CSCI 2021 Project 1: C Programming

Credit for this assignment will be given based on two categories.

• Manual Inspection Criteria (~50%): Each problem has a checklist of things that graders will look for. The checklist is in the spec and often contains hints on what to do. Make sure you have a look at these.
• Automated Testing (~50%): Each problem has tests associated with it along with instructions on how to run those tests from the command line. Tests require that code compiles and runs according to the descriptions given so make sure you verify that these work.

Take the following steps to get started

1. Download the code associated with the project linked at the top of the spec. Unzip it and examine some of the provided code.
2. Examine the overview of the files provided listed in the Download and Setup section. This gives brief descriptions of files that already exist and those that you must create.
3. Pick a problem and read. There is a lot of information and many examples provided for each problem. Reading this will help you write the correct code earlier rather than later.
4. Ask questions: if its not clear how to proceed, put up a Piazza post or visit an office hour.
5. Get coding: don't wait to start for too long as this will greatly increase your stress level an potentially result in late submissions.
6. Familiarize yourself with the late submission policy for assignments so you are not caught off guard. No submissions will be accepted more than 48 hours after the deadline.

A Makefile is provided as part of this project. Building programs in C is a bit tedious and most folks use build systems of which make is the oldest. The instructions and dependencies to create programs are written in a Makefile which is then interpreted by the make program which will run gcc and other commands to create programs.

Use this Makefile by issuing commands like make prob1

> make prob2                    # build problem 2 main program
gcc -Wall -Wno-comment -Werror -g  -c stock_main.c
gcc -Wall -Wno-comment -Werror -g  -c stock_funcs.c
gcc -Wall -Wno-comment -Werror -g  -o stock_main stock_main.o stock_funcs.o

> make clean                    # remove all programs/binary object files
rm -f stock_main stock_demo test_stock_funcs treeset_main  *.o

> make prob3                    # build problem 3 main program
gcc -Wall -Wno-comment -Werror -g  -c treeset_main.c
gcc -Wall -Wno-comment -Werror -g  -c treeset_funcs.c
gcc -Wall -Wno-comment -Werror -g  -o treeset_main treeset_main.o treeset_funcs.o

> make clean                       # remove all programs/binary object files
rm -f stock_main stock_demo test_stock_funcs treeset_main  *.o

> make                             # build all programs/objects for the assignment
gcc -Wall -Wno-comment -Werror -g  -c stock_main.c
gcc -Wall -Wno-comment -Werror -g  -c stock_funcs.c
gcc -Wall -Wno-comment -Werror -g  -o stock_main stock_main.o stock_funcs.o
gcc -Wall -Wno-comment -Werror -g  -c stock_demo.c
gcc -Wall -Wno-comment -Werror -g  -o stock_demo stock_demo.o stock_funcs.o
gcc -Wall -Wno-comment -Werror -g  -o test_stock_funcs test_stock_funcs.c stock_funcs.o
gcc -Wall -Wno-comment -Werror -g  -c treeset_main.c
gcc -Wall -Wno-comment -Werror -g  -c treeset_funcs.c
gcc -Wall -Wno-comment -Werror -g  -o treeset_main treeset_main.o treeset_funcs.o

You are not required to understand all that is in the Makefile (yet) but it is a very useful tool and may be worth your while to inspect.

Running make help will provide a summary of the build/test commands present in the Makefile.

> make help
Typical usage is:
  > make                          # build all programs
  > make clean                    # remove all compiled items
  > make zip                      # create a zip file for submission
  > make prob1                    # built targets associated with problem 1
  > make test                     # run all tests
  > make test-prob2               # run test for problem 2
  > make test-prob2 testnum=5     # run problem 2 test #5 only

Automated tests are included with the code distribution. These tests are known to work on lab machines only but in most cases they should run identically in Linux environments such as the Windows subsystem for Linux or a virtual machine.

The provided Makefile allows automated tests to be run via calls like make test-prob1 to test Problem 1 and make test-prob2 to test Problem 2. See the transcript below.

>> make test-prob1               # run tests for problem 1, compiles required code first
gcc -Wall -Wno-comment -Werror -g  -c stock_funcs.c
gcc -Wall -Wno-comment -Werror -g  -o test_stock_funcs test_stock_funcs.c stock_funcs.o
./testy test_stock1.org 
============================================================
== test_stock1.org : Problem 1 First 3 Functions in stock_funcs.c
== Running 15 / 15 tests
1)  stock_new            : ok
2)  stock_free1          : ok
3)  stock_free2          : ok
4)  stock_free3          : ok
5)  stock_free4          : ok
6)  stock_print1         : ok
7)  stock_print2         : ok
8)  stock_print3         : ok
9)  stock_print4         : ok
10) stock_print5         : ok
11) stock_print_prices_0 : ok
12) stock_print_prices_1 : ok
13) stock_print_prices_2 : ok
14) stock_print_prices_3 : ok
15) stock_print_final    : ok
============================================================
RESULTS: 15 / 15 tests passed

>> make test-prob2                         # run tests for problem 2
gcc -Wall -Wno-comment -Werror -g  -c stock_main.c
gcc -Wall -Wno-comment -Werror -g  -o stock_main stock_main.o stock_funcs.o
./testy test_stock2.org 
============================================================
== test_stock2.org : Problem 2 Remaining Functions in stock_funcs.c
== Running 15 / 15 tests
1)  stock_set_minmax1       : ok
2)  stock_set_minmax2       : ok
3)  stock_set_minmax3       : ok
4)  stock_set_best1         : ok
5)  stock_set_best2 and 3   : ok
6)  stock_set_best4         : ok
7)  count_lines             : ok
8)  stock_load1             : ok
9)  stock_load2 and 3       : ok
10) stock_load pathological : ok
11) stock_plot1             : ok
12) stock_plot2 3 4         : ok
13) stock_plot5 6           : ok
14) stock_main1             : ok
15) stock_main2             : ok
============================================================
RESULTS: 15 / 15 tests passed
 
> make test                         # run tests for all problems
...

