builtin/cat-file.c

   1 /*
   2  * GIT - The information manager from hell
   3  *
   4  * Copyright (C) Linus Torvalds, 2005
   5  */
   6 #include "cache.h"
   7 #include "exec_cmd.h"
   8 #include "tag.h"
   9 #include "tree.h"
  10 #include "builtin.h"
  11 #include "parse-options.h"
  12 #include "diff.h"
  13 #include "userdiff.h"
  14 #include "streaming.h"
  15
  16 static int cat_one_file(int opt, const char *exp_type, const char *obj_name)
  17 {
  18         unsigned char sha1[20];
  19         enum object_type type;
  20         char *buf;
  21         unsigned long size;
  22         struct object_context obj_context;
  23
  24         if (get_sha1_with_context(obj_name, 0, sha1, &obj_context))
  25                 die("Not a valid object name %s", obj_name);
  26
  27         buf = NULL;
  28         switch (opt) {
  29         case 't':
  30                 type = sha1_object_info(sha1, NULL);
  31                 if (type > 0) {
  32                         printf("%s\n", typename(type));
  33                         return 0;
  34                 }
  35                 break;
  36
  37         case 's':
  38                 type = sha1_object_info(sha1, &size);
  39                 if (type > 0) {
  40                         printf("%lu\n", size);
  41                         return 0;
  42                 }
  43                 break;
  44
  45         case 'e':
  46                 return !has_sha1_file(sha1);
  47
  48         case 'p':
  49                 type = sha1_object_info(sha1, NULL);
  50                 if (type < 0)
  51                         die("Not a valid object name %s", obj_name);
  52
  53                 /* custom pretty-print here */
  54                 if (type == OBJ_TREE) {
  55                         const char *ls_args[3] = { NULL };
  56                         ls_args[0] =  "ls-tree";
  57                         ls_args[1] =  obj_name;
  58                         return cmd_ls_tree(2, ls_args, NULL);
  59                 }
  60
  61                 if (type == OBJ_BLOB)
  62                         return stream_blob_to_fd(1, sha1, NULL, 0);
  63                 buf = read_sha1_file(sha1, &type, &size);
  64                 if (!buf)
  65                         die("Cannot read object %s", obj_name);
  66
  67                 /* otherwise just spit out the data */
  68                 break;
  69
  70         case 'c':
  71                 if (!obj_context.path[0])
  72                         die("git cat-file --textconv %s: <object> must be <sha1:path>",
  73                             obj_name);
  74
  75                 if (!textconv_object(obj_context.path, obj_context.mode, sha1, 1, &buf, &size))
  76                         die("git cat-file --textconv: unable to run textconv on %s",
  77                             obj_name);
  78                 break;
  79
  80         case 0:
  81                 if (type_from_string(exp_type) == OBJ_BLOB) {
  82                         unsigned char blob_sha1[20];
  83                         if (sha1_object_info(sha1, NULL) == OBJ_TAG) {
  84                                 enum object_type type;
  85                                 unsigned long size;
  86                                 char *buffer = read_sha1_file(sha1, &type, &size);
  87                                 if (memcmp(buffer, "object ", 7) ||
  88                                     get_sha1_hex(buffer + 7, blob_sha1))
  89                                         die("%s not a valid tag", sha1_to_hex(sha1));
  90                                 free(buffer);
  91                         } else
  92                                 hashcpy(blob_sha1, sha1);
  93
  94                         if (sha1_object_info(blob_sha1, NULL) == OBJ_BLOB)
  95                                 return stream_blob_to_fd(1, blob_sha1, NULL, 0);
  96                         /*
  97                          * we attempted to dereference a tag to a blob
  98                          * and failed; there may be new dereference
  99                          * mechanisms this code is not aware of.
 100                          * fall-back to the usual case.
 101                          */
 102                 }
 103                 buf = read_object_with_reference(sha1, exp_type, &size, NULL);
 104                 break;
 105
 106         default:
 107                 die("git cat-file: unknown option: %s", exp_type);
 108         }
 109
 110         if (!buf)
 111                 die("git cat-file %s: bad file", obj_name);
 112
 113         write_or_die(1, buf, size);
 114         return 0;
 115 }
 116
 117 struct expand_data {
 118         unsigned char sha1[20];
 119         enum object_type type;
 120         unsigned long size;
 121         unsigned long disk_size;
 122         const char *rest;
 123
 124         /*
 125          * If mark_query is true, we do not expand anything, but rather
 126          * just mark the object_info with items we wish to query.
 127          */
 128         int mark_query;
 129
 130         /*
 131          * Whether to split the input on whitespace before feeding it to
 132          * get_sha1; this is decided during the mark_query phase based on
 133          * whether we have a %(rest) token in our format.
