code_app/sdk/webservice/HttpHead.cpp

#include "HttpHead.h"

const unsigned int asciiTableData[256] =
{
	0x004, 0x004, 0x004, 0x004, 0x004, 0x004, 0x004, 0x004,
	0x004, 0x104, 0x104, 0x004, 0x104, 0x104, 0x004, 0x004,
	0x004, 0x004, 0x004, 0x004, 0x004, 0x004, 0x004, 0x004,
	0x004, 0x004, 0x004, 0x004, 0x004, 0x004, 0x004, 0x004,
	0x140, 0x0d0, 0x0d0, 0x0d0, 0x0d0, 0x0d0, 0x0d0, 0x0d0,
	0x0d0, 0x0d0, 0x0d0, 0x0d0, 0x0d0, 0x0d0, 0x0d0, 0x0d0,
	0x459, 0x459, 0x459, 0x459, 0x459, 0x459, 0x459, 0x459,
	0x459, 0x459, 0x0d0, 0x0d0, 0x0d0, 0x0d0, 0x0d0, 0x0d0,
	0x0d0, 0x653, 0x653, 0x653, 0x653, 0x653, 0x653, 0x253,
	0x253, 0x253, 0x253, 0x253, 0x253, 0x253, 0x253, 0x253,
	0x253, 0x253, 0x253, 0x253, 0x253, 0x253, 0x253, 0x253,
	0x253, 0x253, 0x253, 0x0d0, 0x0d0, 0x0d0, 0x0d0, 0x0d0,
	0x0d0, 0x473, 0x473, 0x473, 0x473, 0x473, 0x473, 0x073,
	0x073, 0x073, 0x073, 0x073, 0x073, 0x073, 0x073, 0x073,
	0x073, 0x073, 0x073, 0x073, 0x073, 0x073, 0x073, 0x073,
	0x073, 0x073, 0x073, 0x0d0, 0x0d0, 0x0d0, 0x0d0, 0x004
	/* the upper 128 are all zeroes */
};

static void TrimString(std::string& str)
{
	std::string str1;
	bool add1 = false;
	std::string::const_iterator iter1;
	for (iter1 = str.begin(); iter1 != str.end(); ++iter1)
	{
		int c = (HGByte)(*iter1);
		if (!add1)
		{
			if (!isspace(c))
			{
				str1.push_back(c);
				add1 = true;
			}
		}
		else
		{
			str1.push_back(c);
		}
	}

	if (str1.empty())
	{
		str.clear();
		return;
	}

	std::string str2;
	bool add2 = false;
	std::string::const_reverse_iterator iter2;
	for (iter2 = str1.rbegin(); iter2 != str1.rend(); ++iter2)
	{
		int c = (HGByte)(*iter2);
		if (!add2)
		{
			if (!isspace(c))
			{
				str2.push_back(c);
				add2 = true;
			}
		}
		else
		{
			str2.push_back(c);
		}
	}

	if (str2.empty())
	{
		str.clear();
		return;
	}

	str = std::string(str2.rbegin(), str2.rend());
}

HttpHead::HttpHead()
{

}

HttpHead::~HttpHead()
{

}

bool HttpHead::Parse(const std::string& head)
{
	AnalysisHead(head, m_requestMethod, m_requestURIPath, m_requestURIQueryInfos,
		m_requestURIFragment, m_requestHttpVersion, m_headInfos);
	return true;
}

void HttpHead::Clear()
{
	m_requestMethod.clear();
	m_requestURIPath.clear();
	m_requestURIQueryInfos.clear();
	m_requestURIFragment.clear();
	m_requestHttpVersion.clear();
	m_headInfos.clear();
}

std::string HttpHead::GetRequestMethod() const
{
	return m_requestMethod;
}

std::string HttpHead::GetRequestURIPath() const
{
	return m_requestURIPath;
}

HttpPairs HttpHead::GetRequestURIQueryInfos() const
{
	return m_requestURIQueryInfos;
}

std::string HttpHead::GetRequestURIFragment() const
{
	return m_requestURIFragment;
}

std::string HttpHead::GetRequestHttpVersion() const
{
	return m_requestHttpVersion;
}