Each problem describes specifically how tests can be run and how credit will be assigned.

Note that one can run a single test with the following make invocation which sets testnum.

> make test-prob2 testnum=5

This is useful when debugging to limit the output and time it takes to check program results.

Problems 1 and 2 create a small plotting application that is focused on stock prices. In stock trading, the idea is to buy a stock when it is priced low and sell it at a later time point when the price is high which will net the profit of the difference between each. Of course, this must be done by predicting when prices are at their highest and lowest points and some insight can be garnered from examining historical data for stock prices. The application you will build allows for easy analysis of a simple data file containing times/prices for stocks and display of their information in simple text plots. At the end of problems 1 and 2, you will have an application which produces the following kind of output.

>> ./stock_main 10 data/stock-ascending.txt           # plot a stock-ascending data file
data_file: data/stock-ascending.txt
count: 10
prices: [10.00, 20.00, 30.00, ...]                    # shows the first few stock prices
min_index: 0                                          # calculates time for min/max price
max_index: 9
best_buy:  0                                          # calculates optimal buy/sell time
best_sell: 9
profit:    90.00
max_width: 10                                         # 10 specifed on command line for bar width
range:     90.00                                      # difference between max/min stock
plot step: 9.00                                       # amount represented by each plot hash
                    +----------
  0: B MIN    10.00 |                                 # visual representation of stock
  1:          20.00 |                                 # prices over time
  2:          30.00 |##
  3:          40.00 |###
  4:          50.00 |####
  5:          60.00 |#####
  6:          70.00 |######
  7:          80.00 |#######
  8:          90.00 |########
  9: S MAX   100.00 |##########

>> ./stock_main 20 data/stock-ascending.txt           # same file make the bars bigger:
data_file: data/stock-ascending.txt                   # 20 hashes wide rather than 10 wide
count: 10
prices: [10.00, 20.00, 30.00, ...]
min_index: 0
max_index: 9
best_buy:  0
profit:    90.00
best_sell: 9
max_width: 20                                         # 20 specified on command line which
range:     90.00                                      # makes the bars wider
plot step: 4.50
                    +--------------------
  0: B MIN    10.00 |
  1:          20.00 |##
  2:          30.00 |####
  3:          40.00 |######
  4:          50.00 |########
  5:          60.00 |###########
  6:          70.00 |#############
  7:          80.00 |###############
  8:          90.00 |#################
  9: S MAX   100.00 |####################

>> ./stock_main 20 data/stock-valley.txt              # plot a different file
data_file: data/stock-valley.txt                      # which has a valley shape
count: 12
prices: [100.00, 90.00, 80.00, ...]
min_index: 5
max_index: 11
best_buy:  5
best_sell: 11
profit:    55.00
max_width: 20
range:     55.00
plot step: 2.75
                    +--------------------
  0:         100.00 |##################
  1:          90.00 |##############
  2:          80.00 |##########
  3:          70.00 |#######
  4:          60.00 |###
  5: B MIN    50.00 |
  6:          55.00 |#
  7:          65.00 |#####
  8:          75.00 |#########
  9:          85.00 |############
 10:          95.00 |################
 11: S MAX   105.00 |####################

>> ./stock_main 20 data/stock-jagged.txt              # plot a file with greater
data_file: data/stock-jagged.txt                      # variance in prices
count: 15
prices: [103.00, 250.00, 133.00, ...]
min_index: 8
max_index: 11
best_buy:  8
best_sell: 11
profit:    232.00
max_width: 20
range:     232.00
plot step: 11.60
                    +--------------------
  0:         103.00 |#####
  1:         250.00 |##################
  2:         133.00 |########
  3:         143.00 |#########
  4:         168.00 |###########
  5:          91.00 |####
  6:         234.00 |################
  7:          59.00 |#
  8: B MIN    38.00 |
  9:          45.00 |
 10:         254.00 |##################
 11: S MAX   270.00 |####################
 12:          59.00 |#
 13:          72.00 |##
 14:         107.00 |#####

>> ./stock_main 25 data/stock-min-after-max.txt       # plot a stock with varying
data_file: data/stock-min-after-max.txt               # prices
count: 15
prices: [223.00, 292.00, 27.00, ...]
min_index: 10                                         # minimum price appear after
max_index: 4                                          # the maximum price
best_buy:  2                                          # finding the optimal buy/sell
best_sell: 4                                          # point is harder in this case
profit:    296.00
max_width: 25
range:     309.00
plot step: 12.36
                    +-------------------------
  0:         223.00 |################
  1:         292.00 |######################
  2: B        27.00 |#
  3:          92.00 |######
  4: S MAX   323.00 |#########################
  5:         189.00 |##############
  6:         207.00 |###############
  7:         142.00 |##########
  8:         321.00 |########################
  9:          89.00 |######
 10:   MIN    14.00 |
 11:         182.00 |#############
 12:         164.00 |############
 13:         156.00 |###########
 14:         169.00 |############

>> ./stock_main 9 data/stock-descending.txt           # if the stock is continually 
No viable buy/sell point                              # declining, its best not to buy
data_file: data/stock-descending.txt
count: 10
prices: [100.00, 90.00, 80.00, ...]
min_index: 9
max_index: 0
best_buy:  -1                                         # not worthwhile to buy/sell in
best_sell: -1                                         # this situation
profit:    0.00
max_width: 9
range:     90.00
plot step: 10.00
                    +---------
  0:   MAX   100.00 |#########
  1:          90.00 |########
  2:          80.00 |#######
  3:          70.00 |######
  4:          60.00 |#####
  5:          50.00 |####
  6:          40.00 |###
  7:          30.00 |##
  8:          20.00 |#
  9:   MIN    10.00 |

The file stock_funcs.c will contain most of the support functions for the stock plotting program. An outline of these functions are presented below. Note that each function has the Problem # to which it belongs.

