php中parse_url函数的源码及分析(scheme部分)
前言
看师傅们的文章时发现,parse_url出现的次数较多，单纯parse_url解析漏洞的考题也有很多，在此研究一下源码(太菜了看不懂，待日后再补充Orz)
源码
在ext/standard/url.c文件中
PHPAPI php_url *php_url_parse_ex(char const *str, size_t length)
{
char port_buf[6];
php_url *ret = ecalloc(1, sizeof(php_url));
char const *s, *e, *p, *pp, *ue;
s = str;
ue = s + length;
/* parse scheme */
if ((e = memchr(s, ':', length)) && e != s) {
/* validate scheme */
p = s;
while (p < e) {
/* scheme = 1*[ lowalpha | digit | "+" | "-" | "." ] */
if (!isalpha(*p) && !isdigit(*p) && *p != '+' && *p != '.' && *p != '-') {
if (e + 1 < ue && e < s + strcspn(s, "?#")) {
goto parse_port;
} else if (s + 1 < ue && *s == '/' && *(s + 1) == '/') { /* relative-scheme URL */
s += 2;
e = 0;
goto parse_host;
} else {
goto just_path;
}
}
p++;
}
if (e + 1 == ue) { /* only scheme is available */
ret->scheme = estrndup(s, (e - s));
php_replace_controlchars_ex(ret->scheme, (e - s));
return ret;
}
/*
* certain schemas like mailto: and zlib: may not have any / after them
* this check ensures we support those.
*/
if (*(e+1) != '/') {
/* check if the data we get is a port this allows us to
* correctly parse things like a.com:80
*/
p = e + 1;
while (p < ue && isdigit(*p)) {
p++;
}
if ((p == ue || *p == '/') && (p - e) < 7) {
goto parse_port;
}
ret->scheme = estrndup(s, (e-s));
php_replace_controlchars_ex(ret->scheme, (e - s));
s = e + 1;
goto just_path;
} else {
ret->scheme = estrndup(s, (e-s));
php_replace_controlchars_ex(ret->scheme, (e - s));
if (e + 2 < ue && *(e + 2) == '/') {
s = e + 3;
if (!strncasecmp("file", ret->scheme, sizeof("file"))) {
if (e + 3 < ue && *(e + 3) == '/') {
/* support windows drive letters as in:
file:///c:/somedir/file.txt
*/
if (e + 5 < ue && *(e + 5) == ':') {
s = e + 4;
}
goto just_path;
}
}
} else {
s = e + 1;
goto just_path;
}
}
} else if (e) { /* no scheme; starts with colon: look for port */
parse_port:
p = e + 1;
pp = p;
while (pp < ue && pp - p < 6 && isdigit(*pp)) {
pp++;
}
if (pp - p > 0 && pp - p < 6 && (pp == ue || *pp == '/')) {
zend_long port;
memcpy(port_buf, p, (pp - p));
port_buf[pp - p] = '\0';
port = ZEND_STRTOL(port_buf, NULL, 10);
if (port > 0 && port <= 65535) {
ret->port = (unsigned short) port;
if (s + 1 < ue && *s == '/' && *(s + 1) == '/') { /* relative-scheme URL */
s += 2;
}
} else {
if (ret->scheme) efree(ret->scheme);
efree(ret);
return NULL;
}
} else if (p == pp && pp == ue) {
if (ret->scheme) efree(ret->scheme);
efree(ret);
return NULL;
} else if (s + 1 < ue && *s == '/' && *(s + 1) == '/') { /* relative-scheme URL */
s += 2;
} else {
goto just_path;
}
} else if (s + 1 < ue && *s == '/' && *(s + 1) == '/') { /* relative-scheme URL */
s += 2;
} else {
goto just_path;
}
parse_host:
/* Binary-safe strcspn(s, "/?#") */
e = ue;
if ((p = memchr(s, '/', e - s))) {
e = p;
}
if ((p = memchr(s, '?', e - s))) {
e = p;
}
if ((p = memchr(s, '#', e - s))) {
e = p;
}
/* check for login and password */
if ((p = zend_memrchr(s, '@', (e-s)))) {
if ((pp = memchr(s, ':', (p-s)))) {
ret->user = estrndup(s, (pp-s));
php_replace_controlchars_ex(ret->user, (pp - s));
pp++;
ret->pass = estrndup(pp, (p-pp));
php_replace_controlchars_ex(ret->pass, (p-pp));
} else {
ret->user = estrndup(s, (p-s));
php_replace_controlchars_ex(ret->user, (p-s));
}
s = p + 1;
}
/* check for port */
if (s < ue && *s == '[' && *(e-1) == ']') {
/* Short circuit portscan,
we're dealing with an
