pack.c

php-4.4.7学习linux时下载的源代码

/*
   +----------------------------------------------------------------------+
   | PHP Version 4                                                        |
   +----------------------------------------------------------------------+
   | Copyright (c) 1997-2007 The PHP Group                                |
   +----------------------------------------------------------------------+
   | This source file is subject to version 3.01 of the PHP license,      |
   | that is bundled with this package in the file LICENSE, and is        |
   | available through the world-wide-web at the following url:           |
   | http://www.php.net/license/3_01.txt                                  |
   | If you did not receive a copy of the PHP license and are unable to   |
   | obtain it through the world-wide-web, please send a note to          |
   | [e-mail] so we can mail you a copy immediately.               |
   +----------------------------------------------------------------------+
   | Author: Chris Schneider <[e-mail]>                          |
   +----------------------------------------------------------------------+
 */
/* $Id: pack.c,v 1.40.2.7.2.6 2007/01/01 09:46:48 sebastian Exp $ */

#include "php.h"

#include <stdio.h>
#include <stdlib.h>
#include <errno.h>
#include <sys/types.h>
#include <sys/stat.h>
#include <fcntl.h>
#ifdef PHP_WIN32
#include <windows.h>
#include <winsock.h>
#define O_RDONLY _O_RDONLY
#include "win32/param.h"
#elif defined(NETWARE)
#ifdef USE_WINSOCK
#include <novsock2.h>
#else
#include <sys/socket.h>
#endif
#include <sys/param.h>
#else
#include <sys/param.h>
#endif
#include "ext/standard/head.h"
#include "safe_mode.h"
#include "php_string.h"
#include "pack.h"
#if HAVE_PWD_H
#ifdef PHP_WIN32
#include "win32/pwd.h"
#else
#include <pwd.h>
#endif
#endif
#include "fsock.h"
#if HAVE_NETINET_IN_H
#include <netinet/in.h>
#endif

#define INC_OUTPUTPOS(a,b) \
	if ((a) < 0 || ((INT_MAX - outputpos)/(b)) < (a)) { \
		php_error_docref(NULL TSRMLS_CC, E_WARNING, "Type %c: integer overflow in format string", code); \
		RETURN_FALSE; \
	} \
	outputpos += (a)*(b);

/* Whether machine is little endian */
char machine_little_endian;

/* Mapping of byte from char (8bit) to long for machine endian */
static int byte_map[1];

/* Mappings of bytes from int (machine dependant) to int for machine endian */
static int int_map[sizeof(int)];

/* Mappings of bytes from shorts (16bit) for all endian environments */
static int machine_endian_short_map[2];
static int big_endian_short_map[2];
static int little_endian_short_map[2];

/* Mappings of bytes from longs (32bit) for all endian environments */
static int machine_endian_long_map[4];
static int big_endian_long_map[4];
static int little_endian_long_map[4];

/* {{{ php_pack
 */
static void php_pack(zval **val, int size, int *map, char *output)
{
	int i;
	char *v;

	convert_to_long_ex(val);
	v = (char *) &Z_LVAL_PP(val);

	for (i = 0; i < size; i++) {
		*output++ = v[map[i]];
	}
}
/* }}} */

/* pack() idea stolen from Perl (implemented formats behave the same as there)
 * Implemented formats are A, a, h, H, c, C, s, S, i, I, l, L, n, N, f, d, x, X, @.
 */
/* {{{ proto string pack(string format, mixed arg1 [, mixed arg2 [, mixed ...]])
   Takes one or more arguments and packs them into a binary string according to the format argument */
PHP_FUNCTION(pack)
{
	zval ***argv;
	int argc, i;
	int currentarg;
	char *format;
	int formatlen;
	char *formatcodes;
	int *formatargs;
	int formatcount = 0;
	int outputpos = 0, outputsize = 0;
	char *output;

	argc = ZEND_NUM_ARGS();

	if (argc < 1) {
		WRONG_PARAM_COUNT;
	}

	argv = safe_emalloc(sizeof(zval **), argc, 0);

	if (zend_get_parameters_array_ex(argc, argv) == FAILURE) {
		efree(argv);
		WRONG_PARAM_COUNT;
	}

	convert_to_string_ex(argv[0]);
	format = Z_STRVAL_PP(argv[0]);
	formatlen = Z_STRLEN_PP(argv[0]);

