git-sh-i18n--envsubst: our own envsubst(1) for eval_gettext()
Add a git-sh-i18n--envsubst program which is a stripped-down version of the GNU envsubst(1) program that comes with GNU gettext for use in the eval_gettext() fallback. We need a C helper program because implementing eval_gettext() purely in shell turned out to be unworkable. Digging through the Git mailing list archives will reveal two shell implementations of eval_gettext that are almost good enough, but fail on an edge case which is tested for in the tests which are part of this patch. These are the modifications I made to envsubst.c as I turned it into sh-i18n--envsubst.c: * Added our git-compat-util.h header for xrealloc() and friends. * Removed inclusion of gettext-specific headers. * Removed most of main() and replaced it with my own. The modified version only does option parsing for --variables. That's all it needs. * Modified error() invocations to use our error() instead of error(3). * Replaced the gettext XNMALLOC(n, size) macro with just xmalloc(n). Since XNMALLOC() only allocated char's. * Removed the string_list_destroy function. It's redundant (also in the upstream code). * Replaced the use of stdbool.h (a C99 header) by doing the following replacements on the code: * s/bool/unsigned short int/g * s/true/1/g * s/false/0/g Reported-by: Johannes Sixt <j.sixt@viscovery.net> Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Ævar Arnfjörð Bjarmason committed
May 14, 2011 at 13:47 UTC
ba67aaf2d05d1aa3def2b6eb1dc1d39bdf25bb41
5 files changed
+524
.gitignore
+2
@@ -127,6 +127,8 @@
127
/git-rm
128
/git-send-email
129
/git-send-pack
130
+/git-sh-i18n
131
+/git-sh-i18n--envsubst
132
/git-sh-setup
133
/git-shell
134
/git-shortlog
Documentation/git-sh-i18n--envsubst.txt
new
+26
@@ -0,0 +1,26 @@
1
+git-sh-i18n--envsubst(1)
2
+========================
3
+
4
+NAME
5
+----
6
+git-sh-i18n--envsubst - Git's own envsubst(1) for i18n fallbacks
7
+
8
+DESCRIPTION
9
+-----------
10
+
11
+This is not a command the end user would want to run. Ever.
12
+This documentation is meant for people who are studying the
13
+plumbing scripts and/or are writing new ones.
14
+
15
+git-sh-i18n--envsubst is Git's stripped-down copy of the GNU
16
+`envsubst(1)` program that comes with the GNU gettext package. It's
17
+used internally by linkgit:git-sh-i18n[1] to interpolate the variables
18
+passed to the the `eval_gettext` function.
19
+
20
+No promises are made about the interface, or that this
21
+program won't disappear without warning in the next version
22
+of Git. Don't use it.
23
+
24
+GIT
25
+---
26
+Part of the linkgit:git[1] suite
Makefile
+1
@@ -414,6 +414,7 @@ PROGRAM_OBJS += shell.o
414
PROGRAM_OBJS += show-index.o
415
PROGRAM_OBJS += upload-pack.o
416
PROGRAM_OBJS += http-backend.o
417
+PROGRAM_OBJS += sh-i18n--envsubst.o
418
419
PROGRAMS += $(patsubst %.o,git-%$X,$(PROGRAM_OBJS))
420
sh-i18n--envsubst.c
new
+444
@@ -0,0 +1,444 @@
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 <getopt.h>
55
+#include <stdio.h>
56
+#include <stdlib.h>
57
+#include <string.h>
58
+
59
+/* If true, substitution shall be performed on all variables. */
60
+static unsigned short int all_variables;
61
+
62
+/* Forward declaration of local functions. */
63
+static void print_variables (const char *string);
64
+static void note_variables (const char *string);
65
+static void subst_from_stdin (void);
66
+
67
+int
68
+main (int argc, char *argv[])
69
+{
70
+ /* Default values for command line options. */
71
+ unsigned short int show_variables = 0;
72
+
73
+ switch (argc)
74
+ {
75
+ case 1:
76
+ error ("we won't substitute all variables on stdin for you");
77
+ /*
78
+ all_variables = 1;
79
+ subst_from_stdin ();
80
+ */
81
+ case 2:
82
+ /* echo '$foo and $bar' | git sh-i18n--envsubst --variables '$foo and $bar' */
83
+ all_variables = 0;
84
+ note_variables (argv[1]);
85
+ subst_from_stdin ();
86
+ break;
87
+ case 3:
88
+ /* git sh-i18n--envsubst --variables '$foo and $bar' */
89
+ if (strcmp(argv[1], "--variables"))
90
+ error ("first argument must be --variables when two are given");
91
+ show_variables = 1;
92
+ print_variables (argv[2]);
93
+ break;
94
+ default:
95
+ error ("too many arguments");
96
+ break;
97
+ }
98
+
99
+ /* Close standard error. This is simpler than fwriteerror_no_ebadf, because
100
+ upon failure we don't need an errno - all we can do at this point is to
101
+ set an exit status. */
102
+ errno = 0;
103
+ if (ferror (stderr) || fflush (stderr))
104
+ {
105
+ fclose (stderr);
106
+ exit (EXIT_FAILURE);
107
+ }
108
+ if (fclose (stderr) && errno != EBADF)
109
+ exit (EXIT_FAILURE);
110
+
111
+ exit (EXIT_SUCCESS);
112
+}
113
+
114
+/* Parse the string and invoke the callback each time a $VARIABLE or
115
+ ${VARIABLE} construct is seen, where VARIABLE is a nonempty sequence
116
+ of ASCII alphanumeric/underscore characters, starting with an ASCII
117
+ alphabetic/underscore character.
