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 #include "trace2.h"
  18
  19 /* Substitution of environment variables in shell format strings.
  20    Copyright (C) 2003-2007 Free Software Foundation, Inc.
  21    Written by Bruno Haible <bruno@clisp.org>, 2003.
  22
  23    This program is free software; you can redistribute it and/or modify
  24    it under the terms of the GNU General Public License as published by
  25    the Free Software Foundation; either version 2, or (at your option)
  26    any later version.
  27
  28    This program is distributed in the hope that it will be useful,
  29    but WITHOUT ANY WARRANTY; without even the implied warranty of
  30    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  31    GNU General Public License for more details.
  32
  33    You should have received a copy of the GNU General Public License
  34    along with this program; if not, see <http://www.gnu.org/licenses/>.  */
  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, see <http://www.gnu.org/licenses/>.  */
  51
  52 #include <errno.h>
  53 #include <stdio.h>
  54 #include <stdlib.h>
  55 #include <string.h>
  56
  57 /* If true, substitution shall be performed on all variables.  */
  58 static unsigned short int all_variables;
  59
  60 /* Forward declaration of local functions.  */
  61 static void print_variables (const char *string);
  62 static void note_variables (const char *string);
  63 static void subst_from_stdin (void);
  64
  65 int
  66 cmd_main (int argc, const char *argv[])
  67 {
  68   /* Default values for command line options.  */
  69   /* unsigned short int show_variables = 0; */
  70
  71   trace2_cmd_name("sh-i18n--envsubst");
  72
  73   switch (argc)
  74         {
  75         case 1:
  76           error ("we won't substitute all variables on stdin for you");
  77           break;
  78           /*
  79           all_variables = 1;
  80       subst_from_stdin ();
  81           */
  82         case 2:
  83           /* echo '$foo and $bar' | git sh-i18n--envsubst --variables '$foo and $bar' */
  84           all_variables = 0;
  85           note_variables (argv[1]);
  86       subst_from_stdin ();
  87           break;
  88         case 3:
  89           /* git sh-i18n--envsubst --variables '$foo and $bar' */
  90           if (strcmp(argv[1], "--variables"))
  91                 error ("first argument must be --variables when two are given");
  92           /* show_variables = 1; */
  93       print_variables (argv[2]);
  94           break;
  95         default:
  96           error ("too many arguments");
  97           break;
  98         }
  99
 100   /* Close standard error.  This is simpler than fwriteerror_no_ebadf, because
 101      upon failure we don't need an errno - all we can do at this point is to
 102      set an exit status.  */
 103   errno = 0;
 104   if (ferror (stderr) || fflush (stderr))
 105     {
 106       fclose (stderr);
 107       return (EXIT_FAILURE);
 108     }
 109   if (fclose (stderr) && errno != EBADF)
 110     return (EXIT_FAILURE);
 111
 112   return (EXIT_SUCCESS);
 113 }
 114
 115 /* Parse the string and invoke the callback each time a $VARIABLE or
 116    ${VARIABLE} construct is seen, where VARIABLE is a nonempty sequence
 117    of ASCII alphanumeric/underscore characters, starting with an ASCII
 118    alphabetic/underscore character.