IPv6 embedded address */
p = NULL;
} else {
p = zend_memrchr(s, ':', (e-s));
}
if (p) {
if (!ret->port) {
p++;
if (e-p > 5) { /* port cannot be longer then 5 characters */
if (ret->scheme) efree(ret->scheme);
if (ret->user) efree(ret->user);
if (ret->pass) efree(ret->pass);
efree(ret);
return NULL;
} else if (e - p > 0) {
zend_long port;
memcpy(port_buf, p, (e - p));
port_buf[e - p] = '\0';
port = ZEND_STRTOL(port_buf, NULL, 10);
if (port > 0 && port <= 65535) {
ret->port = (unsigned short)port;
} else {
if (ret->scheme) efree(ret->scheme);
if (ret->user) efree(ret->user);
if (ret->pass) efree(ret->pass);
efree(ret);
return NULL;
}
}
p--;
}
} else {
p = e;
}
/* check if we have a valid host, if we don't reject the string as url */
if ((p-s) < 1) {
if (ret->scheme) efree(ret->scheme);
if (ret->user) efree(ret->user);
if (ret->pass) efree(ret->pass);
efree(ret);
return NULL;
}
ret->host = estrndup(s, (p-s));
php_replace_controlchars_ex(ret->host, (p - s));
if (e == ue) {
return ret;
}
s = e;
just_path:
e = ue;
p = memchr(s, '#', (e - s));
if (p) {
p++;
if (p < e) {
ret->fragment = estrndup(p, (e - p));
php_replace_controlchars_ex(ret->fragment, (e - p));
}
e = p-1;
}
p = memchr(s, '?', (e - s));
if (p) {
p++;
if (p < e) {
ret->query = estrndup(p, (e - p));
php_replace_controlchars_ex(ret->query, (e - p));
}
e = p-1;
}
if (s < e || s == ue) {
ret->path = estrndup(s, (e - s));
php_replace_controlchars_ex(ret->path, (e - s));
}
return ret;
}
/* {{{ proto mixed parse_url(string url, [int url_component])
Parse a URL and return its components */
PHP_FUNCTION(parse_url)
{
char *str;
size_t str_len;
php_url *resource;
zend_long key = -1;
if (zend_parse_parameters(ZEND_NUM_ARGS(), "s|l", &str, &str_len, &key) == FAILURE) {
return;
}
resource = php_url_parse_ex(str, str_len);
if (resource == NULL) {
/* @todo Find a method to determine why php_url_parse_ex() failed */
RETURN_FALSE;
}
if (key > -1) {
switch (key) {
case PHP_URL_SCHEME:
if (resource->scheme != NULL) RETVAL_STRING(resource->scheme);
break;
case PHP_URL_HOST:
if (resource->host != NULL) RETVAL_STRING(resource->host);
break;
case PHP_URL_PORT:
if (resource->port != 0) RETVAL_LONG(resource->port);
break;
case PHP_URL_USER:
if (resource->user != NULL) RETVAL_STRING(resource->user);
break;
case PHP_URL_PASS:
if (resource->pass != NULL) RETVAL_STRING(resource->pass);
break;
case PHP_URL_PATH:
if (resource->path != NULL) RETVAL_STRING(resource->path);
break;
case PHP_URL_QUERY:
if (resource->query != NULL) RETVAL_STRING(resource->query);
break;
case PHP_URL_FRAGMENT:
if (resource->fragment != NULL) RETVAL_STRING(resource->fragment);
break;
default:
php_error_docref(NULL, E_WARNING, "Invalid URL component identifier " ZEND_LONG_FMT, key);
RETVAL_FALSE;
}
goto done;
}
/* allocate an array for return */
array_init(return_value);
/* add the various elements to the array */
if (resource->scheme != NULL)
add_assoc_string(return_value, "scheme", resource->scheme);
if (resource->host != NULL)
add_assoc_string(return_value, "host", resource->host);
if (resource->port != 0)
add_assoc_long(return_value, "port", resource->port);
if (resource->user != NULL)
add_assoc_string(return_value, "user", resource->user);
if (resource->pass != NULL)
add_assoc_string(return_value, "pass", resource->pass);
if (resource->path != NULL)
add_assoc_string(return_value, "path", resource->path);