118
+ We allow only ASCII characters, to avoid dependencies w.r.t. the current
119
+ encoding: While "${\xe0}" looks like a variable access in ISO-8859-1
120
+ encoding, it doesn't look like one in the BIG5, BIG5-HKSCS, GBK, GB18030,
121
+ SHIFT_JIS, JOHAB encodings, because \xe0\x7d is a single character in these
122
+ encodings. */
123
+static void
124
+find_variables (const char *string,
125
+ void (*callback) (const char *var_ptr, size_t var_len))
126
+{
127
+ for (; *string != '\0';)
128
+ if (*string++ == '$')
129
+ {
130
+ const char *variable_start;
131
+ const char *variable_end;
132
+ unsigned short int valid;
133
+ char c;
134
+
135
+ if (*string == '{')
136
+ string++;
137
+
138
+ variable_start = string;
139
+ c = *string;
140
+ if ((c >= 'A' && c <= 'Z') || (c >= 'a' && c <= 'z') || c == '_')
141
+ {
142
+ do
143
+ c = *++string;
144
+ while ((c >= 'A' && c <= 'Z') || (c >= 'a' && c <= 'z')
145
+ || (c >= '0' && c <= '9') || c == '_');
146
+ variable_end = string;
147
+
148
+ if (variable_start[-1] == '{')
149
+ {
150
+ if (*string == '}')
151
+ {
152
+ string++;
153
+ valid = 1;
154
+ }
155
+ else
156
+ valid = 0;
157
+ }
158
+ else
159
+ valid = 1;
160
+
161
+ if (valid)
162
+ callback (variable_start, variable_end - variable_start);
163
+ }
164
+ }
165
+}
166
+
167
+
168
+/* Print a variable to stdout, followed by a newline. */
169
+static void
170
+print_variable (const char *var_ptr, size_t var_len)
171
+{
172
+ fwrite (var_ptr, var_len, 1, stdout);
173
+ putchar ('\n');
174
+}
175
+
176
+/* Print the variables contained in STRING to stdout, each one followed by a
177
+ newline. */
178
+static void
179
+print_variables (const char *string)
180
+{
181
+ find_variables (string, &print_variable);
182
+}
183
+
184
+
185
+/* Type describing list of immutable strings,
186
+ implemented using a dynamic array. */
187
+typedef struct string_list_ty string_list_ty;
188
+struct string_list_ty
189
+{
190
+ const char **item;
191
+ size_t nitems;
192
+ size_t nitems_max;
193
+};
194
+
195
+/* Initialize an empty list of strings. */
196
+static inline void
197
+string_list_init (string_list_ty *slp)
198
+{
199
+ slp->item = NULL;
200
+ slp->nitems = 0;
201
+ slp->nitems_max = 0;
202
+}
203
+
204
+/* Append a single string to the end of a list of strings. */
205
+static inline void
206
+string_list_append (string_list_ty *slp, const char *s)
207
+{
208
+ /* Grow the list. */
209
+ if (slp->nitems >= slp->nitems_max)
210
+ {
211
+ size_t nbytes;
212
+
213
+ slp->nitems_max = slp->nitems_max * 2 + 4;
214
+ nbytes = slp->nitems_max * sizeof (slp->item[0]);
215
+ slp->item = (const char **) xrealloc (slp->item, nbytes);
216
+ }
217
+
218
+ /* Add the string to the end of the list. */
219
+ slp->item[slp->nitems++] = s;
220
+}
221
+
222
+/* Compare two strings given by reference. */
223
+static int
224
+cmp_string (const void *pstr1, const void *pstr2)
225
+{
226
+ const char *str1 = *(const char **)pstr1;
227
+ const char *str2 = *(const char **)pstr2;
228
+
229
+ return strcmp (str1, str2);
230
+}
231
+
232
+/* Sort a list of strings. */
233
+static inline void
234
+string_list_sort (string_list_ty *slp)
235
+{
236
+ if (slp->nitems > 0)
237
+ qsort (slp->item, slp->nitems, sizeof (slp->item[0]), cmp_string);
238
