Implementing a Dynamic Vector (Array) in C(使用c实现动态数组Vector)
An array (vector) is a common-place data type, used to hold and describe a collection of elements. These elements can be fetched at runtime by one or more indices (identifying keys). A distinguishing feature of an array compared to a list is that they allow for constant-time random access lookup, compared to the latters sequential access. Resizable arrays allow for an unspecified upper-bound of collection elements at runtime, and are conceptuality similar to a list. These dynamic arrays are more complicated and less used in introduction to its compatriot list, which is dynamic by nature. Using C as the language of implementation this post will guide you through building a simple vector data-structure. The structure will take advantage of a fixed-size array, with a counter invariant that keeps track of how many elements are currently present. If the underlying array becomes exhausted, the addition operation will re-allocate the contents to a larger size, by way of a copy.
The Make File
‘Make’ is a popular utility used throughout software development to build executable artifacts (programs and libraries) from described source code. Through a simple DSL, associations from descriptive short-names (targets) and a series of related commands to execute are made. Running the ‘make’ command executes the first present target, and this must be considered in the design of the file. Below is a sample Makefile which provides the vector project with simple build, debug and clean targets.
CC=gcc
CFLAGS=
RM=rm -rf
OUT=vector all: build build: main.o vector.o
$(CC) $(CFLAGS) -o $(OUT) main.c vector.c
$(RM) *.o debug: CFLAGS+=-DDEBUG_ON
debug: build main.o: main.c vector.h
$(CC) $(CFLAGS) -c main.c vector.o: vector.c vector.h
$(CC) $(CFLAGS) -c vector.c clean:
$(RM) *.o $(OUT)
Looking at the code example above you will notice a few variables which are used to define specific aspects used when running the targets (such as the compiler command and flags used). To keep things modular the compilation of the ‘main’ and ‘vector’ source-code files has been split, with file dependences specific to each target specified after the short-name. The ‘debug’ target appends a macro definition flag which is used to include any debug information present in the source code.
The Header File
Defining a header file allows the programmer to separate specific aspects of the programs source-code into reusable files. These files commonly contain forward declarations of identifiers and functions. This allows a user to include the codes header file in their own work, separating the definition from the implementation. Including a header file produces the same results as copying the full contents into the callers file. Below shows the header file implemented for the vector example.
#ifndef VECTOR_H
#define VECTOR_H #define VECTOR_INIT_CAPACITY 4 #define VECTOR_INIT(vec) vector vec; vector_init(&vec)
#define VECTOR_ADD(vec, item) vector_add(&vec, (void *) item)
#define VECTOR_SET(vec, id, item) vector_set(&vec, id, (void *) item)
#define VECTOR_GET(vec, type, id) (type) vector_get(&vec, id)
#define VECTOR_DELETE(vec, id) vector_delete(&vec, id)
#define VECTOR_TOTAL(vec) vector_total(&vec)
#define VECTOR_FREE(vec) vector_free(&vec) typedef struct vector {
void **items;
int capacity;
int total;
} vector; void vector_init(vector *);
int vector_total(vector *);
static void vector_resize(vector *, int);
void vector_add(vector *, void *);
void vector_set(vector *, int, void *);
void *vector_get(vector *, int);
void vector_delete(vector *, int);
void vector_free(vector *); #endif
We wrap the contents of this file in a definition condition to make sure that even with multiple inclusion between aggregate source code files, only one inclusion is processed in the result. A ‘vector’ type definition is included which provides access to the capacity and total current elements in the collection. Along with this, a ‘items’ variable with a pointer of void pointers is included, allowing us to insert a heterogeneous collection of elements into the vector. The ‘vector_resize’ method is defined to be ‘static’ resulting in successful execution of the function only occurring in the file it is defined in (accessibility control).
The Implementation File
Using the header file definition, the following file is used to implement these methods. As discussed in the previous section ‘void pointers’ are used to reference the collection elements. Void pointers are pointers which point to some arbitrary data that has no specific type. As a consequence you are unable to directly deference a pointer of this type and must first provide a casting type.
