开扒php内核函数，第三篇 implode-php手册-php.cn

一开始觉得implode挺容易实现，但是写着写着才发现是挺复杂的，不说啦来看看implode的用法吧 1 ? php 2 $arr = array ('Hello','World!','Beautiful','Day!' ); 3 echo implode (" ", $arr ); 4 ? 上面会输出 Hello World! Beautiful Day! 下面的程序的我写

一开始觉得implode挺容易实现，但是写着写着才发现是挺复杂的，不说啦

来看看implode的用法吧

<span>1</span> <span>php
</span><span>2</span> <span>$arr</span> = <span>array</span>('Hello','World!','Beautiful','Day!'<span>);
</span><span>3</span> <span>echo</span> <span>implode</span>(" ",<span>$arr</span><span>);
</span><span>4</span> ?>

로그인 후 복사

上面会输出 Hello World! Beautiful Day!
로그인 후 복사

下面的程序的我写的

<span> 1</span> <span>/*</span><span>字符串翻转</span><span>*/</span>
<span> 2</span> <span>void</span> strover(<span>char</span> *<span> str){
</span><span> 3</span>     <span>int</span> len =<span> strlen(str);
</span><span> 4</span>     <span>//</span><span>int half =  strlen(str)/2;</span>
<span> 5</span>     <span>int</span><span> i,j;
</span><span> 6</span>     <span>char</span><span> tmp;
</span><span> 7</span>     j = len-<span>1</span><span>;
</span><span> 8</span>     <span>for</span>(i=<span>0</span>;i){
<span> 9</span>         tmp =<span> str[j];
</span><span>10</span>         str[j] =<span> str[i];
</span><span>11</span>         str[i] =<span> tmp;
</span><span>12</span>         j--<span>;
</span><span>13</span> <span>    }
</span><span>14</span>     
<span>15</span> 
<span>16</span> <span>}
</span><span>17</span> 
<span>18</span> 
<span>19</span> 
<span>20</span> 
<span>21</span> 
<span>22</span> <span>/*</span>
<span>23</span> <span>    2进制转十进制 要处理正负数啊 涉及到负数啊
</span><span>24</span> <span>    字符串翻转
</span><span>25</span> <span>*/</span>
<span>26</span> <span>char</span> * bin2decimal(<span>int</span><span> number){
</span><span>27</span>    
<span>28</span>    <span>int</span> q = <span>0</span>; <span>//</span><span>商</span>
<span>29</span>     <span>int</span> r = <span>0</span>;<span>//</span><span>余数</span>
<span>30</span>     <span>int</span> i  = <span>0</span><span>;
</span><span>31</span>     <span>int</span> tmp =<span> number;
</span><span>32</span>   <span>int</span> is_negative = <span>0</span><span>;
</span><span>33</span>     <span>char</span> *<span> res;
</span><span>34</span>      res = (<span>char</span> *)malloc(<span>sizeof</span>(<span>char</span>)*<span>5</span>+<span>1</span><span>);
</span><span>35</span>    <span>if</span>(number>=<span>0</span><span>){
</span><span>36</span>       
<span>37</span>    }<span>else</span><span>{
</span><span>38</span>         tmp = -<span>number;
</span><span>39</span>         res[i++] = <span>'</span><span>-</span><span>'</span><span>;
</span><span>40</span>         is_negative = <span>1</span><span>;
</span><span>41</span> <span>   }
</span><span>42</span>   
<span>43</span>          <span>do</span><span>{
