convert.c

   1 #include "cache.h"
   2 #include "attr.h"
   3 #include "run-command.h"
   4
   5 /*
   6  * convert.c - convert a file when checking it out and checking it in.
   7  *
   8  * This should use the pathname to decide on whether it wants to do some
   9  * more interesting conversions (automatic gzip/unzip, general format
  10  * conversions etc etc), but by default it just does automatic CRLF<->LF
  11  * translation when the "auto_crlf" option is set.
  12  */
  13
  14 #define CRLF_GUESS      (-1)
  15 #define CRLF_BINARY     0
  16 #define CRLF_TEXT       1
  17 #define CRLF_INPUT      2
  18
  19 struct text_stat {
  20         /* NUL, CR, LF and CRLF counts */
  21         unsigned nul, cr, lf, crlf;
  22
  23         /* These are just approximations! */
  24         unsigned printable, nonprintable;
  25 };
  26
  27 static void gather_stats(const char *buf, unsigned long size, struct text_stat *stats)
  28 {
  29         unsigned long i;
  30
  31         memset(stats, 0, sizeof(*stats));
  32
  33         for (i = 0; i < size; i++) {
  34                 unsigned char c = buf[i];
  35                 if (c == '\r') {
  36                         stats->cr++;
  37                         if (i+1 < size && buf[i+1] == '\n')
  38                                 stats->crlf++;
  39                         continue;
  40                 }
  41                 if (c == '\n') {
  42                         stats->lf++;
  43                         continue;
  44                 }
  45                 if (c == 127)
  46                         /* DEL */
  47                         stats->nonprintable++;
  48                 else if (c < 32) {
  49                         switch (c) {
  50                                 /* BS, HT, ESC and FF */
  51                         case '\b': case '\t': case '\033': case '\014':
  52                                 stats->printable++;
  53                                 break;
  54                         case 0:
  55                                 stats->nul++;
  56                                 /* fall through */
  57                         default:
  58                                 stats->nonprintable++;
  59                         }
  60                 }
  61                 else
  62                         stats->printable++;
  63         }
  64 }
  65
  66 /*
  67  * The same heuristics as diff.c::mmfile_is_binary()
  68  */
  69 static int is_binary(unsigned long size, struct text_stat *stats)
  70 {
  71
  72         if (stats->nul)
  73                 return 1;
  74         if ((stats->printable >> 7) < stats->nonprintable)
  75                 return 1;
  76         /*
  77          * Other heuristics? Average line length might be relevant,
  78          * as might LF vs CR vs CRLF counts..
  79          *
  80          * NOTE! It might be normal to have a low ratio of CRLF to LF
  81          * (somebody starts with a LF-only file and edits it with an editor
  82          * that adds CRLF only to lines that are added..). But do  we
  83          * want to support CR-only? Probably not.
  84          */
  85         return 0;
  86 }
  87
  88 static void check_safe_crlf(const char *path, int action,
  89                             struct text_stat *stats, enum safe_crlf checksafe)
  90 {
  91         if (!checksafe)
  92                 return;
  93
  94         if (action == CRLF_INPUT || auto_crlf <= 0) {
  95                 /*
  96                  * CRLFs would not be restored by checkout:
  97                  * check if we'd remove CRLFs
  98                  */
  99                 if (stats->crlf) {
 100                         if (checksafe == SAFE_CRLF_WARN)
 101                                 warning("CRLF will be replaced by LF in %s.", path);
 102                         else /* i.e. SAFE_CRLF_FAIL */
 103                                 die("CRLF would be replaced by LF in %s.", path);
 104                 }
 105         } else if (auto_crlf > 0) {
 106                 /*
 107                  * CRLFs would be added by checkout:
 108                  * check if we have "naked" LFs
 109                  */
 110                 if (stats->lf != stats->crlf) {
 111                         if (checksafe == SAFE_CRLF_WARN)
 112                                 warning("LF will be replaced by CRLF in %s", path);
 113                         else /* i.e. SAFE_CRLF_FAIL */
 114                                 die("LF would be replaced by CRLF in %s", path);
 115                 }
 116         }
 117 }
 118
 119 static int crlf_to_git(const char *path, const char *src, size_t len,
 120                        struct strbuf *buf, int action, enum safe_crlf checksafe)
 121 {
 122         struct text_stat stats;
 123         char *dst;
 124
 125         if ((action == CRLF_BINARY) || !auto_crlf || !len)
 126                 return 0;
 127
 128         gather_stats(src, len, &stats);
 129
 130         if (action == CRLF_GUESS) {
 131                 /*
 132                  * We're currently not going to even try to convert stuff
 133                  * that has bare CR characters. Does anybody do that crazy
 134                  * stuff?