// stock_funcs.c: support functions for the stock_main program.

stock_t *stock_new();
// PROBLEM 1: Allocate a new stock struct and initialize its fields.
// Integer fields like 'count' and 'min_index' should be initialied to
// -1.  Pointer fields like 'prices' should be initialized to
// NULL. The stock should be heap-allocated using malloc() and
// returned.  Since this is an allocation function, no use of 'free()'
// should appear.

void stock_free(stock_t *stock);
// PROBLEM 1: Free a stock. Check the 'data_file' and 'prices' fields:
// if they are non-NULL, then free them. Then free the pointer to
// 'stock' itself.

void stock_print(stock_t *stock);
// PROBLEM 1: Prints data about a stock that is passed in via a
// pointer. Uses the syntax ptr->field to access fields of the struct
// pointed by 'stock'.  Output follows the general convention:
//
// data_file: data/stock-jagged.txt                      
// count: 15
// prices: [103.00, 250.00, 133.00, ...]
// min_index: 8
// max_index: 11
// best_buy:  8
// best_sell: 11
// profit:    232.00
//
// Where each line prints a field of the stock_t struct. In all cases,
// floating point numbers are printed with 2 decimal digits of accuracy.
//
// NULLS FOR FIELDS
// The fields 'data_file' and 'prices' may be NULL in which case they
// will be printed specially as in
// 
// data_file: NULL
// prices: NULL
//
// This requires a manual if/else check for NULL values for these pointers.
// 
// PRICES FIELD
// When printing the 'prices' field, if the 'count' field is 0 to 3,
// print the entire array as in
//
// prices: []                        # count == 0
// prices: [70.00]                   # count == 1
// prices: [50.00, 90.00]            # count == 2
// prices: [59.00, 45.00, 103.00]    # count == 3
//
// Otherwise, print the first 3 elements with a ... at the end as in
//
// prices: [10.00, 20.00, 30.00, ...] # count > 3
//
// PROFIT
// There is no 'profit' field in the struct. Instead, calculate the
// profit as the difference between the price at the 'best_sell' index
// and 'best_buy' index.  If these indices are -1 indicating the best
// buy/sell time is not known or not viable, print a proit of 0.0

void stock_set_minmax(stock_t *stock);
// PROBLEM 1: Sets the index of 'min_index' and 'max_index' fields of
// the stock to be the positions in 'prices' of the minimum and
// maximum values present in it. Uses a simple loop over the array
// 'prices' which is 'count' elements long to examine each for
// max/min. If 'count' is zero, makes no changes to 'min_index' and
// 'max_index'.

int stock_set_best(stock_t *stock);
// PROBLEM 2: Sets the 'best_buy' and 'best_sell' fields of 'stock'.
// This corresponds to the pair which produces the best profit. On
// determining the best buy/sell indices which produce a positive
// profit, sets these fields in 'stock' and returns 0. If there is no
// buy/sell point which would result in a positive profit, sets the
// 'best_buy' and 'best_sell' indices to -1 and returns -1. Always
// calculates the earliest buy/sell pair of indices that would get the
// best profit: if 5,8 and 5,9 and 7,10 all give the same, maximal
// profit, the best buy/sell indices are set to 5,7.
// 
// ALGORITHM NOTES
// One intuitive algorithm to compute this is to try every possible
// buy index (outer loop) and every possible sell index after it
// (inner loop) looking for the maximum profit produced in it. This is
// a O(N^2) algorithm with N=count.  Using this algorithm is a good
// start. Some MAKEUP CREDIT will be awarded for implementing a more
// efficient, O(N) algorithm here. See the specification for more details.

int count_lines(char *filename);
// PROBLEM 2: Opens file named 'filename' and counts how many times
// the '\n' newline character appears in it which corresponds to how
// many lines of text are in it.  Makes use of either fscanf() with
// the %c format to read single characters or alternative I/O
// functions like fgetc(). Closes the file before returning a count of
// how many lines are it it.  If for any reason the file cannot be
// opened, prints a message like
//
// Could not open file 'not-there.txt'
//
// and returns -1 to indicate failure.

int stock_load(stock_t *stock, char *filename);
// PROBLEM 2: Loads a stock from file 'filename' into 'stock' filling
// its 'prices' and 'count' fields in. Makes use of the count_lines()
// function to determine how many lines are in the file which will
// dictate 'count' and the length of 'prices'. Allocates space in the
// heap for the stock's 'prices' array, opens the 'filename' and
// iterates through reading prices into the array. The data format for
// prices files is
//
// time_03 133.00
// time_04 143.00
// time_05 168.00
// time_06 91.00
// 
// where each line has a time as as single string and a price which is
// floating point number. The times can be ignored which can be
// accomplished with a fscanf() call with format "%*s" to read a
// string but ignore/discard it.
// 
// Assigns the 'datafile' field to be a duplicated string of
// 'filename' for which 'strdup()' is extremely useful. This string
// must be free()'d later likely in 'stock_free()'
// 
// On successfully loading the stock, returns 0.
//
// If 'filename' cannot be opened, prints the message 
// 
// Unable to open stock file 'some-stock.txt', bailing out
//
// with 'filename' substituted in for the name of the stock and
// returns -1.

