ECE220: Computer Systems & Programming
Machine Problem 5
Using Functions to Draw a Picture
Your task this week is to write a selection of functions that act as simple drawing tools to create a PNG
image. Specifically, you will write functions that can draw lines, rectangles, triangles, parallelograms, and
circles, as well as a color gradient in a rectangular shape. You will then write a function that calls the other
functions to draw a picture of anything you like.
The objective for this week is for you to gain some experience with functions in C, particularly with using
functions as building blocks for more complex functions. This skill is critical when developing software
based on existing libraries and toolkits.
A PNG image is one of several different image
formats for computers. In this MP, it can be
abstracted to a rectangular grid of pixels like the
one shown to the right, where each pixel is
represented by an empty square. Each pixel has a
location in the grid that can be written as a
coordinate pair, (x,y). For the purposes of this MP,
the pixel in the top-left corner of the image has a
position of (0,0). The pixel directly to the right of
that has a position of (1,0), incrementing the
x-position by 1. Following this trend, for an image
that is WIDTH pixels wide, the pixel in the topright corner will have a position of (WIDTH-1,0). The y-position increases as the rows go down, so the row
on the bottom has a y-position of HEIGHT-1 when the image is HEIGHT pixels tall.
Each pixel has a color value associated with it, completely independent of any other pixels. To facilitate a
wide range of colors, the pixel’s color is divided into three channels, representing the relative intensities of
the colors red, green, and blue in the color of that pixel. Each of these channels can have 256 different levels
of intensity. Note that one byte can take on 256 different values, making it a perfect way to represent the
level for each channel.
As might be expected, the greater the number representing the red channel, the more red the pixel will look,
whereas a number close to zero means that the pixel is not red at all. This definition is the same with green
and blue. 255 red, 0 green, and 0 blue represents pure red, while 0 red, 255 green, and 0 blue is pure green,
and 0 red, 0 green, 255 blue is pure blue. 0 for all three channels is a complete absence of color, or black,
while 255 for all three channels is white. A great resource for trying out different colors can be found at:
You will be given the following two C functions to use as tools when you write your code:
int32_t draw_dot(int32_t x, int32_t y);
void set_color(int32_t color);
The draw_dot function the lowest level function in the hierarchy that you must write. It takes two
parameters: x and y. It will change the color of the pixel at the point (x,y) to be whatever color you have
set and return 1. If the pixel given is out of bounds (the image is WIDTH pixels wide and HEIGHT pixels
high), draw_dot returns 0 without drawing anything.
The set_color function changes the color that draw_dot uses for drawing pixels. Once your code
calls set_color, every subsequent use of draw_dot will use the color set by the most recent call to
set_color. At the beginning of execution, this color will be white.
The parameter color is an int32_t, which means it is 32 bits long, or four bytes. The most significant
byte is ignored. The second most significant byte (bits 23-16) represents the red channel of the color. The
third most significant byte (bits 15-8) represents the green channel of the color. The least significant byte
(bits 7-0) represents the blue channel of the color.
Not all functions that you have to write will use either of these two functions, but some of your functions
will require that you make calls to them. Make sure that you are only calling these two functions when
writing functions where you are explicitly told to use either draw_dot or set_color.
Each function that you write returns an int32_t. Some of these functions call draw_dot. For those that
do, the function must return the AND of the return values of all calls to draw_dot. This rule means that
if any of the pixels drawn are outside the bounds of the picture, your function will return 0, and it will return
1 otherwise. However, you are still required to complete all calls to draw_dot, even if the return value is
known to be 0.
Functions that call other functions you have written must find the AND of the return values of each of the
functions called, and return this value. Once again, you are required to complete all function calls, even if
the return value is known to be 0.
The functions you are required to write are as follows:
There are two similar functions for drawing lines called near_horizontal and near_vertical.