 135                  */
 136                 if (stats.cr != stats.crlf)
 137                         return 0;
 138
 139                 /*
 140                  * And add some heuristics for binary vs text, of course...
 141                  */
 142                 if (is_binary(len, &stats))
 143                         return 0;
 144         }
 145
 146         check_safe_crlf(path, action, &stats, checksafe);
 147
 148         /* Optimization: No CR? Nothing to convert, regardless. */
 149         if (!stats.cr)
 150                 return 0;
 151
 152         /* only grow if not in place */
 153         if (strbuf_avail(buf) + buf->len < len)
 154                 strbuf_grow(buf, len - buf->len);
 155         dst = buf->buf;
 156         if (action == CRLF_GUESS) {
 157                 /*
 158                  * If we guessed, we already know we rejected a file with
 159                  * lone CR, and we can strip a CR without looking at what
 160                  * follow it.
 161                  */
 162                 do {
 163                         unsigned char c = *src++;
 164                         if (c != '\r')
 165                                 *dst++ = c;
 166                 } while (--len);
 167         } else {
 168                 do {
 169                         unsigned char c = *src++;
 170                         if (! (c == '\r' && (1 < len && *src == '\n')))
 171                                 *dst++ = c;
 172                 } while (--len);
 173         }
 174         strbuf_setlen(buf, dst - buf->buf);
 175         return 1;
 176 }
 177
 178 static int crlf_to_worktree(const char *path, const char *src, size_t len,
 179                             struct strbuf *buf, int action)
 180 {
 181         char *to_free = NULL;
 182         struct text_stat stats;
 183
 184         if ((action == CRLF_BINARY) || (action == CRLF_INPUT) ||
 185             auto_crlf <= 0)
 186                 return 0;
 187
 188         if (!len)
 189                 return 0;
 190
 191         gather_stats(src, len, &stats);
 192
 193         /* No LF? Nothing to convert, regardless. */
 194         if (!stats.lf)
 195                 return 0;
 196
 197         /* Was it already in CRLF format? */
 198         if (stats.lf == stats.crlf)
 199                 return 0;
 200
 201         if (action == CRLF_GUESS) {
 202                 /* If we have any bare CR characters, we're not going to touch it */
 203                 if (stats.cr != stats.crlf)
 204                         return 0;
 205
 206                 if (is_binary(len, &stats))
 207                         return 0;
 208         }
 209
 210         /* are we "faking" in place editing ? */
 211         if (src == buf->buf)
 212                 to_free = strbuf_detach(buf, NULL);
 213
 214         strbuf_grow(buf, len + stats.lf - stats.crlf);
 215         for (;;) {
 216                 const char *nl = memchr(src, '\n', len);
 217                 if (!nl)
 218                         break;
 219                 if (nl > src && nl[-1] == '\r') {
 220                         strbuf_add(buf, src, nl + 1 - src);
 221                 } else {
 222                         strbuf_add(buf, src, nl - src);
 223                         strbuf_addstr(buf, "\r\n");
 224                 }
 225                 len -= nl + 1 - src;
 226                 src  = nl + 1;
 227         }
 228         strbuf_add(buf, src, len);
 229
 230         free(to_free);
 231         return 1;
 232 }
 233
 234 struct filter_params {
 235         const char *src;
 236         unsigned long size;
 237         const char *cmd;
 238 };
 239
 240 static int filter_buffer(int fd, void *data)
 241 {
 242         /*
 243          * Spawn cmd and feed the buffer contents through its stdin.