 134          */
 135         int split_on_whitespace;
 136
 137         /*
 138          * After a mark_query run, this object_info is set up to be
 139          * passed to sha1_object_info_extended. It will point to the data
 140          * elements above, so you can retrieve the response from there.
 141          */
 142         struct object_info info;
 143 };
 144
 145 static int is_atom(const char *atom, const char *s, int slen)
 146 {
 147         int alen = strlen(atom);
 148         return alen == slen && !memcmp(atom, s, alen);
 149 }
 150
 151 static void expand_atom(struct strbuf *sb, const char *atom, int len,
 152                         void *vdata)
 153 {
 154         struct expand_data *data = vdata;
 155
 156         if (is_atom("objectname", atom, len)) {
 157                 if (!data->mark_query)
 158                         strbuf_addstr(sb, sha1_to_hex(data->sha1));
 159         } else if (is_atom("objecttype", atom, len)) {
 160                 if (data->mark_query)
 161                         data->info.typep = &data->type;
 162                 else
 163                         strbuf_addstr(sb, typename(data->type));
 164         } else if (is_atom("objectsize", atom, len)) {
 165                 if (data->mark_query)
 166                         data->info.sizep = &data->size;
 167                 else
 168                         strbuf_addf(sb, "%lu", data->size);
 169         } else if (is_atom("objectsize:disk", atom, len)) {
 170                 if (data->mark_query)
 171                         data->info.disk_sizep = &data->disk_size;
 172                 else
 173                         strbuf_addf(sb, "%lu", data->disk_size);
 174         } else if (is_atom("rest", atom, len)) {
 175                 if (data->mark_query)
 176                         data->split_on_whitespace = 1;
 177                 else if (data->rest)
 178                         strbuf_addstr(sb, data->rest);
 179         } else
 180                 die("unknown format element: %.*s", len, atom);
 181 }
 182
 183 static size_t expand_format(struct strbuf *sb, const char *start, void *data)
 184 {
 185         const char *end;
 186
 187         if (*start != '(')
 188                 return 0;
 189         end = strchr(start + 1, ')');
 190         if (!end)
 191                 die("format element '%s' does not end in ')'", start);
 192
 193         expand_atom(sb, start + 1, end - start - 1, data);
 194
 195         return end - start + 1;
 196 }
 197
 198 static void print_object_or_die(int fd, const unsigned char *sha1,
 199                                 enum object_type type, unsigned long size)
 200 {
 201         if (type == OBJ_BLOB) {
 202                 if (stream_blob_to_fd(fd, sha1, NULL, 0) < 0)
 203                         die("unable to stream %s to stdout", sha1_to_hex(sha1));
 204         }
 205         else {
 206                 enum object_type rtype;
 207                 unsigned long rsize;
 208                 void *contents;
 209
 210                 contents = read_sha1_file(sha1, &rtype, &rsize);
 211                 if (!contents)
 212                         die("object %s disappeared", sha1_to_hex(sha1));
 213                 if (rtype != type)
 214                         die("object %s changed type!?", sha1_to_hex(sha1));
 215                 if (rsize != size)
 216                         die("object %s change size!?", sha1_to_hex(sha1));
 217
 218                 write_or_die(fd, contents, size);
 219                 free(contents);
 220         }
 221 }
 222
 223 struct batch_options {
 224         int enabled;
 225         int print_contents;
 226         const char *format;
 227 };
 228
 229 static int batch_one_object(const char *obj_name, struct batch_options *opt,
 230                             struct expand_data *data)
 231 {
 232         struct strbuf buf = STRBUF_INIT;
 233
 234         if (!obj_name)
 235            return 1;
 236
 237         if (get_sha1(obj_name, data->sha1)) {
 238                 printf("%s missing\n", obj_name);
 239                 fflush(stdout);
 240                 return 0;
 241         }
 242
 243         if (sha1_object_info_extended(data->sha1, &data->info) < 0) {
 244                 printf("%s missing\n", obj_name);
 245                 fflush(stdout);
 246                 return 0;
 247         }
 248
 249         strbuf_expand(&buf, opt->format, expand_format, data);
 250         strbuf_addch(&buf, '\n');
 251         write_or_die(1, buf.buf, buf.len);
 252         strbuf_release(&buf);
 253
 254         if (opt->print_contents) {
 255                 print_object_or_die(1, data->sha1, data->type, data->size);
 256                 write_or_die(1, "\n", 1);
 257         }
 258         return 0;
 259 }
 260
 261 static int batch_objects(struct batch_options *opt)
 262 {
 263         struct strbuf buf = STRBUF_INIT;
 264         struct expand_data data;
 265
 266         if (!opt->format)
 267                 opt->format = "%(objectname) %(objecttype) %(objectsize)";
 268
 269         /*
 270          * Expand once with our special mark_query flag, which will prime the
 271          * object_info to be handed to sha1_object_info_extended for each
 272          * object.