</span><span>44</span>           q = tmp/<span>10</span><span>;
</span><span>45</span> 
<span>46</span>           r  = tmp%<span>10</span><span>;
</span><span>47</span>          <span>//</span><span> tmp = q;
</span><span>48</span>          <span>//</span><span> c = hex_str[r];</span>
<span>49</span>          res[i++] = <span>'</span><span>0</span><span>'</span>+<span>r;
</span><span>50</span>         tmp =<span> q;
</span><span>51</span>        }<span>while</span><span>(tmp);
</span><span>52</span>    
<span>53</span>    res[i] = <span>'</span><span>\0</span><span>'</span><span>;
</span><span>54</span>   
<span>55</span>     strover(&<span>res[is_negative]);
</span><span>56</span>   <span>return</span><span> res;
</span><span>57</span>    
<span>58</span> 
<span>59</span> <span>}
</span><span>60</span> 
<span>61</span> 
<span>62</span> 
<span>63</span> 
<span>64</span> <span>/*</span>
<span>65</span> <span>    c语言真的太麻烦啦，传数组，但是无法知道数组的长度，只能够手动传入
</span><span>66</span> <span>*/</span>
<span>67</span> <span>char</span> * implode(<span>int</span> *number,<span>int</span> size,<span>char</span> *<span> dem){
</span><span>68</span>     <span>int</span> i = <span>0</span><span>;
</span><span>69</span>     <span>char</span>*<span> c;
</span><span>70</span>     <span>//</span><span>c[1] = '\0';</span>
<span>71</span>     <span>struct</span><span> simple_mem{
</span><span>72</span>         <span>char</span> *<span> res;
</span><span>73</span>         unsigned <span>int</span><span> len;
</span><span>74</span>         unsigned <span>int</span><span> used;
</span><span>75</span> <span>    }test_mem;
</span><span>76</span>     test_mem.res = (<span>char</span> *)malloc(<span>sizeof</span>(<span>char</span>)*<span>20</span><span>);
</span><span>77</span>     test_mem.len = <span>sizeof</span>(<span>char</span>)*<span>20</span><span>;
</span><span>78</span>     test_mem.used = <span>0</span><span>;
</span><span>79</span>     <span>for</span>(;isize;){
<span>80</span>              c=<span> bin2decimal(number[i]);
</span><span>81</span>            memcpy(test_mem.res+<span>test_mem.used,c,strlen(c));
</span><span>82</span>           
<span>83</span>             test_mem.used+=<span>strlen(c);
</span><span>84</span>             <span>if</span>(++isize){
<span>85</span>                  memcpy(test_mem.res+<span>test_mem.used,dem,strlen(dem));
</span><span>86</span>                  test_mem.used+=<span>strlen(dem);
</span><span>87</span> <span>            }
</span><span>88</span>             
<span>89</span> <span>    }
</span><span>90</span>  test_mem.res[test_mem.used] = <span>'</span><span>\0</span><span>'</span><span>;
</span><span>91</span>  printf(<span>"</span><span>%s</span><span>"</span><span>,test_mem.res);
</span><span>92</span> 
<span>93</span> 
<span>94</span> }

