sh-i18n--envsubst.c

   1 /*
   2  * sh-i18n--envsubst.c - a stripped-down version of gettext's envsubst(1)
   3  *
   4  * Copyright (C) 2010 Ævar Arnfjörð Bjarmason
   5  *
   6  * This is a modified version of
   7  * 67d0871a8c:gettext-runtime/src/envsubst.c from the gettext.git
   8  * repository. It has been stripped down to only implement the
   9  * envsubst(1) features that we need in the git-sh-i18n fallbacks.
  10  *
  11  * The "Close standard error" part in main() is from
  12  * 8dac033df0:gnulib-local/lib/closeout.c. The copyright notices for
  13  * both files are reproduced immediately below.
  14  */
  15
  16 #include "git-compat-util.h"
  17
  18 /* Substitution of environment variables in shell format strings.
  19    Copyright (C) 2003-2007 Free Software Foundation, Inc.
  20    Written by Bruno Haible <bruno@clisp.org>, 2003.
  21
  22    This program is free software; you can redistribute it and/or modify
  23    it under the terms of the GNU General Public License as published by
  24    the Free Software Foundation; either version 2, or (at your option)
  25    any later version.
  26
  27    This program is distributed in the hope that it will be useful,
  28    but WITHOUT ANY WARRANTY; without even the implied warranty of
  29    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  30    GNU General Public License for more details.
  31
  32    You should have received a copy of the GNU General Public License
  33    along with this program; if not, write to the Free Software Foundation,
  34    Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.  */
  35
  36 /* closeout.c - close standard output and standard error
  37    Copyright (C) 1998-2007 Free Software Foundation, Inc.
  38
  39    This program is free software; you can redistribute it and/or modify
  40    it under the terms of the GNU General Public License as published by
  41    the Free Software Foundation; either version 2, or (at your option)
  42    any later version.
  43
  44    This program is distributed in the hope that it will be useful,
  45    but WITHOUT ANY WARRANTY; without even the implied warranty of
  46    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  47    GNU General Public License for more details.
  48
  49    You should have received a copy of the GNU General Public License
  50    along with this program; if not, write to the Free Software Foundation,
  51    Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.  */
  52
  53 #include <errno.h>
  54 #include <stdio.h>
  55 #include <stdlib.h>
  56 #include <string.h>
  57
  58 /* If true, substitution shall be performed on all variables.  */
  59 static unsigned short int all_variables;
  60
  61 /* Forward declaration of local functions.  */
  62 static void print_variables (const char *string);
  63 static void note_variables (const char *string);
  64 static void subst_from_stdin (void);
  65
  66 int
  67 main (int argc, char *argv[])
  68 {
  69   /* Default values for command line options.  */
  70   /* unsigned short int show_variables = 0; */
  71
  72   switch (argc)
  73         {
  74         case 1:
  75           error ("we won't substitute all variables on stdin for you");
  76           /*
  77           all_variables = 1;
  78       subst_from_stdin ();
  79           */
  80         case 2:
  81           /* echo '$foo and $bar' | git sh-i18n--envsubst --variables '$foo and $bar' */
  82           all_variables = 0;
  83           note_variables (argv[1]);
  84       subst_from_stdin ();
  85           break;
  86         case 3:
  87           /* git sh-i18n--envsubst --variables '$foo and $bar' */
  88           if (strcmp(argv[1], "--variables"))
  89                 error ("first argument must be --variables when two are given");
  90           /* show_variables = 1; */
  91       print_variables (argv[2]);
  92           break;
  93         default:
  94           error ("too many arguments");
  95           break;
  96         }
  97
  98   /* Close standard error.  This is simpler than fwriteerror_no_ebadf, because
  99      upon failure we don't need an errno - all we can do at this point is to
 100      set an exit status.  */
 101   errno = 0;
 102   if (ferror (stderr) || fflush (stderr))
 103     {
 104       fclose (stderr);
 105       exit (EXIT_FAILURE);
 106     }
 107   if (fclose (stderr) && errno != EBADF)
 108     exit (EXIT_FAILURE);
 109
 110   exit (EXIT_SUCCESS);
 111 }
 112
 113 /* Parse the string and invoke the callback each time a $VARIABLE or
 114    ${VARIABLE} construct is seen, where VARIABLE is a nonempty sequence
 115    of ASCII alphanumeric/underscore characters, starting with an ASCII
 116    alphabetic/underscore character.
