#define THREADS 8
static pthread_t threads[THREADS];
-static void *load_sha1(const unsigned char *sha1, unsigned long *size,
- const char *name);
-static void *load_file(const char *filename, size_t *sz);
-
-enum work_type {WORK_SHA1, WORK_FILE};
-
/* We use one producer thread and THREADS consumer
* threads. The producer adds struct work_items to 'todo' and the
* consumers pick work items from the same array.
*/
struct work_item {
- enum work_type type;
- char *name;
-
- /* if type == WORK_SHA1, then 'identifier' is a SHA1,
- * otherwise type == WORK_FILE, and 'identifier' is a NUL
- * terminated filename.
- */
- void *identifier;
+ struct grep_source source;
char done;
struct strbuf out;
};
pthread_mutex_unlock(&grep_mutex);
}
-/* Used to serialize calls to read_sha1_file. */
-static pthread_mutex_t read_sha1_mutex;
-
-static inline void read_sha1_lock(void)
-{
- if (use_threads)
- pthread_mutex_lock(&read_sha1_mutex);
-}
-
-static inline void read_sha1_unlock(void)
-{
- if (use_threads)
- pthread_mutex_unlock(&read_sha1_mutex);
-}
-
/* Signalled when a new work_item is added to todo. */
static pthread_cond_t cond_add;
static int skip_first_line;
-static void add_work(enum work_type type, char *name, void *id)
+static void add_work(struct grep_opt *opt, enum grep_source_type type,
+ const char *name, const void *id)
{
grep_lock();
pthread_cond_wait(&cond_write, &grep_mutex);
}
- todo[todo_end].type = type;
- todo[todo_end].name = name;
- todo[todo_end].identifier = id;
+ grep_source_init(&todo[todo_end].source, type, name, id);
+ if (opt->binary != GREP_BINARY_TEXT)
+ grep_source_load_driver(&todo[todo_end].source);
todo[todo_end].done = 0;
strbuf_reset(&todo[todo_end].out);
todo_end = (todo_end + 1) % ARRAY_SIZE(todo);
return ret;
}
-static void grep_sha1_async(struct grep_opt *opt, char *name,
- const unsigned char *sha1)
-{
- unsigned char *s;
- s = xmalloc(20);
- memcpy(s, sha1, 20);
- add_work(WORK_SHA1, name, s);
-}
-
-static void grep_file_async(struct grep_opt *opt, char *name,
- const char *filename)
-{
- add_work(WORK_FILE, name, xstrdup(filename));
-}
-
static void work_done(struct work_item *w)
{
int old_done;
write_or_die(1, p, len);
}
- free(w->name);
- free(w->identifier);
+ grep_source_clear(&w->source);
}
if (old_done != todo_done)
break;
opt->output_priv = w;
- if (w->type == WORK_SHA1) {
- unsigned long sz;
- void* data = load_sha1(w->identifier, &sz, w->name);
-
- if (data) {
- hit |= grep_buffer(opt, w->name, data, sz);
- free(data);
- }
- } else if (w->type == WORK_FILE) {
- size_t sz;
- void* data = load_file(w->identifier, &sz);
- if (data) {
- hit |= grep_buffer(opt, w->name, data, sz);
- free(data);
- }
- } else {
- assert(0);
- }
-
+ hit |= grep_source(opt, &w->source);
+ grep_source_clear_data(&w->source);
work_done(w);
}
free_grep_patterns(arg);
int i;
pthread_mutex_init(&grep_mutex, NULL);
- pthread_mutex_init(&read_sha1_mutex, NULL);
+ pthread_mutex_init(&grep_read_mutex, NULL);
pthread_mutex_init(&grep_attr_mutex, NULL);
pthread_cond_init(&cond_add, NULL);
pthread_cond_init(&cond_write, NULL);
pthread_cond_init(&cond_result, NULL);
+ grep_use_locks = 1;
for (i = 0; i < ARRAY_SIZE(todo); i++) {
strbuf_init(&todo[i].out, 0);
}
pthread_mutex_destroy(&grep_mutex);
- pthread_mutex_destroy(&read_sha1_mutex);
+ pthread_mutex_destroy(&grep_read_mutex);
pthread_mutex_destroy(&grep_attr_mutex);
pthread_cond_destroy(&cond_add);
pthread_cond_destroy(&cond_write);
