[Scilab-users] Plotting a New Xpoly Directly Inside a Compound

[RolandB]

Hi,

If one plots several curves into a 2D axes using plot2d(), the curves will
be resembled as polyline children within a compound, which is a child of the
according axes.
Adding further polylines one by one using xpoly() puts these polylines
outside the compound on the same level as the compound is located, as
children of the according axes.

Is there a simple way to have the polyline generated directly within the
existing compound?
I know that I can glue several polylines into a compound, but I would have
to unglue the compound first and then glue all polylines together back into
one compound.
Having many (!) curves makes this quite slow. Especially as glue() reverses
the order of the polylines, so I would have to do a second unglue();glue();
in order to preserve the order they had before ungluing (BTW, is this a bug
or is reversal of the order a desired feature?).

On the other hand, what kind of advantage does the compound have at all,
assuming I wouldn't need the possibility of making all of the curves in the
compound visible or invisible by just modifying the visibility of the
compound?
Ok, I can address all curves by axes.children.children(1:$) without having
to care about the legend, which would be together with the curves in
axes.children(1:$) if all the polylines were not in a compound.

But any additional advantage for using a compound?

Regards,
Roland




--
Sent from: 
http://mailinglists.scilab.org/Scilab-users-Mailing-Lists-Archives-f2602246.html
___
users mailing list
users@lists.scilab.org
http://lists.scilab.org/mailman/listinfo/users

Re: [Scilab-users] Corona modelling

[RolandB]

As there have already been some cases where persons have been infected twice,
you would have to take into account that only a finite percentage of
infected people become immune and that this immunity might also only last
some finite time.

Another thought experiment would be to randomly test some percentage of
people every some days and remove them by isolation in case they are
infected.

Further, after someone becomes ill or is tested positive, you could
determine the people he was in contact with (to a certain percentage,
because they would not remember everyone they had been in contact with) and
have them tested too. 

Aside from the mathematical problem, I think in order to be able to return
to a somewhat normal life without a high risk of provoking another wave
(maybe several of them), we need tests that detect the virus directly
(antigen or DNA tests, as antibodies take too long to rise) and we need them
ridiculously cheap. We need to be able to test everyone at least twice a
week, preferably every other day or even every day.
I think I have seen Influenza rapid antigen tests for about 3,5 Euro in some
tens quantities. If one were to produce them in hundreds of millions per day
(you know how fast the Chinese can ramp up factories if there is a demand),
such a test should be able to be produced for less than an Euro.
Combined with an app that tracks everyone's path (I know, protection of
privacy is important, but there should be a way to ensure privacy while
allowing to be able to detect if one has been in close vicinity of an
infected person), that should allow us to avoid an oscillating cycle of
reinfection waves.



--
Sent from: 
http://mailinglists.scilab.org/Scilab-users-Mailing-Lists-Archives-f2602246.html
___
users mailing list
users@lists.scilab.org
http://lists.scilab.org/mailman/listinfo/users

Re: [Scilab-users] Extraction with $ inside brackets together with other elements

[RolandB]

Hi,

I found  bug 11021   ,
which is about this issue.

Seems it's already known and not easily fixable.

Regards,
Roland



--
Sent from: 
http://mailinglists.scilab.org/Scilab-users-Mailing-Lists-Archives-f2602246.html
___
users mailing list
users@lists.scilab.org
http://lists.scilab.org/mailman/listinfo/users

[Scilab-users] Extraction with $ inside brackets together with other elements

[RolandB]

Hi,

somehow extraction with $ inside brackets seems not to work as I think it
should.

plot(sin(0:0.1:2*%pi*0.5));
axes=gca();
axes.zoom_box;// working
0.1147541  -0.9971014   69.73224   0.9913043  -1.   1.

axes.zoom_box();// working
0.1147541  -0.9971014   69.73224   0.9913043  -1.   1.

axes.zoom_box(1:$);// working
0.1147541  -0.9971014   69.73224   0.9913043  -1.   1.

axes.zoom_box([1:4]);// working
0.1147541  -0.9971014   69.73224   0.9913043

axes.zoom_box([1 3 2 4]);// working
0.1147541   69.73224  -0.9971014   0.9913043

axes.zoom_box([1 3 2 4:6]);// working
0.1147541   69.73224  -0.9971014   0.9913043  -1.   1.

axes.zoom_box([1 3 2 4:$]);// not working
Undefined operation for the given operands.
check or define function %s_c_ip for overloading.

It seems that the $ symbol does only work if it is not combined with other
vector elements that are not directly dependent on it.

Is this how it is intended? Should it be changed? Could it be changed?
Is there another way to do what I want to do without having to first check
the size and assign it to a variable that I would use instead of the dollar
symbol?

I know, zoom_box is always a 6 element vector. I wanted to use the
extraction on data_bounds, which does not always have 6 elements, hence the
use of 4:$ after 1 3 2.