 244          */
 245         struct child_process child_process;
 246         struct filter_params *params = (struct filter_params *)data;
 247         int write_err, status;
 248         const char *argv[] = { "sh", "-c", params->cmd, NULL };
 249
 250         memset(&child_process, 0, sizeof(child_process));
 251         child_process.argv = argv;
 252         child_process.in = -1;
 253         child_process.out = fd;
 254
 255         if (start_command(&child_process))
 256                 return error("cannot fork to run external filter %s", params->cmd);
 257
 258         write_err = (write_in_full(child_process.in, params->src, params->size) < 0);
 259         if (close(child_process.in))
 260                 write_err = 1;
 261         if (write_err)
 262                 error("cannot feed the input to external filter %s", params->cmd);
 263
 264         status = finish_command(&child_process);
 265         if (status)
 266                 error("external filter %s failed %d", params->cmd, -status);
 267         return (write_err || status);
 268 }
 269
 270 static int apply_filter(const char *path, const char *src, size_t len,
 271                         struct strbuf *dst, const char *cmd)
 272 {
 273         /*
 274          * Create a pipeline to have the command filter the buffer's
 275          * contents.
 276          *
 277          * (child --> cmd) --> us
 278          */
 279         int ret = 1;
 280         struct strbuf nbuf;
 281         struct async async;
 282         struct filter_params params;
 283
 284         if (!cmd)
 285                 return 0;
 286
 287         memset(&async, 0, sizeof(async));
 288         async.proc = filter_buffer;
 289         async.data = &params;
 290         params.src = src;
 291         params.size = len;
 292         params.cmd = cmd;
 293
 294         fflush(NULL);
 295         if (start_async(&async))
 296                 return 0;       /* error was already reported */
 297
 298         strbuf_init(&nbuf, 0);
 299         if (strbuf_read(&nbuf, async.out, len) < 0) {
 300                 error("read from external filter %s failed", cmd);
 301                 ret = 0;
 302         }
 303         if (close(async.out)) {
 304                 error("read from external filter %s failed", cmd);
 305                 ret = 0;
 306         }
 307         if (finish_async(&async)) {
 308                 error("external filter %s failed", cmd);
 309                 ret = 0;
 310         }
 311
 312         if (ret) {
 313                 strbuf_swap(dst, &nbuf);
 314         }
 315         strbuf_release(&nbuf);
 316         return ret;
 317 }
 318
 319 static struct convert_driver {
 320         const char *name;
 321         struct convert_driver *next;
 322         char *smudge;
 323         char *clean;
 324 } *user_convert, **user_convert_tail;
 325
 326 static int read_convert_config(const char *var, const char *value)
 327 {
 328         const char *ep, *name;
 329         int namelen;
 330         struct convert_driver *drv;
 331
 332         /*
 333          * External conversion drivers are configured using
 334          * "filter.<name>.variable".
 335          */
 336         if (prefixcmp(var, "filter.") || (ep = strrchr(var, '.')) == var + 6)
 337                 return 0;
 338         name = var + 7;
 339         namelen = ep - name;
 340         for (drv = user_convert; drv; drv = drv->next)
 341                 if (!strncmp(drv->name, name, namelen) && !drv->name[namelen])
 342                         break;
 343         if (!drv) {
 344                 drv = xcalloc(1, sizeof(struct convert_driver));
 345                 drv->name = xmemdupz(name, namelen);
 346                 *user_convert_tail = drv;
 347                 user_convert_tail = &(drv->next);
 348         }
 349
 350         ep++;
 351
 352         /*
 353          * filter.<name>.smudge and filter.<name>.clean specifies
 354          * the command line:
 355          *
 356          *      command-line
 357          *
 358          * The command-line will not be interpolated in any way.