#include <stdio.h>
#include <stdlib.h> #include "vector.h" void vector_init(vector *v)
{
v->capacity = VECTOR_INIT_CAPACITY;
v->total = ;
v->items = malloc(sizeof(void *) * v->capacity);
} int vector_total(vector *v)
{
return v->total;
} static void vector_resize(vector *v, int capacity)
{
#ifdef DEBUG_ON
printf("vector_resize: %d to %d\n", v->capacity, capacity);
#endif void **items = realloc(v->items, sizeof(void *) * capacity);
if (items) {
v->items = items;
v->capacity = capacity;
}
} void vector_add(vector *v, void *item)
{
if (v->capacity == v->total)
vector_resize(v, v->capacity * );
v->items[v->total++] = item;
} void vector_set(vector *v, int index, void *item)
{
if (index >= && index < v->total)
v->items[index] = item;
} void *vector_get(vector *v, int index)
{
if (index >= && index < v->total)
return v->items[index];
return NULL;
} void vector_delete(vector *v, int index)
{
if (index < || index >= v->total)
return; v->items[index] = NULL; for (int i = index; i < v->total - ; i++) {
v->items[i] = v->items[i + ];
v->items[i + ] = NULL;
} v->total--; if (v->total > && v->total == v->capacity / )
vector_resize(v, v->capacity / );
} void vector_free(vector *v)
{
free(v->items);
}
Looking at the code example above you will notice that the ‘vector_resize’ function is called if certain conditions are met on addition or deletion. If the current vector capacity has been exhausted when an addition has been requested the size is doubled and the vector contents re-allocated. In a similar fashion, upon deletion, if the vector is a quarter full the contents is reallocated to a vector of half the current size. These conditions for resizing work well in practice to balance memory capacity and computation time required to fulfill each resize.
The Test Case
With all the pieces put in place we are now able to test case the implementation. Below shows an example using the direct functions, adding a few strings (character sequences) to a collection, printing the contents, modifying the contents and then printing it out again. One unfortunate use-case detail that can not be avoided with the use of void pointers is the necessary cast.
#include <stdio.h>
#include <stdlib.h> #include "vector.h" int main(void)
{
int i; vector v;
vector_init(&v); vector_add(&v, "Bonjour");
vector_add(&v, "tout");
vector_add(&v, "le");
vector_add(&v, "monde"); for (i = ; i < vector_total(&v); i++)
printf("%s ", (char *) vector_get(&v, i));
printf("\n"); vector_delete(&v, );
vector_delete(&v, );
vector_delete(&v, ); vector_set(&v, , "Hello");
vector_add(&v, "World"); for (i = ; i < vector_total(&v); i++)
printf("%s ", (char *) vector_get(&v, i));
printf("\n"); vector_free(&v);
}
To simplify the use of the vector implementation the header file defines a few macro functions which can be used in place of the base function calls. Below highlights these definition in practice, removing some of the verbosity present in the previous example.
#include <stdio.h>
#include <stdlib.h> #include "vector.h" int main(void)
{
int i; VECTOR_INIT(v); VECTOR_ADD(v, "Bonjour");
VECTOR_ADD(v, "tout");
VECTOR_ADD(v, "le");
VECTOR_ADD(v, "monde"); for (i = ; i < VECTOR_TOTAL(v); i++)
printf("%s ", VECTOR_GET(v, char*, i));
printf("\n"); VECTOR_DELETE(v, );
VECTOR_DELETE(v, );
VECTOR_DELETE(v, ); VECTOR_SET(v, , "Hello");
VECTOR_ADD(v, "World"); for (i = ; i < VECTOR_TOTAL(v); i++)
printf("%s ", VECTOR_GET(v, char*, i));
printf("\n"); VECTOR_FREE(v);
}
Despite still having to provide a casting data type when retrieving a collection element, the macros clean-up and simplify the process a great deal.
Resources
- Why use Pointers? Dynamic Memory Allocation
- Void Pointers in C
- Implementation of a Vector data structure in C
- What does “static” mean in a C program?
Implementing a Dynamic Vector (Array) in C(使用c实现动态数组Vector)的更多相关文章
- 动态数组 - vector
#include <iostream> #include <vector> // 头文件 using namespace std; int main() { vector< ...
- 【模板】c++动态数组vector
相信大家都知道$C$++里有一个流弊的$STL$模板库.. 今天我们就要谈一谈这里面的一个容器:动态数组$vector$. $vector$实际上类似于$a[]$这个东西,也就是说它重载了$[]$运算 ...