The function near_horizontal should only be called for lines that have a slope between -1 and 1
inclusive. Such slopes look shallow to a human eye. On the other hand, near_vertical should be called
on lines that have a slope larger than 1 or smaller than -1. Given two pixels at coordinates (x1,y1) and (x2,y2),
the slope is calculated by ୷ଶି୷ଵ
. Note that you do not need to perform the division to check which type
should be drawn. When (x1,y1) = (x2,y2), call near_vertical.
int32_t near_horizontal(int32_t x_start, int32_t y_start,
int32_t x_end, int32_t y_end);
For this function, you will use point-slope form to select a y-coordinate that corresponds to each
x-coordinate between x_start and x_end, including those two points. Begin by deciding which point is
(x1,y1) and which is (x2,y2). Depending on how you choose to order the two points, one part of your code
will be simpler, while another will be more complex. Think carefully about the loop structure and the
algebra. You will then use draw_dot to fill in the pixel at each location. Point-slope is as follows:
𝑦 − 𝑦ଵ =
(𝑦ଶ − 𝑦ଵ
)(𝑥 − 𝑥ଵ)
(𝑥ଶ − 𝑥ଵ)
However, we want the lines to be smooth, so we need to round off the integer division. We first multiply
both numerator and denominator by two, then add another half of the denominator to the numerator. Since
most processors round towards zero, however, we need to “add” the extra half in the right direction. In
particular, we need a factor of sgn(𝑦ଶ − 𝑦ଵ), the sign of y2 – y1, which is 1 when y2 > y1, 0 when y2 = y1,
and -1 when y2 < y1. You do not need to implement the 0 case, however. We obtain:
𝑦 − 𝑦ଵ =
2(𝑦ଶ − 𝑦ଵ
)(𝑥 − 𝑥ଵ
) + sgn(𝑦ଶ − 𝑦ଵ)(𝑥ଶ − 𝑥ଵ)
2(𝑥ଶ − 𝑥ଵ)
This equation can be rearranged to be:
(2(𝑦ଶ − 𝑦ଵ)(𝑥 − 𝑥ଵ) + sgn(𝑦ଶ − 𝑦ଵ)(𝑥ଶ − 𝑥ଵ))
(2(𝑥ଶ − 𝑥ଵ)) ൨ + 𝑦ଵ
It is imperative that the function evaluate these expressions using the parentheses as specified. This
expression will generate correct intermediate values that will yield a final output identical to the one
generated in the gold code.
int32_t near_vertical(int32_t x_start, int32_t y_start,
int32_t x_end, int32_t y_end);
Unlike the previous function, for this one, you will use point-form to select an x-coordinate that corresponds
to each y-coordinate between y_start and y_end, including those two points. As before, you must first
choose which point is (x1,y1) and which is (x2,y2). Also unlike the previous function, the two points may
be identical, in which case your function should draw a single dot (it is up to you to guarantee that
near_horizontal does not receive identical points when called from your code). You will then use
draw_dot to fill in the pixel at each location. You can use the equation in this form:
𝑥 = ቈ
(2(𝑥ଶ − 𝑥ଵ)(𝑦 − 𝑦ଵ) + sgn(𝑥ଶ − 𝑥ଵ)(𝑦ଶ − 𝑦ଵ
(2(𝑦ଶ − 𝑦ଵ
Again, it is imperative that you evaluate the expression using the exact parentheses specified in this
int32_t draw_line(int32_t x_start, int32_t y_start,
int32_t x_end, int32_t y_end);
This function must call one of near_horizontal or near_vertical depending on the slope of the
line between (x_start, y_start) and (x_end, y_end). Remember that given two pixels at
coordinates (x1,y1) and (x2,y2), the slope is calculated by ୷ଶି୷ଵ
. Also, remember that near_horizontal
should only be called for lines that have a slope between -1 and 1 inclusive, and near_vertical should
be called on lines that have a slope larger than 1 or smaller than -1. For lines for which the slope is not
defined (that is, for which the two points are the same), your code must call near_vertical.
int32_t draw_rect(int32_t x, int32_t y, int32_t w, int32_t h);
This function draws…you guessed it, a rectangle! It draws a rectangle of a height and width specified in the
parameters as h and w, respectively, with x and y being the coordinates of the top left corner of the rectangle.
