code_scanner/common/referer.cpp

#include "referer.h"

#include <iconv.h>

#include "log_util.h"


/////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
// refer
refer::refer() : ref_(1)
{
	on_born();
}
refer::~refer()
{
	on_dead();
}

void refer::on_born(void)
{}
void refer::on_dead(void)
{}

int32_t refer::add_ref(void)
{
	LOCKER lock(mutex_);

	return ++ref_;
}
int32_t refer::release(void)
{
	int32_t ref = 0;

	{
		LOCKER lock(mutex_);
		ref = --ref_;
	}

	if (ref == 0)
		delete this;

	return ref;
}











/////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
// chronograph
#include <sys/time.h>


chronograph::chronograph()
{
	reset();
}
chronograph::~chronograph()
{}

bool chronograph::now(struct timeval* tv)
{
	struct timezone	tz = { 0 };

	return gettimeofday(tv, &tz) == 0;
}
bool chronograph::now(uint64_t* seconds, uint64_t* u_seconds)
{
	struct timeval	tv = { 0 };
	struct timezone	tz = { 0 };

	if (gettimeofday(&tv, &tz) == 0)
	{
		if (seconds)
			*seconds = tv.tv_sec;
		if (u_seconds)
			*u_seconds = tv.tv_usec;

		return true;
	}
	else
	{
		return false;
	}
}
std::string chronograph::now(bool with_ms/*whether with milliseconds*/)            // return '2022-11-30 10:38:42.123', no '.123' if with_ms was false
{
	struct timeval	tv = { 0 };

	if (!chronograph::now(&tv))
		return "";

	char	buf[40] = { 0 };
	time_t	t = tv.tv_sec;
	struct tm* l = localtime(&t);

	if (with_ms)
		sprintf(buf, "%04d-%02d-%02d %02d:%02d:%02d.%06d", l->tm_year + 1900, l->tm_mon + 1, l->tm_mday
			, l->tm_hour, l->tm_min, l->tm_sec, tv.tv_usec);
	else
		sprintf(buf, "%04d-%02d-%02d %02d:%02d:%02d", l->tm_year + 1900, l->tm_mon + 1, l->tm_mday
			, l->tm_hour, l->tm_min, l->tm_sec);

	return buf;
}

uint64_t chronograph::elapse_s(void)
{
	struct timeval	tv = { 0 };

	chronograph::now(&tv);

	return tv.tv_sec - bgn_.tv_sec;
}
uint64_t chronograph::elapse_ms(void)
{
	struct timeval	tv = { 0 };
	uint64_t		dif = 0;

	chronograph::now(&tv);
	dif = SEC_2_MS(tv.tv_sec - bgn_.tv_sec);
	dif += tv.tv_usec / MSEC_2_US(1);
	dif -= bgn_.tv_usec / MSEC_2_US(1);

	return dif;
}
uint64_t chronograph::elapse_us(void)
{
	struct timeval	tv = { 0 };
	uint64_t		dif = 0;

	chronograph::now(&tv);
	dif = SEC_2_US(tv.tv_sec - bgn_.tv_sec);
	dif += tv.tv_usec;
	dif -= bgn_.tv_usec;

	return dif;
}
void chronograph::reset()
{
	chronograph::now(&bgn_);
}





/////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
// sys utility
#include <dirent.h>
#include <unistd.h>
#include <sys/stat.h>
#include <string.h>

namespace sys_util
{
	static bool find_module(const char* path, bool is_dir, void* param)
	{
		std::string* para = (std::string*)param;

		if (para[0].empty())
		{
			const char* now = getenv("PWD");
			bool found = strstr(path, now) == path;

			if(found)
			{
				found = path[strlen(now)] == '/' && strstr(path + strlen(now) + 1, "/") == nullptr;
			}
			if(found || para[1].empty())
				para[1] = path;

			return !found;
		}
		else
		{
			const char* name = strrchr(path, '/');
			if (name++ == nullptr)
				name = path;
			if (strstr(name, para[0].c_str()))
			{
				para[1] = path;