void stock_plot(stock_t *stock, int max_width);
// PROBLEM 2: Plots a graphical representation of stock
// information. First calculates and prints plot which is in the
// following format:
//
// max_width: 25
// range:     40.00
// plot step: 1.60
//
// The prints a header line and iterates through each stock data on it
// along with a bar which varies in width with the stock price. A
// sample format is as follows with some annotations.
// 
//....20 spaces.......+ ...max_width dashes.....
//01234567890123456789
//                    +-------------------------
//  0:         223.00 |################
//  1:         292.00 |######################
//  2: B        27.00 |#
//  3:          92.00 |######
//  4: S MAX   323.00 |#########################
//  5:         189.00 |##############
//  6:         207.00 |###############
//  7:         142.00 |##########
//  8:         321.00 |########################
//  9:          89.00 |######
// 10:   MIN    14.00 |
// 11:         182.00 |#############
// 12:         164.00 |############
// 13:         156.00 |###########
// 14:         169.00 |############
//01234567890123456789
//|    | |     |       |
//|    | |     |       +-> Each bar is (price-min)/plot_step hashes wide
//|    | |     +-> Stock price printed with format %8.2f
//|    | +-> Print MIN or MAX if the stock is at min_index/max_index
//|    +-> Print B or S if the stock is at the best_buy/best_sell index
//+--> Index in the array printed with format %3d

The project code contains a skeleton version of stock_funcs.c which you should fill in with definitions. With this skeleton version, you can immediately start testing your code by typing make test-prob1. Without changes, you will get failures for all tests as in

>> make test-prob1
gcc -Wall -Wno-comment -Werror -g  -o test_stock_funcs test_stock_funcs.c stock_funcs.o
./testy test_stock1.org 
============================================================
== test_stock1.org : Problem 1 First 3 Functions in stock_funcs.c
== Running 15 / 15 tests
1)  stock_new            : FAIL -> results in file 'test-results/prob1-01-result.tmp'
2)  stock_free1          : FAIL -> results in file 'test-results/prob1-02-result.tmp'
3)  stock_free2          : FAIL -> results in file 'test-results/prob1-03-result.tmp'
...
============================================================
RESULTS: 0 / 15 tests passed

However, the ability to run tests on your code an see progress is extremely important. Your first goal when starting a new project should be to see some results or running the program which is much easier if some benevolent dictator has provided a bunch of tests.

You can also use the invocation

> make test-prob1 testnum=5

to run a single test and see its results in the terminal.

Failed tests generate results files which can be viewed in any text editor or in the terminal. The cat command can show such results in the terminal via cat test-results/some-file.txt.

Below the results for the first test are shown and from the comparison and Valgrind report, it appears that there is some sort of Memory problem with the stock_new() function.

>> cat test-results/prob1-01-result.tmp 

* (TEST 1) stock_new
COMMENTS:
** program: ./test_stock_funcs stock_new

** --- Failure messages --- 
- FAILURE(139) due to SIGSEGV (segmentation fault) from OS
- FAILURE: Output Mismatch at lines marked

** --- Side by Side Differences ---
- Expect output in: test-results/raw/prob1-01-expect.tmp
- Actual output in: test-results/raw/prob1-01-actual.tmp
- Differing lines have a character like '|' '>' or '<' in the middle
#+BEGIN_SRC sbs-diff
==== EXPECT ====                                                      ==== ACTUAL ====                                                   
{                                                                     {
    // Tests stock_new() function and whether it initializes fields       // Tests stock_new() function and whether it initializes fields
    // correctly before returning a stock.                                // correctly before returning a stock.

    stock_t *stock = stock_new(); // call function to allocate/init       stock_t *stock = stock_new(); // call function to allocate/init

    printf("stock->data_file: %p\n" , stock->data_file);                  printf("stock->data_file: %p\n" , stock->data_file);
    printf("stock->count:     %d\n" , stock->count);                      printf("stock->count:     %d\n" , stock->count);
    printf("stock->prices:    %p\n" , stock->prices);                     printf("stock->prices:    %p\n" , stock->prices);
    printf("stock->min_index: %d\n" , stock->min_index);                  printf("stock->min_index: %d\n" , stock->min_index);
    printf("stock->max_index: %d\n" , stock->max_index);                  printf("stock->max_index: %d\n" , stock->max_index);
    printf("stock->best_buy:  %d\n" , stock->best_buy);                   printf("stock->best_buy:  %d\n" , stock->best_buy);
    printf("stock->best_sell: %d\n" , stock->best_sell);                  printf("stock->best_sell: %d\n" , stock->best_sell);

    free(stock);                // de-allocate manually                   free(stock);                // de-allocate manually
}                                                                     }
stock->data_file: (nil)                                             | Segmentation Fault
stock->count:     -1                                                <
stock->prices:    (nil)                                             <
stock->min_index: -1                                                <
stock->max_index: -1                                                <
stock->best_buy:  -1                                                <
stock->best_sell: -1                                                <
#+END_SRC

** --- Line Differences ---
EXPECT:  17) stock->data_file: (nil)
EXPECT:  18) stock->count:     -1
EXPECT:  19) stock->prices:    (nil)
EXPECT:  20) stock->min_index: -1
EXPECT:  21) stock->max_index: -1
EXPECT:  22) stock->best_buy:  -1
EXPECT:  23) stock->best_sell: -1
ACTUAL:  17) Segmentation Fault