	/* We have a maximum of <formatlen> format codes to deal with */
	formatcodes = safe_emalloc(formatlen, sizeof(*formatcodes), 0);
	formatargs = safe_emalloc(formatlen, sizeof(*formatargs), 0);
	currentarg = 1;

	/* Preprocess format into formatcodes and formatargs */
	for (i = 0; i < formatlen; formatcount++) {
		char code = format[i++];
		int arg = 1;

		/* Handle format arguments if any */
		if (i < formatlen) {
			char c = format[i];

			if (c == '*') {
				arg = -1;
				i++;
			}
			else if (c >= '0' && c <= '9') {
				arg = atoi(&format[i]);
		  
				while (format[i] >= '0' && format[i] <= '9' && i < formatlen) {
					i++;
				}
			}
		}

		/* Handle special arg '*' for all codes and check argv overflows */
		switch ((int) code) {
			/* Never uses any args */
			case 'x': 
			case 'X':	
			case '@':
				if (arg < 0) {
					php_error_docref(NULL TSRMLS_CC, E_WARNING, "Type %c: '*' ignored", code);
					arg = 1;
				}
				break;

			/* Always uses one arg */
			case 'a': 
			case 'A': 
			case 'h': 
			case 'H':
				if (currentarg >= argc) {
					efree(argv);
					efree(formatcodes);
					efree(formatargs);
					php_error_docref(NULL TSRMLS_CC, E_WARNING, "Type %c: not enough arguments", code);
					RETURN_FALSE;
				}

				if (arg < 0) {
					arg = Z_STRLEN_PP(argv[currentarg]);
				}

				currentarg++;
				break;

			/* Use as many args as specified */
			case 'c': 
			case 'C': 
			case 's': 
			case 'S': 
			case 'i': 
			case 'I':
			case 'l': 
			case 'L': 
			case 'n': 
			case 'N': 
			case 'v': 
			case 'V':
			case 'f': 
			case 'd': 
				if (arg < 0) {
					arg = argc - currentarg;
				}

				currentarg += arg;

				if (currentarg > argc) {
					efree(argv);
					efree(formatcodes);
					efree(formatargs);
					php_error_docref(NULL TSRMLS_CC, E_WARNING, "Type %c: too few arguments", code);
					RETURN_FALSE;
				}
				break;

			default:
				efree(argv);
				efree(formatcodes);
				efree(formatargs);
				php_error_docref(NULL TSRMLS_CC, E_WARNING, "Type %c: unknown format code", code);
				RETURN_FALSE;
		}

		formatcodes[formatcount] = code;
		formatargs[formatcount] = arg;
	}

	if (currentarg < argc) {
		php_error_docref(NULL TSRMLS_CC, E_WARNING, "%d arguments unused", (argc - currentarg));
	}

	/* Calculate output length and upper bound while processing*/
	for (i = 0; i < formatcount; i++) {
	    int code = (int) formatcodes[i];
		int arg = formatargs[i];

		switch ((int) code) {
			case 'h': 
			case 'H': 
				INC_OUTPUTPOS((arg + (arg % 2)) / 2,1)	/* 4 bit per arg */
				break;

			case 'a': 
			case 'A':
			case 'c': 
			case 'C':
			case 'x':
				INC_OUTPUTPOS(arg,1)		/* 8 bit per arg */
				break;

			case 's': 
			case 'S': 
			case 'n': 
			case 'v':
				INC_OUTPUTPOS(arg,2)		/* 16 bit per arg */
				break;

			case 'i': 
			case 'I':
				INC_OUTPUTPOS(arg,sizeof(int))
				break;

			case 'l': 
			case 'L': 
			case 'N': 
			case 'V':
				INC_OUTPUTPOS(arg,4)		/* 32 bit per arg */
				break;

			case 'f':
				INC_OUTPUTPOS(arg,sizeof(float))
				break;

			case 'd':
				INC_OUTPUTPOS(arg,sizeof(double))
				break;

			case 'X':
				outputpos -= arg;

				if (outputpos < 0) {
					php_error_docref(NULL TSRMLS_CC, E_WARNING, "Type %c: outside of string", code);
					outputpos = 0;
				}
				break;

			case '@':
				outputpos = arg;
				break;
		}

		if (outputsize < outputpos) {
			outputsize = outputpos;
		}
	}

	output = emalloc(outputsize + 1);
	outputpos = 0;
	currentarg = 1;

