libiberty/lrealpath.c

   1 /* Libiberty realpath.  Like realpath, but more consistent behavior.
   2    Based on gdb_realpath from GDB.
   3
   4    Copyright (C) 2003-2024 Free Software Foundation, Inc.
   5
   6    This file is part of the libiberty library.
   7
   8    This program is free software; you can redistribute it and/or modify
   9    it under the terms of the GNU General Public License as published by
  10    the Free Software Foundation; either version 2 of the License, or
  11    (at your option) any later version.
  12
  13    This program is distributed in the hope that it will be useful,
  14    but WITHOUT ANY WARRANTY; without even the implied warranty of
  15    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  16    GNU General Public License for more details.
  17
  18    You should have received a copy of the GNU General Public License
  19    along with this program; if not, write to the Free Software
  20    Foundation, Inc., 51 Franklin Street - Fifth Floor,
  21    Boston, MA 02110-1301, USA.  */
  22
  23 /*
  24
  25 @deftypefn Replacement {const char*} lrealpath (const char *@var{name})
  26
  27 Given a pointer to a string containing a pathname, returns a canonical
  28 version of the filename.  Symlinks will be resolved, and ``.'' and ``..''
  29 components will be simplified.  The returned value will be allocated using
  30 @code{malloc}, or @code{NULL} will be returned on a memory allocation error.
  31
  32 @end deftypefn
  33
  34 */
  35
  36 #include "config.h"
  37 #include "ansidecl.h"
  38 #include "libiberty.h"
  39
  40 #ifdef HAVE_LIMITS_H
  41 #include <limits.h>
  42 #endif
  43 #ifdef HAVE_STDLIB_H
  44 #include <stdlib.h>
  45 #endif
  46 #ifdef HAVE_UNISTD_H
  47 #include <unistd.h>
  48 #endif
  49 #ifdef HAVE_STRING_H
  50 #include <string.h>
  51 #endif
  52
  53 /* On GNU libc systems the declaration is only visible with _GNU_SOURCE.  */
  54 #if defined(HAVE_CANONICALIZE_FILE_NAME) \
  55     && defined(NEED_DECLARATION_CANONICALIZE_FILE_NAME)
  56 extern char *canonicalize_file_name (const char *);
  57 #endif
  58
  59 #if defined(HAVE_REALPATH)
  60 # if defined (PATH_MAX)
  61 #  define REALPATH_LIMIT PATH_MAX
  62 # else
  63 #  if defined (MAXPATHLEN)
  64 #   define REALPATH_LIMIT MAXPATHLEN
  65 #  endif
  66 # endif
  67 #else
  68   /* cygwin has realpath, so it won't get here.  */
  69 # if defined (_WIN32)
  70 #  define WIN32_LEAN_AND_MEAN
  71 #  include <windows.h> /* for GetFullPathName/GetFinalPathNameByHandle/
  72                           CreateFile/CloseHandle */
  73 #  define WIN32_REPLACE_SLASHES(_ptr, _len) \
  74      for (unsigned i = 0; i != (_len); ++i) \
  75        if ((_ptr)[i] == '\\') (_ptr)[i] = '/';
  76
  77 #  define WIN32_UNC_PREFIX "//?/UNC/"
  78 #  define WIN32_UNC_PREFIX_LEN (sizeof(WIN32_UNC_PREFIX)-1)
  79 #  define WIN32_IS_UNC_PREFIX(ptr) \
  80   (0 == memcmp(ptr, WIN32_UNC_PREFIX, WIN32_UNC_PREFIX_LEN))
  81
  82 #  define WIN32_NON_UNC_PREFIX "//?/"
  83 #  define WIN32_NON_UNC_PREFIX_LEN (sizeof(WIN32_NON_UNC_PREFIX)-1)
  84 #  define WIN32_IS_NON_UNC_PREFIX(ptr) \
  85   (0 == memcmp(ptr, WIN32_NON_UNC_PREFIX, WIN32_NON_UNC_PREFIX_LEN))
  86
  87 /* Get full path name without symlinks resolution.
