PyNGL and PyNIO graphics exercises
- Basic graphical exercises
- Color exercises
- Title exercises
- XY plot exercises (set 1)
- XY plot exercises (set 2)
- Contour plot exercises
- Contours over map exercises (set 1)
- Contours over map exercises (set 2)
- Vector plot exercises
- Primitives exercises (set 1)
- Primitives exercises (set 2)
- Primitives exercises (set 3)
- Paneling exercises
Basic graphical exercises
The basic_ex.py script generates dummy data to draw a simple line plot (also known as an "XY plot"). Use this script for to start the following exercises:
- The graphical output is currently going to an X11 window. Change
the graphical output to go to a PostScript or PDF file called
"basic_ex01.ps" or "basic_ex01.pdf".
[Hint] [Answer] [Image] - Note how the image in the previous example is rotated ("landscape"). (The reason for
this will be covered in the next exercise.) Using this script, force
the orientation of the plot to be right side up ("
portrait").
[Hint] [Answer] [Image] - Using the basic_ex.py script,
set the nglMaximize resource to False to
see what this does to the plot.
[Hint] [Answer] [Image] - Modify the basic_ex.py script to
generate a second set of points using the "cos" function. Use these
points to create a second XY plot.
Try it first with the output going to an X11 window, and then with the output going to a PostScript file.
[Hint] [Answer] [Image 1] [Image 2] - Using the script from the previous exercise, set the nglFrame resource to False for the first plot
only, to see what happens to the plots.
[Hint] [Answer] [Image] - Take the script from the previous exercise and set the nglFrame resource to False for both plots so they
are both drawn in the same frame. Draw the two plots in different
locations on the frame so they don't overlap.
In order for a valid graphical image to be created, the frame needs to be advanced manually (since we turned off the automatic frame advance), so after both plots are drawn, add a frame advance call.
[Hint] [Answer] [Image]
Color exercises
- Write a script to draw the default color table.
[Hint] [Answer] [Image] - Using the above script you wrote, change the color table to "BlGrYeOrReVi200".
[Hint] [Answer] [Image]Try drawing some other predefined color tables. You can see the full list of available color tables in the color table gallery.
- Keeping the "BlGrYeOrReVi200" color table, add gray to it.
[Hint] [Answer] [Image]Try adding other colors, using different RGB triplets.
- Create your own color table using
RGB triplets.
[Hint] [Answer] [Image] - Create your own color table using named
colors.
[Hint] [Answer] [Image]
Title exercises
Look at the title_ex.py script that sets a main title. Use this script for the following exercises:
- Add two title
resources to change the font and the color of the title. Try using
both a named color and a color map index value.
[Hint] [Answer] [Image] - Add titles to the X and Y axes, and increase the font height
of both.
[Hint] [Answer] [Image] - Add a Y title to the right axis and change the orientation of the
text so the letters are right side up.
[Hint] [Answer] [Image]
XY plot exercises (set 1)
Use the xy_ex.py XY plot script as a base for the following exercises:
- Add an xy
resource to change the line color. Try using both a named color and
a color map index value.
[Hint] [Answer] [Image] - Add an "xy" resource to make the line four-and-a-half times as
thick.
[Hint] [Answer] [Image] - Add three "xy" resources to 1) change to "marker" mode, 2) make the
marker a filled dot, and 3) change the marker color to "red".
[Hint] [Answer] [Image] - Using the script from the previous exercise, use Ngl.new_marker to define your own XY marker,
and use this instead. Subscript the "y" data array so that only every
third value is plotted.
[Hint] [Answer] [Image]
XY plot exercises (set 2)
You will need to download the xy2.txt ASCII file to do these exercises.
- Modify the xy2_ex.py XY plot
script to add an X array with 500 points that goes from -100 to 100.
Change the Ngl.y call to Ngl.xy and pass it this new X array.
[Hint] [Answer] [Image] - Using the previous script, plot the full "y" array as four
separate curves. In order to do this, the array needs to be
dimensioned ncurves (4) x
npts (500), which means it needs to be reordered first.
[Hint] [Answer] [Image] - Modify the script from the previous exercise so the curves have
different line thicknesses and colors. Also, use a special "ngl"
resource to point the tickmarks outward.
[Hint] [Answer] [Image] - Using the previous script, change the limits of the X and Y axes
so that the X axis goes from min(x) to max(x), and the Y axis goes
from -500 to 1100.
[Hint] [Answer] [Image] - Using the previous script, add a legend to the XY plot.
[Hint] [Answer] [Image] - Note the strange labels in the legend. These are default labels
used by PyNGL.
Using the script from the previous exercise, change the labels in the legend to something else, and also make the legend box smaller and move it inside the plot.
[Hint] [Answer] [Image]
Contour plot exercises
Use the contour_ex.py contour plot script as a base for the following examples:
- Add some contouring resources to change the contour levels to
start at 5 and go to 95 in steps of 5. Also, use a special "ngl"
resource to point the tickmarks outward.
[Hint] [Answer] [Image] - Using the script from the previous example, make each contour line
3 times as thick, and each a different color.