- C++ STL之动态数组vector(⽮量)的使⽤
写再最前面:摘录于柳神的笔记: 之前C语⾔⾥⾯⽤ int arr[] 定义数组,它的缺点是数组的⻓度不能随⼼所欲的改变,⽽C++⾥⾯有⼀个能完全替代数组的动态数组 vector (有的书⾥⾯把它翻 ...
- 越努力越幸运--动态数组vector
最近回忆山哥写的stl,觉得很好用,也写了一份. 感谢群里的大佬帮忙review,还是很多的问题的. code:https://github.com/HellsingAshen/vector_c.gi ...
- C++ vector动态数组
#include<vector>头文件 vector类称作向量类 百度百科的解释:https://baike.baidu.com/item/vector/3330482 我喜欢把知识点拿出 ...
- C++向量 vector动态数组
需要包含头文件, #include <vector> using namespace std; vector 容器与数组相比其优点在于它能够根据需要随时自动调整自身的大小以便容下所 ...
- vector:动态数组
vector是C++标准模板库中的部分内容,中文偶尔译作“容器”,但并不准确.它是一个多功能的,能够操作多种数据结构和算法的模板类和函数库.vector之所以被认为是一个容器,是因为它能够像容器一样存 ...
- vector & array
private static const NUM_LOOPS:int = 15; public function VectorTest():void { var vector:Vector.<i ...
- 数组Array和列表集合ArrayList、LinkedList和Vector的区别
一.ArrayList和Vector的区别 ArrayList与Vector主要从以下方面来说. 1.同步性: Vector是线程安全的,也就是说是同步的,而ArrayList是线程序不安全的,不是同 ...
随机推荐
- Redis03——Redis架构
Redis架构 1.1.问题 redis是单线程,单实例,为什么并发那么多,依旧很快呢? 回答:因为调用了系统内核的epoll 1.2.Linux的早期版本 Linux有Linux kernal,我们 ...
- Ubuntu 18.04 手动升级内核
一般情况下,系统正常更新,会自动升级内核到可用的最新版. 查看已安装的内核 $ sudo dpkg -l | grep linux-image 查看当前使用的内核 $ sudo uname -r 查看 ...
- Resource ResourceLoader
DefaultResourceLoader -- > ResourceLoader 方法 ResourceLoader getResource(String location); Class ...
- win10 sql server2012 连接navicat 无法使用
修改 连接navicat时报如下错误,按照网上办法,各种删除navicat的客户端,行不通. 点击sqlserver管理器,提示评估期已过.有关如何升级的测试版软件的信息,请访问http://www. ...
- Eclipse里导入Mybatis源码工程
打开Eclipse,在前两天的记录里我已经把Maven什么的都配置好了,还有Mybatis的源码也下载下来了,不相信的话可以去看一下我之前的记录:) OK. Mybatis源码解压之后是一个标准的Ma ...
- 一些C++编码规范
1.成员变量是引用类型,头文件只需向前声明对应类,不需包含类头文件,在实现文件中需要包含: 2.头文件声明变量和函数按照pulic.protected.private顺序: 3.成员变量声明,加 &q ...
- jenkins汉化
插件: Localization: Chinese (Simplified) locale plugin(或者是这个版本不一样,名字不一样) 可以直接安装这个插件,然后走最后一步设置即可. 由于安装失 ...
- include和require的区别(PHP版本7)
亲自测试了一下,发现include有条件包含require无条件包含这个区别在PHP7版本中(据说PHP5以后)是不存在的了,也就是在if(false){ } 中都不会执行:还有一个返回值的问题,测试 ...
- LibreOffice/Calc:带条件判断的求和
本文适用于LibreOffice Calc 5.1.6.2 + Ubuntu 16.04,熊猫帮帮主@cnblogs 2018/3/7 以下图为例,假设要根据C列对D列中被选中单元进行求和,即对D列中 ...
- R-ts()
概述 ts(gm,frequency=12,start=c(1975,1)) 这个命令表示: frequency=12表明时间单位为年,而且在每一个时间单位中有12个均匀间隔的观察值. 因此gm是 ...