CS100 Homework5

hw5.md 10/25/2019

2 / 8 Problem 1. Data Analyser Description In this Problem, we provide a base class named DataAnalyser. In this class, we've already implemented the constructor, destructor, and a virtual function called calculate, as shown below.

class DataAnalyser

{

public:

 DataAnalyser() {};

virtual ~DataAnalyser() {};

virtual float calculate(float* data, size_t size)

{

std::cout << "ERROR: Try to access virtual function in base

class\n";

return ERRORCODE;

 };

};

Explainations of these functions The constructor of base class ~ The destructor of base class _ A virtual function which takes 2 parameters, the first one, data, is an array that need to be calculated, and the second one, size, is the size of the array, returns the calculated result. You are required to implement 5 classes called MaximumAnalyser, MinimumAnalyser,AverageAnalyser,

MedianAnalyser and StdDevAnalyser inherits from the given DataAnalyser (i.e. They are child classes of DataAnalyser), you need to overload the function calculate in these child classes. For MaximumAnalyser, calling function calculate will return the maximum value in data. For MinimumAnalyser, calling function calculate will return the minimum value in data. For AverageAnalyser, calling function calculate will return the average(mean) of data. Here is the formula to calculate the average of data: Let Then,

hw5.md 10/25/2019

3 / 8 For MedianAnalyser, calling function calculate will return the median of data. For StdDevAnalyser, calling function calculate will return the standard deviation of data. Here is the formula to calculate the standard deviation of data: Let Then, where is the average number of data. Note that, if your implementation is correct, no "ERROR" information will be printed when running your program. Requirements You are required to implement 5 different child classes inherits from DataAnalyser, each child class can calculate one property of given data. You need to overload calculate function, to give different performance in child classes. You CANNOT modify the name or arguments list of the calculate

function or the name of these classes, or compile errors will occur. If you want to add your own functions and classes, feel free to add them. For more details, see the template code given to you. Debugging & Testing To make you debug easier, we provide a some useful codes for you, you can modify it as you want. The following codes should be added to your main function, so, DO NOT submit them to OJ.

/* You can change these numbers to whatever you want*/

float arr[] = {0.3, -100, 1, 2, 5, 6, 9, 12, 2}; 

 DataAnalyser* foo = new MinimumAnalyser(); /* it can be any of the

required 5 child classes */

std::cout << foo->calculate(arr, 9) << std::endl; /* should be -100 in

this case */

delete foo;

Submission When submitting it to online judge, you need to submit the base class and 5 child classes you write, If you add additional includes, classes, functions, variables or definitions, please also submit them to OJ.

DO NOT submit the main() function, or compile errors will occur.

hw5.md 10/25/2019

4 / 8 Problem 2. Structured outputing in console Description Assume you are one of TAs of CS999. Oneday, Homework?? was released. It was a huge project which contains m different student informations, like name, email, or scores for each problem, etc. After the deadline, you need to collect all n students' information and send the report to professor. The original report you downloaded from OJ contains n lines, matching to n students. For each line, it contains m numbers separated by a single whitespace (i.e. " "), representing someone's informations (i.e. "Madoka Madoka@shanghaitech.edu.cn 99 100 81"). For instance, n = 3 and m = 4 represents there are 3 students and 4 informations per student, a sample score sheet is shown below.

Umi 20 aa 43

Honoka 43253 65789 87912

Kotori 1.7 foo 44

But the professor of CS999 is extremely strict with formats of reports, here are the requirements. The first line of the report should starts with a "/", ends with a "\", interiors are filled with "-". (e.g. "/------------------------\") The last line of the report should start with a "\", ends with a "/", interiors are filled with "-". (e.g. "\------------------------/") For middle lines, it can be (a)."line contains numbers" or (b)."separate-line", they are all start and end whith a bar-separator ("|"), and occurs alternately (i.e. abababa). line contains information: these lines only contain bar-separators ("|"), informations (e.g. 100, GeZiWang, 1.233, OOP), and whitespaces (" "). There are at least one whitespaces between each "information" and "bar-separators", and numbers are left-aligned. (e.g. "| 100 | 200 | 300 |") separate-line: these lines only contain bar-separators ("|") and minus signs ("-"). The "|"s should be aligned with other lines. (e.g. "|-----|-----|-----|") Note: All bar-separators should be aligned, all informations are left-aligned Note: All lines you output should have the same length (same number of characters). The students should be sorted alphabetically by the column (starts from 0) of their information (e.g. If sorting by column 0, Honoka > Kotori > Umi in case shown above).

std::string provides you an operator <, so you will not need to write your own string comparison.

hw5.md 10/25/2019

5 / 8 If you want to swap student a and student b, you should swap the whole row, not only the given column, otherwise the informations will be messed up. e.g. if the data is

b 12

a 123

you should sort it to this (by column 0):

a 123

b 12

std::sort may be very useful for this requirement. Hint: you can store all the informations as strings, instead of using multiple types. For instance, the following report is the structured report of report shown above (sorting by col 0).