				return false;
			}

			return true;
		}
	}
	typedef struct _enum_proc_cb
	{
		bool process;
		bool(*on_found)(uint64_t pid, const char* path_name, void* param);
		void* param;
	}ENPROCCB, * LPENPROCCB;
	static bool found_process(const char* path, bool is_dir, void* param)
	{
		LPENPROCCB	cb = (LPENPROCCB)param;
		const char* id = strrchr(path, '/');
		uint64_t 	pid = 0;
		std::string path_name("");

		if (id++ == nullptr)
		{
			id = path;
		}
		while (*id)
		{
			if (*id < '0' || *id > '9')
				break;

			pid *= 10;
			pid += *id - '0';
			id++;
		}
		if (*id || pid == 0)
			return true;

		if (cb->process)
		{
			path_name = get_module_path(nullptr, pid);
			id = path_name.c_str();
		}
		else
		{
			// get start address of pid to path_name
			id = 0;
		}

		return cb->on_found(pid, id, cb->param);
	}
	static bool on_stack_line_read(char* line, void* param)
	{
		LPENPROCCB cb = (LPENPROCCB)param;
		std::string m(line);
		uint64_t 	off = 0;
		size_t		pos = m.find("]");

		if (pos++ != std::string::npos)
			m.erase(0, pos);
		trim_left(m);
		pos = m.find("+0x");
		if (pos != std::string::npos)
		{
			off = from_hex_str(m.c_str() + pos + 3);
			m.erase(pos);
		}

		return cb->on_found(off, m.c_str(), cb->param);
	}

	int32_t enum_modules(bool(*on_found)(const char* path_module_name, bool is_dir, void* param),// return false to stop enumeratin
		void* param,                                                            // user defined data, passed into callback on_found
		unsigned pid                                                            // process id, -1 is self
	)                                                                           // return errno
	{
		char	path[128] = { 0 };

		if (pid == -1)
			pid = getpid();
		sprintf(path, "/proc/%u/map_files/", pid);

		return enum_files(path, on_found, param, false);
	}

	int32_t enum_files(const char* dir,                                             // dir path
		bool(*on_found)(const char* path_name, bool is_dir, void* param),                    // return false to stop enumeratin
		void* param                                                             // user defined data, passed into callback on_found
		, bool recursive
	)                                                                           // return errno
	{
		int32_t	 ret = 0;
		DIR* pdir = nullptr;
		struct dirent* ent = nullptr;

		pdir = opendir(dir);
		if (!pdir)
			return errno;

		while ((ent = readdir(pdir)))
		{
			if (strcmp(ent->d_name, ".") == 0 || strcmp(ent->d_name, "..") == 0)
				continue;

			std::string file(dir);

			file += "/";
			file += ent->d_name;
			if ((ent->d_type & DT_DIR) == 0)
			{
				file = read_link(file.c_str());
				if(file.empty())
					file = std::string(dir) + "/" + ent->d_name;
			}
			if (!on_found(file.c_str(), ent->d_type & DT_DIR, param))
			{
				ret = 0x5e17;
				break;
			}

			if ((ent->d_type & DT_DIR) && recursive)
			{
				std::string	sub(dir);
				sub += "/";
				sub += ent->d_name;
				ret = enum_files(sub.c_str(), on_found, param, recursive);
				if (ret == 0x5e17)
					break;
			}
		}
		closedir(pdir);

		return ret == 0x5e17 ? 0 : ret;
	}

	int32_t enum_processes(bool(*on_found)(uint64_t pid, const char* path_name, void* param), void* param)
	{
		ENPROCCB cb;

		cb.on_found = on_found;
		cb.param = param;
		cb.process = true;

		return enum_files("/proc", found_process, &cb, false);
	}
	uint32_t enum_threads(uint64_t pid, bool(*on_found)(uint64_t tid, void* start_addr, void* param), void* param)
	{
		ENPROCCB	cb;