로그인 후 복사

我们写的implode写的函数是针对整形数组，php的当然什么类型都支持啊，c语言也可以实现泛型，但毕竟比较麻烦的，上面的程序还是比较多问题的，优化的地方有很多，但是我们是抱着学习的态度来的

<span> 1</span> <span>int</span><span> main(){
</span><span> 2</span>     <span>//</span><span>char * res = bin2hex("a");
</span><span> 3</span>     <span>//</span><span>printf("hex a=%s",res);
</span><span> 4</span>     <span>//</span><span>char * res = hex2bin("6578616d706c65206865782064617461");</span>
<span> 5</span>     <span>int</span> integer[<span>3</span>] = {<span>1</span>,-<span>24</span>,<span>3</span><span>};
</span><span> 6</span>     implode(integer,<span>sizeof</span>(integer)/<span>sizeof</span>(<span>int</span>),<span>"</span><span>*</span><span>"</span><span>);
</span><span> 7</span> 
<span> 8</span>     <span>//</span><span>bin2decimal(-1234);</span>
<span> 9</span>     
<span>10</span>     <span>return</span> <span>0</span><span>;
</span><span>11</span> }

로그인 후 복사

先说说思路吧

1,主要是算法是2进制转10进制字符串显示，当然我们要注意负数啦，还有字符串翻转

2 内存分配，因为我们没有限制数组的长度，所以我们要动态去分配，其实我们可以有一样可以确定的是整形的范围 0到65535 就是说一个整形最多占5个字符，

3 其他就没什么啦

来看看php的吧

<span> 1</span> <span>/*</span>
<span> 2</span> <span> * Convert num to its decimal format.
</span><span> 3</span> <span> * Return value:
</span><span> 4</span> <span> *   - a pointer to a string containing the number (no sign)
</span><span> 5</span> <span> *   - len contains the length of the string
</span><span> 6</span> <span> *   - is_negative is set to TRUE or FALSE depending on the sign
</span><span> 7</span> <span> *     of the number (always set to FALSE if is_unsigned is TRUE)
</span><span> 8</span> <span> *
</span><span> 9</span> <span> * The caller provides a buffer for the string: that is the buf_end argument
</span><span>10</span> <span> * which is a pointer to the END of the buffer + 1 (i.e. if the buffer
</span><span>11</span> <span> * is declared as buf[ 100 ], buf_end should be &buf[ 100 ])
</span><span>12</span>  <span>*/</span>
<span>13</span> <span>/*</span><span> char * ap_php_conv_10() {{{ </span><span>*/</span>
<span>14</span> <span>char</span> *<span> ap_php_conv_10(register wide_int num, register bool_int is_unsigned,
</span><span>15</span>        register bool_int * is_negative, <span>char</span> *buf_end, register <span>int</span> *<span>len)
</span><span>16</span> <span>{
</span><span>17</span>     register <span>char</span> *p =<span> buf_end;
</span><span>18</span> <span>    register u_wide_int magnitude;
</span><span>19</span> 
<span>20</span>     <span>if</span><span> (is_unsigned) {
</span><span>21</span>         magnitude =<span> (u_wide_int) num;
</span><span>22</span>         *is_negative =<span> FALSE;
</span><span>23</span>     } <span>else</span><span> {
</span><span>24</span>         *is_negative = (num 0<span>);
</span><span>25</span> 
<span>26</span>         <span>/*</span>
<span>27</span> <span>         * On a 2's complement machine, negating the most negative integer
</span><span>28</span> <span>         * results in a number that cannot be represented as a signed integer.
</span><span>29</span> <span>         * Here is what we do to obtain the number's magnitude:
</span><span>30</span> <span>         *      a. add 1 to the number
</span><span>31</span> <span>         *      b. negate it (becomes positive)
</span><span>32</span> <span>         *      c. convert it to unsigned
</span><span>33</span> <span>         *      d. add 1
</span><span>34</span>          <span>*/</span>
<span>35</span>         <span>if</span> (*<span>is_negative) {
</span><span>36</span>             wide_int t = num + <span>1</span><span>;
</span><span>37</span>             magnitude = ((u_wide_int) - t) + <span>1</span><span>;
</span><span>38</span>         } <span>else</span><span> {
</span><span>39</span>             magnitude =<span> (u_wide_int) num;
</span><span>40</span> <span>        }
</span><span>41</span> <span>    }
</span><span>42</span> 
<span>43</span>     <span>/*</span>
<span>44</span> <span>     * We use a do-while loop so that we write at least 1 digit
</span><span>45</span>      <span>*/</span>
<span>46</span>     <span>do</span><span> {
</span><span>47</span>         register u_wide_int new_magnitude = magnitude / <span>10</span><span>;
</span><span>48</span> 
<span>49</span>         *--p = (<span>char</span>)(magnitude - new_magnitude * <span>10</span> + <span>'</span><span>0</span><span>'</span><span>);
</span><span>50</span>         magnitude =<span> new_magnitude;
</span><span>51</span> <span>    }
</span><span>52</span>     <span>while</span><span> (magnitude);
</span><span>53</span> 
<span>54</span>     *len = buf_end -<span> p;
</span><span>55</span>     <span>return</span><span> (p);
</span><span>56</span> }