pthread_cond_destroy(&cond_result);
+ grep_use_locks = 0;
return hit;
}
#else /* !NO_PTHREADS */
-#define read_sha1_lock()
-#define read_sha1_unlock()
static int wait_all(void)
{
{
void *data;
- read_sha1_lock();
+ grep_read_lock();
data = read_sha1_file(sha1, type, size);
- read_sha1_unlock();
- return data;
-}
-
-static void *load_sha1(const unsigned char *sha1, unsigned long *size,
- const char *name)
-{
- enum object_type type;
- void *data = lock_and_read_sha1_file(sha1, &type, size);
-
- if (!data)
- error(_("'%s': unable to read %s"), name, sha1_to_hex(sha1));
-
+ grep_read_unlock();
return data;
}
const char *filename, int tree_name_len)
{
struct strbuf pathbuf = STRBUF_INIT;
- char *name;
if (opt->relative && opt->prefix_length) {
quote_path_relative(filename + tree_name_len, -1, &pathbuf,
strbuf_addstr(&pathbuf, filename);
}
- name = strbuf_detach(&pathbuf, NULL);
-
#ifndef NO_PTHREADS
if (use_threads) {
- grep_sha1_async(opt, name, sha1);
+ add_work(opt, GREP_SOURCE_SHA1, pathbuf.buf, sha1);
+ strbuf_release(&pathbuf);
return 0;
} else
#endif
{
+ struct grep_source gs;
int hit;
- unsigned long sz;
- void *data = load_sha1(sha1, &sz, name);
- if (!data)
- hit = 0;
- else
- hit = grep_buffer(opt, name, data, sz);
-
- free(data);
- free(name);
- return hit;
- }
-}
-static void *load_file(const char *filename, size_t *sz)
-{
- struct stat st;
- char *data;
- int i;
+ grep_source_init(&gs, GREP_SOURCE_SHA1, pathbuf.buf, sha1);
+ strbuf_release(&pathbuf);
+ hit = grep_source(opt, &gs);
- if (lstat(filename, &st) < 0) {
- err_ret:
- if (errno != ENOENT)
- error(_("'%s': %s"), filename, strerror(errno));
- return NULL;
- }
- if (!S_ISREG(st.st_mode))
- return NULL;
- *sz = xsize_t(st.st_size);
- i = open(filename, O_RDONLY);
- if (i < 0)
- goto err_ret;
- data = xmalloc(*sz + 1);
- if (st.st_size != read_in_full(i, data, *sz)) {
- error(_("'%s': short read %s"), filename, strerror(errno));
- close(i);
- free(data);
- return NULL;
+ grep_source_clear(&gs);
+ return hit;
}
- close(i);
- data[*sz] = 0;
- return data;
}
static int grep_file(struct grep_opt *opt, const char *filename)
{
struct strbuf buf = STRBUF_INIT;
- char *name;
if (opt->relative && opt->prefix_length)
quote_path_relative(filename, -1, &buf, opt->prefix);
else
strbuf_addstr(&buf, filename);
- name = strbuf_detach(&buf, NULL);
#ifndef NO_PTHREADS
if (use_threads) {
- grep_file_async(opt, name, filename);
+ add_work(opt, GREP_SOURCE_FILE, buf.buf, filename);
+ strbuf_release(&buf);
return 0;
} else
#endif
{
+ struct grep_source gs;
int hit;
- size_t sz;
- void *data = load_file(filename, &sz);
- if (!data)
- hit = 0;
- else
- hit = grep_buffer(opt, name, data, sz);
- free(data);
- free(name);
+ grep_source_init(&gs, GREP_SOURCE_FILE, buf.buf, filename);
+ strbuf_release(&buf);
+ hit = grep_source(opt, &gs);
+
+ grep_source_clear(&gs);
return hit;
}
}
struct strbuf base;
int hit, len;
- read_sha1_lock();
+ grep_read_lock();
data = read_object_with_reference(obj->sha1, tree_type,
&size, NULL);
- read_sha1_unlock();
+ grep_read_unlock();
if (!data)
die(_("unable to read tree (%s)"), sha1_to_hex(obj->sha1));
use_threads = 0;
#endif
- opt.use_threads = use_threads;
-
#ifndef NO_PTHREADS
if (use_threads) {
if (!(opt.name_only || opt.unmatch_name_only || opt.count)
}
#ifndef NO_PTHREADS
+int grep_use_locks;
+
/*
* This lock protects access to the gitattributes machinery, which is
* not thread-safe.