/--------------------------------\

| Honoka | 43253 | 65789 | 87912 |

|--------|-------|-------|-------|

| Kotori | 1.7 | foo | 44 |

|--------|-------|-------|-------|

| Umi | 20 | aa | 43 |

\--------------------------------/

So, you want to write a class called ReportParser to structure the report, and give the structured one. We also provide a template for you (as shown).

class ReportParser

{

public:

// The constructor

 ReportParser(int numStudents, int numInfos);

// The destructor

 ~ReportParser();

// read & write functions

void readReport();

void writeStructuredReport(int sortOption);

// Add your own functions and variables here

private:

// Add your own functions and variables here

};

hw5.md 10/25/2019

6 / 8 Some functions are provided for you. The constructor of ReportParser, numStudents is the number of students, numInfos is the number of informations. ~ The destructor of ReportParser Read the input report from the console (stdin) Write the structured report to the console (stdout) Requirements The input contains n+1 lines, the first line contains 3 numbers m, n and sortOption, represents the info number per line, total student number, and the result should be sorted by which column (sortOption is an integer in ). For the following n lines, each line contains m informations seperated by " ", representing m informations for this student. You need to implement the ReportParser class, read the input report and structured it into the required format. Be careful about the requirements of structured report shown above, especially the bar￾separators alignment, infos alignment and whitespaces. Your output is a structured report.

main() function is also provided to you in the template code, you don't need to modify it.

Submission When submitting it to online judge, you should submit all your code except main(). If you add additional includes, classes, functions, variables or definitions, please also submit them to OJ.

DO NOT submit the main() function, or compile errors will occur.

hw5.md 10/25/2019

7 / 8 Problem 3. Lookup Table Description In this problem, you are required to implement a simple lookup table. This is, users has some variable , and a function , they want to know what's the result of . A simple idea is, we pre-calculate for all s, but that's not possible for a continuous function. But we can use sampling to discrete it, and use a vector to store these discrete (x, y) points. i.e. values of which start from , ends at and increments by can be stored as

 x | 0 | 1 | 2 | 3 | 4 | 5 

 --|---|---|---|---|----|---

 y | 0 | 1 | 4 | 9 | 16 | 25

In the beginning (class constructor), we will give the start position , end position , and an increment . You need to calculate the value ranges from to with an increment Then, users may input some arguments that they want to find the value of . For example, for , if some user input a number , you will output a number because . Here are some notes you need to pay attention to: If user wants to lookup some out of the range , for example, if someone try to find , which is not in the input range shown above, your program should call an error handling function

InputOutOfRangeError that provides to you, and just returns 0 for error. If someone tries to lookup a value in the input range, but there are no actually values matching, you should use a linear approximation to give the user an approximate value. That is, if someone try to find , but there are no stored in your database. The two s which have least difference with is and , then the value of is given by the following formula: For example, if someone tries to lookup 2.3 in the database shown above, then, we can get by the following formula: which equals to 5.5, respectively.

hw5.md 10/25/2019

8 / 8 We provide a pure virtual base class for you to do this job, you need to complete 3 classes inherits from this base class. The base class is like this:

class LookupTable

{

public:

// constructor, start: start position; end: end position; increment:

the increment "dx"

// You should calculate the value in range [start, end], with

increment "increment"

 LookupTable(double start, double end, double increment);

// virtual destructor

virtual ~LookupTable() = 0;

// get the value f(x) of the given x

virtual double getValue(double x) = 0;

};

Get the value of , where is the given argument double x. You need to return the value by searching in your database. Your job is to implement 3 child classes (inherit from LookupTable) named SquareLookupTable,

SinLookupTable, and LogLookupTable. In their constructors, you should calculate the values of f(x) from start to end, with increment

increment using functions from , and store it in a vector. Calling the function getValue(a) in SquareLookupTable, SinLookupTable, and LogLookupTable

will return , , . // is the same as Requirements In this problem, you are required to implement a lookup table. Your job is to complete 3 child classes shown above.

Submission When submitting it to online judge, you need to submit your LookupTable class and all child classes. If you add additional includes, classes, functions, variables or definitions, please also submit them to OJ.

DO NOT submit the main() function, or compile errors will occur.