builtin/patch-id.c

   1 #include "builtin.h"
   2 #include "config.h"
   3 #include "diff.h"
   4 #include "gettext.h"
   5 #include "hash.h"
   6 #include "hex.h"
   7 #include "parse-options.h"
   8 #include "setup.h"
   9
  10 static void flush_current_id(int patchlen, struct object_id *id, struct object_id *result)
  11 {
  12         if (patchlen)
  13                 printf("%s %s\n", oid_to_hex(result), oid_to_hex(id));
  14 }
  15
  16 static int remove_space(char *line)
  17 {
  18         char *src = line;
  19         char *dst = line;
  20         unsigned char c;
  21
  22         while ((c = *src++) != '\0') {
  23                 if (!isspace(c))
  24                         *dst++ = c;
  25         }
  26         return dst - line;
  27 }
  28
  29 static int scan_hunk_header(const char *p, int *p_before, int *p_after)
  30 {
  31         static const char digits[] = "0123456789";
  32         const char *q, *r;
  33         int n;
  34
  35         q = p + 4;
  36         n = strspn(q, digits);
  37         if (q[n] == ',') {
  38                 q += n + 1;
  39                 *p_before = atoi(q);
  40                 n = strspn(q, digits);
  41         } else {
  42                 *p_before = 1;
  43         }
  44
  45         if (n == 0 || q[n] != ' ' || q[n+1] != '+')
  46                 return 0;
  47
  48         r = q + n + 2;
  49         n = strspn(r, digits);
  50         if (r[n] == ',') {
  51                 r += n + 1;
  52                 *p_after = atoi(r);
  53                 n = strspn(r, digits);
  54         } else {
  55                 *p_after = 1;
  56         }
  57         if (n == 0)
  58                 return 0;
  59
  60         return 1;
  61 }
  62
  63 static int get_one_patchid(struct object_id *next_oid, struct object_id *result,
  64                            struct strbuf *line_buf, int stable, int verbatim)
  65 {
  66         int patchlen = 0, found_next = 0;
  67         int before = -1, after = -1;
  68         int diff_is_binary = 0;
  69         char pre_oid_str[GIT_MAX_HEXSZ + 1], post_oid_str[GIT_MAX_HEXSZ + 1];
  70         git_hash_ctx ctx;
  71
  72         the_hash_algo->init_fn(&ctx);
  73         oidclr(result, the_repository->hash_algo);
  74
  75         while (strbuf_getwholeline(line_buf, stdin, '\n') != EOF) {
  76                 char *line = line_buf->buf;
  77                 const char *p = line;
  78                 int len;
  79
  80                 /* Possibly skip over the prefix added by "log" or "format-patch" */
  81                 if (!skip_prefix(line, "commit ", &p) &&
  82                     !skip_prefix(line, "From ", &p) &&
  83                     starts_with(line, "\\ ") && 12 < strlen(line)) {
  84                         if (verbatim)
  85                                 the_hash_algo->update_fn(&ctx, line, strlen(line));
  86                         continue;
  87                 }
  88
  89                 if (!get_oid_hex(p, next_oid)) {
  90                         found_next = 1;
  91                         break;
  92                 }
  93
  94                 /* Ignore commit comments */
  95                 if (!patchlen && !starts_with(line, "diff "))
  96                         continue;
  97
  98                 /* Parsing diff header?  */
  99                 if (before == -1) {
 100                         if (starts_with(line, "GIT binary patch") ||
 101                             starts_with(line, "Binary files")) {
 102                                 diff_is_binary = 1;
 103                                 before = 0;
 104                                 the_hash_algo->update_fn(&ctx, pre_oid_str,
 105                                                          strlen(pre_oid_str));
 106                                 the_hash_algo->update_fn(&ctx, post_oid_str,
 107                                                          strlen(post_oid_str));
 108                                 if (stable)
 109                                         flush_one_hunk(result, &ctx);
 110                                 continue;
 111                         } else if (skip_prefix(line, "index ", &p)) {
 112                                 char *oid1_end = strstr(line, "..");
 113                                 char *oid2_end = NULL;
 114                                 if (oid1_end)
 115                                         oid2_end = strstr(oid1_end, " ");
 116                                 if (!oid2_end)
 117                                         oid2_end = line + strlen(line) - 1;
 118                                 if (oid1_end != NULL && oid2_end != NULL) {
 119                                         *oid1_end = *oid2_end = '\0';
 120                                         strlcpy(pre_oid_str, p, GIT_MAX_HEXSZ + 1);
 121                                         strlcpy(post_oid_str, oid1_end + 2, GIT_MAX_HEXSZ + 1);
 122                                 }
 123                                 continue;
 124                         } else if (starts_with(line, "--- "))
 125                                 before = after = 1;
 126                         else if (!isalpha(line[0]))
