo-users] NBOpenGL on Pharo5

[Richard Wettel]

Hi
I was wondering what the status of NBOpenGL on Pharo5 was. As far as I know, it 
does not work. I would like to have CodeCity work on Pharo5.
Is anybody else (for instance the Roassal3d team) affected by this?
CheersRicky

Re: [Pharo-users] Voyage, Magritte descriptions, and nil exceptions

[Norbert Hartl]

> Am 09.03.2016 um 09:35 schrieb Sabine Manaa :
> 
> you are right.
> 
> I tried to introduce a parent class "Numbers" as superclass of One, Two and 
> Three and defined the relation like this
> 
> mongoLink
>  
>   ^ VOMongoToOneDescription new
>   attributeName: 'link';
>   kind: Numbers;
>   yourself
> 
> But it did not work.
> 
> I get a system error...I don't have so much time but I keep thinking and I am 
> interested in your results.

I think you can always expect problems when using magritte relationships that 
aren't homogenous. Voyage is capable of finding objects that are not of the 
same class. But they need to be in the same inheritance chain. Meaning if you 
request Numbers than all objects being found need to be kindOf Numbers. The 
second aspect is that the collection in mongo is derived from the class name so 
it cannot work because queries over multiple collections are not possible in 
mongo. You need to do


^ VOMongoContainer new
collectionName: 'Numbers';
yourself

in all subclasses of Numbers. This way every object is stored in the same 
collection, a query will find them all and will istantiate them because they 
have the right relationship.

Norbert
> 
> My model is so, that I always have the same types in one relationship, so it 
> did not occur in my project.
> 
> 2016-03-08 19:58 GMT+01:00 Sabine Manaa  >:
> I don't think it is a bug, there is missing the description of the 
> relationship between one and two:
> If I define the following method, it works.
> 
> Thomasso, do you know this documentation?
> 
> https://ci.inria.fr/pharo-contribution/job/EnterprisePharoBook/ws/book-result/Voyage/Voyage.pdf
>  
> 
> 
> 
> One class>>mongoLink
>  
>   ^ VOMongoToOneDescription new
>   attributeName: 'link';
>   kind: Two;
>   yourself
> 
> regards
> Sabine
> 
> 2016-03-08 16:31 GMT+01:00 EstebanLM [via Smalltalk] <[hidden email] 
> >:
> Hi Tommaso, 
> 
> that might be a bug… which version are you using?
> if latest, can you fill a bug here: 
> https://github.com/estebanlm/voyage/issues 
>  (so I do not forget it)?
> 
> thanks,
> Esteban
> 
>> On 08 Mar 2016, at 16:25, Tommaso Dal Sasso <[hidden email] 
>> > wrote:
>> 
>> Hi,
>> 
>> thanks for your answer, I also tried with One>>mongoContainer, but sadly 
>> that does not solve the issue.
>> I think it works in your case because you have the One object cached from 
>> the previous query, but if you inspect the "link" field, it should be nil.
>> 
>> If you reset the repository connection, it should give the exception again.
>> 
>> I'll keep trying :-)
>> Tommaso
>> 
>> 
>> 
>> On 07/03/16 14:07, Sabine Manaa wrote:
>>> Hi Tommaso,
>>> 
>>> you did not define a One>>mongoContainer. I defined it and then it seems to 
>>> work.
>>> 
>>> Regards
>>> 
>>> 2016-03-06 1:42 GMT+01:00 Tommaso DS [via Smalltalk] <>> moz-do-not-send="true" 
>>> href="x-msg://133/user/SendEmail.jtp?type=node&node=4882948&i=0" 
>>> target="_top" rel="nofollow" link="external" class="">[hidden email]>:
>>> Hi, 
>>> 
>>> I am having troubles with Voyage and Mongo. 
>>> I have some Classes that I save in Mongo, say "One", "Two", and "Three". 
>>> All three classes and are voyage roots, so the objects from each class 
>>> are stored in different collections. 
>>> One objects has a reference to Two or Three objects. 
>>> 
>>> The problem arises when I use a Magritte description and define a Mongo 
>>> container to optimize the storage: since each class is stored in its own 
>>> collection, there is no point in storing the #instanceOf attribute as well. 
>>> 
>>> Two>>mongoContainer 
>>>   
>>> 
>>>  ^ VOMongoContainer new 
>>>  kind: Two; 
>>>  yourself 
>>> 
>>> If I add this method, when I try to load an object One, I get the 
>>> exception: "KeyNotFound: key #nil not found in SystemDictionary". 
>>> It looks like Voyage is not able to understand the type of the objects 
>>> stored in the collection Two, when loaded from a reference. 
>>> 
>>> Am I missing something? Do I need to specify something more to make it 
>>> work? 
>>> 
>>> I reproduced the scenario in an image that can be downloaded here: 
>>> https://dl.dropboxusercontent.com/u/11506662/VoyageMongo-Link-Test.zip 
>>> 
>>> 
>>> Do you have any idea? Should I report this on the issue tracker? 
>>> 
>>> Thanks! 
>>> Tommaso 
>>> 
>>> 
>>> 
>>> If you reply to this email, your message will be added to the discussion 
>>> below:
>>> http://forum.world.st/Voyage-Magritte-descriptions-and-nil-exceptions-tp4882726.html
>>>

Re: [Pharo-users] Voyage, Magritte descriptions, and nil exceptions

[Sabine Manaa]

It would be interesting to hear the opinion of esteban.