--- Valgrind Log from: test-results/raw/prob1-01-valgrd.tmp ---
==150464== Memcheck, a memory error detector
==150464== Copyright (C) 2002-2017, and GNU GPL'd, by Julian Seward et al.
==150464== Using Valgrind-3.17.0 and LibVEX; rerun with -h for copyright info
==150464== Command: ./test_stock_funcs stock_new
==150464== Parent PID: 150463
==150464== 
==150464== Invalid read of size 8
==150464==    at 0x109354: main (test_stock_funcs.c:33)
==150464==  Address 0x0 is not stack'd, malloc'd or (recently) free'd
==150464== 
==150464== 
==150464== Process terminating with default action of signal 11 (SIGSEGV): dumping core
==150464==  Access not within mapped region at address 0x0
==150464==    at 0x109354: main (test_stock_funcs.c:33)
==150464==  If you believe this happened as a result of a stack
==150464==  overflow in your program's main thread (unlikely but
==150464==  possible), you can try to increase the size of the
==150464==  main thread stack using the --main-stacksize= flag.
==150464==  The main thread stack size used in this run was 10022912.
==150464== 
==150464== HEAP SUMMARY:
==150464==     in use at exit: 0 bytes in 0 blocks
==150464==   total heap usage: 0 allocs, 0 frees, 0 bytes allocated
==150464== 
==150464== All heap blocks were freed -- no leaks are possible
==150464== 
==150464== For lists of detected and suppressed errors, rerun with: -s
==150464== ERROR SUMMARY: 1 errors from 1 contexts (suppressed: 0 from 0)

This problem implements further functions to complete the Stock Plotting application. These functions are somewhat more complex and utilize functions and techniques that are part of Problem 1. When this problem is complete, all functions in stock_funcs.c will be working which will allow the provided stock_main.c program to be compiled and run on the command line.

The remainder of this section gives some notes and implementation hints on how to handle the various functions.

void stock_set_minmax(stock_t *stock)

This function should be implemented as a simple "scan" across the prices[] array. Each step will check an element of prices[] and adjust a running min/max. Make use of similar syntax here as you used in stock_print() in order to access the elements of prices and subsequently set the min_index and max_index fields.

int stock_set_best(stock_t *stock);

In many cases it is best to buy at a minimum price and sell at a maximum price. This approach fails if the maximum price occurs before the minimum price. The job of stock_set_best() is to run an algorithm to locate the best possible buy/sell pair and set the best_buy / best_sell fields of a stock. As the documentation for the function indicates, one can use a simple "brute force" approach to this by simply trying every buy index paired with every sell index (which must occur later in the array than the buy index). This approach leads to the following pattern of iteration through the prices array which uses a doubly nested pair of loops.

set best_buy and best_sell to -1
set best_profit to 0.0
for every buy index I in prices
  for every sell index J later than I in prices
    if prices[J] - prices[I] > best_profit
      set best_buy,best_sell, to I,J
      set best_profit to prices[J] - prices[I]
  done
done

Start by adapting this approach in the function and make sure it works. Then move on to solve other problems here.

For those that want a bit more fun, revisit this function later to implement a more efficient version of it as described in the Optional MAKEUP CREDIT Section.

Data on stock prices over time are stored in several example files in the data/ directory such as data/stock-ascending.txt and data/stock-jagged.txt.

>> ls data/stock-*     # list all the sample stock data
data/stock-1only.txt	   data/stock-empty.txt		   data/stock-TSLA-08-02-2021.txt
data/stock-2only.txt	   data/stock-FB-08-02-2021.txt    data/stock-TSLA-08-12-2021.txt
data/stock-3only.txt	   data/stock-GOOG-08-02-2021.txt  data/stock-valley.txt
data/stock-ascending.txt   data/stock-jagged.txt
data/stock-descending.txt  data/stock-min-after-max.txt

Each of has a simple data format which looks like the following.

>> cat data/stock-jagged.txt 
time_01 103.00
time_02 250.00
time_03 133.00
time_04 143.00
time_05 168.00
time_06 91.00
time_07 234.00
time_08 59.00
time_09 38.00
time_10 45.00
time_11 254.00
time_12 270.00
time_13 59.00
time_14 72.00
time_15 107.00

• Each time/price pair appears on its own line so that the number of lines in the file is the number of time/price pairs
• The first item on each line is a time at which a price appears: this will be ignored.
• The second item that appears on each line is a floating point number which is the stock price at that time.

Since the number of lines in stock files corresponded to the number of prices, it is useful to have a utility function to count lines in a file which is required.

int count_lines(char *filename)

As the documentation for the function indicates, this function will open the named file and count the lines in it. This can be done most easily via treating the file as a long array of characters and looking for the special \n newline character which designates a line break. Character comparisons are done in C like other equality comparisons via syntax like

if(somevar == '\n'){ ... }

if(somevar == 'X'){ ... }

To read a single character from a file, a common approach is to use the fscanf() function with the %c format specifier as in

char c;
int ret = fscanf(fin, "%c", &c);

Keep in mind that the fscanf() function will fail to read a character when it reaches the end of a file. In these cases it will return the special return value EOF which should be checked. A common structure is

int ret = fscanf(fin, ..., ...);
if(ret == EOF){
  // take action for end of file, like breaking out of a loop
  // or returning
}
else{
  // fscanf() completed normally so variables should contain 
  // new values
}

Experiment with these elements to complete the count_lines() function.

Keep in mind that count_lines() should fail gracefully if a file cannot be opened. In such cases, the fopen() function returns NULL. Make sure to check for this, print an error message as indicated in the function documentation, and return -1 from the function to indicate a failure has occurred.

int stock_load(stock_t *stock, char *filename)

This function takes an existing stock and fills in its price and count fields based on the contents of the given file. An example:

stock_t *stock = new_stock();
// all fields of stock have default values like -1 and NULL

int ret = stock_load(stock, "data/stock-ascending.txt");
// Specified data file as 10 lines so
// stock->count is now 10 
// stock->prices is an array of 10 values read from the file
// If the load failed, -1 would be returned

Make use of the count_lines() function to determine how many lines are in the stock file. This will allow you to immediately allocate enough space for a double array for all of the prices in the file.