 273          */
 274         memset(&data, 0, sizeof(data));
 275         data.mark_query = 1;
 276         strbuf_expand(&buf, opt->format, expand_format, &data);
 277         data.mark_query = 0;
 278
 279         /*
 280          * We are going to call get_sha1 on a potentially very large number of
 281          * objects. In most large cases, these will be actual object sha1s. The
 282          * cost to double-check that each one is not also a ref (just so we can
 283          * warn) ends up dwarfing the actual cost of the object lookups
 284          * themselves. We can work around it by just turning off the warning.
 285          */
 286         warn_on_object_refname_ambiguity = 0;
 287
 288         while (strbuf_getline(&buf, stdin, '\n') != EOF) {
 289                 int error;
 290
 291                 if (data.split_on_whitespace) {
 292                         /*
 293                          * Split at first whitespace, tying off the beginning
 294                          * of the string and saving the remainder (or NULL) in
 295                          * data.rest.
 296                          */
 297                         char *p = strpbrk(buf.buf, " \t");
 298                         if (p) {
 299                                 while (*p && strchr(" \t", *p))
 300                                         *p++ = '\0';
 301                         }
 302                         data.rest = p;
 303                 }
 304
 305                 error = batch_one_object(buf.buf, opt, &data);
 306                 if (error)
 307                         return error;
 308         }
 309
 310         return 0;
 311 }
 312
 313 static const char * const cat_file_usage[] = {
 314         N_("git cat-file (-t|-s|-e|-p|<type>|--textconv) <object>"),
 315         N_("git cat-file (--batch|--batch-check) < <list_of_objects>"),
 316         NULL
 317 };
 318
 319 static int git_cat_file_config(const char *var, const char *value, void *cb)
 320 {
 321         if (userdiff_config(var, value) < 0)
 322                 return -1;
 323
 324         return git_default_config(var, value, cb);
 325 }
 326
 327 static int batch_option_callback(const struct option *opt,
 328                                  const char *arg,
 329                                  int unset)
 330 {
 331         struct batch_options *bo = opt->value;
 332
 333         if (unset) {
 334                 memset(bo, 0, sizeof(*bo));
 335                 return 0;
 336         }
 337
 338         bo->enabled = 1;
 339         bo->print_contents = !strcmp(opt->long_name, "batch");
 340         bo->format = arg;
 341
 342         return 0;
 343 }
 344
 345 int cmd_cat_file(int argc, const char **argv, const char *prefix)
 346 {
 347         int opt = 0;
 348         const char *exp_type = NULL, *obj_name = NULL;
 349         struct batch_options batch = {0};
 350
 351         const struct option options[] = {
 352                 OPT_GROUP(N_("<type> can be one of: blob, tree, commit, tag")),
 353                 OPT_SET_INT('t', NULL, &opt, N_("show object type"), 't'),
 354                 OPT_SET_INT('s', NULL, &opt, N_("show object size"), 's'),
 355                 OPT_SET_INT('e', NULL, &opt,
 356                             N_("exit with zero when there's no error"), 'e'),
 357                 OPT_SET_INT('p', NULL, &opt, N_("pretty-print object's content"), 'p'),
 358                 OPT_SET_INT(0, "textconv", &opt,
 359                             N_("for blob objects, run textconv on object's content"), 'c'),
 360                 { OPTION_CALLBACK, 0, "batch", &batch, "format",
 361                         N_("show info and content of objects fed from the standard input"),
 362                         PARSE_OPT_OPTARG, batch_option_callback },
 363                 { OPTION_CALLBACK, 0, "batch-check", &batch, "format",
 364                         N_("show info about objects fed from the standard input"),
 365                         PARSE_OPT_OPTARG, batch_option_callback },
 366                 OPT_END()
 367         };
 368
 369         git_config(git_cat_file_config, NULL);
 370
 371         if (argc != 3 && argc != 2)
 372                 usage_with_options(cat_file_usage, options);
 373
 374         argc = parse_options(argc, argv, prefix, options, cat_file_usage, 0);
 375
 376         if (opt) {
 377                 if (argc == 1)
 378                         obj_name = argv[0];
 379                 else
 380                         usage_with_options(cat_file_usage, options);
 381         }
 382         if (!opt && !batch.enabled) {
 383                 if (argc == 2) {
 384                         exp_type = argv[0];
 385                         obj_name = argv[1];
 386                 } else
 387                         usage_with_options(cat_file_usage, options);
 388         }
 389         if (batch.enabled && (opt || argc)) {
 390                 usage_with_options(cat_file_usage, options);
 391         }
 392
 393         if (batch.enabled)
 394                 return batch_objects(&batch);
 395
 396         return cat_one_file(opt, exp_type, obj_name);
 397 }