로그인 후 복사

> php5ts_debug.dll!ap_php_conv_10(__int64 num=-278, int is_unsigned=0, int * is_negative=0x00c3e154, char * buf_end=0x00c3e9c0, int * len=0x00c3ea64) 行320 C
php5ts_debug.dll!format_converter(buf_area * odp=0x00c3eb9c, const char * fmt=0x105d799e, char * ap=0x00c3ecc0) 行869 + 0x34 字节 C
php5ts_debug.dll!strx_printv(int * ccp=0x00c3eca0, char * buf=0x00c3ee90, unsigned int len=12, const char * format=0x105d799c, char * ap=0x00c3ecbc) 行1213 + 0x11 字节 C
php5ts_debug.dll!ap_php_slprintf(char * buf=0x00c3ee90, unsigned int len=12, const char * format=0x105d799c, ...) 行1229 + 0x19 字节 C
php5ts_debug.dll!php_implode(_zval_struct * delim=0x030dffd8, _zval_struct * arr=0x030dff88, _zval_struct * return_value=0x030e0028, void * * * tsrm_ls=0x00353040) 行1154 + 0x1b 字节 C
php5ts_debug.dll!zif_implode(int ht=2, _zval_struct * return_value=0x030e0028, _zval_struct * * return_value_ptr=0x00000000, _zval_struct * this_ptr=0x00000000, int return_value_used=1, void * * * tsrm_ls=0x00353040) 行1250 + 0x15 字节 C
php5ts_debug.dll!zend_do_fcall_common_helper_SPEC(_zend_execute_data * execute_data=0x030c20d8, void * * * tsrm_ls=0x00353040) 行643 + 0x62 字节 C
php5ts_debug.dll!ZEND_DO_FCALL_SPEC_CONST_HANDLER(_zend_execute_data * execute_data=0x030c20d8, void * * * tsrm_ls=0x00353040) 行2234 C
php5ts_debug.dll!execute(_zend_op_array * op_array=0x030dfa40, void * * * tsrm_ls=0x00353040) 行410 + 0x11 字节 C
php5ts_debug.dll!zend_execute_scripts(int type=8, void * * * tsrm_ls=0x00353040, _zval_struct * * retval=0x00000000, int file_count=3, ...) 行1329 + 0x21 字节 C
php5ts_debug.dll!php_execute_script(_zend_file_handle * primary_file=0x00c3fcf4, void * * * tsrm_ls=0x00353040) 行2502 + 0x1b 字节 C
php.exe!do_cli(int argc=2, char * * argv=0x00352fa0, void * * * tsrm_ls=0x00353040) 行989 + 0x10 字节 C
php.exe!main(int argc=2, char * * argv=0x00352fa0) 行1365 + 0x11 字节 C

调用堆栈如上

<span>1</span>  <span>do</span><span> {
</span><span>2</span>          register u_wide_int new_magnitude = magnitude / <span>10</span><span>;
</span><span>3</span>  
<span>4</span>          *--p = (<span>char</span>)(magnitude - new_magnitude * <span>10</span> + <span>'</span><span>0</span><span>'</span><span>);
</span><span>5</span>          magnitude =<span> new_magnitude;
</span><span>6</span> <span>     }
</span><span>7</span>      <span>while</span> (magnitude);

로그인 후 복사

关键是这段代码，作者没有像我们用取余去计算，而是把它乘，举个例子吧

magnitude = 283

new_magnitude = 283/10 = 28

*--p = 283 - 28*10+'0' = '3'

magnitude = new_magnitude = 28

然后继续上面的步骤啦

取余考虑和乘法考虑那个高，不知道作者的想法是怎样的，有时间用汇编证明一下，那个用的指令比较多

第二个的就是作者用了倒序字符复制 *--p,这就要读读内存的代码啦

<strong>*len = buf_end -</strong><span><strong> p;  字符串长度可以这样计算的，指针的作用就是不错啊<br></strong><br>回到调用的地方如下<br><br></span>

로그인 후 복사

<span> 1</span>     s = ap_php_conv_10(i_num, (*fmt) == <span>'</span><span>u</span><span>'</span>, &<span>is_negative,
</span><span> 2</span>                                 &num_buf[NUM_BUF_SIZE], &<span>s_len);
</span><span> 3</span> <span>                    FIX_PRECISION(adjust_precision, precision, s, s_len);
</span><span> 4</span> 
<span> 5</span>                     <span>if</span> (*fmt != <span>'</span><span>u</span><span>'</span><span>) {
</span><span> 6</span>                         <span>if</span><span> (is_negative) {
</span><span> 7</span>                             prefix_char = <span>'</span><span>-</span><span>'</span><span>;
</span><span> 8</span>                         } <span>else</span> <span>if</span><span> (print_sign) {
</span><span> 9</span>                             prefix_char = <span>'</span><span>+</span><span>'</span><span>;
</span><span>10</span>                         } <span>else</span> <span>if</span><span> (print_blank) {
</span><span>11</span>                             prefix_char = <span>'</span> <span>'</span><span>;
</span><span>12</span> <span>                        }
</span><span>13</span> <span>                    }
</span><span>14</span>                     <span>break</span>;

로그인 후 복사

num_buf[NUM_BUF_SIZE] 这个东西长度为2048，不知道为什么要分配这么多的内存<br><br>上面的判断就是看看是不是负数，然后就 赋给修饰符<br><br>