*/
pthread_mutex_t grep_attr_mutex;
-static inline void grep_attr_lock(struct grep_opt *opt)
+static inline void grep_attr_lock(void)
{
- if (opt->use_threads)
+ if (grep_use_locks)
pthread_mutex_lock(&grep_attr_mutex);
}
-static inline void grep_attr_unlock(struct grep_opt *opt)
+static inline void grep_attr_unlock(void)
{
- if (opt->use_threads)
+ if (grep_use_locks)
pthread_mutex_unlock(&grep_attr_mutex);
}
+
+/*
+ * Same as git_attr_mutex, but protecting the thread-unsafe object db access.
+ */
+pthread_mutex_t grep_read_mutex;
+
#else
-#define grep_attr_lock(opt)
-#define grep_attr_unlock(opt)
+#define grep_attr_lock()
+#define grep_attr_unlock()
#endif
-static int match_funcname(struct grep_opt *opt, const char *name, char *bol, char *eol)
+static int match_funcname(struct grep_opt *opt, struct grep_source *gs, char *bol, char *eol)
{
xdemitconf_t *xecfg = opt->priv;
if (xecfg && !xecfg->find_func) {
- struct userdiff_driver *drv;
- grep_attr_lock(opt);
- drv = userdiff_find_by_path(name);
- grep_attr_unlock(opt);
- if (drv && drv->funcname.pattern) {
- const struct userdiff_funcname *pe = &drv->funcname;
+ grep_source_load_driver(gs);
+ if (gs->driver->funcname.pattern) {
+ const struct userdiff_funcname *pe = &gs->driver->funcname;
xdiff_set_find_func(xecfg, pe->pattern, pe->cflags);
} else {
xecfg = opt->priv = NULL;
return 0;
}
-static void show_funcname_line(struct grep_opt *opt, const char *name,
- char *buf, char *bol, unsigned lno)
+static void show_funcname_line(struct grep_opt *opt, struct grep_source *gs,
+ char *bol, unsigned lno)
{
- while (bol > buf) {
+ while (bol > gs->buf) {
char *eol = --bol;
- while (bol > buf && bol[-1] != '\n')
+ while (bol > gs->buf && bol[-1] != '\n')
bol--;
lno--;
if (lno <= opt->last_shown)
break;
- if (match_funcname(opt, name, bol, eol)) {
- show_line(opt, bol, eol, name, lno, '=');
+ if (match_funcname(opt, gs, bol, eol)) {
+ show_line(opt, bol, eol, gs->name, lno, '=');
break;
}
}
}
-static void show_pre_context(struct grep_opt *opt, const char *name, char *buf,
+static void show_pre_context(struct grep_opt *opt, struct grep_source *gs,
char *bol, char *end, unsigned lno)
{
unsigned cur = lno, from = 1, funcname_lno = 0;
int funcname_needed = !!opt->funcname;
- if (opt->funcbody && !match_funcname(opt, name, bol, end))
+ if (opt->funcbody && !match_funcname(opt, gs, bol, end))
funcname_needed = 2;
if (opt->pre_context < lno)
from = opt->last_shown + 1;
/* Rewind. */
- while (bol > buf &&
+ while (bol > gs->buf &&
cur > (funcname_needed == 2 ? opt->last_shown + 1 : from)) {
char *eol = --bol;
- while (bol > buf && bol[-1] != '\n')
+ while (bol > gs->buf && bol[-1] != '\n')
bol--;
cur--;
- if (funcname_needed && match_funcname(opt, name, bol, eol)) {
+ if (funcname_needed && match_funcname(opt, gs, bol, eol)) {
funcname_lno = cur;
funcname_needed = 0;
}
/* We need to look even further back to find a function signature. */
if (opt->funcname && funcname_needed)
- show_funcname_line(opt, name, buf, bol, cur);
+ show_funcname_line(opt, gs, bol, cur);
/* Back forward. */
while (cur < lno) {
while (*eol != '\n')
eol++;
- show_line(opt, bol, eol, name, cur, sign);
+ show_line(opt, bol, eol, gs->name, cur, sign);