[Hint] [Answer] [Image] - Remove all resources from the previous script except for the one
that points the tickmarks outward, and add new resources to turn on
contour full (the full color map will be spanned by default), and turn
off contour lines and line labels.
[Hint] [Answer] [Image] - Using the script from the previous example, add a main title,
change the font color, and change the font to times-bold
[Hint] [Answer] [Image] - Using the script from the previous example, make the
labelbar horizontal, and turn off the labelbar box lines.
[Hint] [Answer] [Image] - Using the script from the previous example, change the major
tickmark spacing on the Y axis to 5, turn off the top tickmarks, and
explicitly label the bottom axis tickmarks at values 5, 10, 20, 25,
and 35 with the labels "Jan", "Feb", "Mar", "Apr", and "May".
[Hint] [Answer] [Image] - Using the script from the previous example, make the labelbar
width slightly longer, the height slightly shorter, and move the
labelbar closer to the plot.
[Hint] [Answer] [Image]
Contours over maps exercises (set 1)
- Change the contourmap_ex.py script so that
the contours are overlaid on a cylindrical equidistant map. You will
notice that the contours are overlaid on a small portion of the map.
We'll fix this in the next exercise.
[Hint] [Answer] [Image] - Set two scalar
field resources in order to get the contours overlaid
correctly on the map.
[Hint] [Answer] [Image] - Using what you learned from the contour plot exercises, edit the
script from the previous exercise to do the following:
- Manually set the contour levels to go from 195 to 328 in steps of 5.
- Turn on contour fill, and turn off contour lines and line labels.
- Using the script from the previous exercise, change the labelbar
orientation to horizontal, remove the labelbar box lines, and move
the labelbar closer to the plot.
[Hint] [Answer script] [Image] - Using the script from the previous exercise, make it so
there are fewer labelbar labels. This will cause the labels to
get bigger, so resize them to make them smaller.
[Hint] [Answer script] [Image] - Using the script from the previous exercise, set some
map resources to turn off the lat/lon grid lines, turn on
map fill such that the oceans and major inland water areas
are filled in with "tan", and the land is not filled at all.
[Hint] [Answer script] [Image]
Contours over maps exercises (set 2)
- Run the contourmap2_ex.py and notice
the output from the "print" of the file. There
are variables "t", "lat", and "lon" on the file. The lat, lon arrays
are not coordinate arrays of "t", but we happen to know that they
are indeed supposed to be coordinate arrays.
Use the lat, lon arrays to correctly overlay the contours on a map. Note that these lat, lon arrays only cover the area over the United State. We'll deal with this in the next exercise.
Note that "t" contains missing value and that a resource is being set to avoid plotting these values.
Set the main title to "January 1996 Snow Storm".
- Using the script from the previous example, zoom in on the area
defined by the latitude/longitude arrays (it's an area that
encompasses the United States).
[Hint] [Answer] [Image] - Using the script from the previous example and resources that you
learned from previous examples, 1) change the color map to "temp1", 2)
turn on contour fill, 3) turn off the contour lines and labels, 4) set
the contour level spacing to 2.25, 5) make the labelbar horizontal,
and 6) change the labelbar labels to label only every 5th box.
[Hint] [Answer] [Image] - Using the script from the previous example, change the projection
to "LambertConformal". This will require tweaking the center
latitude/longitude, and setting the map limit mode.
[Hint] [Answer] [Image] - Using the script from the previous example, set some map resources
to change the draw order of the grid lines to "predraw", and to set a
dash pattern, color, thickness, and spacing for the grid lines.
Vector plot exercises
Use the vector_ex.py vector plot script as a base for the following examples:
- Add a single vector resource to make the vectors multi-colored.
[Hint] [Answer] [Image] - Using the script from the previous example, set the reference
magnitude and increase the length of it.
[Hint] [Answer] [Image] - Using the script from the previous example, change
the color map to "rainbow+gray", and set some resources to
start at color index 16, and to not include the last color
in the color map (which is gray). Point the tickmarks outward.
[Hint] [Answer] [Image] - Set some vector resources to double the thickness of the line
arrows, and to move the little reference annotation box to inside the
plot area, flush left and bottom of the plot.
[Hint] [Answer] [Image] - Set some resources to set the vector levels to a fixed array,
and to set the main title to a concatenation of the
"reftime" on the file, and the appropriate "timestep".
[Hint] [Answer] [Image] - Set some vector field resources to set the lat, lon arrays. See
what this does to your X/Y axes values.
Make the labelbar bar thinner and the labels smaller.
- Tweak the previous example to generate a vector plot at
each timestep, but skip over the fields that are all
missing values in either u or v. Make sure the title
changes with each timestep.
[Hint] [Answer] [Animation]
For more vector plot examples, see the vector applications page.
Primitives exercises (set 1)
- Using the poly_ex.py script
that draws a single-curve XY plot, add a marker (also called a
"polymarker") at the point (-20,-300). To do this, you will need to
turn off the automatic frame advance so the page won't advance
before you draw the marker.
[Hint] [Answer] [Image] - Note that it is hard to see the marker from the previous exercise.
Using this same script, change the marker to a purple filled dot that
is larger than the default size.