 117    We allow only ASCII characters, to avoid dependencies w.r.t. the current
 118    encoding: While "${\xe0}" looks like a variable access in ISO-8859-1
 119    encoding, it doesn't look like one in the BIG5, BIG5-HKSCS, GBK, GB18030,
 120    SHIFT_JIS, JOHAB encodings, because \xe0\x7d is a single character in these
 121    encodings.  */
 122 static void
 123 find_variables (const char *string,
 124                 void (*callback) (const char *var_ptr, size_t var_len))
 125 {
 126   for (; *string != '\0';)
 127     if (*string++ == '$')
 128       {
 129         const char *variable_start;
 130         const char *variable_end;
 131         unsigned short int valid;
 132         char c;
 133
 134         if (*string == '{')
 135           string++;
 136
 137         variable_start = string;
 138         c = *string;
 139         if ((c >= 'A' && c <= 'Z') || (c >= 'a' && c <= 'z') || c == '_')
 140           {
 141             do
 142               c = *++string;
 143             while ((c >= 'A' && c <= 'Z') || (c >= 'a' && c <= 'z')
 144                    || (c >= '0' && c <= '9') || c == '_');
 145             variable_end = string;
 146
 147             if (variable_start[-1] == '{')
 148               {
 149                 if (*string == '}')
 150                   {
 151                     string++;
 152                     valid = 1;
 153                   }
 154                 else
 155                   valid = 0;
 156               }
 157             else
 158               valid = 1;
 159
 160             if (valid)
 161               callback (variable_start, variable_end - variable_start);
 162           }
 163       }
 164 }
 165
 166
 167 /* Print a variable to stdout, followed by a newline.  */
 168 static void
 169 print_variable (const char *var_ptr, size_t var_len)
 170 {
 171   fwrite (var_ptr, var_len, 1, stdout);
 172   putchar ('\n');
 173 }
 174
 175 /* Print the variables contained in STRING to stdout, each one followed by a
 176    newline.  */
 177 static void
 178 print_variables (const char *string)
 179 {
 180   find_variables (string, &print_variable);
 181 }
 182
 183
 184 /* Type describing list of immutable strings,
 185    implemented using a dynamic array.  */
 186 typedef struct string_list_ty string_list_ty;
 187 struct string_list_ty
 188 {
 189   const char **item;
 190   size_t nitems;
 191   size_t nitems_max;
 192 };
 193
 194 /* Initialize an empty list of strings.  */
 195 static inline void
 196 string_list_init (string_list_ty *slp)
 197 {
 198   slp->item = NULL;
 199   slp->nitems = 0;
 200   slp->nitems_max = 0;
 201 }
 202
 203 /* Append a single string to the end of a list of strings.  */
 204 static inline void
 205 string_list_append (string_list_ty *slp, const char *s)
 206 {
 207   /* Grow the list.  */
 208   if (slp->nitems >= slp->nitems_max)
 209     {
 210       size_t nbytes;
 211
 212       slp->nitems_max = slp->nitems_max * 2 + 4;
 213       nbytes = slp->nitems_max * sizeof (slp->item[0]);
 214       slp->item = (const char **) xrealloc (slp->item, nbytes);
 215     }
 216
 217   /* Add the string to the end of the list.  */
 218   slp->item[slp->nitems++] = s;
 219 }
 220
 221 /* Compare two strings given by reference.  */
 222 static int
 223 cmp_string (const void *pstr1, const void *pstr2)
 224 {
 225   const char *str1 = *(const char **)pstr1;
 226   const char *str2 = *(const char **)pstr2;
 227
 228   return strcmp (str1, str2);
 229 }
 230
 231 /* Sort a list of strings.  */
 232 static inline void
 233 string_list_sort (string_list_ty *slp)
 234 {
 235   if (slp->nitems > 0)
 236     qsort (slp->item, slp->nitems, sizeof (slp->item[0]), cmp_string);
 237 }
 238
 239 /* Test whether a string list contains a given string.  */
 240 static inline int
 241 string_list_member (const string_list_ty *slp, const char *s)
 242 {
 243   size_t j;
 244
 245   for (j = 0; j < slp->nitems; ++j)
 246     if (strcmp (slp->item[j], s) == 0)
 247       return 1;
 248   return 0;
 249 }
 250
 251 /* Test whether a sorted string list contains a given string.  */
 252 static int
 253 sorted_string_list_member (const string_list_ty *slp, const char *s)
 254 {
 255   size_t j1, j2;
 256
 257   j1 = 0;
 258   j2 = slp->nitems;
 259   if (j2 > 0)
 260     {
 261       /* Binary search.  */
 262       while (j2 - j1 > 1)
 263         {
 264           /* Here we know that if s is in the list, it is at an index j
 265              with j1 <= j < j2.  */
 266           size_t j = (j1 + j2) >> 1;
 267           int result = strcmp (slp->item[j], s);
 268
 269           if (result > 0)
 270             j2 = j;
 271           else if (result == 0)
 272             return 1;
 273           else
 274             j1 = j + 1;
 275         }
 276       if (j2 > j1)
 277         if (strcmp (slp->item[j1], s) == 0)
 278           return 1;
 279     }
 280   return 0;
 281 }
 282
 283
 284 /* Set of variables on which to perform substitution.
