X-Git-Url: https://git.octo.it/?a=blobdiff_plain;f=builtin-grep.c;h=5fac5701e6ccebdbe83b00d2e2df19343d3aea0e;hb=fd7e9fb7ae206a64b87c7faecfc88716e98a7577;hp=4be1514a4e5ba386ddd0d9c37bbe6e60000504dd;hpb=1362671f6a716e1a34abbcab43adfb68c7e2fd3b;p=git.git diff --git a/builtin-grep.c b/builtin-grep.c index 4be1514a..5fac5701 100644 --- a/builtin-grep.c +++ b/builtin-grep.c @@ -12,6 +12,7 @@ #include "builtin.h" #include #include +#include /* * git grep pathspecs are somewhat different from diff-tree pathspecs; @@ -26,7 +27,7 @@ static int pathspec_matches(const char **paths, const char *name) for (i = 0; paths[i]; i++) { const char *match = paths[i]; int matchlen = strlen(match); - const char *slash, *cp; + const char *cp, *meta; if ((matchlen <= namelen) && !strncmp(name, match, matchlen) && @@ -38,54 +39,110 @@ static int pathspec_matches(const char **paths, const char *name) if (name[namelen-1] != '/') continue; - /* We are being asked if the name directory is worth + /* We are being asked if the directory ("name") is worth * descending into. * * Find the longest leading directory name that does * not have metacharacter in the pathspec; the name * we are looking at must overlap with that directory. */ - for (cp = match, slash = NULL; cp - match < matchlen; cp++) { + for (cp = match, meta = NULL; cp - match < matchlen; cp++) { char ch = *cp; - if (ch == '/') - slash = cp; - if (ch == '*' || ch == '[') + if (ch == '*' || ch == '[' || ch == '?') { + meta = cp; break; + } } - if (!slash) - slash = match; /* toplevel */ - else - slash++; - if (namelen <= slash - match) { + if (!meta) + meta = cp; /* fully literal */ + + if (namelen <= meta - match) { /* Looking at "Documentation/" and * the pattern says "Documentation/howto/", or - * "Documentation/diff*.txt". + * "Documentation/diff*.txt". The name we + * have should match prefix. */ if (!memcmp(match, name, namelen)) return 1; + continue; } - else { + + if (meta - match < namelen) { /* Looking at "Documentation/howto/" and - * the pattern says "Documentation/h*". + * the pattern says "Documentation/h*"; + * match up to "Do.../h"; this avoids descending + * into "Documentation/technical/". */ - if (!memcmp(match, name, slash - match)) + if (!memcmp(match, name, meta - match)) return 1; + continue; } } return 0; } -struct grep_opt { +struct grep_pat { + struct grep_pat *next; + const char *origin; + int no; const char *pattern; regex_t regexp; +}; + +struct grep_opt { + struct grep_pat *pattern_list; + struct grep_pat **pattern_tail; + regex_t regexp; unsigned linenum:1; unsigned invert:1; unsigned name_only:1; + unsigned unmatch_name_only:1; + unsigned count:1; + unsigned word_regexp:1; + unsigned fixed:1; +#define GREP_BINARY_DEFAULT 0 +#define GREP_BINARY_NOMATCH 1 +#define GREP_BINARY_TEXT 2 + unsigned binary:2; int regflags; unsigned pre_context; unsigned post_context; }; +static void add_pattern(struct grep_opt *opt, const char *pat, + const char *origin, int no) +{ + struct grep_pat *p = xcalloc(1, sizeof(*p)); + p->pattern = pat; + p->origin = origin; + p->no = no; + *opt->pattern_tail = p; + opt->pattern_tail = &p->next; + p->next = NULL; +} + +static void compile_patterns(struct grep_opt *opt) +{ + struct grep_pat *p; + for (p = opt->pattern_list; p; p = p->next) { + int err = regcomp(&p->regexp, p->pattern, opt->regflags); + if (err) { + char errbuf[1024]; + char where[1024]; + if (p->no) + sprintf(where, "In '%s' at %d, ", + p->origin, p->no); + else if (p->origin) + sprintf(where, "%s, ", p->origin); + else + where[0] = 0; + regerror(err, &p->regexp, errbuf, 1024); + regfree(&p->regexp); + die("%s'%s': %s", where, p->pattern, errbuf); + } + } +} + static char *end_of_line(char *cp, unsigned long *left) { unsigned long l = *left; @@ -97,13 +154,45 @@ static char *end_of_line(char *cp, unsigned long *left) return cp; } +static int word_char(char ch) +{ + return isalnum(ch) || ch == '_'; +} + static void show_line(struct grep_opt *opt, const char *bol, const char *eol, const char *name, unsigned lno, char sign) { printf("%s%c", name, sign); if (opt->linenum) printf("%d%c", lno, sign); - printf("%.*s\n", eol-bol, bol); + printf("%.*s\n", (int)(eol-bol), bol); +} + +/* + * NEEDSWORK: share code with diff.c + */ +#define FIRST_FEW_BYTES 8000 +static int buffer_is_binary(const char *ptr, unsigned long size) +{ + if (FIRST_FEW_BYTES < size) + size = FIRST_FEW_BYTES; + if (memchr(ptr, 0, size)) + return 1; + return 0; +} + +static int fixmatch(const char *pattern, char *line, regmatch_t *match) +{ + char *hit = strstr(line, pattern); + if (!hit) { + match->rm_so = match->rm_eo = -1; + return REG_NOMATCH; + } + else { + match->rm_so = hit - line; + match->rm_eo = match->rm_so + strlen(pattern); + return 0; + } } static int grep_buffer(struct grep_opt *opt, const char *name, @@ -118,7 +207,22 @@ static int grep_buffer(struct grep_opt *opt, const char *name, } *prev = NULL, *pcl; unsigned last_hit = 0; unsigned last_shown = 0; + int binary_match_only = 0; const char *hunk_mark = ""; + unsigned count = 0; + + if (buffer_is_binary(buf, size)) { + switch (opt->binary) { + case GREP_BINARY_DEFAULT: + binary_match_only = 1; + break; + case GREP_BINARY_NOMATCH: + return 0; /* Assume unmatch */ + break; + default: + break; + } + } if (opt->pre_context) prev = xcalloc(opt->pre_context, sizeof(*prev)); @@ -128,23 +232,70 @@ static int grep_buffer(struct grep_opt *opt, const char *name, while (left) { regmatch_t pmatch[10]; char *eol, ch; - int hit; + int hit = 0; + struct grep_pat *p; eol = end_of_line(bol, &left); ch = *eol; *eol = 0; - hit = !regexec(&opt->regexp, bol, ARRAY_SIZE(pmatch), - pmatch, 0); + for (p = opt->pattern_list; p; p = p->next) { + if (!opt->fixed) { + regex_t *exp = &p->regexp; + hit = !regexec(exp, bol, ARRAY_SIZE(pmatch), + pmatch, 0); + } + else { + hit = !fixmatch(p->pattern, bol, pmatch); + } + + if (hit && opt->word_regexp) { + /* Match beginning must be either + * beginning of the line, or at word + * boundary (i.e. the last char must + * not be alnum or underscore). + */ + if ((pmatch[0].rm_so < 0) || + (eol - bol) <= pmatch[0].rm_so || + (pmatch[0].rm_eo < 0) || + (eol - bol) < pmatch[0].rm_eo) + die("regexp returned nonsense"); + if (pmatch[0].rm_so != 0 && + word_char(bol[pmatch[0].rm_so-1])) + hit = 0; + if (pmatch[0].rm_eo != (eol-bol) && + word_char(bol[pmatch[0].rm_eo])) + hit = 0; + } + if (hit) + break; + } + /* "grep -v -e foo -e bla" should list lines + * that do not have either, so inversion should + * be done outside. + */ if (opt->invert) hit = !hit; + if (opt->unmatch_name_only) { + if (hit) + return 0; + goto next_line; + } if (hit) { + count++; + if (binary_match_only) { + printf("Binary file %s matches\n", name); + return 1; + } if (opt->name_only) { printf("%s\n", name); return 1; } /* Hit at this line. If we haven't shown the * pre-context lines, we would need to show them. + * When asked to do "count", this still show + * the context which is nonsense, but the user + * deserves to get that ;-). */ if (opt->pre_context) { unsigned from; @@ -166,7 +317,8 @@ static int grep_buffer(struct grep_opt *opt, const char *name, } if (last_shown && lno != last_shown + 1) printf(hunk_mark); - show_line(opt, bol, eol, name, lno, ':'); + if (!opt->count) + show_line(opt, bol, eol, name, lno, ':'); last_shown = last_hit = lno; } else if (last_hit && @@ -185,11 +337,29 @@ static int grep_buffer(struct grep_opt *opt, const char *name, prev->bol = bol; prev->eol = eol; } + + next_line: *eol = ch; bol = eol + 1; + if (!left) + break; left--; lno++; } + + if (opt->unmatch_name_only) { + /* We did not see any hit, so we want to show this */ + printf("%s\n", name); + return 1; + } + + /* NEEDSWORK: + * The real "grep -c foo *.c" gives many "bar.c:0" lines, + * which feels mostly useless but sometimes useful. Maybe + * make it another option? For now suppress them. + */ + if (opt->count && count) + printf("%s:%u\n", name, count); return !!last_hit; } @@ -240,12 +410,158 @@ static int grep_file(struct grep_opt *opt, const char *filename) return i; } +static int exec_grep(int argc, const char **argv) +{ + pid_t pid; + int status; + + argv[argc] = NULL; + pid = fork(); + if (pid < 0) + return pid; + if (!pid) { + execvp("grep", (char **) argv); + exit(255); + } + while (waitpid(pid, &status, 0) < 0) { + if (errno == EINTR) + continue; + return -1; + } + if (WIFEXITED(status)) { + if (!WEXITSTATUS(status)) + return 1; + return 0; + } + return -1; +} + +#define MAXARGS 1000 +#define ARGBUF 4096 +#define push_arg(a) do { \ + if (nr < MAXARGS) argv[nr++] = (a); \ + else die("maximum number of args exceeded"); \ + } while (0) + +static int external_grep(struct grep_opt *opt, const char **paths, int cached) +{ + int i, nr, argc, hit, len; + const char *argv[MAXARGS+1]; + char randarg[ARGBUF]; + char *argptr = randarg; + struct grep_pat *p; + + len = nr = 0; + push_arg("grep"); + if (opt->fixed) + push_arg("-F"); + if (opt->linenum) + push_arg("-n"); + if (opt->regflags & REG_EXTENDED) + push_arg("-E"); + if (opt->regflags & REG_ICASE) + push_arg("-i"); + if (opt->word_regexp) + push_arg("-w"); + if (opt->name_only) + push_arg("-l"); + if (opt->unmatch_name_only) + push_arg("-L"); + if (opt->count) + push_arg("-c"); + if (opt->post_context || opt->pre_context) { + if (opt->post_context != opt->pre_context) { + if (opt->pre_context) { + push_arg("-B"); + len += snprintf(argptr, sizeof(randarg)-len, + "%u", opt->pre_context); + if (sizeof(randarg) <= len) + die("maximum length of args exceeded"); + push_arg(argptr); + argptr += len; + } + if (opt->post_context) { + push_arg("-A"); + len += snprintf(argptr, sizeof(randarg)-len, + "%u", opt->post_context); + if (sizeof(randarg) <= len) + die("maximum length of args exceeded"); + push_arg(argptr); + argptr += len; + } + } + else { + push_arg("-C"); + len += snprintf(argptr, sizeof(randarg)-len, + "%u", opt->post_context); + if (sizeof(randarg) <= len) + die("maximum