辅导 CS111 2025、C/C++程序设计讲解
CS111 2025 Spring Homework 2:
Task 0
For this program, the source code files appear in different folders: Algorithms, Core, IO, Utils, and each of them has its own Test folder containing a testing file.
Compare this organization with putting all files in a single folder.
•Which approach is better for a project with many files, and why?
•Consider aspects such as maintainability, compilation speed, teamwork, and ease of navigation.
Answer:
Task 1: Answer the question.
In the Utils folder, you see files with .h, .hpp, .cpp, and .c extensions. Explain the typical use of each extension in a C++ project. When should you use .h vs .hpp, and .cpp vs .c?
Task 2: Answer the question.
In the Utils folder, according to the Makefile, the file c_utils.o (created by the C compiler gcc) will be linked with test_c_util.o (created by the C++ compiler g++).
•How can this be done successfully?
•What must be considered in the header files to ensure correct linkage?
Task 3: Answer the question.
There are three header files in the Utils folder:
•c_and_cpp_utils.hpp
•c_only_utils.h
•cpp_only_utils.hpp
What are the different intended uses of each file, and why?
Task 4
•In bit_stream.hpp, there is a line:
BitStream() = default;
Is this line needed? Why or why not?
•Consider another similar case, doubly_linked_list.hpp, there is a line:
DoublyLinkedList() = default;
Is this line needed? Why or why not?
Task 5
In bit_stream.hpp, there is a line:
size_t size() const noexcept;
•Is the noexcept part proper or helpful here? Why or why not?
•Discuss the implications of using noexcept regarding performance, correctness, and code design.
Task 6
In the file, there are two lines:
class BitProxy;
BitProxy operator[](size_t index);
Why is the first line (class BitProxy;) needed?
Why does operator[] return a BitProxy object instead of a bool& or bool?
Compare this design to how operator[] works in standard containers like std::vector or arrays.
Task 7
Regarding the design of the BitStream as shown in the file bit_stream.hpp, why not simply use an array of bool to replace it? Explain.
Regarding the design of the BitStream as shown in the file bit_stream.hpp, why not simply use an array of bool to replace it? Explain.
•Discuss the advantages and disadvantages of using an array of bool versus a BitStream.
•Consider aspects such as memory usage, performance, and practical implementation details, and relevance to the project.
Task 8
In the file bit_stream.hpp, you will find the following declaration:
std::vector data;
Suppose we replace uint8_t with char, i.e.,
std::vector data;
•Would this change be appropriate? Why or why not?
•Discuss the implications of using char instead of uint8_t for storing raw bit data, considering type safety, portability, and clarity.
Task 9
In bit_stream.cpp, consider the function definition of append_internal.
a.) Explain the meaning of the following code:
if (byte_index >= data.size()) {
data.push_back(0);
}
b.) Provide the missing statements in the following code, and briefly explain what your code does:
if (bit) {
// code missing
} else {
// code missing
}
Task 10
In the file bit_stream.cpp, in the definition of the function write_to_file, there are two lines:
const uint64_t bit_count = end_pos - start_pos;
file.write(reinterpret_cast(&bit_count), sizeof(bit_count));
Explain these two lines of statements.
Task 11
In the file bit_stream.cpp, in the function read_from_file, there is a loop to read the bits one by one from the file.
for (uint8_t byte : buffer) {
for (int bit_pos = 7; bit_pos >= 0 && bits_added < bit_count; --bit_pos) {
append_internal((byte >> bit_pos) & 1);
bits_added++;
}
}
Is this necessary? Why not simply read a sequence of uint8_t bytes from the file and save them into the BitStream object? Why?
Task 12
In doubly_linked_list.hpp, there are lines to assign values to members that are not part of a constructor, like:
Node* head = nullptr;
Node* tail = nullptr;
size_t len = 0;
What is the meaning of these, and why are these allowed? How does this differ from initializing these members in the constructor’s initializer list?
Task 13
In doubly_linked_list.hpp, the structure Node is declared in the public section of the class. Compared to another choice, putting this declaration in a private section of the class, what are the advantages and disadvantages of these two choices? Which is better for this project, and why?
Task 14
In doubly_linked_list.hpp, consider the constructor of Node:
explicit Node(const T&& val) : ...
What is the meaning of this? When does the method of the DoublyLinkedList class call this constructor of Node?
Task 15
In doubly_linked_list.hpp, in the function link_node, the code part for inserting a new node before an existing node is missing. Provide the missing code.
•You can put it in both places, here in this report file, and in the missing-code place in doubly_linked_list.hpp.
Task 16
In doubly_linked_list, there is a template function:
template
void insert_back(U&& value) {
link_node(new Node(std::forward(value)), tail, false);
}
What is the meaning of calling the forward function here?