+}
239
+
240
+/* Test whether a string list contains a given string. */
241
+static inline int
242
+string_list_member (const string_list_ty *slp, const char *s)
243
+{
244
+ size_t j;
245
+
246
+ for (j = 0; j < slp->nitems; ++j)
247
+ if (strcmp (slp->item[j], s) == 0)
248
+ return 1;
249
+ return 0;
250
+}
251
+
252
+/* Test whether a sorted string list contains a given string. */
253
+static int
254
+sorted_string_list_member (const string_list_ty *slp, const char *s)
255
+{
256
+ size_t j1, j2;
257
+
258
+ j1 = 0;
259
+ j2 = slp->nitems;
260
+ if (j2 > 0)
261
+ {
262
+ /* Binary search. */
263
+ while (j2 - j1 > 1)
264
+ {
265
+ /* Here we know that if s is in the list, it is at an index j
266
+ with j1 <= j < j2. */
267
+ size_t j = (j1 + j2) >> 1;
268
+ int result = strcmp (slp->item[j], s);
269
+
270
+ if (result > 0)
271
+ j2 = j;
272
+ else if (result == 0)
273
+ return 1;
274
+ else
275
+ j1 = j + 1;
276
+ }
277
+ if (j2 > j1)
278
+ if (strcmp (slp->item[j1], s) == 0)
279
+ return 1;
280
+ }
281
+ return 0;
282
+}
283
+
284
+
285
+/* Set of variables on which to perform substitution.
286
+ Used only if !all_variables. */
287
+static string_list_ty variables_set;
288
+
289
+/* Adds a variable to variables_set. */
290
+static void
291
+note_variable (const char *var_ptr, size_t var_len)
292
+{
293
+ char *string = xmalloc (var_len + 1);
294
+ memcpy (string, var_ptr, var_len);
295
+ string[var_len] = '\0';
296
+
297
+ string_list_append (&variables_set, string);
298
+}
299
+
300
+/* Stores the variables occurring in the string in variables_set. */
301
+static void
302
+note_variables (const char *string)
303
+{
304
+ string_list_init (&variables_set);
305
+ find_variables (string, ¬e_variable);
306
+ string_list_sort (&variables_set);
307
+}
308
+
309
+
310
+static int
311
+do_getc (void)
312
+{
313
+ int c = getc (stdin);
314
+
315
+ if (c == EOF)
316
+ {
317
+ if (ferror (stdin))
318
+ error ("error while reading standard input");
319
+ }
320
+
321
+ return c;
322
+}
323
+
324
+static inline void
325
+do_ungetc (int c)
326
+{
327
+ if (c != EOF)
328
+ ungetc (c, stdin);
329
+}
330
+
331
+/* Copies stdin to stdout, performing substitutions. */
332
+static void
333
+subst_from_stdin (void)
334
+{
335
+ static char *buffer;
336
+ static size_t bufmax;
337
+ static size_t buflen;
338
+ int c;
339
+
340
+ for (;;)
341
+ {
342
+ c = do_getc ();
343
+ if (c == EOF)
344
+ break;
345
+ /* Look for $VARIABLE or ${VARIABLE}. */
346
+ if (c == '$')
347
+ {
348
+ unsigned short int opening_brace = 0;
349
+ unsigned short int closing_brace = 0;
350
+
351
+ c = do_getc ();
352
+ if (c == '{')
353
+ {
354
+ opening_brace = 1;
355
+ c = do_getc ();
356
+ }
357
+ if ((c >= 'A' && c <= 'Z') || (c >= 'a' && c <= 'z') || c == '_')
358
+ {
359
+ unsigned short int valid;
360
+
361
+ /* Accumulate the VARIABLE in buffer. */
362
+ buflen = 0;
363
+ do
364
+ {
365
+ if (buflen >= bufmax)
366
+ {
367
+ bufmax = 2 * bufmax + 10;
368
+ buffer = xrealloc (buffer, bufmax);
369
+ }
370
+ buffer[buflen++] = c;
371
+
372
+ c = do_getc ();
373
+ }
374
+ while ((c >= 'A' && c <= 'Z') || (c >= 'a' && c <= 'z')
375
+ || (c >= '0' && c <= '9') || c == '_');
376
+
377
+ if (opening_brace)
378
+ {
379
+ if (c == '}')
380
+ {
381
+ closing_brace = 1;