 119    We allow only ASCII characters, to avoid dependencies w.r.t. the current
 120    encoding: While "${\xe0}" looks like a variable access in ISO-8859-1
 121    encoding, it doesn't look like one in the BIG5, BIG5-HKSCS, GBK, GB18030,
 122    SHIFT_JIS, JOHAB encodings, because \xe0\x7d is a single character in these
 123    encodings.  */
 124 static void
 125 find_variables (const char *string,
 126                 void (*callback) (const char *var_ptr, size_t var_len))
 127 {
 128   for (; *string != '\0';)
 129     if (*string++ == '$')
 130       {
 131         const char *variable_start;
 132         const char *variable_end;
 133         unsigned short int valid;
 134         char c;
 135
 136         if (*string == '{')
 137           string++;
 138
 139         variable_start = string;
 140         c = *string;
 141         if ((c >= 'A' && c <= 'Z') || (c >= 'a' && c <= 'z') || c == '_')
 142           {
 143             do
 144               c = *++string;
 145             while ((c >= 'A' && c <= 'Z') || (c >= 'a' && c <= 'z')
 146                    || (c >= '0' && c <= '9') || c == '_');
 147             variable_end = string;
 148
 149             if (variable_start[-1] == '{')
 150               {
 151                 if (*string == '}')
 152                   {
 153                     string++;
 154                     valid = 1;
 155                   }
 156                 else
 157                   valid = 0;
 158               }
 159             else
 160               valid = 1;
 161
 162             if (valid)
 163               callback (variable_start, variable_end - variable_start);
 164           }
 165       }
 166 }
 167
 168
 169 /* Print a variable to stdout, followed by a newline.  */
 170 static void
 171 print_variable (const char *var_ptr, size_t var_len)
 172 {
 173   fwrite (var_ptr, var_len, 1, stdout);
 174   putchar ('\n');
 175 }
 176
 177 /* Print the variables contained in STRING to stdout, each one followed by a
 178    newline.  */
 179 static void
 180 print_variables (const char *string)
 181 {
 182   find_variables (string, &print_variable);
 183 }
 184
 185
 186 /* Type describing list of immutable strings,
 187    implemented using a dynamic array.  */
 188 typedef struct string_list_ty string_list_ty;
 189 struct string_list_ty
 190 {
 191   const char **item;
 192   size_t nitems;
 193   size_t nitems_max;
 194 };
 195
 196 /* Initialize an empty list of strings.  */
 197 static inline void
 198 string_list_init (string_list_ty *slp)
 199 {
 200   slp->item = NULL;
 201   slp->nitems = 0;
 202   slp->nitems_max = 0;
 203 }
 204
 205 /* Append a single string to the end of a list of strings.  */
 206 static inline void
 207 string_list_append (string_list_ty *slp, const char *s)
 208 {
 209   /* Grow the list.  */
 210   if (slp->nitems >= slp->nitems_max)
 211     {
 212       slp->nitems_max = slp->nitems_max * 2 + 4;
 213       REALLOC_ARRAY(slp->item, slp->nitems_max);
 214     }
 215
 216   /* Add the string to the end of the list.  */
 217   slp->item[slp->nitems++] = s;
 218 }
 219
 220 /* Compare two strings given by reference.  */
 221 static int
 222 cmp_string (const void *pstr1, const void *pstr2)
 223 {
 224   const char *str1 = *(const char **)pstr1;
 225   const char *str2 = *(const char **)pstr2;
 226
 227   return strcmp (str1, str2);
 228 }
 229
 230 /* Sort a list of strings.  */
 231 static inline void
 232 string_list_sort (string_list_ty *slp)
 233 {
 234   QSORT(slp->item, slp->nitems, cmp_string);
 235 }
 236
 237 /* Test whether a sorted string list contains a given string.  */
 238 static int
 239 sorted_string_list_member (const string_list_ty *slp, const char *s)
 240 {
 241   size_t j1, j2;
 242
 243   j1 = 0;
 244   j2 = slp->nitems;
 245   if (j2 > 0)
 246     {
 247       /* Binary search.  */
 248       while (j2 - j1 > 1)
 249         {
 250           /* Here we know that if s is in the list, it is at an index j
 251              with j1 <= j < j2.  */
 252           size_t j = j1 + ((j2 - j1) >> 1);
 253           int result = strcmp (slp->item[j], s);
 254
 255           if (result > 0)
 256             j2 = j;
 257           else if (result == 0)
 258             return 1;
 259           else
 260             j1 = j + 1;
 261         }
 262       if (j2 > j1)
 263         if (strcmp (slp->item[j1], s) == 0)
 264           return 1;
 265     }
 266   return 0;
 267 }
 268
 269
 270 /* Set of variables on which to perform substitution.