The upper right corner is at (x+h,y), and so forth. If either the height or the width is negative, the function
must immediately return 0 without altering the image in any way. The sides of the rectangle must be drawn
by calling the draw_line functions.
int32_t draw_triangle(int32_t x_A, int32_t y_A,
int32_t x_B, int32_t y_B,
int32_t x_C, int32_t y_C);
A triangle is any shape that has three sides. For this function, you will be given three x-values and three
y-values with the suffixes A, B, and C corresponding to the three vertices of the triangle. You are expected
to use the draw_line function to connect each of the three pairs of vertices with a line. This technique
will form a triangle.
int32_t draw_parallelogram(int32_t x_A, int32_t y_A,
int32_t x_B, int32_t y_B,
int32_t x_C, int32_t y_C);
A parallelogram is a four-sided shape such that opposite sides are parallel, as shown below.
For this function, you will be given three points of the parallelogram in the form of three sets of x-values
and three sets of y-values with the suffixes A, B, and C to show which point they correspond to. The sides
AB and BC are part of the parallelogram, but it is important to note that AC is not. It will be up to your
function to determine the location of the final point, D, and use it to draw the two remaining sides, CD, and
DA. You must use draw_line to accomplish this task. If any of the pixels in the parallelogram lie out of
bounds, you must return 0. Otherwise, return 1.
An important piece of information to note while drawing your
parallelogram is that for any side, the relationship in space between the
two end points is exactly the same as the relationship between end
points on the opposite side. For example, in the parallelogram to the
right, the point on the left is three squares down and three squares to
the left of the point on the top. Likewise, the point on the bottom is
three squares down and three squares to the left of the point on the
int32_t draw_circle(int32_t x, int32_t y,
int32_t inner_r, int32_t outer_r);
This function draws a circle centered at (x,y). The function must first verify that the inner radius given is
greater than or equal to 0. The outer radius must be greater than or equal to the inner radius. If either of
these constraints do not hold, the function must return 0 without altering the image in any way. The circle
must use draw_dot to fill in every pixel in between the inner radius and outer radius (inclusive on both
sides). To determine whether a pixel is in the given range, use the Pythagorean theorem/distance formula:
(xଶ − 𝑥ଵ)
ଶ + (𝑦ଶ − 𝑦ଵ)
ଶ = distanceଶ
You must verify that the pixel to be filled has a distance that is less than or equal to the outer radius and
greater than or equal to the inner radius.
int32_t rect_gradient(int32_t x, int32_t y, int32_t w, int32_t h,
int32_t start_color, int32_t end_color);
For this function, like draw_rect, you are given the coordinates
of the pixel at the top-left corner and the height and width of the
rectangle to draw. If the height is less than 0, or the width is less
than 1, this function must immediately return 0 without altering the
image in any way. This function is unique, because instead of using
draw_line to draw the sides of the rectangle, you will be using a
combination of draw_dot and set_color to create a rectangular
region in the image that forms a gradient from start_color to
end_color from left to right in the rectangle. The image to the left
shows a rectangular region with a gradient from black to white. (This image also has a black border for
clarity. Your function should not draw a border.)
In order to draw the gradient, you will need find the intensities of the three channels in each of the colors
given, then use the following equation to determine for each pixel in the image the amount of red, green,
and blue will be in the color.