HttpPairs HttpHead::GetHeadInfos() const
{
	return m_headInfos;
}

int HttpHead::GetContentLength() const
{
	int len = 0;
	for (int i = 0; i < (int)m_headInfos.size(); ++i)
	{
#if defined(HG_CMP_MSC)
		if (0 == _stricmp("Content-Length", m_headInfos[i].first.c_str()))
#else
		if (0 == strcasecmp("Content-Length", m_headInfos[i].first.c_str()))
#endif
		{
			len = atoi(m_headInfos[i].second.c_str());
			break;
		}
	}

	return len;
}

std::string HttpHead::GetContentType() const
{
	std::string type;
	for (int i = 0; i < (int)m_headInfos.size(); ++i)
	{
#if defined(HG_CMP_MSC)
		if (0 == _stricmp("Content-Type", m_headInfos[i].first.c_str()))
#else
		if (0 == strcasecmp("Content-Type", m_headInfos[i].first.c_str()))
#endif
		{
			type = m_headInfos[i].second.c_str();
			break;
		}
	}

	return type;
}

std::string HttpHead::GetValue(const HttpPairs& infos, const std::string& key)
{
	std::string value;
	for (int i = 0; i < (int)infos.size(); ++i)
	{
		if (key == infos[i].first)
		{
			value = infos[i].second;
			break;
		}
	}

	return value;
}

void HttpHead::AnalysisURIQuery(const std::string& query, HttpPairs& queryInfos)
{
	std::vector<std::string> queryList;

	char* p = new char[query.size() + 1];
	strcpy(p, query.c_str());
	char* pStr = strtok(p, "&");
	if (NULL != pStr)
		queryList.push_back(pStr);
	while (1)
	{
		pStr = strtok(NULL, "&");
		if (NULL == pStr)
			break;
		queryList.push_back(pStr);
	}
	delete[] p;

	queryInfos.clear();
	for (int i = 0; i < (int)queryList.size(); ++i)
	{
		p = new char[queryList[i].size() + 1];
		strcpy(p, queryList[i].c_str());

		std::pair <std::string, std::string> pr;
		pStr = strtok(p, "=");
		if (NULL != pStr)
			pr.first = AnalyURIString(pStr);
		pStr = strtok(NULL, "=");
		if (NULL != pStr)
			pr.second = AnalyURIString(pStr);

		queryInfos.push_back(pr);
		delete[] p;
	}
}

void HttpHead::AnalysisURI(const std::string& uri, std::string& path, HttpPairs& queryInfos, std::string& fragment)
{
	size_t pathPos = uri.find('/');
	size_t queryPos = uri.find('?');
	size_t fragmentPos = uri.find('#');

	path.clear();
	if (std::string::npos != pathPos)
	{
		size_t count = std::string::npos;
		if (queryPos != std::string::npos)
		{
			assert(queryPos > pathPos);
			count = queryPos - pathPos;
		}
		else if (fragmentPos != std::string::npos)
		{
			assert(fragmentPos > pathPos);
			count = fragmentPos - pathPos;
		}

		path = AnalyURIString(uri.substr(pathPos, count));
	}

	queryInfos.clear();
	if (std::string::npos != queryPos)
	{
		size_t count = std::string::npos;
		if (fragmentPos != std::string::npos)
		{
			assert(fragmentPos > queryPos);
			count = fragmentPos - queryPos;
		}

		std::string query = uri.substr(queryPos + 1, count - 1);
		AnalysisURIQuery(query, queryInfos);
	}

	fragment.clear();
	if (std::string::npos != fragmentPos)
	{
		fragment = AnalyURIString(uri.substr(fragmentPos + 1));
	}
}

void HttpHead::AnalysisHead(const std::string& head, std::string& requestMethod, std::string& requestURIPath,
	HttpPairs& requestURIQueryInfos, std::string& requestURIFragment, std::string& httpVersion, HttpPairs& headInfos)
{
	requestMethod.clear();
	requestURIPath.clear();
	requestURIQueryInfos.clear();
	requestURIFragment.clear();
	httpVersion.clear();
	headInfos.clear();