 271    Used only if !all_variables.  */
 272 static string_list_ty variables_set;
 273
 274 /* Adds a variable to variables_set.  */
 275 static void
 276 note_variable (const char *var_ptr, size_t var_len)
 277 {
 278   char *string = xmemdupz (var_ptr, var_len);
 279
 280   string_list_append (&variables_set, string);
 281 }
 282
 283 /* Stores the variables occurring in the string in variables_set.  */
 284 static void
 285 note_variables (const char *string)
 286 {
 287   string_list_init (&variables_set);
 288   find_variables (string, &note_variable);
 289   string_list_sort (&variables_set);
 290 }
 291
 292
 293 static int
 294 do_getc (void)
 295 {
 296   int c = getc (stdin);
 297
 298   if (c == EOF)
 299     {
 300       if (ferror (stdin))
 301         error ("error while reading standard input");
 302     }
 303
 304   return c;
 305 }
 306
 307 static inline void
 308 do_ungetc (int c)
 309 {
 310   if (c != EOF)
 311     ungetc (c, stdin);
 312 }
 313
 314 /* Copies stdin to stdout, performing substitutions.  */
 315 static void
 316 subst_from_stdin (void)
 317 {
 318   static char *buffer;
 319   static size_t bufmax;
 320   static size_t buflen;
 321   int c;
 322
 323   for (;;)
 324     {
 325       c = do_getc ();
 326       if (c == EOF)
 327         break;
 328       /* Look for $VARIABLE or ${VARIABLE}.  */
 329       if (c == '$')
 330         {
 331           unsigned short int opening_brace = 0;
 332           unsigned short int closing_brace = 0;
 333
 334           c = do_getc ();
 335           if (c == '{')
 336             {
 337               opening_brace = 1;
 338               c = do_getc ();
 339             }
 340           if ((c >= 'A' && c <= 'Z') || (c >= 'a' && c <= 'z') || c == '_')
 341             {
 342               unsigned short int valid;
 343
 344               /* Accumulate the VARIABLE in buffer.  */
 345               buflen = 0;
 346               do
 347                 {
 348                   if (buflen >= bufmax)
 349                     {
 350                       bufmax = 2 * bufmax + 10;
 351                       buffer = xrealloc (buffer, bufmax);
 352                     }
 353                   buffer[buflen++] = c;
 354
 355                   c = do_getc ();
 356                 }
 357               while ((c >= 'A' && c <= 'Z') || (c >= 'a' && c <= 'z')
 358                      || (c >= '0' && c <= '9') || c == '_');
 359
 360               if (opening_brace)
 361                 {
 362                   if (c == '}')
 363                     {
 364                       closing_brace = 1;
 365                       valid = 1;
 366                     }
 367                   else
 368                     {
 369                       valid = 0;
 370                       do_ungetc (c);
 371                     }
 372                 }
 373               else
 374                 {
 375                   valid = 1;
 376                   do_ungetc (c);
 377                 }
 378
 379               if (valid)
 380                 {
 381                   /* Terminate the variable in the buffer.  */
 382                   if (buflen >= bufmax)
 383                     {
 384                       bufmax = 2 * bufmax + 10;
 385                       buffer = xrealloc (buffer, bufmax);
 386                     }
 387                   buffer[buflen] = '\0';
 388
 389                   /* Test whether the variable shall be substituted.  */
 390                   if (!all_variables
 391                       && !sorted_string_list_member (&variables_set, buffer))
 392                     valid = 0;
 393                 }
 394
 395               if (valid)
 396                 {
 397                   /* Substitute the variable's value from the environment.  */
 398                   const char *env_value = getenv (buffer);
 399
 400                   if (env_value != NULL)
 401                     fputs (env_value, stdout);
 402                 }
 403               else
 404                 {
 405                   /* Perform no substitution at all.  Since the buffered input
 406                      contains no other '$' than at the start, we can just
 407                      output all the buffered contents.  */
 408                   putchar ('$');
 409                   if (opening_brace)
 410                     putchar ('{');
 411                   fwrite (buffer, buflen, 1, stdout);
 412                   if (closing_brace)
 413                     putchar ('}');
 414                 }
 415             }
 416           else
 417             {
 418               do_ungetc (c);
 419               putchar ('$');
 420               if (opening_brace)
 421                 putchar ('{');
 422             }
 423         }
 424       else
 425         putchar (c);
 426     }
 427 }