  88    It also converts all forward slashes to back slashes.
  89 */
  90 char* get_full_path_name(const char *filename) {
  91   DWORD len;
  92   char *buf, *ptr, *res;
  93
  94   /* determining the required buffer size.
  95      from the man: `If the lpBuffer buffer is too small to contain
  96      the path, the return value is the size, in TCHARs, of the buffer
  97      that is required to hold the path _and_the_terminating_null_character_`
  98   */
  99   len = GetFullPathName(filename, 0, NULL, NULL);
 100
 101   if ( len == 0 )
 102     return strdup(filename);
 103
 104   buf = (char *)malloc(len);
 105
 106   /* no point to check the result again */
 107   len = GetFullPathName(filename, len, buf, NULL);
 108   buf[len] = 0;
 109
 110   /* replace slashes */
 111   WIN32_REPLACE_SLASHES(buf, len);
 112
 113   /* calculate offset based on prefix type */
 114   len = WIN32_IS_UNC_PREFIX(buf)
 115     ? (WIN32_UNC_PREFIX_LEN - 2)
 116     : WIN32_IS_NON_UNC_PREFIX(buf)
 117       ? WIN32_NON_UNC_PREFIX_LEN
 118       : 0
 119   ;
 120
 121   ptr = buf + len;
 122   if ( WIN32_IS_UNC_PREFIX(buf) ) {
 123     ptr[0] = '/';
 124     ptr[1] = '/';
 125   }
 126
 127   res = strdup(ptr);
 128
 129   free(buf);
 130
 131   return res;
 132 }
 133
 134 # if _WIN32_WINNT >= 0x0600
 135
 136 /* Get full path name WITH symlinks resolution.
 137    It also converts all forward slashes to back slashes.
 138 */
 139 char* get_final_path_name(HANDLE fh) {
 140   DWORD len;
 141   char *buf, *ptr, *res;
 142
 143   /* determining the required buffer size.
 144      from the  man: `If the function fails because lpszFilePath is too
 145      small to hold the string plus the terminating null character,
 146      the return value is the required buffer size, in TCHARs. This
 147      value _includes_the_size_of_the_terminating_null_character_`.
 148      but in my testcase I have path with 26 chars, the function
 149      returns 26 also, ie without the trailing zero-char...
 150   */
 151   len = GetFinalPathNameByHandle(
 152      fh
 153     ,NULL
 154     ,0
 155     ,FILE_NAME_NORMALIZED | VOLUME_NAME_DOS
 156   );
 157
 158   if ( len == 0 )
 159     return NULL;
 160
 161   len += 1; /* for zero-char */
 162   buf = (char *)malloc(len);
 163
 164   /* no point to check the result again */
 165   len = GetFinalPathNameByHandle(
 166      fh
 167     ,buf
 168     ,len
 169     ,FILE_NAME_NORMALIZED | VOLUME_NAME_DOS
 170   );
 171   buf[len] = 0;
 172
 173   /* replace slashes */
 174   WIN32_REPLACE_SLASHES(buf, len);
 175
 176   /* calculate offset based on prefix type */
 177   len = WIN32_IS_UNC_PREFIX(buf)
 178     ? (WIN32_UNC_PREFIX_LEN - 2)
 179     : WIN32_IS_NON_UNC_PREFIX(buf)
 180       ? WIN32_NON_UNC_PREFIX_LEN
 181       : 0
 182   ;
 183
 184   ptr = buf + len;
 185   if ( WIN32_IS_UNC_PREFIX(buf) ) {
 186     ptr[0] = '/';
 187     ptr[1] = '/';
 188   }
 189
 190   res = strdup(ptr);
 191
 192   free(buf);
 193
 194   return res;
 195 }
 196
 197 # endif // _WIN32_WINNT >= 0x0600
 198
 199 # endif // _WIN32
 200 #endif
 201
 202 char *
 203 lrealpath (const char *filename)
 204 {
 205   /* Method 1: The system has a compile time upper bound on a filename
 206      path.  Use that and realpath() to canonicalize the name.  This is