Task 17
Since the insert_back function appears in the template class DoublyLinkedList, it is automatically a template function. So, maybe the template type name can be ignored and shared with the template type name T of the class. Then, can we use the following function? Why?
void insert_back(T&& value) {
link_node(new Node(std::forward(value)), tail, false);
}
Task 18
How about writing the insert_back function as follows:
template
void insert_back(T&& value) {
link_node(new Node(std::forward(value)), tail, false);
}
Is it ok? Why?
Task 19
In the file linked_list_and_priority_queue_tests.cpp, we see a range-based for loop:
for (auto& item : list) {
forward.push_back(item);
std::cout << item << " " << "pushed back" << " forwardly " <}
Here list is an object of the DoublyLinkedList class. Why is the range-based for loop supported here?
Alternatively, for a class X to support range-based for loop, what are the requirements of X?
Task 20
In doubly_linked_list.hpp, inside the iterator class,
there is a statement:
using iterator_category = std::bidirectional_iterator_tag;
Which header file is needed for this statement?
How about a typedef in this line, like:
typedef std::bidirectional_iterator_tag iterator_category ;
Which is better, and why?
Task 21
The type alias iterator_category, which is defined in doubly_linked_list.hpp, is not used in the program. Why is it needed or useful?
Task 22
In the file priority_queue.hpp, the template class PriorityQueue is declared as follows:
template <
typename T,
typename Compare = std::greater,
//template class Container = VectorHeapContainer
// Use the VectorHeapContainer for faster performance.
template class Container = OrderedLinkedListContainer
>
class PriorityQueue {
//...
};
By observing this code snippet, we should have learned something. Answer the following questions:
•When a template class has a template parameter that is also another template class, how do you express the code?
•How do you specify the default choice of a template parameter?
Task 23
In priority_queue.hpp, by observing, the PriorityQueue class can use any template class that satisfies some specific requirements, working as the container. Briefly describe what these requirements are.
Task 24
The requirements for a possible container class are similar to the concepts of interface and the is-a relationship, which the public inheritance mechanism can describe.
Sketch some design: a Container class that describes these requirements, so that any derived class of the Container class can be used by the PriorityQueue, which is adjusted accordingly.
Bonus Points if the adjusted program can compile and run correctly.
Task 25
In priority_queue.hpp, in the class OrderedLinkedListContainer, in the insert function, there is a statement.
static_assert(
std::is_convertible::type, T>::value, // C++11 syntax
"calling OrderedLinkedListContainer::insert() with data of type U. Type U must be implicitly convertible to T");
What is the meaning of this statement? Explain in terms of the usage of:
•is_convertible
•std::decay
•and std::static_assert;
Is it similar to some validation function defined in this program? What are the differences?
Task 26
In the file priority_queue.hpp, inside the class OrderedLinkedListcontainer, in the insert function, there is a statement missing, which is to locate the iterator it to the first place where the value is larger than the data of it or it reaches the end.
Provide the missing code (write the code here and in the . hpp file)
Task 27
In doubly_linked_list.hpp, in the class DoublyLinkedList, there are two iterator classes, iterator and const_iterator, defining the same overloaded operators. Why do we need these iterator classes ? Alternatively, asking, what would happen if only one iterator type were provided?
Task 28
In doubly_linked_list.hpp, the arrow operator -> will return a pointer. Explain how it works. Alternatively, explain, given the following sketch of statements:
iterator it = ...;
it->some_member;
What is an expression equivalent to it->some_member?
Task 29
The -- operator has a prefix version and a postfix version. How are the two operator functions declared differently, as shown in doubly_linked_list.hpp? A similar mechanism can overload the ++ operators.
Task 30
In the file link_list_and_priority_queue_tests.cpp. Two macros, TEST_FUNCTION and END_TEST, are defined. Some tests are carried out and the number of passing and failure are recorded. This mechanism is similarly used in the huffman_tests.cpp in the Algorithms folder. Describe this pattern of writing and testing code. What is the advantage of doing so?
Task 31:
In the file IO/Bit_stream/bit_stream.hpp
For the class BitStream, we observe that the BitProxy class has no copy constructor or copy assignment.
•Is defining these special methods (with copy semantics) problematic for this class? Why?
•If we want to turn off using these methods for this class explicitly, how do we do it?
–Discuss the legacy way and/or a modern C++ way.
Task 32:
In the file huffman_tree.hpp. The two concrete classes LeafNode and InternalNode share the same base class Node. What are the advantages and usefulness of this design? Consider the print_huffman_tree function defined in the huffman_tree.cpp as an example for the explanation.
