forums.silverfrost.com

[aebolzan]

I was trying to run your example program but using the option of x_array, with x_array changing from calc1 to calc2 subroutines.., so I have two data sets, x-y and x-z, so the program code is:

[aebolzan]

Found the problem!!!.....although this is not clear in the notes (at least for me), seems that one has to provide as many as x_array/y_array pairs as needed, so that, the last statement in the previous code should be:

[PaulLaidler]

Agustin

Yes. Maybe there was an error in my sample code.

When using the options [independent,x_array], the x and y arrays are in pairs otherwise there is just one x array.

[DanRRight]

[PaulLaidler]

Dan

The is nothing useful to return from winop@.

[DanRRight]

Not all i=winio@ are useful either but some are. Formally these subroutines with CALL and functions with i=winio@ look and feel like they do the same things.

Why lose the possibility to have potential respond from function? This i=winio@
could be useful for something in future like generally useful is i=winio@

Besides CALLs break the style and confuse

On early stage of development it is possible to change things. Later this will be almost impossible.

[PaulLaidler]

OK. I will look into that.

[aebolzan]

Dear Paul,

I have found some problems with the locations of tick labels. As I am trying to implement the new %pl for plotting different types of data sets during the run of the program (they can be Current-Time, Potential-Time, Current-Potential, etc.), I have implemented a Function Legend that changes the Axis labels depending on what is plotted at a certain time. The result is that the tick labels can appear correctly located below the x-axis, for instance, but once I plotted a second data set, they move upwards (the same for the y-axis) and remain there for the rest of the run. See this code, for instance:

[code]WINAPP
MODULE mydata
USE clrwin
INTEGER n_points(2)
REAL*8 xstart(2),dx(2)
INTEGER,PARAMETER::n=1000
REAL*8:: y(n)=0.0,z(n)=0.0,w(n)
integer*4 :: ancho,alto
logical*4 :: plot=.false.

CONTAINS

INTEGER FUNCTION legend()
integer ancho2,alto2,alto3,ancho3
ancho2=ancho/2.0
alto2=alto-20
ancho3=ancho*0.05
alto3=alto/2.0
if(plot) then
CALL draw_characters@("Time/s", ancho2, alto2, 0)
CALL rotate_font@(90.0D0)
CALL draw_characters@("Current/A", ancho3, alto3, 0)
CALL rotate_font@(0.0D0)
endif
legend = 2
END FUNCTION legend

SUBROUTINE calc1(k)
INTEGER i,x,k
DOUBLE PRECISION p1,p2,p3
p1=1.5d0
p2=150.0d0
p3=15d0
x=0
plot=.true.
DO i=1,k
y(i)=p1*exp(-x/p2)
x=x+1
ENDDO
END SUBROUTINE calc1

SUBROUTINE calc2(k)
INTEGER i,x,k
DOUBLE PRECISION p1,p2,p3
p1=1.0d0
p2=150.0d0
p3=20d0
x=0
plot=.true.
DO i=1,k
z(i)=p1*exp(-x/p2)
x=x+1
ENDDO
END SUBROUTINE calc2

SUBROUTINE calc3(k)
INTEGER i,x,k
DOUBLE PRECISION p1,p2,p3
p1=2.0d0
p2=150.0d0
p3=20d0
x=0
plot=.true.
DO i=1,k
w(i)=p1*exp(-x/p2)
x=x+1
ENDDO
END SUBROUTINE calc3

INTEGER FUNCTION draw1()
CALL calc1(n_points(1))
CALL simpleplot_redraw@()
draw1 = 2
END FUNCTION draw1


INTEGER FUNCTION draw2()
CALL calc2(n_points(2))
CALL simpleplot_redraw@()
draw2 = 2
END FUNCTION draw2

INTEGER FUNCTION clear()
y = 0.0d0
z = 0.0d0
w = 0.0d0
plot=.false.
CALL simpleplot_redraw@()
clear = 2
END FUNCTION clear

INTEGER FUNCTION change()
INTEGER i
if(n_points(2) == 500)then
n_points(2) = 1000
dx(2) = 1.0d0
CALL calc3(1000)
i = CHANGE_GRAPH_STYLE@(2,0,0,255)
i = CHANGE_GRAPH_DATA@(2,0.0d0,w)
else
n_points(2) = 500
dx(2) = 2.0d0
CALL calc2(500)
i = CHANGE_GRAPH_STYLE@(2,2,4,RGB@(0,0,255))
i = CHANGE_GRAPH_DATA@(2,0.0d0,z)
endif
CALL simpleplot_redraw@()
change = 2
END FUNCTION change
END MODULE mydata

PROGRAM main
USE mydata
external legend
ancho=0.8*clearwin_info@('SCREEN_WIDTH')
alto=0.8*clearwin_info@('SCREEN_DEPTH')

n_points(1) = 1000
n_points(2) = 500
xstart = 0.0d0
dx(1) = 1.0d0
dx(2) = 2.0d0
i=winio@('%ww[no_border]%ca[Damped wave]%pv&')
i=winio@('%mn[Edit[Draw first,Draw second,Change second,Clear

[PaulLaidler]

Thanks. I will take a look at it.

[aebolzan]

Ooopppss....seems that the code is incomplete. This is the main program:

[DanRRight]

What I also missed in %pl and had no explanation why it was not included was keyword Y_MAX=Value. It has Y_MIN=Value but not Y_MAX.

[aebolzan]

what I have not checked is if [x_array] and [xmin,dx] are exclusive options.....although it seems so.....more interesting is the fact that the location of ticks follows a very funny code: it is impossible to know what the plot will show....until it appears...

Using [x_array] you can select somehow the xmin and xmax limits of the data in a plot, but I also found some strange consequences of this (in the tick labels) that I am still investigating....

Agustin

[DanRRight]

I'd be patient, making good universal utility is time consuming, lot and lot of adjustments will still be done I am sure.

Great that INDEPENDENT was introduced but this will complicate the development.

I already see improvements even in current %pl, it became more stable, before it crashed at values > 1e22, now it goes much higher and is still OK.
Though if you add into settings parameter list by mistake something not supported (like X_min=1. for example, probably needs to be added too if it was not) it might still crash

[aebolzan]

I think that this new tool needs more symbols for scatter plots and more control on the axes, first...then maybe the possibility of drawing axis on top and right....it should also be noted that the scheme is rather rigid, i.e. one cannot change line style or colour on the fly, it should be set at the very beginning......but I think I can live with this.......at least for now.....

Agustin

[PaulLaidler]

Thanks for the feedback.

1) Y_MIN and Y_MAX appear in the original documentation.

2) [x_array] sets a TRUE/FALSE flag for all the graphs.

3) CHANGE_GRAPH_STYLE@ can be called to change the line style, the pen style and the colour.

What would the extra plot symbols look like?