You will need to skip the first string on each line in prices files:

time_03 133.00
time_04 143.00
|       |
|       +---> Read these into prices
+---> Skip these times

An easy way to accomplish this is to use the C format modifier * which reads a field but does not place it in memory anywhere. For example:

fscanf(fin, "%*s");

would read a string but ignore it thus one does not have to provide a memory address to store the string.

One requirement of stock_load() is to set the data_file field to be a copy of the filename parameter. While this can be done manually with malloc() and a loop, the strdup() function makes this easy and is encouraged here. Look up documentation for it either in the Unix man pages (type man strdup in a terminal) or via an internet search.

Ensure that at the end of stock_load() the data_file, prices, and count fields have been set according to the data from the file. Do not modify any other fields in the stock.

void stock_plot(stock_t *stock, int max_width);

The ultimate goal of the Stock Plot application is fulfilled in this function which is meant to print a graphical representation of stock data. A sample of output is below and will be discussed in more detail.

max_width: 25
range:     40.00
plot step: 1.60
                   +-------------------------
 0:         223.00 |################
 1:         292.00 |######################
 2: B        27.00 |#
 3:          92.00 |######
 4: S MAX   323.00 |#########################
 5:         189.00 |##############
 6:         207.00 |###############
 7:         142.00 |##########
 8:         321.00 |########################
 9:          89.00 |######
10:   MIN    14.00 |
11:         182.00 |#############
12:         164.00 |############
13:         156.00 |###########
14:         169.00 |############

The output comes from the following code fragment which is present in the provided file stock_demo.c.

{
  stock_t *stock = stock_new();
  stock_load(stock, "data/stock-min-after-max.txt");
  stock_set_minmax(stock);
  stock_set_best(stock);
  stock_plot(stock, 25);
}

The initial output is

max_width: 25
range:     40.00
plot step: 1.60

and shows the calculation needed to compute the "plot step". This relies on knowing the Minimum and Maximum price (found via stock_set_minmax()) to first get the Range (difference between min/max). The range is then divided by the max_width parameter. max_width dictates how wide the widest "bar" of the plot should be. The reported "plot step" is 1.60 in this case so that each of the # has marks corresponds to 1.60 in price in the below bars.

Next an axis line appears as in

                   +-------------------------
....20 spaces....   ....max_width dashes.....

The left area has a fixed number of spaces followed by a + then a number of dashes equal to the max_width parameter. Use a loop to print out the proper number of dashes.

The remainder of the plot has one line per stock price. Examples of the first few lines:

//  0:         223.00 |################
//  1:         292.00 |######################

Note that the initial index at the left is printed with format %3d which leaves spaces on the left for single and double digit numbers.

Next appears the price for the stock at that time. This is printed using the format specifier %8.2f which leaves 2 digits beyond the decimal place and again, may add some space on the left for smaller prices.

At the right, each "bar" has a length which can be computed by the formula

int width = (price[i] - minimum_price) / plot_step

This will convert a double value into an integer which can then be used in a loop to print out the proper number of # marks.

The space in between the index and the price is used to denote the optimal buy/sell time and the min/max stock prices. As a couple examples here are possible outputs that combine some of these.

  2: B        27.00 |#
  4: S MAX   323.00 |#########################
 10:   MIN    14.00 |

• Use a B or S at the index that matches best_buy / best_sell. Print this as a single character string OR print a space here if the stock is not at the best buy/sell index.
• Print MAX or MIN if the stock is at the min/max index or print 3 spaces if not. In either case follow with 1 more space.

A common tactic to solve such problems is to use a conditional structure like the following which re-assigns a string based on whether a particular stock price is one of the best buy/sell or min/max.

char *min_or_max_str = "   ";
if(i == stock->min_index){
  min_or_max_str = "MIN";
}
if(i == stock->max_index){
  ...
}
printf("%s ",min_or_max_string);

The same process can be executed to print out either a blank space or B/S for the best buy/sell designator as well.

It may take some time to get the output for the stock_plot() correct but this can be expedited through careful inspection of test feedback.

The "brute force" approach initially suggested to find best_buy / best_sell indices compares all possible buy/sell pairs to determine the best profit possible. This is a Quadratic algorithm with complexity O(N^2) with N being the length of the prices[] array.

One can improve upon this algorithm to get an O(N) linear algorithm. This takes some cleverness or some research. For example, the problem is directly related to the Maximum Subarray Problem and its various solutions. These are described here:

• https://en.wikipedia.org/wiki/Maximum_subarray_problem
• https://medium.com/@rsinghal757/kadanes-algorithm-dynamic-programming-how-and-why-does-it-work-3fd8849ed73d
• https://cs.gmu.edu/~kauffman/cs310/01-intro.pdf

Note that adapting these approaches to the current setting will require some care: the Maximum Subarray problem assumes elements in the array are added together whereas in our Stock setting, it is the difference between elements that matter. However, buy absorbing the tricks used the linear time Maximum Subarray algorithm, one can get the same effect for the Stocks, an O(N) algorithm to set best_buy / best_sell.

Code that implements a Linear Time algorithm for stock_set_best() will be awarded 10 POINTS of PROJECT MAKEUP CREDIT.

Makeup Credit will go into the general pool of points earned for projects to make up for points missed on projects elsewhere. If a student scores 110/100 on Project 1 and 90 / 100 on Project 2, the 10 points from Project 1 will make up for the credit lost on Project 2. Your total score on All Projects cannot exceed 100% so any points beyond the overall total will simply be dropped.