 127                                 break;
 128                 }
 129
 130                 if (diff_is_binary) {
 131                         if (starts_with(line, "diff ")) {
 132                                 diff_is_binary = 0;
 133                                 before = -1;
 134                         }
 135                         continue;
 136                 }
 137
 138                 /* Looking for a valid hunk header?  */
 139                 if (before == 0 && after == 0) {
 140                         if (starts_with(line, "@@ -")) {
 141                                 /* Parse next hunk, but ignore line numbers.  */
 142                                 scan_hunk_header(line, &before, &after);
 143                                 continue;
 144                         }
 145
 146                         /* Split at the end of the patch.  */
 147                         if (!starts_with(line, "diff "))
 148                                 break;
 149
 150                         /* Else we're parsing another header.  */
 151                         if (stable)
 152                                 flush_one_hunk(result, &ctx);
 153                         before = after = -1;
 154                 }
 155
 156                 /* If we get here, we're inside a hunk.  */
 157                 if (line[0] == '-' || line[0] == ' ')
 158                         before--;
 159                 if (line[0] == '+' || line[0] == ' ')
 160                         after--;
 161
 162                 /* Add line to hash algo (possibly removing whitespace) */
 163                 len = verbatim ? strlen(line) : remove_space(line);
 164                 patchlen += len;
 165                 the_hash_algo->update_fn(&ctx, line, len);
 166         }
 167
 168         if (!found_next)
 169                 oidclr(next_oid, the_repository->hash_algo);
 170
 171         flush_one_hunk(result, &ctx);
 172
 173         return patchlen;
 174 }
 175
 176 static void generate_id_list(int stable, int verbatim)
 177 {
 178         struct object_id oid, n, result;
 179         int patchlen;
 180         struct strbuf line_buf = STRBUF_INIT;
 181
 182         oidclr(&oid, the_repository->hash_algo);
 183         while (!feof(stdin)) {
 184                 patchlen = get_one_patchid(&n, &result, &line_buf, stable, verbatim);
 185                 flush_current_id(patchlen, &oid, &result);
 186                 oidcpy(&oid, &n);
 187         }
 188         strbuf_release(&line_buf);
 189 }
 190
 191 static const char *const patch_id_usage[] = {
 192         N_("git patch-id [--stable | --unstable | --verbatim]"), NULL
 193 };
 194
 195 struct patch_id_opts {
 196         int stable;
 197         int verbatim;
 198 };
 199
 200 static int git_patch_id_config(const char *var, const char *value,
 201                                const struct config_context *ctx, void *cb)
 202 {
 203         struct patch_id_opts *opts = cb;
 204
 205         if (!strcmp(var, "patchid.stable")) {
 206                 opts->stable = git_config_bool(var, value);
 207                 return 0;
 208         }
 209         if (!strcmp(var, "patchid.verbatim")) {
 210                 opts->verbatim = git_config_bool(var, value);
 211                 return 0;
 212         }
 213
 214         return git_default_config(var, value, ctx, cb);
 215 }
 216
 217 int cmd_patch_id(int argc, const char **argv, const char *prefix)
 218 {
 219         /* if nothing is set, default to unstable */
 220         struct patch_id_opts config = {0, 0};
 221         int opts = 0;
 222         struct option builtin_patch_id_options[] = {
 223                 OPT_CMDMODE(0, "unstable", &opts,
 224                     N_("use the unstable patch-id algorithm"), 1),
 225                 OPT_CMDMODE(0, "stable", &opts,
 226                     N_("use the stable patch-id algorithm"), 2),
 227                 OPT_CMDMODE(0, "verbatim", &opts,
 228                         N_("don't strip whitespace from the patch"), 3),
 229                 OPT_END()
 230         };
 231
 232         git_config(git_patch_id_config, &config);
 233
 234         /* verbatim implies stable */
 235         if (config.verbatim)
 236                 config.stable = 1;
 237
 238         argc = parse_options(argc, argv, prefix, builtin_patch_id_options,
 239                              patch_id_usage, 0);
 240
 241         /*
 242          * We rely on `the_hash_algo` to compute patch IDs. This is dubious as
 243          * it means that the hash algorithm now depends on the object hash of
 244          * the repository, even though git-patch-id(1) clearly defines that
 245          * patch IDs always use SHA1.
 246          *
 247          * NEEDSWORK: This hack should be removed in favor of converting
 248          * the code that computes patch IDs to always use SHA1.
 249          */
 250         if (!the_hash_algo)
 251                 repo_set_hash_algo(the_repository, GIT_HASH_SHA1);
 252
 253         generate_id_list(opts ? opts > 1 : config.stable,
 254                          opts ? opts == 3 : config.verbatim);
 255         return 0;
 256 }