		cb.process = false;
		cb.param = param;
		*(void**)&cb.on_found = *(void**)&on_found;

		return enum_files(("/proc/" + std::to_string(pid) + "/task").c_str(), found_process, &cb, false);
	}
	uint32_t get_thread_callstack(uint64_t pid, uint64_t tid, bool(*on_found)(uint64_t off, const char* module, void* param), void* param)
	{
		ENPROCCB	cb;

		cb.process = false;
		cb.param = param;
		*(void**)&cb.on_found = *(void**)&on_found;

		return read_line(("/proc/" + std::to_string(pid) + "/task/" + std::to_string(tid) + "/stack").c_str(), on_stack_line_read, &cb);
	}
	uint32_t read_line(const char* file, bool(*on_line)(char* line, void* param), void* param)
	{
		FILE* src = fopen(file, "rb");
		int		err = 0;

		if (!src)
			return errno;

		size_t ll = 1024;
		char* line = new char[ll];
		while (fgets(line, ll - 1, src))
		{
			if (!on_line(line, param))
				break;

			memset(line, 0, ll);
		}
		err = errno;
		delete[] line;
		fclose(src);

		return err;
	}

	std::string get_module_path(const char* module_name, unsigned pid)			// get module full path, nullptr is for main-exe
	{
		std::string	param[] = { module_name ? module_name : "", "" };

		enum_modules(find_module, param, pid);

		return param[1];
	}
	std::string read_link(const char* lnk)
	{
		char	path[512] = { 0 };

		readlink(lnk, path, sizeof(path) - 1);

		return path;
	}
	size_t get_page_size(void)
	{
		size_t size = sysconf(_SC_PAGESIZE);

		if (size < 1024 || (size & 0x0fe0000ff))	// nKB && < 16MB
			size = getpagesize();

		return size;
	}
	bool create_folder(const char* dir)
	{
		bool	ret = mkdir(dir, S_IREAD | S_IWRITE | S_IEXEC) == 0 || errno == EEXIST;

		if(errno == ENOENT)
		{
			std::string	path(dir), cur("");
			size_t		pos = path.find("/", 1);

			while(pos != std::string::npos)
			{
				ret = mkdir(path.substr(0, pos).c_str(), S_IREAD | S_IWRITE | S_IEXEC) == 0 || errno == EEXIST;
				if(!ret)
				{
					printf("mkdir(%s) = %d(%s)\n", path.substr(0, pos).c_str(), errno, strerror(errno));
					break;
				}
				pos = path.find("/", pos + 1);
			}
			if(ret)
				ret = mkdir(path.c_str(), S_IREAD | S_IWRITE | S_IEXEC) == 0;
		}

		return ret;
	}

	int32_t get_memory_info(uint64_t* total, uint64_t* available)
	{
		if (!total && !available)
			return 0;

		char	line[128] = { 0 };
		FILE* src = fopen("/proc/meminfo", "rb");
		int32_t		count = total && available ? 2 : 1;
		unsigned long val = 0;

		if (!src)
			return log_cls::log_when_err(-1, "fopen('/proc/meminfo', 'rb')", LOG_LEVEL_FATAL);

		while (fgets(line, sizeof(line) - 1, src))
		{
			if (sscanf(line, "MemTotal: %ld", &val))
			{
				if (total)
				{
					*total = val * 1024;
					if (--count == 0)
						break;
				}
			}
			else if (sscanf(line, "MemFree: %ld", &val))
			{
				if (available)
				{
					*available = val * 1024;
					if (--count == 0)
						break;
				}
			}
		}

		fclose(src);

		return 0;
	}
	std::string format_readable_bytes(uint64_t bytes)
	{
		std::string	str("\0", 80);

		if (bytes >= SIZE_GB(1))
		{
			double v = bytes * 1.0f / (SIZE_GB(1));
			size_t	pos = 0;