 207      the most common case.  Note that, if there isn't a compile time
 208      upper bound, you want to avoid realpath() at all costs.  */
 209 #if defined(REALPATH_LIMIT)
 210   {
 211     char buf[REALPATH_LIMIT];
 212     const char *rp = realpath (filename, buf);
 213     if (rp == NULL)
 214       rp = filename;
 215     return strdup (rp);
 216   }
 217 #endif /* REALPATH_LIMIT */
 218
 219   /* Method 2: The host system (i.e., GNU) has the function
 220      canonicalize_file_name() which malloc's a chunk of memory and
 221      returns that, use that.  */
 222 #if defined(HAVE_CANONICALIZE_FILE_NAME)
 223   {
 224     char *rp = canonicalize_file_name (filename);
 225     if (rp == NULL)
 226       return strdup (filename);
 227     else
 228       return rp;
 229   }
 230 #endif
 231
 232   /* Method 3: Now we're getting desperate!  The system doesn't have a
 233      compile time buffer size and no alternative function.  Query the
 234      OS, using pathconf(), for the buffer limit.  Care is needed
 235      though, some systems do not limit PATH_MAX (return -1 for
 236      pathconf()) making it impossible to pass a correctly sized buffer
 237      to realpath() (it could always overflow).  On those systems, we
 238      skip this.  */
 239 #if defined (HAVE_REALPATH) && defined (HAVE_UNISTD_H)
 240   {
 241     /* Find out the max path size.  */
 242     long path_max = pathconf ("/", _PC_PATH_MAX);
 243     if (path_max > 0)
 244       {
 245         /* PATH_MAX is bounded.  */
 246         char *buf, *rp, *ret;
 247         buf = (char *) malloc (path_max);
 248         if (buf == NULL)
 249           return NULL;
 250         rp = realpath (filename, buf);
 251         ret = strdup (rp ? rp : filename);
 252         free (buf);
 253         return ret;
 254       }
 255   }
 256 #endif
 257
 258   /* The MS Windows method */
 259 #if defined (_WIN32)
 260   {
 261     char *res;
 262
 263     /* For Windows Vista and greater */
 264 #if _WIN32_WINNT >= 0x0600
 265
 266     /* For some reason the function receives just empty `filename`, but not NULL.
 267        What should we do in that case?
 268        According to `strdup()` implementation
 269          (https://elixir.bootlin.com/glibc/latest/source/string/strdup.c)
 270        it will alloc 1 byte even for empty but non NULL string.
 271        OK, will use `strdup()` for that case.
 272     */
 273     if ( 0 == strlen(filename) )
 274       return strdup(filename);
 275
 276     HANDLE fh = CreateFile(
 277        filename
 278       ,FILE_READ_ATTRIBUTES
 279       ,FILE_SHARE_READ | FILE_SHARE_WRITE | FILE_SHARE_DELETE
 280       ,NULL
 281       ,OPEN_EXISTING
 282       ,FILE_FLAG_BACKUP_SEMANTICS
 283       ,NULL
 284     );
 285
 286     if ( fh == INVALID_HANDLE_VALUE ) {
 287       res = get_full_path_name(filename);
 288     } else {
 289       res = get_final_path_name(fh);
 290       CloseHandle(fh);
 291
 292       if ( !res )
 293         res = get_full_path_name(filename);
 294     }
 295
 296 #else
 297
 298     /* For Windows XP */
 299     res = get_full_path_name(filename);
 300
 301 #endif // _WIN32_WINNT >= 0x0600
 302
 303     return res;
 304   }
 305 #endif // _WIN32
 306 }