 359          */
 360
 361         if (!strcmp("smudge", ep)) {
 362                 if (!value)
 363                         return config_error_nonbool(var);
 364                 drv->smudge = strdup(value);
 365                 return 0;
 366         }
 367
 368         if (!strcmp("clean", ep)) {
 369                 if (!value)
 370                         return config_error_nonbool(var);
 371                 drv->clean = strdup(value);
 372                 return 0;
 373         }
 374         return 0;
 375 }
 376
 377 static void setup_convert_check(struct git_attr_check *check)
 378 {
 379         static struct git_attr *attr_crlf;
 380         static struct git_attr *attr_ident;
 381         static struct git_attr *attr_filter;
 382
 383         if (!attr_crlf) {
 384                 attr_crlf = git_attr("crlf", 4);
 385                 attr_ident = git_attr("ident", 5);
 386                 attr_filter = git_attr("filter", 6);
 387                 user_convert_tail = &user_convert;
 388                 git_config(read_convert_config);
 389         }
 390         check[0].attr = attr_crlf;
 391         check[1].attr = attr_ident;
 392         check[2].attr = attr_filter;
 393 }
 394
 395 static int count_ident(const char *cp, unsigned long size)
 396 {
 397         /*
 398          * "$Id: 0000000000000000000000000000000000000000 $" <=> "$Id$"
 399          */
 400         int cnt = 0;
 401         char ch;
 402
 403         while (size) {
 404                 ch = *cp++;
 405                 size--;
 406                 if (ch != '$')
 407                         continue;
 408                 if (size < 3)
 409                         break;
 410                 if (memcmp("Id", cp, 2))
 411                         continue;
 412                 ch = cp[2];
 413                 cp += 3;
 414                 size -= 3;
 415                 if (ch == '$')
 416                         cnt++; /* $Id$ */
 417                 if (ch != ':')
 418                         continue;
 419
 420                 /*
 421                  * "$Id: ... "; scan up to the closing dollar sign and discard.
 422                  */
 423                 while (size) {
 424                         ch = *cp++;
 425                         size--;
 426                         if (ch == '$') {
 427                                 cnt++;
 428                                 break;
 429                         }
 430                 }
 431         }
 432         return cnt;
 433 }
 434
 435 static int ident_to_git(const char *path, const char *src, size_t len,
 436                         struct strbuf *buf, int ident)
 437 {
 438         char *dst, *dollar;
 439
 440         if (!ident || !count_ident(src, len))
 441                 return 0;
 442
 443         /* only grow if not in place */
 444         if (strbuf_avail(buf) + buf->len < len)
 445                 strbuf_grow(buf, len - buf->len);
 446         dst = buf->buf;
 447         for (;;) {
 448                 dollar = memchr(src, '$', len);
 449                 if (!dollar)
 450                         break;
 451                 memcpy(dst, src, dollar + 1 - src);
 452                 dst += dollar + 1 - src;
 453                 len -= dollar + 1 - src;
 454                 src  = dollar + 1;
 455
 456                 if (len > 3 && !memcmp(src, "Id:", 3)) {
 457                         dollar = memchr(src + 3, '$', len - 3);
 458                         if (!dollar)
 459                                 break;
 460                         memcpy(dst, "Id$", 3);
 461                         dst += 3;
 462                         len -= dollar + 1 - src;
 463                         src  = dollar + 1;
 464                 }
 465         }
 466         memcpy(dst, src, len);
 467         strbuf_setlen(buf, dst + len - buf->buf);
 468         return 1;
 469 }
 470
 471 static int ident_to_worktree(const char *path, const char *src, size_t len,
 472                              struct strbuf *buf, int ident)
 473 {
 474         unsigned char sha1[20];
 475         char *to_free = NULL, *dollar;
 476         int cnt;
 477
 478         if (!ident)
 479                 return 0;
 480
 481         cnt = count_ident(src, len);
 482         if (!cnt)
 483                 return 0;
 484
 485         /* are we "faking" in place editing ? */
 486         if (src == buf->buf)
 487                 to_free = strbuf_detach(buf, NULL);
 488         hash_sha1_file(src, len, "blob", sha1);
 489
 490         strbuf_grow(buf, len + cnt * 43);
 491         for (;;) {
 492                 /* step 1: run to the next '$' */
 493                 dollar = memchr(src, '$', len);
 494                 if (!dollar)
 495                         break;
 496                 strbuf_add(buf, src, dollar + 1 - src);
 497                 len -= dollar + 1 - src;
 498                 src  = dollar + 1;
 499
 500                 /* step 2: does it looks like a bit like Id:xxx$ or Id$ ? */
 501                 if (len < 3 || memcmp("Id", src, 2))