Regards,
Roland



--
Sent from: 
http://mailinglists.scilab.org/Scilab-users-Mailing-Lists-Archives-f2602246.html
___
users mailing list
users@lists.scilab.org
http://lists.scilab.org/mailman/listinfo/users

Re: [Scilab-users] Next Scilab releaes?

[RolandB]

Hi,

as the current master branch is often referred to as Scilab 6.1.0, I assume
it will not just be a minor patch release.

I would */highly/* recommend that a potential release date be announced and
that the users be encouraged to test the current master build(s) on their
scripts in order to find (and hopefully be able to fix) as many bugs as
possible before the release.

If you guys don't want to announce a release date, maybe you can announce a
date where you recommend to file bug reports in order to at least still have
a chance to get them fixed before the release.
I remember reading in quite a few bug reports that the pre-release testing
has to be more thoroughly with the next release. I hope that were not only
empty words.

Regards,
Roland 



--
Sent from: 
http://mailinglists.scilab.org/Scilab-users-Mailing-Lists-Archives-f2602246.html
___
users mailing list
users@lists.scilab.org
http://lists.scilab.org/mailman/listinfo/users

Re: [Scilab-users] Problem with event handler.

[RolandB]

Hi,

your problem could be realted to this bug:

http://bugzilla.scilab.org/show_bug.cgi?id=15374

And/or this bug:

http://bugzilla.scilab.org/show_bug.cgi?id=15996

Regards,
Roland





--
Sent from: 
http://mailinglists.scilab.org/Scilab-users-Mailing-Lists-Archives-f2602246.html
___
users mailing list
users@lists.scilab.org
http://lists.scilab.org/mailman/listinfo/users

Re: [Scilab-users] Get duplicates of a vector

[RolandB]

Hi,


B=unique(A)(members(unique(A), A)==2);

Regards,
Roland



--
Sent from: 
http://mailinglists.scilab.org/Scilab-users-Mailing-Lists-Archives-f2602246.html
___
users mailing list
users@lists.scilab.org
http://lists.scilab.org/mailman/listinfo/users

Re: [Scilab-users] Parallel Run

[RolandB]

Hi,

Which version of Scilab do you use?

AFAIK parallel run does not work in Scilab 6.x

It only worked on Scilab 5 and there it would also only work under Linux.

Regards
Roland



--
Sent from: 
http://mailinglists.scilab.org/Scilab-users-Mailing-Lists-Archives-f2602246.html
___
users mailing list
users@lists.scilab.org
http://lists.scilab.org/mailman/listinfo/users

Re: [Scilab-users] callback and output variables

[RolandB]

Hi,

in the callback function you can access the calling GUI element with gcbo.
As those elements have an userdata property, you can assign an array or list
to that property and then access this array or list from elsewhere.

In the callback function:

my_array_or_list(1)=uigetfile();

my_array_or_list(2)="what_else_you_need";

set(gcbo, "userdata", my_array_or_list);

And if you want to access the data:

my_array_or_list=get(findobj("tag";"the_tag_you_had_defined_for_your_button"),
"userdata");







--
Sent from: 
http://mailinglists.scilab.org/Scilab-users-Mailing-Lists-Archives-f2602246.html
___
users mailing list
users@lists.scilab.org
http://lists.scilab.org/mailman/listinfo/users

Re: [Scilab-users] transparently overlapping histograms

[RolandB]

Hi,

Actually there is a possibility for some kind of transparency in Scilab. 

See here:

https://help.scilab.org/docs/6.0.2/en_US/pixel_drawing_mode.html

It would be nicer if the bitwise operations could be done with the number
that finally selects a color from the color table, as that would enable to
achieve much better pseudo-transparency, but the present possibilities are
still better than nothing.

Roland





--
Sent from: 
http://mailinglists.scilab.org/Scilab-users-Mailing-Lists-Archives-f2602246.html
___
users mailing list
users@lists.scilab.org
http://lists.scilab.org/mailman/listinfo/users

Re: [Scilab-users] question on graphic children order

[RolandB]

Hi,

> After the first interesting Stéphane's attempt, i am also afraid that
> changing the order now would break a lot of code. 

It would definitely break a lot of my code.
So if that behaviour would be changed, I would ask for an announcement
lng before it is taking effect.

If you want to index the children in the order they have been created, you
may use ($+1-creationindex).

IMO also the order different children are drawn (polylines...) depends on
their index. They are drawn from $ to 1 so that child 1, which is the most
recent one, is drawn on top.

Regards,
Roland




--
Sent from: 
http://mailinglists.scilab.org/Scilab-users-Mailing-Lists-Archives-f2602246.html
___
users mailing list
users@lists.scilab.org
http://lists.scilab.org/mailman/listinfo/users

Re: [Scilab-users] (no subject)

[RolandB]

This is a test. Please excuse me for hijacking this empty thread.



--
Sent from: 
http://mailinglists.scilab.org/Scilab-users-Mailing-Lists-Archives-f2602246.html
___
users mailing list
users@lists.scilab.org
http://lists.scilab.org/mailman/listinfo/users