This problem implements a rudimentary "tree set" application in C along with a program that uses it. The architecture is similar to a HW problem involving linked lists so reviewing that code can provide some insight.

Basic Binary Search Trees are covered in most introductory data structure classes such as CSCI 1913 and 1933. A "Tree Set" is simply a binary search tree which only allows one copy of each unique item in it. When searching for an item, the organization of the binary search tree is exploited to speed up the location of the time or determine that it is not present. When adding items, duplicate items are discarded without altering the tree.

The intent of this problem is to develop a small application which behaves as the following demo indicates. Study this demo carefully as it illustrates most of the capabilities of the application with each capability being checked by automated tests.

> make treeset_main  # build with Makefile
gcc -Wall -Wno-comment -Werror -g  -c treeset_main.c
gcc -Wall -Wno-comment -Werror -g  -c treeset_funcs.c
gcc -Wall -Wno-comment -Werror -g  -o treeset_main treeset_main.o treeset_funcs.o

> ./treeset_main                   # run main program
Treeset Main
Commands:
  print:          shows contents of the tree in reverse sorted order
  size:           prints the number of nodes currently in the tree
  clear:          eliminates all elements from the tree
  quit:           exit the program
  add <name>:     inserts the given string into the tree, duplicates ignored
  find <name>:    prints FOUND if the name is in the tree, NOT FOUND otherwise
  preorder:       prints contents of the tree in pre-order which is how it will be saved
  save <file>:    writes the contents of the tree in pre-order to the given file
  load <file>:    clears the current tree and loads the one in the given file
TS#> add Lucas                  # add some data
TS#> add Mike
TS#> add Dustin
TS#> add Will
TS#> add El
TS#> size                       # show number of nodes in the tree
5 nodes
TS#> print                      # print tree, reverse order, indentation indicates depth/structure
    Will                        # root->right->right
  Mike                          # root->right
Lucas                           # root
    El                          # root->left->right
  Dustin                        # root->left
TS#> preorder                   # show tree in preorder, root first
Lucas                           # root
  Dustin                        # root->left
    El                          # root
  Mike                          # root->right
    Will                        # root->right->right
TS#> find Mike
FOUND
TS#> find Nancy
NOT FOUND
TS#> add Nancy                  # add Nancy
TS#> size                       # show size of tree has increased
6 nodes
TS#> print
    Will
      Nancy                     # root->right->right->left
  Mike
Lucas
    El
  Dustin
TS#> add Mike                   # adding duplicates is ignored
duplicate element ignored
TS#> add El
duplicate element ignored
TS#> find Max
NOT FOUND
TS#> add Max                    # add max
TS#> print
    Will
      Nancy
  Mike                          # root->right->left
    Max
Lucas
    El
  Dustin
TS#> find Max
FOUND
TS#> find Barb
NOT FOUND
TS#> add Barb                   # add Barb
TS#> print
    Will
      Nancy
  Mike
    Max
Lucas
    El
  Dustin
    Barb                        # root->left->left

TS#> save stranger.tree         # save tree into named file
TS#> clear                      # eliminate tree
TS#> size                       # show tree is empty now
0 nodes
TS#> print                      # print shows nothing: empty tree
TS#> add Demigorgon             # add Demigorgon to empty
TS#> print
Demigorgon                      # root
TS#> load stranger.tree         # clear tree, replace with contents of file
TS#> size                       # 8 nodes loaded from file
8 nodes
TS#> print                      # tree is restored
    Will
      Nancy
  Mike
    Max
Lucas
    El
  Dustin
    Barb
TS#> add Jim                    # add Jim
TS#> add Joyce                  # add Joyce
TS#> print                      # show tree
    Will
      Nancy
  Mike
    Max
Lucas                           # root
        Joyce                   # root->left->right->right->right
      Jim                       # root->left->right->right
    El
  Dustin
    Barb
TS#> load stranger.tree         # clear and reload from file
TS#> print                      # restored to previous contents
    Will
      Nancy
  Mike
    Max
Lucas
    El
  Dustin
    Barb
TS#> quit                       # quit program

Examine the header file tree.h to see the two C structs which will be used to represent trees.

// Nodes that go into the tree
typedef struct tsnode {
  char name[128];               // tsnode data
  struct tsnode *left;          // left branch,  NULL if not present
  struct tsnode *right;         // right branch, NULL if not present
} tsnode_t;

// Type of tree itself
typedef struct {
  tsnode_t *root;               // root of tree, NULL if tree empty
  int size;                     // number of tsnodes in tree
} treeset_t;

The tree set will track unique strings each of which is stored in a tsnode_t struct. The treeset_t struct tracks the root of the tree along with the total number of nodes in the tree.

Standard operations are supported by the tree.

• Inserting item
• Searching for an element
• Clearing the entire tree
• Printing tree in reverse sorted order which is handy for visualization
• Printing the tree in pre-order (root first, then children) which is useful for saving and loading to files
• Saving and loading from a file

These are broken down into the following C functions which you will need to implement in the file treeset_funcs.c. Keep in mind that to honor the binary search tree property (left less than / right greater than), you will need to use the return values of strcmp() function calls as string data is present in the tree.

Below is an outline of the functions that should be present in treeset_funcs.c along with guiding documentation about how to implement them.