			sprintf(&str[0], "%.2fGB", v);
			pos = str.find(".");
			while (pos > 3)
			{
				pos -= 3;
				str.insert(pos, ",");
			}
		}
		else if (bytes >= SIZE_MB(1))
		{
			double v = bytes * 1.0f / (SIZE_MB(1));
			sprintf(&str[0], "%.2fMB", v);
		}
		else if (bytes >= SIZE_KB(1))
		{
			double v = bytes * 1.0f / (SIZE_KB(1));
			sprintf(&str[0], "%.2fKB", v);
		}
		else
		{
			sprintf(&str[0], "%uB", (unsigned)bytes);
		}

		return str;
	}
	std::string get_command_output(const char* cmd, uint16_t max_line_len, bool one_line)
	{
		FILE* src = popen(cmd, "r");
		std::string	ret("");

		if (src)
		{
			char* buf = new char[max_line_len + 4];
			if (buf)
			{
				memset(buf, 0, max_line_len + 4);
				fgets(buf, max_line_len, src);
				ret = buf;
				while (!one_line && fgets(buf, max_line_len, src))
					ret += "\n" + std::string(buf);

				delete[] buf;
			}

			pclose(src);
		}
			
		return ret;
	}

	static bool is_char_in(const char* str, char ch)
	{
		if (ch == 0)
			return false;

		while (*str)
		{
			if (*str++ == ch)
				return true;
		}

		return false;
	}
	bool trim_left(std::string& str, const char* space)
	{
		int		off = 0;

		for (; off < str.length(); ++off)
		{
			if (!is_char_in(space, str[off]))
				break;
		}
		if (off)
			str.erase(0, off);

		return off > 0;
	}
	bool trim_right(std::string& str, const char* space)
	{
		int		off = str.length() - 1;

		for (; off >= 0; --off)
		{
			if (!is_char_in(space, str[off]))
				break;
		}
		if (off < str.length() - 1)
		{
			str.erase(off + 1);
			return true;
		}

		return false;
	}
	uint64_t from_hex_str(const char* hex, const char** end)
	{
		uint64_t val = 0;

		for (int i = 0; i < 16; ++i)
		{
			if (*hex >= '0' && *hex <= '9')
			{
				val <<= 4;
				val += *hex - '0';
			}
			else if (*hex >= 'a' && *hex <= 'f')
			{
				val <<= 4;
				val += *hex - 'a' + 10;
			}
			else if (*hex >= 'A' && *hex <= 'F')
			{
				val <<= 4;
				val += *hex - 'A' + 10;
			}
			else
			{
				break;
			}
			hex++;
		}

		if (end)
			*end = hex;

		return val;
	}

	static uint64_t from_dec_str(const char* str, const char** end)
	{
		uint64_t	val = 0;

		for(int i = 0; *str && i < 19; ++i, ++str)
		{
			if(*str >= '0' && *str <= '9')
			{
				val *= 10;
				val += *str - '0';
			}
			else
				break;
		}
		if(end)
			*end = str;
		
		return val;
	}
	static uint64_t from_oct_str(const char* str, const char** end)
	{
		uint64_t	val = 0;

		for(int i = 0; *str && i < 21; ++i, ++str)
		{
			if(*str >= '0' && *str <= '7')
			{
				val *= 8;
				val += *str - '0';
			}
			else
				break;
		}
		if(end)
			*end = str;
		
		return val;
	}
	static uint64_t from_bin_str(const char* str, const char** end)
	{
		uint64_t	val = 0;

		for(int i = 0; *str && i < 64; ++i, ++str)
		{
			if(*str >= '0' && *str <= '1')
			{
				val *= 2;
				val += *str - '0';
			}
			else
				break;
		}
		if(end)
			*end = str;
		
		return val;
	}
	uint64_t to_int(const char* str, const char** end)
	{
		if(!str || *str == 0)
		{
			if(end)
				*end = str;
			
			return 0;
		}

		if(strstr(str, "0x") == str)
			return from_hex_str(str + 2, end);
		