 502                         continue;
 503
 504                 /* step 3: skip over Id$ or Id:xxxxx$ */
 505                 if (src[2] == '$') {
 506                         src += 3;
 507                         len -= 3;
 508                 } else if (src[2] == ':') {
 509                         /*
 510                          * It's possible that an expanded Id has crept its way into the
 511                          * repository, we cope with that by stripping the expansion out
 512                          */
 513                         dollar = memchr(src + 3, '$', len - 3);
 514                         if (!dollar) {
 515                                 /* incomplete keyword, no more '$', so just quit the loop */
 516                                 break;
 517                         }
 518
 519                         len -= dollar + 1 - src;
 520                         src  = dollar + 1;
 521                 } else {
 522                         /* it wasn't a "Id$" or "Id:xxxx$" */
 523                         continue;
 524                 }
 525
 526                 /* step 4: substitute */
 527                 strbuf_addstr(buf, "Id: ");
 528                 strbuf_add(buf, sha1_to_hex(sha1), 40);
 529                 strbuf_addstr(buf, " $");
 530         }
 531         strbuf_add(buf, src, len);
 532
 533         free(to_free);
 534         return 1;
 535 }
 536
 537 static int git_path_check_crlf(const char *path, struct git_attr_check *check)
 538 {
 539         const char *value = check->value;
 540
 541         if (ATTR_TRUE(value))
 542                 return CRLF_TEXT;
 543         else if (ATTR_FALSE(value))
 544                 return CRLF_BINARY;
 545         else if (ATTR_UNSET(value))
 546                 ;
 547         else if (!strcmp(value, "input"))
 548                 return CRLF_INPUT;
 549         return CRLF_GUESS;
 550 }
 551
 552 static struct convert_driver *git_path_check_convert(const char *path,
 553                                              struct git_attr_check *check)
 554 {
 555         const char *value = check->value;
 556         struct convert_driver *drv;
 557
 558         if (ATTR_TRUE(value) || ATTR_FALSE(value) || ATTR_UNSET(value))
 559                 return NULL;
 560         for (drv = user_convert; drv; drv = drv->next)
 561                 if (!strcmp(value, drv->name))
 562                         return drv;
 563         return NULL;
 564 }
 565
 566 static int git_path_check_ident(const char *path, struct git_attr_check *check)
 567 {
 568         const char *value = check->value;
 569
 570         return !!ATTR_TRUE(value);
 571 }
 572
 573 int convert_to_git(const char *path, const char *src, size_t len,
 574                    struct strbuf *dst, enum safe_crlf checksafe)
 575 {
 576         struct git_attr_check check[3];
 577         int crlf = CRLF_GUESS;
 578         int ident = 0, ret = 0;
 579         char *filter = NULL;
 580
 581         setup_convert_check(check);
 582         if (!git_checkattr(path, ARRAY_SIZE(check), check)) {
 583                 struct convert_driver *drv;
 584                 crlf = git_path_check_crlf(path, check + 0);
 585                 ident = git_path_check_ident(path, check + 1);
 586                 drv = git_path_check_convert(path, check + 2);
 587                 if (drv && drv->clean)
 588                         filter = drv->clean;
 589         }
 590
 591         ret |= apply_filter(path, src, len, dst, filter);
 592         if (ret) {
 593                 src = dst->buf;
 594                 len = dst->len;
 595         }
 596         ret |= crlf_to_git(path, src, len, dst, crlf, checksafe);
 597         if (ret) {
 598                 src = dst->buf;
 599                 len = dst->len;
 600         }
 601         return ret | ident_to_git(path, src, len, dst, ident);
 602 }
 603
 604 int convert_to_working_tree(const char *path, const char *src, size_t len, struct strbuf *dst)
 605 {
 606         struct git_attr_check check[3];
 607         int crlf = CRLF_GUESS;
 608         int ident = 0, ret = 0;
 609         char *filter = NULL;
 610
 611         setup_convert_check(check);
 612         if (!git_checkattr(path, ARRAY_SIZE(check), check)) {
 613                 struct convert_driver *drv;
 614                 crlf = git_path_check_crlf(path, check + 0);
 615                 ident = git_path_check_ident(path, check + 1);
 616                 drv = git_path_check_convert(path, check + 2);
 617                 if (drv && drv->smudge)
 618                         filter = drv->smudge;
 619         }
 620
 621         ret |= ident_to_worktree(path, src, len, dst, ident);
 622         if (ret) {
 623                 src = dst->buf;
 624                 len = dst->len;
 625         }
 626         ret |= crlf_to_worktree(path, src, len, dst, crlf);
 627         if (ret) {
 628                 src = dst->buf;
 629                 len = dst->len;
 630         }
 631         return ret | apply_filter(path, src, len, dst, filter);
 632 }