𝑙𝑒𝑣𝑒𝑙 = ቈ
(2(𝑥 − 𝑥ଵ
)(𝑙𝑒𝑣𝑒𝑙ଶ − 𝑙𝑒𝑣𝑒𝑙ଵ
) + sgn(𝑙𝑒𝑣𝑒𝑙ଶ − 𝑙𝑒𝑣𝑒𝑙ଵ)(𝑥ଶ − 𝑥ଵ
(2(𝑥ଶ − 𝑥ଵ
Here, x is the x-location of the pixel for which you are finding the color, x2 is the x-location of the right
edge of the rectangle, and x1 is the x-location of the left edge of the rectangle. level refers to the value 0-255
representing the intensity of one of the three color channels at this pixel. Similarly, level1 and level2
represent the corresponding levels at the left and right edges of the rectangle, respectively. Using this
equation to calculate the level for all three channels at a pixel will allow you to form the correct color for
that pixel and pass it to set_color, and then you can use draw_dot to draw a pixel of the correct color
to the image. Using this approach for every pixel in the rectangle will result in a smooth transition from
start_color to end_color. As before, the function must evaluate the expression using the
parentheses specified in order to generate the proper output.
draw_picture takes no parameters and may call any or all of the other functions to draw a picture of
your choice. When you write this function, you have two choices. The first choice is that you may call the
other functions you have written to draw a five-or-more-letter word in the image (but not “IllINI”—sorry!).
The second choice is to draw a picture of your own design that has greater or equal complexity compared
with a five-or-more-letter word. Use your best judgement when adopting this choice; at least one member
of the course staff must agree in advance that the picture is of suitable complexity in order for it to be judged
Your program will consist of a total of two files:
mp5.h This header file provides function declarations and a brief description of the
functions that you must write for this assignment.
mp5.c The main source file for your code (you must write it yourself). Include the
mp5.h header file and be sure that your function matches the one in the header.
Two other files are also provided to you:
main.c A source file that interprets commands and calls your function.
Makefile A file that allows you to use make commands to compile your code.
You need not read these files, although you are welcome to do so.
You should read the description of the functions in the header file before you begin coding.
Your code must be written in C and must be contained in the mp5.c file in the mp/mp5 subdirectory
of your repository. We will NOT grade any other files. Changes made to any other files WILL
BE IGNORED during grading. If your code does not work properly without such changes, you
are likely to receive 0 credit.
The image given is 624 pixels wide and 320 pixels high. There are two #define statements in
mp5.h that define WIDTH to be 624 and HEIGHT to be 320. Use these names in your code if
Each function that you write will return 1 if all of the pixels it draws are within the bounds of the
image and 0 if any of the pixels are out of bounds.
o For functions that call draw_dot directly, you will return the AND of all calls to
draw_dot. This rule means that if any call to draw_dot passes a pixel location that is
out of bounds, the return value for that call will be 0, making the total return value 0. But
if all pixels are in the image, the total return value will be 1. You will not return from the
function until all calls to draw_dot have been made, so even if one of the calls return 0,
you must still make all of the remaining calls. For example, if draw_circle draws a
circle which has a right half that is out of bounds and your code determines that the final
return value is 0 before it has drawn any of the bottom half of the circle, you must still draw
the rest of the circle. This approach will result in the appearance that whatever image you
have drawn is completely drawn, but clipped off the side of the screen.
o Functions that do not call draw_dot will make calls to other functions that you have
written. These functions will return 1 or 0. The return value of the function that calls them
will be the AND of each of the functions called. You are still required to call all of the
functions and only return after attempting to draw all parts of the shape.
o For near_vertical, your function must handle the case of two identical points by
drawing a single dot (using one call to draw_dot).
o For draw_rect and rect_gradient, if the given height or width is negative, or if the
width is 0, your function must return 0 and should not modify the image in any way.
o For draw_circle, inner_r must be greater than or equal to 0. Likewise, outer_r
must be greater than or equal to inner_r. If either of these two relations is not observed,
your function must return 0 and should not modify the image in any way.
The top-left pixel in the image has coordinates (0,0).
o Pixels increase in x-location as they go to the right.
o Pixels increase in y-location as they go down.
The image begins with every pixel being white.
The color of pixels drawn with draw_dot will initially be white. Use set_color to change the
color of pixels drawn.