length of args exceeded"); + push_arg(argptr); + argptr += len; + } + } + for (p = opt->pattern_list; p; p = p->next) { + push_arg("-e"); + push_arg(p->pattern); + } + + /* + * To make sure we get the header printed out when we want it, + * add /dev/null to the paths to grep. This is unnecessary + * (and wrong) with "-l" or "-L", which always print out the + * name anyway. + * + * GNU grep has "-H", but this is portable. + */ + if (!opt->name_only && !opt->unmatch_name_only) + push_arg("/dev/null"); + + hit = 0; + argc = nr; + for (i = 0; i < active_nr; i++) { + struct cache_entry *ce = active_cache[i]; + char *name; + if (ce_stage(ce) || !S_ISREG(ntohl(ce->ce_mode))) + continue; + if (!pathspec_matches(paths, ce->name)) + continue; + name = ce->name; + if (name[0] == '-') { + int len = ce_namelen(ce); + name = xmalloc(len + 3); + memcpy(name, "./", 2); + memcpy(name + 2, ce->name, len + 1); + } + argv[argc++] = name; + if (argc < MAXARGS) + continue; + hit += exec_grep(argc, argv); + argc = nr; + } + if (argc > nr) + hit += exec_grep(argc, argv); + return 0; +} + static int grep_cache(struct grep_opt *opt, const char **paths, int cached) { int hit = 0; int nr; read_cache(); +#ifdef __unix__ + /* + * Use the external "grep" command for the case where + * we grep through the checked-out files. It tends to + * be a lot more optimized + */ + if (!cached) { + hit = external_grep(opt, paths, cached); + if (hit >= 0) + return hit; + } +#endif + for (nr = 0; nr < active_nr; nr++) { struct cache_entry *ce = active_cache[nr]; if (ce_stage(ce) || !S_ISREG(ntohl(ce->ce_mode))) @@ -264,11 +580,9 @@ static int grep_tree(struct grep_opt *opt, const char **paths, struct tree_desc *tree, const char *tree_name, const char *base) { - unsigned mode; int len; int hit = 0; - const char *path; - const unsigned char *sha1; + struct name_entry entry; char *down; char *path_buf = xmalloc(PATH_MAX + strlen(tree_name) + 100); @@ -283,36 +597,32 @@ static int grep_tree(struct grep_opt *opt, const char **paths, } len = strlen(path_buf); - while (tree->size) { - int pathlen; - sha1 = tree_entry_extract(tree, &path, &mode); - pathlen = strlen(path); - strcpy(path_buf + len, path); + while (tree_entry(tree, &entry)) { + strcpy(path_buf + len, entry.path); - if (S_ISDIR(mode)) + if (S_ISDIR(entry.mode)) /* Match "abc/" against pathspec to * decide if we want to descend into "abc" * directory. */ - strcpy(path_buf + len + pathlen, "/"); + strcpy(path_buf + len + entry.pathlen, "/"); if (!pathspec_matches(paths, down)) ; - else if (S_ISREG(mode)) - hit |= grep_sha1(opt, sha1, path_buf); - else if (S_ISDIR(mode)) { + else if (S_ISREG(entry.mode)) + hit |= grep_sha1(opt, entry.sha1, path_buf); + else if (S_ISDIR(entry.mode)) { char type[20]; struct tree_desc sub; void *data; - data = read_sha1_file(sha1, type, &sub.size); + data = read_sha1_file(entry.sha1, type, &sub.size); if (!data) die("unable to read tree (%s)", - sha1_to_hex(sha1)); + sha1_to_hex(entry.sha1)); sub.buf = data; hit |= grep_tree(opt, paths, &sub, tree_name, down); free(data); } - update_tree_entry(tree); } return hit; } @@ -344,22 +654,31 @@ static const char builtin_grep_usage[] = int cmd_grep(int argc, const char **argv, char **envp) { - int err; int hit = 0; - int no_more_flags = 0; - int seen_noncommit = 0; int cached = 0; + int seen_dashdash = 0; struct grep_opt opt; struct object_list *list, **tail, *object_list = NULL; const char *prefix = setup_git_directory(); const char **paths = NULL; + int i; memset(&opt, 0, sizeof(opt)); + opt.pattern_tail = &opt.pattern_list; opt.regflags = REG_NEWLINE; /* - * No point using rev_info, really. + * If there is no -- then the paths must exist in the working + * tree. If there is no explicit pattern specified with -e or + * -f, we take the first unrecognized non option to be the + * pattern, but then what follows it must be zero or more + * valid refs up to the -- (if exists), and then existing + * paths. If there is an explicit pattern, then the first + * unrecocnized non option is the beginning of the refs list + * that continues up to the -- (if exists), and then paths. */ + + tail = &object_list; while (1 < argc) { const char *arg = argv[1]; argc--; argv++; @@ -367,11 +686,20 @@ int cmd_grep(int argc, const char **argv, char **envp) cached = 1; continue; } + if (!strcmp("-a", arg) || + !strcmp("--text", arg)) { + opt.binary = GREP_BINARY_TEXT; + continue; + } if (!strcmp("-i", arg) || !strcmp("--ignore-case", arg)) { opt.regflags |= REG_ICASE; continue; } + if (!strcmp("-I", arg)) { + opt.binary = GREP_BINARY_NOMATCH; + continue; + } if (!strcmp("-v", arg) || !strcmp("--invert-match", arg)) { opt.invert = 1; @@ -382,6 +710,11 @@ int cmd_grep(int argc, const char **argv, char **envp) opt.regflags |= REG_EXTENDED; continue; } + if (!strcmp("-F", arg) || + !strcmp("--fixed-strings", arg)) { + opt.fixed = 1; + continue; + } if (!strcmp("-G", arg) || !strcmp("--basic-regexp", arg)) { opt.regflags &= ~REG_EXTENDED; @@ -402,18 +735,49 @@ int cmd_grep(int argc, const char **argv, char **envp) opt.name_only = 1; continue; } - if (!strcmp("-A", arg) || - !strcmp("-B", arg) || - !strcmp("-C", arg)) { + if (!strcmp("-L", arg) || + !strcmp("--files-without-match", arg)) { + opt.unmatch_name_only = 1; + continue; + } + if (!strcmp("-c", arg) || + !strcmp("--count", arg)) { + opt.count = 1; + continue; + } + if (!strcmp("-w", arg) || + !strcmp("--word-regexp", arg)) { + opt.word_regexp = 1; + continue; + } + if (!strncmp("-A", arg, 2) || + !strncmp("-B", arg, 2) || + !strncmp("-C", arg, 2) || + (arg[0] == '-' && '1' <= arg[1] && arg[1] <= '9')) { unsigned num; - if (argc <= 1 || - sscanf(*++argv, "%u", &num) != 1) + const char *scan; + switch (arg[1]) { + case 'A': case 'B': case 'C': + if (!arg[2]) { + if (argc <= 1) + usage(builtin_grep_usage); + scan = *++argv; + argc--; + } + else + scan = arg + 2; + break; + default: + scan = arg + 1; + break; + } + if (sscanf(scan, "%u", &num) != 1) usage(builtin_grep_usage); - argc--; switch (arg[1]) { case 'A': opt.post_context = num; break; + default: case 'C': opt.post_context = num; case 'B': @@ -422,66 +786,96 @@ int cmd_grep(int argc, const char **argv, char **envp) } continue; } + if (!strcmp("-f", arg)) { + FILE *patterns; + int lno = 0; + char buf[1024]; + if (argc <= 1) + usage(builtin_grep_usage); + patterns = fopen(argv[1], "r"); + if (!patterns) + die("'%s': %s", argv[1], strerror(errno)); + while (fgets(buf, sizeof(buf), patterns)) { + int len = strlen(buf); + if (buf[len-1] == '\n') + buf[len-1] = 0; + /* ignore empty line like grep does */ + if (!buf[0]) + continue; + add_pattern(&opt, strdup(buf), argv[1], ++lno); + } + fclose(patterns); + argv++; + argc--; + continue; + } if (!strcmp("-e", arg)) { if (1 < argc) { - /* We probably would want to do - * -e pat1 -e pat2 as well later... - */ - if (opt.pattern) - die("more than one pattern?"); - opt.pattern = *++argv; + add_pattern(&opt, argv[1], "-e option", 0); + argv++; argc--; continue; } usage(builtin_grep_usage); } - if (!strcmp("--", arg)) { - no_more_flags = 1; - continue; - } - /* Either unrecognized option or a single pattern */ - if (!no_more_flags && *arg == '-') + if (!strcmp("--", arg)) + break; + if (*arg == '-') usage(builtin_grep_usage); - if (!opt.pattern) { - opt.pattern = arg; + + /* First unrecognized non-option token */ + if (!opt.pattern_list) { + add_pattern(&opt, arg, "command line", 0); break; } else { /* We are looking at the first path or rev; - * it is found at argv[0] after leaving the + * it is found at argv[1] after leaving the * loop. */ argc++; argv--; break; } } - if (!opt.pattern) + + if (!opt.pattern_list) die("no pattern given."); - err = regcomp(&opt.regexp, opt.pattern, opt.regflags); - if (err) { - char errbuf[1024]; - regerror(err, &opt.regexp, errbuf, 1024); - regfree(&opt.regexp); - die("'%s': %s", opt.pattern, errbuf); - } - tail = &object_list; - while (1 < argc) { - struct object *object; - struct object_list *elem; - const char *arg = argv[1]; + if ((opt.regflags != REG_NEWLINE) && opt.fixed) + die("cannot mix --fixed-strings and regexp"); + if (!opt.fixed) + compile_patterns(&opt); + + /* Check revs and then paths */ + for (i = 1; i < argc; i++) { + const char *arg = argv[i]; unsigned char sha1[20]; - if (get_sha1(arg, sha1) < 0) - break; - object = parse_object(sha1); - if (!object) - die("bad object %s", arg); - elem = object_list_insert(object, tail); - elem->name = arg; - tail = &elem->next; - argc--; argv++; + /* Is it a rev? */ + if (!get_sha1(arg, sha1)) { + struct object *object = parse_object(sha1); + struct object_list *elem; + if (!object) + die("bad object %s", arg); + elem = object_list_insert(object, tail); + elem->name = arg; + tail = &elem->next; + continue; + } + if (!strcmp(arg, "--")) { + i++; + seen_dashdash = 1; + } + break; } - if (1 < argc) - paths = get_pathspec(prefix, argv + 1); + + /* The rest are paths */ + if (!seen_dashdash) { + int j; + for (j = i; j < argc; j++) + verify_filename(prefix, argv[j]); + } + + if (i < argc) + paths = get_pathspec(prefix, argv + i); else if (prefix) { paths = xcalloc(2, sizeof(const char *)); paths[0] = prefix; @@ -490,21 +884,9 @@ int cmd_grep(int argc, const char **argv, char **envp) if (!object_list) return !grep_cache(&opt, paths, cached); - /* - * Do not walk "grep -e foo master next pu -- Documentation/" - * but do walk "grep -e foo master..next -- Documentation/". - * Ranged request mixed with a blob or tree object, like - * "grep -e foo v1.0.0:Documentation/ master..next" - * so detect that and complain. - */ - for (list = object_list; list; list = list->next) { - struct object *real_obj; - real_obj = deref_tag(list->item, NULL, 0); - if (strcmp(real_obj->type, commit_type)) - seen_noncommit = 1; - } + if (cached) - die("both --cached and revisions given."); + die("both --cached and trees are given."); for (list = object_list; list; list = list->next) { struct object *real_obj;