CSE2425, C programming lab, course 2020-2021
Final assignment: Hash map
1 Introduction
In this final assignment you will implement a hash map
1
. A hash map is a data
structure that associates a key with a value (a chunk of data). Most hash maps
are implemented as an array of so-called buckets. A hash function translates
a given key (e.g., a name) to an index in the array, where the corresponding
bucket is stored.
Below we will specify the data structures that you have to provide, and the
functions that you have to implement. This assignment includes two bonus
functions that can raise your score from pass (C) to good (B) to excellent (A).
2 Testing
The first part of the assignment consist of implementing a test set for the hash
map. We have created a number of incorrect hash map implementations. The
goal is to create a test set on which these incorrect implementations fail. When
you have finished creating this test set, you can use this test set to test your own
implementation by copy&pasting it into the my tests of the Hashmap assignment
in Weblab.
3 Hash map structure
Define a type HashMap, which represents the hash map data structure.
Note: Use typedef such that a HashMap structure can be used without using
the struct keyword, i.e. the following construction should be possible:
HashMap *hm;
4 Creating a hash map
1. Implement a function create_hashmap that returns a pointer to the newly
constructed HashMap structure and has parameter
ˆ key_space, a size_t
2
that represents the number of buckets in the hash
map.
1http://en.wikipedia.org/wiki/Hashmap
2http://en.wikipedia.org/wiki/Size_t
1CSE2425, C programming lab, course 2020-2021
This function should allocate enough memory to fit key_space buckets, and the
allocated memory should be zeroed (i.e., NULLed).
2. A hash function maps a string (i.e. an array of chars ending with a null
character) to an index, so it returns a unsigned int. The parameter of a hash
function is simply a
ˆ key, a null-terminated string of characters.
As the hash map can only hold up to key_space buckets, using the hash function
–for example to lookup a mapping– requires some care; apply modulo key_space
to the result such that the value will be in the available bucket range.
3. A default hash function named hash should be implemented. This function
should sum all ASCII values of the characters of the key.
For example:
char *key = "AC";
unsigned int h = hash(key);
=> h = 1**
5 Inserting data
Implement a function insert_data that has parameters
ˆ hm, a pointer to a hash map;
ˆ key, a null-terminated string of characters;
ˆ data, a void pointer to the source data;
ˆ resolve_collision, a ResolveCollisionCallback (see below).
The function should store the data pointer and a copy of the key in the bucket
that can be found by applying the hash function on the key. In case of a
collision, i.e. when there already is data with the same key in the hash map, the
resolve_collision function should be called with the the previously stored
data and data as arguments and the returned void pointer should be stored in
the bucket instead.
ResolveCollisionCallback, a pointer to a function that returns a void pointer
and has two parameters:
ˆ old_data, a void pointer to the previously stored data;
ˆ new_data, a void pointer to the data that is being newly inserted.
The function should determine what data is stored in the has map in case of a
key collision by returning the void pointer to the data that is to be stored.
2CSE2425, C programming lab, course 2020-2021
6 Retrieving data
Implement a function get_data that has parameters
ˆ hm, a pointer to a hash map;
ˆ key, a null-terminated string of characters.
The function should return the data pointer (a void pointer) in the hash map
that is associated with the key. If the key is not present in the hash map, NULL
should be returned.
7 Iterator
Implement a function iterate that has parameters
ˆ hm, a pointer to a hash map;
ˆ callback, a pointer to a function that returns nothing (i.e. void) and has
two parameters:
– key, a null-terminated string of characters;
– data, a void pointer to the data.
This function should iterate over the entire hash map. For each data element
it finds, the callback function should be called with the two members of the
element.
8 Removing data
Implement a function remove_data that has parameters
ˆ hm, a pointer to a hash map;
ˆ key, a null-terminated string of characters.
ˆ destroy_data, a DestroyDataCallback (see below).
This function should remove the element in the hash map that is associated with
the given key. If the destroy_data parameter is non-NULL it should be called
with the data pointer of the element as argument. If the key is not present, the
hash map should remain untouched. As the remove_data function cannot fail,
its return type is void.
DestroyDataCallback, a pointer to to a function that returns nothing (i.e.
void) and has one parameter:
ˆ data, a void pointer.
The function should clean up the data (e.g. free allocated memory).
3CSE2425, C programming lab, course 2020-2021
9 Deleting a hash map
Implement a function delete_hashmap that has parameters
ˆ hm, a pointer to the hash map that is to be deleted;
ˆ destroy_data, a DestroyDataCallback (see 8).
The function should deallocate all memory that was allocated by the hash map.
If the destroy_data parameter is non-NULL it should be called for every data
element that is stored in the hash map with the data pointer of the element as
argument.
10 Bonus: New hash function
Implement a function set_hash_function that has parameters
ˆ hm, a pointer to a hash map;
ˆ hash_function, a pointer to a hash function that returns a unsigned int
and a single parameter:
– key, a null-terminated string of characters.
This function should set hash_function as the new hash function of the hash
map hm. Changing the hash function means that a particular key may now be
hashed to different bucket than it was with the previous hash function. The
hash map must be updated (rehashed) to reflect this so that all data in the
hash map can still be retrieved with their corresponding keys.
11 Bonus: Counting Words
Implement a function count_words that has parameters
ˆ stream, a pointer to a FILE.
This function should count the number of times each word in the stream occurs
using the hash map you implemented. A word is defined as a sequence of one or
more alphanumeric characters (case sensitive). You may use fscanf
3
to read a
particular set of characters from a stream but other solutions are also accepted.
The data stored in the hash map should be properly allocated and deallocated,
do not simply store an integer that is cast to a pointer type. The return type
of the function is void.
3http://en.cppreference.com/w/c/io/fscanf
4CSE2425, C programming lab, course 2020-2021
Given the input:
foo bar_, foo!
bar "baz".
foo?
The program should write the following to the standard output:
bar: 2
baz: 1
foo: 3
The order in which the output is printed is not important.
12 Submission
The assignment should be implemented on Weblab.
ˆ All test code should be located in the Testing assignment.
ˆ All hash map code should be located in the Hashmap assignment.
ˆ Put all the word count source code inside the Wordcount assignment;
ˆ If you have implemented the first bonus exercise, add the following macro
to your Hashamp submission:
#define NEW_HASH
ˆ Do not include a main function. (We will use our own test driver, just like
the example test provided.)
Submissions violating the above requirements will be automatically rejected by
the Weblab system.
請加QQ：99515681 郵箱：
[email protected] WX：codinghelp