// treeset_funcs.c: Provides a small library of functions that operate
// on binary search trees of unique strings.

void treeset_init(treeset_t *tree);
// Initialize the given tree to have a null root and have size 0.

int treeset_insert(treeset_t *tree, char name[]);
// Inserts given name into a binary search tree. Uses an ITERATIVE
// (loopy) approach to insertion which starts a pointer at the root of
// the tree and changes its location until the correct insertion point
// is located. Returns 1 if a new tsnode is created and 0 if a duplicate
// name is found in the tree already in which case the name is not
// added.

int treeset_find(treeset_t *tree, char name[]);
// Determines whether name is present or not in the tree using binary
// search. Returns 1 if name is present and 0 otherwise. Uses an
// ITERATIVE (loopy) approach which starts a pointer at the root of
// the tree and changes it until the search name is found or
// determined to not be in the tree.

void treeset_clear(treeset_t *tree);
// Eliminate all tsnodes in the tree setting its contents empty. Uses
// recursive tsnode_remove_all() function to free memory for all tsnodes.

void tsnode_remove_all(tsnode_t *cur);
// Recursive helper function which visits all tsnodes in a tree and
// frees the memory associated with them. This requires a post-order
// traversal: visit left tree, visit right tree, then free the cur
// tsnode.

void treeset_print_revorder(treeset_t *tree);
// Prints the elements of the tree in reverse order at differing
// levels of indentation which shows all elements and their structure
// in the tree. Visually the tree can be rotated clockwise to see its
// structure.

void tsnode_print_revorder(tsnode_t *cur, int indent);
// Recursive helper function which prints all elements in the tree
// rooted at cur in reverse order: traverses right subtree, prints cur
// tsnode, traverses left tree. Parameter 'indent' indicates how far to
// indent (2 spaces per indent level).

void treeset_print_preorder(treeset_t *tree);
// Print all the data in the tree in pre-order with indentation
// corresponding to the depth of the tree. Makes use of
// tsnode_write_preorder() for this.

void treeset_save(treeset_t *tree, char *fname);
// Saves the tree by opening the named file, writing the tree to it in
// pre-order with tsnode_write_preorder(), then closing the file.

void tsnode_write_preorder(tsnode_t *cur, FILE *out, int depth);
// Recursive helper function which writes/prints the tree in pre-order
// to the given open file handle. The parameter depth gives how far to
// indent tsnode data, 2 spaces per unit depth. Depth increases by 1 on
// each recursive call. The function prints the cur tsnode data,
// traverses the left tree, then traverses the right tree.

int treeset_load(treeset_t *tree, char *fname );
// Clears the given tree then loads new elements to it from the
// named. Repeated calls to treeset_insert() are used to add strings read
// from the file.  If the tree is stored in pre-order in the file, its
// exact structure will be restored.  Returns 1 if the tree is loaded
// successfully and 0 if opening the named file fails in which case no
// changes should be made to the tree.

In treeset_main.c implement an interactive program which allows users to type commands to manipulate a treeset. Most of these commands will dispatch directly to an analogous function in treeset_funcs.c. Example: the add command will likely call treeset_insert().

Once the files treeset_funcs.c and treeset_main.c are present, the provided Makefile should compile the complete program and allow it to run as follows:

>> make treeset_main
gcc -Wall -Wno-comment -Werror -g  -c treeset_main.c
gcc -Wall -Wno-comment -Werror -g  -c treeset_funcs.c
gcc -Wall -Wno-comment -Werror -g  -o treeset_main treeset_main.o treeset_funcs.o

>> ./treeset_main
Treeset Main
Commands:
  print:          shows contents of the tree in reverse sorted order
  size:           prints the number of nodes currently in the tree
  clear:          eliminates all elements from the tree
  quit:           exit the program
  add <name>:     inserts the given string into the tree, duplicates ignored
  find <name>:    prints FOUND if the name is in the tree, NOT FOUND otherwise
  preorder:       prints contents of the tree in pre-order which is how it will be saved
  save <file>:    writes the contents of the tree in pre-order to the given file
  load <file>:    clears the current tree and loads the one in the given file
TS#> 

The following sections provide some implementation details.

Some of the pictures below mention 'Assignment' which is now 'Project' and may mention some files that are not part of the current project. The process of uploading submission is otherwise the same.

1. In a terminal, change to your project code directory and type make zip which will create a zip file of your code. A session should look like this:

      > cd Desktop/2021/p1-code      # location of assignment code
   
      > ls 
      Makefile    stock_main.c  stock_funcs.c  treeset_main.c 
      ...
   
      > make zip                     # create a zip file using Makefile target
      rm -f p1-code.zip
      cd .. && zip "p1-code/p1-code.zip" -r "p1-code"
        adding: p1-code/ (stored 0%)
        adding: p1-code/Makefile (deflated 68%)
        adding: p1-code/test_prob1.org (deflated 69%)
        adding: p1-code/treeset_main.c (deflated 71%)
        ...
      Zip created in p1-code.zip
   
      > ls p1-code.zip
      p1-code.zip
   

2. Log into Gradescope and locate and click 'Project 1' which will open up submission

   

3. Click on the 'Drag and Drop' text which will open a file selection dialog; locate and choose your p1-code.zip file

   

4. This will show the contents of the Zip file and should include your C source files along with testing files and directories.

   

5. Click 'Upload' which will show progress uploading files. It may take a few seconds before this dialog closes to indicate that the upload is successful. Note: there is a limit of 256 files per upload; normal submissions are not likely to have problems with this but you may want to make sure that nothing has gone wrong such as infinite loops creating many files or incredibly large files.

   WARNING: There is a limit of 256 files per zip. Doing make zip will warn if this limit is exceeded but uploading to Gradescope will fail without any helpful messages if you upload more the 256 files in a zip.

   

6. Once files have successfully uploaded, the Autograder will begin running the command line tests and recording results. These are the same tests that are run via make test.

   

7. When the tests have completed, results will be displayed summarizing scores along with output for each batch of tests.

   

You may wish to review the policy on late project submission which will cost 1 Engagement Point per day late. No projects will be accepted more than 48 hours after the deadline.

https://www-users.cs.umn.edu/~kauffman/2021/syllabus.html