	std::vector<std::string> headList;

	char* p = new char[head.size() + 1];
	strcpy(p, head.c_str());
	char* pStr = strtok(p, "\r\n");
	if (NULL != pStr)
		headList.push_back(pStr);
	while (1)
	{
		pStr = strtok(NULL, "\r\n");
		if (NULL == pStr)
			break;
		headList.push_back(pStr);
	}
	delete[] p;

	if (headList.size() < 1)
	{
		return;
	}

	std::string requestURI;

	// 解析请求行
	p = new char[headList[0].size() + 1];
	strcpy(p, headList[0].c_str());
	pStr = strtok(p, " ");
	if (NULL != pStr)
		requestMethod = pStr;
	pStr = strtok(NULL, " ");
	if (NULL != pStr)
		requestURI = pStr;
	pStr = strtok(NULL, " ");
	if (NULL != pStr)
		httpVersion = pStr;
	delete[] p;

	// 解析URI
	AnalysisURI(requestURI, requestURIPath, requestURIQueryInfos, requestURIFragment);

	// 解析请求头
	for (int i = 1; i < (int)headList.size(); ++i)
	{
		p = new char[headList[i].size() + 1];
		strcpy(p, headList[i].c_str());

		std::pair <std::string, std::string> pr;
		pStr = strtok(p, ":");
		if (NULL != pStr)
			pr.first = pStr;
		pStr = strtok(NULL, ":");
		if (NULL != pStr)
			pr.second = pStr;

		TrimString(pr.first);
		TrimString(pr.second);
		headInfos.push_back(pr);

		delete[] p;
	}
}

/*判断ascii码是否是数字0-9*/
static bool asciiIsDigit(char c)
{
	/*字符的ascii码&8 结果为0-127，则是数字*/
	return asciiTableData[(unsigned char)c & (1 << 3)];
}

static int asciiDigitValue(char c)
{
	if (asciiIsDigit(c))
		return c - '0';
	return -1;
}

static int asciiXdigitValue(char c)
{
	//printf("-->%c\n",c);
	if (c >= 'A' && c <= 'F')
		return c - 'A' + 10;//(A B C D E F)->(10 11 12 13 14 15)
	if (c >= 'a' && c <= 'f')
		return c - 'a' + 10;

	return asciiDigitValue(c);//('0'...'9')->(0...9)
}

static int unescapeCharacter(const char* scanner)
{
	int first = asciiXdigitValue(scanner[0]);
	if (first < 0)
		return -1;

	int second = asciiXdigitValue(scanner[1]);
	if (second < 0)
		return -1;

	return (first << 4) | second; //== (first*16 | second) == (first*16 + second)
}

static char* unescapeUriString(const char* uriString, bool asciiEscape)
{
	if (NULL == uriString)
		return NULL;

	int strLen = (int)strlen(uriString);
	char* result = (char*)malloc(strLen + 1);//可推测解码后的长度<=原长度
	char* out = result;

	const char* in, * end;
	for (in = uriString, end = in + strLen; in < end; ++in)
	{
		int c = *in;

		//遇到了'%'才去解析
		if ('%' == c)
		{
			if (in + 3 > end)
				break;
			//获取%后2个字符的解码值
			c = unescapeCharacter(in + 1);
			if (c <= 0)
				break;

			if (asciiEscape && c <= 0x7F)
				break;

			in += 2;//一般的格式为%后加两个ascii码字符
		}

		*out++ = c;//存储转义结果
	}

	*out = '\0';

	if (in != end)
	{
		free(result);
		return NULL;
	}

	return result;
}

std::string HttpHead::AnalyURIString(const std::string& str)
{
	std::string ret;
	char* decodeStr = unescapeUriString(str.c_str(), false);
	if (NULL != decodeStr)
	{
		ret = decodeStr;
		free(decodeStr);
	}

	return ret;
}