382
+ valid = 1;
383
+ }
384
+ else
385
+ {
386
+ valid = 0;
387
+ do_ungetc (c);
388
+ }
389
+ }
390
+ else
391
+ {
392
+ valid = 1;
393
+ do_ungetc (c);
394
+ }
395
+
396
+ if (valid)
397
+ {
398
+ /* Terminate the variable in the buffer. */
399
+ if (buflen >= bufmax)
400
+ {
401
+ bufmax = 2 * bufmax + 10;
402
+ buffer = xrealloc (buffer, bufmax);
403
+ }
404
+ buffer[buflen] = '\0';
405
+
406
+ /* Test whether the variable shall be substituted. */
407
+ if (!all_variables
408
+ && !sorted_string_list_member (&variables_set, buffer))
409
+ valid = 0;
410
+ }
411
+
412
+ if (valid)
413
+ {
414
+ /* Substitute the variable's value from the environment. */
415
+ const char *env_value = getenv (buffer);
416
+
417
+ if (env_value != NULL)
418
+ fputs (env_value, stdout);
419
+ }
420
+ else
421
+ {
422
+ /* Perform no substitution at all. Since the buffered input
423
+ contains no other '$' than at the start, we can just
424
+ output all the buffered contents. */
425
+ putchar ('$');
426
+ if (opening_brace)
427
+ putchar ('{');
428
+ fwrite (buffer, buflen, 1, stdout);
429
+ if (closing_brace)
430
+ putchar ('}');
431
+ }
432
+ }
433
+ else
434
+ {
435
+ do_ungetc (c);
436
+ putchar ('$');
437
+ if (opening_brace)
438
+ putchar ('{');
439
+ }
440
+ }
441
+ else
442
+ putchar (c);
443
+ }
444
+}
t/t0201-gettext-fallbacks.sh
new
+51
@@ -0,0 +1,51 @@
1
+#!/bin/sh
2
+#
3
+# Copyright (c) 2010 Ævar Arnfjörð Bjarmason
4
+#
5
+
6
+test_description='Gettext Shell fallbacks'
7
+
8
+. ./test-lib.sh
9
+. "$GIT_BUILD_DIR"/git-sh-i18n
10
+
11
+test_expect_success 'gettext: our gettext() fallback has pass-through semantics' '
12
+ printf "test" >expect &&
13
+ gettext "test" >actual &&
14
+ test_i18ncmp expect actual &&
15
+ printf "test more words" >expect &&
16
+ gettext "test more words" >actual &&
17
+ test_i18ncmp expect actual
18
+'
19
+
20
+test_expect_success 'eval_gettext: our eval_gettext() fallback has pass-through semantics' '
21
+ printf "test" >expect &&
22
+ eval_gettext "test" >actual &&
23
+ test_i18ncmp expect actual &&
24
+ printf "test more words" >expect &&
25
+ eval_gettext "test more words" >actual &&
26
+ test_i18ncmp expect actual
27
+'
28
+
29
+test_expect_success 'eval_gettext: our eval_gettext() fallback can interpolate variables' '
30
+ printf "test YesPlease" >expect &&
31
+ GIT_INTERNAL_GETTEXT_TEST_FALLBACKS=YesPlease eval_gettext "test \$GIT_INTERNAL_GETTEXT_TEST_FALLBACKS" >actual &&
32
+ test_i18ncmp expect actual
33
+'
34
+
35
+test_expect_success 'eval_gettext: our eval_gettext() fallback can interpolate variables with spaces' '
36
+ cmdline="git am" &&
37
+ export cmdline;
38
+ printf "When you have resolved this problem run git am --resolved." >expect &&
39
+ eval_gettext "When you have resolved this problem run \$cmdline --resolved." >actual
40
+ test_i18ncmp expect actual
41
+'
42
+
43
+test_expect_success 'eval_gettext: our eval_gettext() fallback can interpolate variables with spaces and quotes' '
44
+ cmdline="git am" &&
45
+ export cmdline;
46
+ printf "When you have resolved this problem run \"git am --resolved\"." >expect &&
47
+ eval_gettext "When you have resolved this problem run \"\$cmdline --resolved\"." >actual
48
+ test_i18ncmp expect actual
49
+'
50
+
51
+test_done