		if(str[strlen(str) - 1] == 'h' || str[strlen(str) - 1] == 'H')
			return from_hex_str(str, end);
		if(str[strlen(str) - 1] == 'o' || str[strlen(str) - 1] == 'O')
			return from_oct_str(str, end);
		if(str[strlen(str) - 1] == 'b' || str[strlen(str) - 1] == 'B')
			return from_bin_str(str, end);
		
		bool hex = false;
		for(int i = 0; str[i]; ++i)
		{
			if(str[i] < '0')
				break;
			if(str[i] > '9')
			{
				if((str[i] >= 'a' && str[i] <= 'f') ||
					(str[i] >= 'A' && str[i] <= 'F'))
					hex = true;
				break;
			}
		}

		return hex ? from_hex_str(str, end) : from_dec_str(str, end);
	}

	/* iconv_open language options:

		Europe:
			ASCII,    ISO-8859-{1,2,3,4,5,7,9,10,13,14,15,16},    KOI8-R,    KOI8-U,   KOI8-RU,
			CP{1250,1251,1252,1253,1254,1257},                                     CP{850,866},
			Mac{Roman,CentralEurope,Iceland,Croatian,Romania},
			Mac{Cyrillic,Ukraine,Greek,Turkish}, Macintosh

       	Semitic:
            ISO-8859-{6,8}, CP{1255,1256}, CP862, Mac{Hebrew,Arabic}

       	Janpanese:
			EUC-JP, SHIFT-JIS, CP932, ISO-2022-JP, ISO-2022-JP-2, ISO-2022-JP-1

       	Chinese:
			EUC-CN,  HZ,  GBK,  GB18030,  EUC-TW,   BIG5,   CP950,   BIG5-HKSCS,   ISO-2022-CN, ISO-2022-CN-EXT

       	Korean:
			EUC-KR, CP949, ISO-2022-KR, JOHAB

       	Armenian:
            ARMSCII-8

       	Georgian:
            Georgian-Academy, Georgian-PS

       	Thai:   
			TIS-620, CP874, MacThai

       	Laos:
			MuleLao-1, CP1133

       	Vietnam:
			VISCII, TCVN, CP1258

       	Special:
            HP-ROMAN8, NEXTSTEP

       	Full Unicode:
			UTF-8
			UCS-2, UCS-2BE, UCS-2LE
			UCS-4, UCS-4BE, UCS-4LE
			UTF-16, UTF-16BE, UTF-16LE
			UTF-32, UTF-32BE, UTF-32LE
			UTF-7
			JAVA
	*/
	std::string transform_between_gbk_and_utf8(const char* in, bool to_utf8, int *err)
	{
		size_t 	len = strlen(in) + 8, ol = len * 2;
		char	*buf = (char*)malloc(len), *oper = buf, *out = nullptr, *oper1 = nullptr;
		iconv_t conv;

		memset(buf, 0, len);
		strcpy(buf, in);
		if(to_utf8)
			conv = iconv_open("UTF-8", "GBK");
		else
			conv = iconv_open("GBK", "UTF-8");
		if(conv == (iconv_t)-1)
		{
			if(err)
				*err = errno;
			
			free(buf);

			return "";
		}

		oper1 = out = (char*)malloc(ol);
		memset(out, 0, ol);
		len -= 8;
		if(iconv(conv, &oper, &len, &oper1, &ol))
		{
			if(err)
				*err = errno;
		}
		else if(err)
			*err = 0;
		
		std::string ret(out);

		free(buf);
		free(out);
		iconv_close(conv);

		return std::move(ret);
	}
	std::string transform_between_utf16_and_utf8(const char* in, size_t bytes, bool to_utf8)
	{
		static const unsigned char firstByteMark[7] = { 0x00, 0x00, 0xC0, 0xE0, 0xF0, 0xF8, 0xFC };
		std::string	ret("");

		if(to_utf8)
		{
			unsigned short *uc = *(unsigned short**)&in, val = 0;
			for(size_t i = 0; i < bytes / 2; ++i, ++uc)
			{
				val = *uc;
				if ((*uc>=0xDC00 && *uc<=0xDFFF) || *uc==0)	break;	/* check for invalid.	*/