 285    Used only if !all_variables.  */
 286 static string_list_ty variables_set;
 287
 288 /* Adds a variable to variables_set.  */
 289 static void
 290 note_variable (const char *var_ptr, size_t var_len)
 291 {
 292   char *string = xmalloc (var_len + 1);
 293   memcpy (string, var_ptr, var_len);
 294   string[var_len] = '\0';
 295
 296   string_list_append (&variables_set, string);
 297 }
 298
 299 /* Stores the variables occurring in the string in variables_set.  */
 300 static void
 301 note_variables (const char *string)
 302 {
 303   string_list_init (&variables_set);
 304   find_variables (string, &note_variable);
 305   string_list_sort (&variables_set);
 306 }
 307
 308
 309 static int
 310 do_getc (void)
 311 {
 312   int c = getc (stdin);
 313
 314   if (c == EOF)
 315     {
 316       if (ferror (stdin))
 317         error ("error while reading standard input");
 318     }
 319
 320   return c;
 321 }
 322
 323 static inline void
 324 do_ungetc (int c)
 325 {
 326   if (c != EOF)
 327     ungetc (c, stdin);
 328 }
 329
 330 /* Copies stdin to stdout, performing substitutions.  */
 331 static void
 332 subst_from_stdin (void)
 333 {
 334   static char *buffer;
 335   static size_t bufmax;
 336   static size_t buflen;
 337   int c;
 338
 339   for (;;)
 340     {
 341       c = do_getc ();
 342       if (c == EOF)
 343         break;
 344       /* Look for $VARIABLE or ${VARIABLE}.  */
 345       if (c == '$')
 346         {
 347           unsigned short int opening_brace = 0;
 348           unsigned short int closing_brace = 0;
 349
 350           c = do_getc ();
 351           if (c == '{')
 352             {
 353               opening_brace = 1;
 354               c = do_getc ();
 355             }
 356           if ((c >= 'A' && c <= 'Z') || (c >= 'a' && c <= 'z') || c == '_')
 357             {
 358               unsigned short int valid;
 359
 360               /* Accumulate the VARIABLE in buffer.  */
 361               buflen = 0;
 362               do
 363                 {
 364                   if (buflen >= bufmax)
 365                     {
 366                       bufmax = 2 * bufmax + 10;
 367                       buffer = xrealloc (buffer, bufmax);
 368                     }
 369                   buffer[buflen++] = c;
 370
 371                   c = do_getc ();
 372                 }
 373               while ((c >= 'A' && c <= 'Z') || (c >= 'a' && c <= 'z')
 374                      || (c >= '0' && c <= '9') || c == '_');
 375
 376               if (opening_brace)
 377                 {
 378                   if (c == '}')
 379                     {
 380                       closing_brace = 1;
 381                       valid = 1;
 382                     }
 383                   else
 384                     {
 385                       valid = 0;
 386                       do_ungetc (c);
 387                     }
 388                 }
 389               else
 390                 {
 391                   valid = 1;
 392                   do_ungetc (c);
 393                 }
 394
 395               if (valid)
 396                 {
 397                   /* Terminate the variable in the buffer.  */
 398                   if (buflen >= bufmax)
 399                     {
 400                       bufmax = 2 * bufmax + 10;
 401                       buffer = xrealloc (buffer, bufmax);
 402                     }
 403                   buffer[buflen] = '\0';
 404
 405                   /* Test whether the variable shall be substituted.  */
 406                   if (!all_variables
 407                       && !sorted_string_list_member (&variables_set, buffer))
 408                     valid = 0;
 409                 }
 410
 411               if (valid)
 412                 {
 413                   /* Substitute the variable's value from the environment.  */
 414                   const char *env_value = getenv (buffer);
 415
 416                   if (env_value != NULL)
 417                     fputs (env_value, stdout);
 418                 }
 419               else
 420                 {
 421                   /* Perform no substitution at all.  Since the buffered input
 422                      contains no other '$' than at the start, we can just
 423                      output all the buffered contents.  */
 424                   putchar ('$');
 425                   if (opening_brace)
 426                     putchar ('{');
 427                   fwrite (buffer, buflen, 1, stdout);
 428                   if (closing_brace)
 429                     putchar ('}');
 430                 }
 431             }
 432           else
 433             {
 434               do_ungetc (c);
 435               putchar ('$');
 436               if (opening_brace)
 437                 putchar ('{');
 438             }
 439         }
 440       else
 441         putchar (c);
 442     }
 443 }