I geht the error:

***System error handling failed ***
Original error: VOMongoError: Lazy reference not found Numbers: OID(...)

2016-03-09 9:35 GMT+01:00 Sabine Manaa :

> you are right.
>
> I tried to introduce a parent class "Numbers" as superclass of One, Two
> and Three and defined the relation like this
>
> mongoLink
>  
> ^ VOMongoToOneDescription new
> attributeName: 'link';
> kind: Numbers;
> yourself
>
> But it did not work.
>
> I get a system error...I don't have so much time but I keep thinking and I
> am interested in your results.
>
> My model is so, that I always have the same types in one relationship, so
> it did not occur in my project.
>
> 2016-03-08 19:58 GMT+01:00 Sabine Manaa :
>
>> I don't think it is a bug, there is missing the description of the
>> relationship between one and two:
>> If I define the following method, it works.
>>
>> Thomasso, do you know this documentation?
>>
>>
>> https://ci.inria.fr/pharo-contribution/job/EnterprisePharoBook/ws/book-result/Voyage/Voyage.pdf
>>
>>
>> One class>>mongoLink
>>  
>> ^ VOMongoToOneDescription new
>> attributeName: 'link';
>> kind: Two;
>> yourself
>>
>> regards
>> Sabine
>>
>> 2016-03-08 16:31 GMT+01:00 EstebanLM [via Smalltalk] <[hidden email]
>> >:
>>
>>> Hi Tommaso,
>>>
>>> that might be a bug… which version are you using?
>>> if latest, can you fill a bug here:
>>> https://github.com/estebanlm/voyage/issues (so I do not forget it)?
>>>
>>> thanks,
>>> Esteban
>>>
>>> On 08 Mar 2016, at 16:25, Tommaso Dal Sasso <[hidden email]
>>> > wrote:
>>>
>>> Hi,
>>>
>>> thanks for your answer, I also tried with One>>mongoContainer, but sadly
>>> that does not solve the issue.
>>> I think it works in your case because you have the One object cached
>>> from the previous query, but if you inspect the "link" field, it should be
>>> nil.
>>>
>>> If you reset the repository connection, it should give the exception
>>> again.
>>>
>>> I'll keep trying :-)
>>> Tommaso
>>>
>>>
>>>
>>> On 07/03/16 14:07, Sabine Manaa wrote:
>>>
>>> Hi Tommaso,
>>>
>>> you did not define a One>>mongoContainer. I defined it and then it seems
>>> to work.
>>>
>>> Regards
>>>
>>> 2016-03-06 1:42 GMT+01:00 Tommaso DS [via Smalltalk] <>> moz-do-not-send="true"
>>> href="x-msg://133/user/SendEmail.jtp?type=node&node=4882948&i=0"
>>> target="_top" rel="nofollow" link="external" class="">[hidden email]>:
>>>
 Hi,

 I am having troubles with Voyage and Mongo.
 I have some Classes that I save in Mongo, say "One", "Two", and
 "Three".
 All three classes and are voyage roots, so the objects from each class
 are stored in different collections.
 One objects has a reference to Two or Three objects.

 The problem arises when I use a Magritte description and define a Mongo
 container to optimize the storage: since each class is stored in its
 own
 collection, there is no point in storing the #instanceOf attribute as
 well.

 Two>>mongoContainer
  

  ^ VOMongoContainer new
  kind: Two;
  yourself

 If I add this method, when I try to load an object One, I get the
 exception: "KeyNotFound: key #nil not found in SystemDictionary".
 It looks like Voyage is not able to understand the type of the objects
 stored in the collection Two, when loaded from a reference.

 Am I missing something? Do I need to specify something more to make it
 work?

 I reproduced the scenario in an image that can be downloaded here:
 https://dl.dropboxusercontent.com/u/11506662/VoyageMongo-Link-Test.zip

 Do you have any idea? Should I report this on the issue tracker?

 Thanks!
 Tommaso



 --
 If you reply to this email, your message will be added to the
 discussion below:

 http://forum.world.st/Voyage-Magritte-descriptions-and-nil-exceptions-tp4882726.html
 To start a new topic under Pharo Smalltalk Users, email >>> moz-do-not-send="true"
 href="x-msg://133/user/SendEmail.jtp?type=node&node=4882948&i=1"
 target="_top" rel="nofollow" link="external" class="">[hidden email]
 To unsubscribe from Pharo Smalltalk Users, click here.
 NAML
 

>>>
>>>
>>> --
>>> View this message in context: Re: Voyage, Magritte descriptions, and
>>> nil exceptions
>>> 

Re: [Pharo-users] Voyage, Magritte descriptions, and nil exceptions

[Sabine Manaa]

you are right.