You must be conscientious of these edge cases.
o A line with a start point and end point that are the same point should just draw that singular
o A rectangle with 0 width or 0 height will just be a line.
o A rectangle with 0 width and 0 height will just draw a single dot.
o A triangle where all three given points are collinear (lie on the same line) will just draw a
o A parallelogram where all three given points are collinear (lie on the same line) will just
draw a line.
o A circle with an inner radius and outer radius of 0 will just draw a single dot (the center).
o You may assume that we will not call rect_gradient with w (width) equal to 0.
Each function may only call functions directly specified in the documentation, and must call these
Your routines’ return values and outputs must be correct.
Each function in mp5.c is given to you in the following manner (using near_horizontal as
near_horizontal(int32_t x_start, int32_t y_start, int32_t x_end,
The autograder relies on near_horizontal being at the beginning of the line as shown here.
You must not alter this part of the code. Only modify the file in the function bodies. Do not add
additional subroutines to implement your functions.
Your code must be well-commented. Follow the commenting style of the code examples provided
in class and in the textbook.
Library Needed for the Fall 2020 VM Provided by TAs
If you want to use the VM provided by the TAs, you must first install the PNG library by typing:
sudo apt -y install libpng-dev
in a terminal window.
Compiling and Executing Your Program
Once you have written the functions into the mp5.c file, you can compile your code by typing one of:
Either command will compile the program and make an executable called mp5. You can also type:
This command will delete all files generated by the compiler. These make commands are specified by the
Makefile, which you can view if you are curious.
The mp5 program takes one command line argument:
The argument specifies one of eight prebuilt tests that can use to test each of your functions individually.
These tests are as follows:
These commands each produce an output called out#.png where # is replaced with the number you
entered. There will be files called gold#.png available for you to compare to, to see if you have produced
the correct output. These files are located in the gold_out directory.
If no parameter is given, or the given parameter is not 1-8, the program will call draw_picture.
Note that the minimum required for the program to compile is for all of the functions to be defined in
mp5.c and for them all to return something. If you want to test one of your functions before you have
written any or all of the other ones, you can simply set the others to return 1, and they will be valid functions
from the compiler’s point of view, allowing you to test the function you are writing.
Note that these tests are not comprehensive, and you are strongly urged to write your own test code.
To test your code further, you can use the gold executable provided to you to run any of the required
functions. To do this, type ./gold <r1> <r2> <r3> <r4> <r5> <r6> <r7>, where <r1>
represents a number 1-8 to indicate which function you would like to test (using the same key as the test
outputs for mp5). The remaining arguments to the executable are the arguments to whichever function you
would like to call. For functions that only have four arguments, <r6> and <r7> are ignored and are not
necessary. The output will be put in a file called gold_out.png.
We put a fair amount of emphasis on style and clarity in this class, as reflected in the rubric below.
8% – near_horizontal produces the correct output
8% – near_vertical produces the correct output
4% – draw_line produces the correct output
4% – draw_triangle produces the correct output
4% – draw_rect produces the correct output
4% – draw_parallelogram produces the correct output
10% – draw_circle produces the correct output
12% – rect_gradient produces the correct output
12% – draw_picture produces an acceptable output (as specified in the function explanation)
4% – all functions have proper return values
15% – All functions call the functions they are required to call and do not call any of the other
Comments, Clarity, and Write-up (15%)
5% – introductory paragraph explaining what you did (even if it’s just the required work)
10% – code is clear and well-commented, and compilation generates no warnings (note: any
warning means 0 points here)
Note that some categories in the rubric may depend on other categories and/or criteria. For example, if you
code does not compile, you will receive no functionality points. As always, your functions must be able to
be called many times and produce the correct results, so we suggest that you avoid using any static storage
(or you may lose most/all of your functionality points).
This picture shows all of the functions that you will write or interact with in this program. An arrow from
one function to another means that the function at the source of the arrow is required to call the function at
the destination of the arrow. draw_picture, being open-ended, may call any of the functions defined in