로그인 후 복사

	　　　　　　　　if (prefix_char != NUL) {
				*--s = prefix_char;
				s_len++;
			}

로그인 후 복사

应该很容易吧

下面来看下一层的调用

<span> 1</span> <span>while</span> (zend_hash_get_current_data_ex(Z_ARRVAL_P(arr), (<span>void</span> **) &tmp, &pos) ==<span> SUCCESS) {
</span><span> 2</span>         <span>switch</span> ((*tmp)-><span>type) {
</span><span> 3</span>             <span>case</span><span> IS_STRING:
</span><span> 4</span>                 smart_str_appendl(&<span>implstr, Z_STRVAL_PP(tmp), Z_STRLEN_PP(tmp));
</span><span> 5</span>                 <span>break</span><span>;
</span><span> 6</span> 
<span> 7</span>             <span>case</span><span> IS_LONG: {
</span><span> 8</span>                 <span>char</span> stmp[MAX_LENGTH_OF_LONG + <span>1</span><span>];
</span><span> 9</span>                 str_len = slprintf(stmp, <span>sizeof</span>(stmp), <span>"</span><span>%ld</span><span>"</span><span>, Z_LVAL_PP(tmp));
</span><span>10</span>                 smart_str_appendl(&<span>implstr, stmp, str_len);
</span><span>11</span> <span>            }
</span><span>12</span>                 <span>break</span><span>;
</span><span>13</span> 
<span>14</span>             <span>case</span><span> IS_BOOL:
</span><span>15</span>                 <span>if</span> (Z_LVAL_PP(tmp) == <span>1</span><span>) {
</span><span>16</span>                     smart_str_appendl(&implstr, <span>"</span><span>1</span><span>"</span>, <span>sizeof</span>(<span>"</span><span>1</span><span>"</span>)-<span>1</span><span>);
</span><span>17</span> <span>                }
</span><span>18</span>                 <span>break</span><span>;
</span><span>19</span> 
<span>20</span>             <span>case</span><span> IS_NULL:
</span><span>21</span>                 <span>break</span><span>;
</span><span>22</span> 
<span>23</span>             <span>case</span><span> IS_DOUBLE: {
</span><span>24</span>                 <span>char</span> *<span>stmp;
</span><span>25</span>                 str_len = spprintf(&stmp, <span>0</span>, <span>"</span><span>%.*G</span><span>"</span>, (<span>int</span><span>) EG(precision), Z_DVAL_PP(tmp));
</span><span>26</span>                 smart_str_appendl(&<span>implstr, stmp, str_len);
</span><span>27</span> <span>                efree(stmp);
</span><span>28</span> <span>            }
</span><span>29</span>                 <span>break</span><span>;
</span><span>30</span> 
<span>31</span>             <span>case</span><span> IS_OBJECT: {
</span><span>32</span>                 <span>int</span><span> copy;
</span><span>33</span> <span>                zval expr;
</span><span>34</span>                 zend_make_printable_zval(*tmp, &expr, &<span>copy);
</span><span>35</span>                 smart_str_appendl(&<span>implstr, Z_STRVAL(expr), Z_STRLEN(expr));
</span><span>36</span>                 <span>if</span><span> (copy) {
</span><span>37</span>                     zval_dtor(&<span>expr);
</span><span>38</span> <span>                }
</span><span>39</span> <span>            }
</span><span>40</span>                 <span>break</span><span>;
</span><span>41</span> 
<span>42</span>             <span>default</span><span>:
</span><span>43</span>                 tmp_val = **<span>tmp;
</span><span>44</span>                 zval_copy_ctor(&<span>tmp_val);
</span><span>45</span>                 convert_to_string(&<span>tmp_val);
</span><span>46</span>                 smart_str_appendl(&<span>implstr, Z_STRVAL(tmp_val), Z_STRLEN(tmp_val));
</span><span>47</span>                 zval_dtor(&<span>tmp_val);
</span><span>48</span>                 <span>break</span><span>;
</span><span>49</span> 
<span>50</span>         }

로그인 후 복사

我们是在这段代码

case IS_LONG: {
char stmp[MAX_LENGTH_OF_LONG + 1];
str_len = slprintf(stmp, sizeof(stmp), "%ld", Z_LVAL_PP(tmp));
smart_str_appendl(&implstr, stmp, str_len);
}