I tried to introduce a parent class "Numbers" as superclass of One, Two and
Three and defined the relation like this

mongoLink
 
^ VOMongoToOneDescription new
attributeName: 'link';
kind: Numbers;
yourself

But it did not work.

I get a system error...I don't have so much time but I keep thinking and I
am interested in your results.

My model is so, that I always have the same types in one relationship, so
it did not occur in my project.

2016-03-08 19:58 GMT+01:00 Sabine Manaa :

> I don't think it is a bug, there is missing the description of the
> relationship between one and two:
> If I define the following method, it works.
>
> Thomasso, do you know this documentation?
>
>
> https://ci.inria.fr/pharo-contribution/job/EnterprisePharoBook/ws/book-result/Voyage/Voyage.pdf
>
>
> One class>>mongoLink
>  
> ^ VOMongoToOneDescription new
> attributeName: 'link';
> kind: Two;
> yourself
>
> regards
> Sabine
>
> 2016-03-08 16:31 GMT+01:00 EstebanLM [via Smalltalk] <[hidden email]
> >:
>
>> Hi Tommaso,
>>
>> that might be a bug… which version are you using?
>> if latest, can you fill a bug here:
>> https://github.com/estebanlm/voyage/issues (so I do not forget it)?
>>
>> thanks,
>> Esteban
>>
>> On 08 Mar 2016, at 16:25, Tommaso Dal Sasso <[hidden email]
>> > wrote:
>>
>> Hi,
>>
>> thanks for your answer, I also tried with One>>mongoContainer, but sadly
>> that does not solve the issue.
>> I think it works in your case because you have the One object cached from
>> the previous query, but if you inspect the "link" field, it should be nil.
>>
>> If you reset the repository connection, it should give the exception
>> again.
>>
>> I'll keep trying :-)
>> Tommaso
>>
>>
>>
>> On 07/03/16 14:07, Sabine Manaa wrote:
>>
>> Hi Tommaso,
>>
>> you did not define a One>>mongoContainer. I defined it and then it seems
>> to work.
>>
>> Regards
>>
>> 2016-03-06 1:42 GMT+01:00 Tommaso DS [via Smalltalk] <> moz-do-not-send="true"
>> href="x-msg://133/user/SendEmail.jtp?type=node&node=4882948&i=0"
>> target="_top" rel="nofollow" link="external" class="">[hidden email]>:
>>
>>> Hi,
>>>
>>> I am having troubles with Voyage and Mongo.
>>> I have some Classes that I save in Mongo, say "One", "Two", and "Three".
>>> All three classes and are voyage roots, so the objects from each class
>>> are stored in different collections.
>>> One objects has a reference to Two or Three objects.
>>>
>>> The problem arises when I use a Magritte description and define a Mongo
>>> container to optimize the storage: since each class is stored in its own
>>> collection, there is no point in storing the #instanceOf attribute as
>>> well.
>>>
>>> Two>>mongoContainer
>>>  
>>>
>>>  ^ VOMongoContainer new
>>>  kind: Two;
>>>  yourself
>>>
>>> If I add this method, when I try to load an object One, I get the
>>> exception: "KeyNotFound: key #nil not found in SystemDictionary".
>>> It looks like Voyage is not able to understand the type of the objects
>>> stored in the collection Two, when loaded from a reference.
>>>
>>> Am I missing something? Do I need to specify something more to make it
>>> work?
>>>
>>> I reproduced the scenario in an image that can be downloaded here:
>>> https://dl.dropboxusercontent.com/u/11506662/VoyageMongo-Link-Test.zip
>>>
>>> Do you have any idea? Should I report this on the issue tracker?
>>>
>>> Thanks!
>>> Tommaso
>>>
>>>
>>>
>>> --
>>> If you reply to this email, your message will be added to the discussion
>>> below:
>>>
>>> http://forum.world.st/Voyage-Magritte-descriptions-and-nil-exceptions-tp4882726.html
>>> To start a new topic under Pharo Smalltalk Users, email >> moz-do-not-send="true"
>>> href="x-msg://133/user/SendEmail.jtp?type=node&node=4882948&i=1"
>>> target="_top" rel="nofollow" link="external" class="">[hidden email]
>>> To unsubscribe from Pharo Smalltalk Users, click here.
>>> NAML
>>> 
>>>
>>
>>
>> --
>> View this message in context: Re: Voyage, Magritte descriptions, and nil
>> exceptions
>> 
>> Sent from the Pharo Smalltalk Users mailing list archive
>>  at Nabble.com
>> .
>>
>>
>>
>>
>>
>> --
>> If you reply to this email, your message will be added to the discussion
>> below:
>>
>> http://forum.world.st/Voyage-Magritte-descriptions-and-nil-exceptions