	/* Do actual packing */
	for (i = 0; i < formatcount; i++) {
	    int code = (int) formatcodes[i];
		int arg = formatargs[i];
		zval **val;

		switch ((int) code) {
			case 'a': 
			case 'A': 
				memset(&output[outputpos], (code == 'a') ? '\0' : ' ', arg);
				val = argv[currentarg++];
				convert_to_string_ex(val);
				memcpy(&output[outputpos], Z_STRVAL_PP(val),
					   (Z_STRLEN_PP(val) < arg) ? Z_STRLEN_PP(val) : arg);
				outputpos += arg;
				break;

			case 'h': 
			case 'H': {
				int nibbleshift = (code == 'h') ? 0 : 4;
				int first = 1;
				char *v;

				val = argv[currentarg++];
				convert_to_string_ex(val);
				v = Z_STRVAL_PP(val);
				outputpos--;
				if(arg > Z_STRLEN_PP(val)) {
					php_error_docref(NULL TSRMLS_CC, E_WARNING, "Type %c: not enough characters in string", code);
					arg = Z_STRLEN_PP(val);
				}

				while (arg-- > 0) {
					char n = *v++;

					if (n >= '0' && n <= '9') {
						n -= '0';
					} else if (n >= 'A' && n <= 'F') {
						n -= ('A' - 10);
					} else if (n >= 'a' && n <= 'f') {
						n -= ('a' - 10);
					} else {
						php_error_docref(NULL TSRMLS_CC, E_WARNING, "Type %c: illegal hex digit %c", code, n);
						n = 0;
					}

					if (first--) {
						output[++outputpos] = 0;
					} else {
					  first = 1;
					}

					output[outputpos] |= (n << nibbleshift);
					nibbleshift = (nibbleshift + 4) & 7;
				}

				outputpos++;
				break;
			}

			case 'c': 
			case 'C':
				while (arg-- > 0) {
					php_pack(argv[currentarg++], 1, byte_map, &output[outputpos]);
					outputpos++;
				}
				break;

			case 's': 
			case 'S': 
			case 'n': 
			case 'v': {
				int *map = machine_endian_short_map;

				if (code == 'n') {
					map = big_endian_short_map;
				} else if (code == 'v') {
					map = little_endian_short_map;
				}

				while (arg-- > 0) {
					php_pack(argv[currentarg++], 2, map, &output[outputpos]);
					outputpos += 2;
				}
				break;
			}

			case 'i': 
			case 'I': 
				while (arg-- > 0) {
					php_pack(argv[currentarg++], sizeof(int), int_map, &output[outputpos]);
					outputpos += sizeof(int);
				}
				break;

			case 'l': 
			case 'L': 
			case 'N': 
			case 'V': {
				int *map = machine_endian_long_map;

				if (code == 'N') {
					map = big_endian_long_map;
				} else if (code == 'V') {
					map = little_endian_long_map;
				}

				while (arg-- > 0) {
					php_pack(argv[currentarg++], 4, map, &output[outputpos]);
					outputpos += 4;
				}
				break;
			}

			case 'f': {
				float v;

				while (arg-- > 0) {
					val = argv[currentarg++];
					convert_to_double_ex(val);
					v = (float) Z_DVAL_PP(val);
					memcpy(&output[outputpos], &v, sizeof(v));
					outputpos += sizeof(v);
				}
				break;
			}

			case 'd': {
				double v;

				while (arg-- > 0) {
					val = argv[currentarg++];
					convert_to_double_ex(val);
					v = (double) Z_DVAL_PP(val);
					memcpy(&output[outputpos], &v, sizeof(v));
					outputpos += sizeof(v);
				}
				break;
			}

			case 'x':
				memset(&output[outputpos], '\0', arg);
				outputpos += arg;
				break;

			case 'X':
				outputpos -= arg;

				if (outputpos < 0) {
					outputpos = 0;
				}
				break;

			case '@':
				if (arg > outputpos) {
					memset(&output[outputpos], '\0', arg - outputpos);
				}
				outputpos = arg;
				break;
		}