+ &implstr 0x00c3ef04 {c=0x030e0100 "1.5-" len=4 a=78 } smart_str *
+ stmp 0x00c3ee90 "-278" char [12]
str_len 4 int
+ tmp 0x030e0924 _zval_struct * *

php数字默认类型是长整形的，从上面可知道，stmp="-278",strlen = 4,

我们接下来看看implstr是这样处理的，首先他的结构是这样的

<span>1</span> typedef <span>struct</span><span> {
</span><span>2</span>     <span>char</span> *<span>c; 指向一段内存
</span><span>3</span> <span>    size_t len; 已经用了多小
</span><span>4</span> <span>    size_t a; 总共有多小
</span><span>5</span> } smart_str;

로그인 후 복사

smart_str_appendl 的定义是这样的<br><br>

로그인 후 복사

<span>1</span> <span>#define</span> smart_str_appendl_ex(dest, src, nlen, what) do {            \
<span>2</span> <span>    register size_t __nl;                                            \
</span><span>3</span>     smart_str *__dest = (smart_str *<span>) (dest);                        \
</span><span>4</span> <span>                                                                    \
</span><span>5</span> <span>    smart_str_alloc4(__dest, (nlen), (what), __nl);                    \
</span><span>6</span>     memcpy(__dest->c + __dest-><span>len, (src), (nlen));                    \
</span><span>7</span>     __dest->len =<span> __nl;                                                \
</span><span>8</span> } <span>while</span> (<span>0</span>)

로그인 후 복사

复制字符串用了memcpy

smart_str_alloc4这个定义如下

<span> 1</span> <span>#define</span> smart_str_alloc4(d, n, what, newlen) do {                    \
<span> 2</span>     <span>if</span> (!(d)-><span>c) {                                                    \
</span><span> 3</span>         (d)->len = <span>0</span><span>;                                                \
</span><span> 4</span>         newlen =<span> (n);                                                \
</span><span> 5</span>         (d)->a = newlen  SMART_STR_START_SIZE                         \
<span> 6</span>                 ?<span> SMART_STR_START_SIZE                                 \
</span><span> 7</span>                 : newlen +<span> SMART_STR_PREALLOC;                        \
</span><span> 8</span> <span>        SMART_STR_DO_REALLOC(d, what);                                \
</span><span> 9</span>     } <span>else</span><span> {                                                        \
</span><span>10</span>         newlen = (d)->len +<span> (n);                                    \
</span><span>11</span>         <span>if</span> (newlen >= (d)-><span>a) {                                        \
</span><span>12</span>             (d)->a = newlen +<span> SMART_STR_PREALLOC;                    \
</span><span>13</span> <span>            SMART_STR_DO_REALLOC(d, what);                            \
</span><span>14</span> <span>        }                                                            \
</span><span>15</span> <span>    }                                                                \
</span><span>16</span> } <span>while</span> (<span>0</span>)

로그인 후 복사

这个很清楚啦流程啦

如果implstr 没有分配过的，那么闲分配一段内存

如果implstr分配过，并且当前的空间不够容纳新的字符在这基础上扩展啦 SMART_STR_PREALLOC =78 不知道为什么是78

看下定义吧

1 #define SMART_STR_DO_REALLOC(d, what) \

2 (d)->c = SMART_STR_REALLOC((d)->c, (d)->a + 1, (what))

1 #define SMART_STR_REALLOC(a,b,c) perealloc((a),(b),(c))

1 #define perealloc(ptr, size, persistent) ((persistent)?__zend_realloc((ptr), (size)):erealloc((ptr), (size)))

<span>1</span> inline <span>static</span> <span>void</span> * __zend_realloc(<span>void</span> *<span>p, size_t len)
</span><span>2</span> <span>{
</span><span>3</span>     p =<span> realloc(p, len);
</span><span>4</span>     <span>if</span><span> (p) {
</span><span>5</span>         <span>return</span><span> p;
</span><span>6</span> <span>    }
</span><span>7</span>     fprintf(stderr, <span>"</span><span>Out of memory\n</span><span>"</span><span>);
</span><span>8</span>     exit(<span>1</span><span>);
</span><span>9</span> }

로그인 후 복사

最终是调用了 c语言的realloc函数，这样就大概明白了吧

到最后加上分割符号

1 if (++i != numelems) {

2 smart_str_appendl(&implstr, Z_STRVAL_P(delim), Z_STRLEN_P(delim));

3 }

上面是分析了整数的implode，至于浮点数，对象，字符串的implode大家可以用上面的方法去研究下