1  /*
       2                              __  __            _
       3                           ___\ \/ /_ __   __ _| |_
       4                          / _ \\  /| '_ \ / _` | __|
       5                         |  __//  \| |_) | (_| | |_
       6                          \___/_/\_\ .__/ \__,_|\__|
       7                                   |_| XML parser
       8  
       9     Copyright (c) 1997-2000 Thai Open Source Software Center Ltd
      10     Copyright (c) 2000      Clark Cooper <coopercc@users.sourceforge.net>
      11     Copyright (c) 2001-2004 Fred L. Drake, Jr. <fdrake@users.sourceforge.net>
      12     Copyright (c) 2002-2009 Karl Waclawek <karl@waclawek.net>
      13     Copyright (c) 2016-2017 Sebastian Pipping <sebastian@pipping.org>
      14     Copyright (c) 2017      Rhodri James <rhodri@wildebeest.org.uk>
      15     Copyright (c) 2017      Franek Korta <fkorta@gmail.com>
      16     Licensed under the MIT license:
      17  
      18     Permission is  hereby granted,  free of charge,  to any  person obtaining
      19     a  copy  of  this  software   and  associated  documentation  files  (the
      20     "Software"),  to  deal in  the  Software  without restriction,  including
      21     without  limitation the  rights  to use,  copy,  modify, merge,  publish,
      22     distribute, sublicense, and/or sell copies of the Software, and to permit
      23     persons  to whom  the Software  is  furnished to  do so,  subject to  the
      24     following conditions:
      25  
      26     The above copyright  notice and this permission notice  shall be included
      27     in all copies or substantial portions of the Software.
      28  
      29     THE  SOFTWARE  IS  PROVIDED  "AS  IS",  WITHOUT  WARRANTY  OF  ANY  KIND,
      30     EXPRESS  OR IMPLIED,  INCLUDING  BUT  NOT LIMITED  TO  THE WARRANTIES  OF
      31     MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN
      32     NO EVENT SHALL THE AUTHORS OR  COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM,
      33     DAMAGES OR  OTHER LIABILITY, WHETHER  IN AN  ACTION OF CONTRACT,  TORT OR
      34     OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
      35     USE OR OTHER DEALINGS IN THE SOFTWARE.
      36  */
      37  
      38  #include <sys/types.h>
      39  #include <sys/stat.h>
      40  #include <fcntl.h>
      41  #include <stdlib.h>
      42  #include <stdio.h>
      43  
      44  /* Functions close(2) and read(2) */
      45  #if ! defined(_WIN32) && ! defined(_WIN64)
      46  #  include <unistd.h>
      47  #endif
      48  
      49  /* Function "read": */
      50  #if defined(_MSC_VER)
      51  #  include <io.h>
      52  /* https://msdn.microsoft.com/en-us/library/wyssk1bs(v=vs.100).aspx */
      53  #  define _EXPAT_read _read
      54  #  define _EXPAT_read_count_t int
      55  #  define _EXPAT_read_req_t unsigned int
      56  #else /* POSIX */
      57  /* http://pubs.opengroup.org/onlinepubs/009695399/functions/read.html */
      58  #  define _EXPAT_read read
      59  #  define _EXPAT_read_count_t ssize_t
      60  #  define _EXPAT_read_req_t size_t
      61  #endif
      62  
      63  #ifndef S_ISREG
      64  #  ifndef S_IFREG
      65  #    define S_IFREG _S_IFREG
      66  #  endif
      67  #  ifndef S_IFMT
      68  #    define S_IFMT _S_IFMT
      69  #  endif
      70  #  define S_ISREG(m) (((m)&S_IFMT) == S_IFREG)
      71  #endif /* not S_ISREG */
      72  
      73  #ifndef O_BINARY
      74  #  ifdef _O_BINARY
      75  #    define O_BINARY _O_BINARY
      76  #  else
      77  #    define O_BINARY 0
      78  #  endif
      79  #endif
      80  
      81  #include "xmltchar.h"
      82  #include "filemap.h"
      83  
      84  int
      85  filemap(const tchar *name,
      86          void (*processor)(const void *, size_t, const tchar *, void *arg),
      87          void *arg) {
      88    size_t nbytes;
      89    int fd;
      90    _EXPAT_read_count_t n;
      91    struct stat sb;
      92    void *p;
      93  
      94    fd = topen(name, O_RDONLY | O_BINARY);
      95    if (fd < 0) {
      96      tperror(name);
      97      return 0;
      98    }
      99    if (fstat(fd, &sb) < 0) {
     100      tperror(name);
     101      close(fd);
     102      return 0;
     103    }
     104    if (! S_ISREG(sb.st_mode)) {
     105      ftprintf(stderr, T("%s: not a regular file\n"), name);
     106      close(fd);
     107      return 0;
     108    }
     109    if (sb.st_size > XML_MAX_CHUNK_LEN) {
     110      close(fd);
     111      return 2; /* Cannot be passed to XML_Parse in one go */
     112    }
     113  
     114    nbytes = sb.st_size;
     115    /* malloc will return NULL with nbytes == 0, handle files with size 0 */
     116    if (nbytes == 0) {
     117      static const char c = '\0';
     118      processor(&c, 0, name, arg);
     119      close(fd);
     120      return 1;
     121    }
     122    p = malloc(nbytes);
     123    if (! p) {
     124      ftprintf(stderr, T("%s: out of memory\n"), name);
     125      close(fd);
     126      return 0;
     127    }
     128    n = _EXPAT_read(fd, p, (_EXPAT_read_req_t)nbytes);
     129    if (n < 0) {
     130      tperror(name);
     131      free(p);
     132      close(fd);
     133      return 0;
     134    }
     135    if (n != (_EXPAT_read_count_t)nbytes) {
     136      ftprintf(stderr, T("%s: read unexpected number of bytes\n"), name);
     137      free(p);
     138      close(fd);
     139      return 0;
     140    }
     141    processor(p, nbytes, name, arg);
     142    free(p);
     143    close(fd);
     144    return 1;
     145  }