Task 33:
In huffman_tree.hpp, in the classInternalNode, its two members left and right are unique pointers:
const std::unique_ptr left;
const std::unique_ptr right;
Why are these unique pointers proper? Comparing using raw pointers for left and right, what are the advantages/disadvantages?
Task 34:
In huffman_tree.cpp, in the print_huffman_tree definition, the code in the else block is missing, which should handle the recursive case of the function. It will print the left subtree, the internal node (the root of the subtree), and the right subtree. Provide the missing code.
•Write missing statements in the .cpp file, and also paste them here.
Task 35:
In huffman_algorithm.cpp, in the function string_to_frequence_map, a part is missing, which is to compute the count of each character in a string text.
Provide the missing code in the .cpp file and paste the lines here in this report file.
Task 36:
In huffman_tree.cpp, in the constructor of HuffmanTree, which builds a tree based on a frequency map (the parameter), there are missing parts of code.
•Provide code for both (a) inserting leaf nodes into the priority queue and (b) combining nodes to build the tree.
–The comments in the file are helpful.
–Put the missing code in the .cpp file and here in the report document.
•Explain in Part (b) why is needed when combining nodes?
Task 37:
In huffman_algorithm.cpp, in the function buildwere_table, why need to handle the special case that, if the prefix is empty, push afalseinto theprefix`? What would happen if this special case was not handled?
Task 38:
In huffman_algorithm.cpp, for the class HuffmanEncoder, the body is missing for the method encod. Write the missing code of the body.
•Hint: The comments in the function are helpful.
•Paste the code in the .cpp file and also here in this report file.
Task 39:
In huffman_algorithm.cpp, for the class HuffmanDecoder, the method decode, a loop is missing. Provide the missing code.
•Hint: the comments in the function are helpful.
•Paste the code in the .cpp file and also here in this report file.
Task 40:
Based on your code,
•During the encoding process, what will happen if there is a character in the string that is not in a leaf of the tree, therefore is not included in the encoding table?
•Any idea to solve the problem?
•Bonus , modify the code so that this problem be solved.
Task 41:
In huffman_tree.cpp, in the constructor of HuffmanTree, we see a line:
PriorityQueue, CompareNodes> pq;
So, the HuffmanTree object will always choose the default container choice of PriorityQueue. That is, there is no way to have two HuffmanTree objects simultaneously using different containers for their PriorityQueue.
How can the HuffmanTree declaration be modified so that different kinds of containers can possibly be used? Describe your idea.
Bonus: Compile and test the modified code.
Task 42
Based on a Makefile that you possibly want to use, explain the meaning of the following two rules:
1.
bit_stream.o: ../../IO/Bit_stream/bit_stream.cpp
$(CXX) $(CXXFLAGS) -c $< -o $@
1.
bit_stream_tests: bit_stream_tests.o bit_stream.o
$(CXX) $(CXXFLAGS) -o $@ $^
1.
all_exe: huffman_tests linked_list_and_priority_queue_tests \
bit_stream_tests c_utils_tests
4.
clean_o:
rm -f *.o
Task 43: Compilation and Execution
•Compile the program using the provided Makefile for your OS (Linux/MAC/Windows).
•Run and test all your generated executable files.
–If there are some errors and you can not fix them, describe the errors here.
Deliverable:
a) Paste a screenshot of your terminal showing:
•Simply type your name, and the terminal will ignore it as a wrong command
•The compilation command that uses a Makefile.
–You may use make or mingw32-make.
Paste the image in this report.
A sample screenshot could show something like:
\Huffman_code_program\Build\Windows_g++> my name is LI, BAI
'my' is not recognized as an internal or external command,
operable program or batch file.
\Huffman_code_program\Build\Windows_g++> mingw32-make
g++ -c ../../IO/Bit_stream/bit_stream.cpp -o bit_stream.o
g++ -o bit_stream_tests bit_stream_tests.o bit_stream.o
...
b) a screen shot of running an exectuable file. Paste the image in this report.
c) [optional]: redirect output to a file, like:
\Huffman_code_program\Build\Windows_g++>huffman_tests.exe > huffman_tests_output.txt
Upload test ouptut .txt files together with the report file.
Task 44
[optional]
Describe any other extra work that worth bonus points.
Task 45
Summarize your experience, what you have learned in this project. You may think in the aspects like:
•The specific C++ features you found valuable: design patterns, skills of debugging, testing and compiling skills ...
•How will this experience help you in future programming projects?
•How this project can assist a future Data Structure and algorithm course?
Submission Instructions:
Submit the folloing files:
1.The completed report file (Word/PDF) with answers
2.A .zip file made by compressing the whole program folder, which contain all subfolders as distributed by this homework. The folder should contain all your modified program files.
3.Other supporting files, like
•the executing output recording files.
•...