if (resource->query != NULL)
add_assoc_string(return_value, "query", resource->query);
if (resource->fragment != NULL)
add_assoc_string(return_value, "fragment", resource->fragment);
done:
php_url_free(resource);
}
代码中遇到的问题解决
函数定义部分
PHP_FUNCTION(parse_url)
{
char *str;
size_t str_len;
php_url *resource;
zend_long key = -1;
if (zend_parse_parameters(ZEND_NUM_ARGS(), "s|l", &str, &str_len, &key) == FAILURE) {
return;
}
resource = php_url_parse_ex(str, str_len);
if (resource == NULL) {
/* @todo Find a method to determine why php_url_parse_ex() failed */
RETURN_FALSE;
}
引用这篇文章的内容http://www.nowamagic.net/librarys/veda/detail/1467
b Boolean
l Integer 整型
d Floating point 浮点型
s String 字符串
r Resource 资源
a Array 数组
o Object instance 对象
O Object instance of a specified type 特定类型的对象
z Non-specific zval 任意类型
Z zval**类型
f 表示函数、方法名称
那么其中的"s|l"表示parse_url需要两个参数，一个字符串型，一个整型
php_url类型的声明在ext/standard/url.h中
typedef struct php_url {
char *scheme;
char *user;
char *pass;
char *host;
unsigned short port;
char *path;
char *query;
char *fragment;
} php_url;
问题
- parse_url只有两个参数，不知道strlen这个参数哪里去了……？还有他的值到底是怎么获得的……
函数内部实现部分
使用php_url_parse_ex函数来处理我们传过去的url，先暂定str_len为str的长度……
if ((e = memchr(s, ':', length)) && e != s) {
/* validate scheme */
p = s;
while (p < e) {
/* scheme = 1*[ lowalpha | digit | "+" | "-" | "." ] */
if (!isalpha(*p) && !isdigit(*p) && *p != '+' && *p != '.' && *p != '-') {
if (e + 1 < ue && e < s + strcspn(s, "?#")) {
goto parse_port;
} else if (s + 1 < ue && *s == '/' && *(s + 1) == '/') { /* relative-scheme URL */
s += 2;
e = 0;
goto parse_host;
} else {
goto just_path;
}
}
p++;
}
if (e + 1 == ue) { /* only scheme is available */
ret->scheme = estrndup(s, (e - s));
php_replace_controlchars_ex(ret->scheme, (e - s));
return ret;
}
/*
* certain schemas like mailto: and zlib: may not have any / after them
* this check ensures we support those.
*/
if (*(e+1) != '/') {
/* check if the data we get is a port this allows us to
* correctly parse things like a.com:80
*/
p = e + 1;
while (p < ue && isdigit(*p)) {
p++;
}
if ((p == ue || *p == '/') && (p - e) < 7) {
goto parse_port;
}
ret->scheme = estrndup(s, (e-s));
php_replace_controlchars_ex(ret->scheme, (e - s));
s = e + 1;
goto just_path;
} else {
ret->scheme = estrndup(s, (e-s));
php_replace_controlchars_ex(ret->scheme, (e - s));
if (e + 2 < ue && *(e + 2) == '/') {
s = e + 3;
if (!strncasecmp("file", ret->scheme, sizeof("file"))) {
if (e + 3 < ue && *(e + 3) == '/') {
/* support windows drive letters as in:
file:///c:/somedir/file.txt
*/
if (e + 5 < ue && *(e + 5) == ':') {
s = e + 4;
}
goto just_path;
}
}
} else {
s = e + 1;
goto just_path;
}
}
} else if (e) { /* no scheme; starts with colon: look for port */
如果s中含有冒号则e指向冒号
且同时如果冒号不在s的开头，p指向s
当p不指向冒号向循环，p指向下一位
如果p指向的值是字母或者数字或者是+,-,.则指针指向下一位，这就代表冒号前面的值其实是任意的字母、数字、+、-、.
如果冒号所在位置小于str，且?#在冒号后面(如果有的话)，就跳转到port解析部分
如果str的长度大于1且str的前两个字符是//，s指向//后面的一个字符，e变为0，跳转到host解析
如果冒号是最后一位字符，则冒号前面的东西会当作scheme返回
如果冒号后面不是/，则p指向冒号后面一位
当p小于str且p指向的为数字字符，p一直指向后一位，直到p指向str末尾或者p指向的字符为/，同时冒号后面的数字位数小于6位，跳转到port解析
如果冒号后面不是纯数字或数字后面有一个/，那么冒号前面的内容就当作scheme，放在ret的scheme参数中，s指向冒号后一位，跳转到path解析
如果冒号后面是/，那么冒号前面的内容就当作scheme，放在ret的scheme参数中。如果下面一位也是/，那么s指向//后面一位，如果scheme为file，那么判断接下来一位是不是/，如果是，判断冒号后是否有五个字符，如果有那么第五个字符是不是冒号(为了处理file:///c:)，s指向///后的一位字符，跳转到path解析
如果冒号后面不是三个/，s指向冒号后面一位，之后跳转到path解析
如果冒号在str开头，那么进行port解析
姿势
- 只要请求的url里不含有冒号(:)就会被当成path解析