[Hint] [Answer] [Image] - Use the script from the previous exerise. Create three new markers
using the Ng.new_marker function and
selecting markers from the weather2 font table.
Draw three groups of markers using these new markers, each one a different color and size. Also, draw multiple markers for each group.
- Use the script from the previous exercise. Instead of
using Ngl.polymarker,
use Ngl.add_polymarker
to attach the markers to the plot.
When you do this, the markers become part of the plot, and won't be drawn until the plot is drawn. This is useful if you later plan to resize the plot, or pass a series of these plots to Ngl.panel.
Primitives exercises (set 2)
- Using the poly2_ex.py script
that draws a zoomed in view of a cylindrical equidistant map, add some
lines (also called "polylines") to make a box at the lat/lon
coordinates (30,-90), (30,-45), (0,-45), and (0,-90). Make the lines
red and twice as thick as the default.
[Hint] [Answer] [Image] - Using the same code from the previous example, change the lines to
a "firebrick" (named color) filled polygon.
[Hint] [Answer] [Image] - Note that the polygon in the previous example has the map outlines
drawn on top of it. Using this script, add two map resources so that
the map outlines and grid lines are masked by the polygon fill.
- Use the script from the previous exerise. Change the solid fill to
a purple pattern fill, and increase the density of the fill pattern.
- Using the previous exercise and the one from example 1, draw the
polygon and then outline the polygon with lines. Add markers at all
four corner of the boxes. Try playing with "gs" resources to change
things like color, sizes, thickness, and fill patterns.
- Using the script from the previous exercise, add a text box in the
middle of the polygon.
Primitives exercises (set 3)
- Using the poly3_ex.py script
that draws a simple XY plot, change each Y value to be
a bar that goes from the bottom X axis to the Y value. The bar
should straddle the X value that it is associated with, and
be a width of 30% of the distance between the X values.
The best way to do this is to loop through each Y value, and call Ngl.xy for each bar. Use a function to return the X,Y coordinates of each bar.
- Using the script from the previous exercise, remove the tickmarks and their labels for the bottom, top, and right axes.
- Using the script from the previous exercise, define a new color map using the named colors: ["white", "black", "FireBrick", "Red", "Orange", "Green", "Navy", "Blue", "SkyBlue","SlateBlue"]. Fill each bar in a different color, starting with "FireBrick".
- Using the script from the previous exercise, move and resize the plot so that the upper left corner is at y=0.95, the width is 0.70, and the height is 0.80.
- To get ready for adding some text and filled boxes outside the plot area, add a call Ngl.draw_ndc_grid to the previous script to draw an NDC grid.
- Using the script from the previous exercise, add a box outline (in color "firebrick" and 3 times as thick as the default) with an upper left corner of (0.2,0.12) and a width of 0.07 and a height of 0.05.
- Using the script from the previous exercise, replace the single
box outline with eight box outlines with the same width and height,
and with upper left corners of:
xpos = [0.20,0.40,0.60,0.80,0.20, 0.40, 0.60, 0.80] ypos = [0.13,0.13,0.13,0.13,0.055,0.055,0.055,0.055]
Outline each box in a different color, starting with "firebrick".
- Using the script from the previous exercise, add a main title of
"Estimated Cancer Deaths for 2002". You will need to slightly decrease
the value of vpYF and vpHeightF to make room for the title.
Also add text strings ["Lung", "Colon", "Breast", "Prostate", "Pancreas", "Lymphoma", "Leukemias", "Ovary"] to the right of each box.
- Using the script from the previous exercise, fill each box in the same color used for the lines, and then outline the box in black. Also outline each bar in the plot in black.
- Using the script from the previous exercise, make the Y tickmarks length 0 (effectively removing them), and draw horizontal dashed grid lines at Y major tickmark locations. Remove the call to Ngl.draw_ndc_grid.
Paneling exercises
This suite of exercises shows how to put multiple plots (of the same size) on a single frame.- Using the panel_ex.py script
that draws a contour plot using dummy data, create three more
individual contour plots, using Ngl.generate_2d_array to generate more
(unique) dummy data.
[Hint] [Answer] [Image] (only first image shown)
- Modify the script from the previous example to panel the four
plots into 2 rows and 2 columns in one frame.
[Hint] [Answer] [Image] (only last image shown)
- Modify the script from the previous example to turn off the
drawing of the individual plots (leave the paneled plots as is).
- The labelbars in the previous script all have different
ranges. Modify the script from the previous example
to make the labelbars use the same range, by setting a minimum and
maximum contour level, and a spacing.
- Modify the script from the previous example to set the contour
levels via an explicit array of levels that are not equally spaced.
- Modify the script from the previous example to turn off the
individual labelbars for each plot, and then set a common labelbar in
the paneled plots.
- Modify the script from the previous example to add a single figure
string to each plot, and to decrease the amount of white space between
the two rows of plots.
- Modify the script from the previous example to point the tickmarks
outward in the individual plots, and turn off the top and right
tickmarks. It will no longer be necessary to use the nglPanel resource
you set in the previous script to decrease the white space.
- Modify the script from the previous example to add a bold
title at the top of the panelled plots.