				if (*uc>=0xD800 && *uc<=0xDBFF)	/* UTF16 surrogate pairs.	*/
				{
					unsigned short uc2 = uc[1];
					if (uc2 < 0xDC00 || uc2 > 0xDFFF)		break;	/* invalid second-half of surrogate.	*/
					val = 0x10000 + (((*uc & 0x3FF) << 10) | (uc2 & 0x3FF));
				}

				int len = 4;
				char buf[4] = {0}, *ptr2 = buf;
				if (val<0x80) len=1;else if (val<0x800) len=2;else if (val<0x10000) len=3; ptr2+=len;
				
				switch (len) {
					case 4: *--ptr2 =((val | 0x80) & 0xBF); val >>= 6;
					case 3: *--ptr2 =((val | 0x80) & 0xBF); val >>= 6;
					case 2: *--ptr2 =((val | 0x80) & 0xBF); val >>= 6;
					case 1: *--ptr2 =(val | firstByteMark[len]);
				}
				ret += std::string(buf, len);
			}
		}
		else
		{
			char* unic = (char*)malloc(strlen(in) * 3 + 8);
			unsigned char	* cur = (unsigned char*)unic;

			while (*cur = *in++)
			{
				if ((*cur & 0x0f0) == 0x0e0)
				{
					if (((unsigned char)in[0] & 0x0c0) == 0x80 &&
						((unsigned char)in[1] & 0x0c0) == 0x80)
					{
						char* hex = "0123456789ABCDEF";
						unsigned short us = *cur & 0x0f;
						us <<= 6;
						us += in[0] & 0x3f;
						us <<= 6;
						us += in[1] & 0x3f;

						*cur++ = '\\';
						*cur++ = 'u';
						cur[3] = hex[us & 0x0f];
						us >>= 4;
						cur[2] = hex[us & 0x0f];
						us >>= 4;
						cur[1] = hex[us & 0x0f];
						us >>= 4;
						cur[0] = hex[us & 0x0f];
						cur += 3;
						in += 2;
					}
				}
				cur++;
			}
			*cur++ = 0;
			ret = std::string(unic, (char*)cur - unic);
			free(unic);
		}

		return std::move(ret);
	}

    std::string to_abs_path(const char* base, const char* rel_path)
	{
		if(*rel_path == '/')
			return rel_path;

		std::string b(base), r(rel_path);
		size_t pos = 0;

		while(b.length() && b[b.length() - 1] == '/')
			b.erase(b.length() - 1);
		
		while(r[0] == '.')
		{
			if(r[1] == '/')
				r.erase(0, 2);
			else if(r[1] == '.' && r[2] == '/')
			{
				pos = b.rfind('/');
				if(pos == std::string::npos)
					break;
				b.erase(pos);
				r.erase(0, 3);
			}
			else
				break;
		}

		b += "/" + r;

		return std::move(b);
	}
    std::string to_rel_path(const char* base, const char* abs_path)
	{
		if(*abs_path != '/')
			return abs_path;

		std::string rel(""), b(base), a(abs_path);
		size_t		pos = 0;

		while(b.length() && b[b.length() - 1] == '/')
			b.erase(b.length() - 1);

		while(a.find(b) == std::string::npos)
		{
			rel += "../";
			pos = b.rfind('/');
			if(pos == 0)
				break;
			b.erase(pos - 1);
		}
		if(rel.empty())
			rel = "./";
	
		return std::move(rel + a);
	}

	const char* pick_simple_block(const char* head, char end)
	{
		const char* tail = head + 1;
		int cnt = 1;

		while(*tail)
		{
			if(*tail == end)
			{
				cnt--;
				if(cnt == 0)
					break;
			}
			else if(*tail == *head)
			{
				cnt++;
			}
			else if(*tail == '\\')
			{
				tail++;
				if(*tail == 0)
					break;
			}
			tail++;
		}

		return cnt == 0 ? tail : nullptr;
	}
}