bol = eol + 1;
cur++;
}
fwrite(buf, size, 1, stdout);
}
-static int grep_buffer_1(struct grep_opt *opt, const char *name,
- char *buf, unsigned long size, int collect_hits)
+static int grep_source_1(struct grep_opt *opt, struct grep_source *gs, int collect_hits)
{
- char *bol = buf;
- unsigned long left = size;
+ char *bol;
+ unsigned long left;
unsigned lno = 1;
unsigned last_hit = 0;
int binary_match_only = 0;
switch (opt->binary) {
case GREP_BINARY_DEFAULT:
- if (buffer_is_binary(buf, size))
+ if (grep_source_is_binary(gs))
binary_match_only = 1;
break;
case GREP_BINARY_NOMATCH:
- if (buffer_is_binary(buf, size))
+ if (grep_source_is_binary(gs))
return 0; /* Assume unmatch */
break;
case GREP_BINARY_TEXT:
try_lookahead = should_lookahead(opt);
+ if (grep_source_load(gs) < 0)
+ return 0;
+
+ bol = gs->buf;
+ left = gs->size;
while (left) {
char *eol, ch;
int hit;
if (opt->status_only)
return 1;
if (opt->name_only) {
- show_name(opt, name);
+ show_name(opt, gs->name);
return 1;
}
if (opt->count)
goto next_line;
if (binary_match_only) {
opt->output(opt, "Binary file ", 12);
- output_color(opt, name, strlen(name),
+ output_color(opt, gs->name, strlen(gs->name),
opt->color_filename);
opt->output(opt, " matches\n", 9);
return 1;
* pre-context lines, we would need to show them.
*/
if (opt->pre_context || opt->funcbody)
- show_pre_context(opt, name, buf, bol, eol, lno);
+ show_pre_context(opt, gs, bol, eol, lno);
else if (opt->funcname)
- show_funcname_line(opt, name, buf, bol, lno);
- show_line(opt, bol, eol, name, lno, ':');
+ show_funcname_line(opt, gs, bol, lno);
+ show_line(opt, bol, eol, gs->name, lno, ':');
last_hit = lno;
if (opt->funcbody)
show_function = 1;
goto next_line;
}
- if (show_function && match_funcname(opt, name, bol, eol))
+ if (show_function && match_funcname(opt, gs, bol, eol))
show_function = 0;
if (show_function ||
(last_hit && lno <= last_hit + opt->post_context)) {
/* If the last hit is within the post context,
* we need to show this line.
*/
- show_line(opt, bol, eol, name, lno, '-');
+ show_line(opt, bol, eol, gs->name, lno, '-');
}
next_line:
return 0;
if (opt->unmatch_name_only) {
/* We did not see any hit, so we want to show this */
- show_name(opt, name);
+ show_name(opt, gs->name);
return 1;
}
*/
if (opt->count && count) {
char buf[32];
- output_color(opt, name, strlen(name), opt->color_filename);
+ output_color(opt, gs->name, strlen(gs->name), opt->color_filename);
output_sep(opt, ':');
snprintf(buf, sizeof(buf), "%u\n", count);
opt->output(opt, buf, strlen(buf));
}
}
-int grep_buffer(struct grep_opt *opt, const char *name, char *buf, unsigned long size)
+int grep_source(struct grep_opt *opt, struct grep_source *gs)
{
/*
* we do not have to do the two-pass grep when we do not check
* buffer-wide "all-match".
*/
if (!opt->all_match)
- return grep_buffer_1(opt, name, buf, size, 0);
+ return grep_source_1(opt, gs, 0);
/* Otherwise the toplevel "or" terms hit a bit differently.
* We first clear hit markers from them.
*/
clr_hit_marker(opt->pattern_expression);
- grep_buffer_1(opt, name, buf, size, 1);
+ grep_source_1(opt, gs, 1);
if (!chk_hit_marker(opt->pattern_expression))
return 0;
- return grep_buffer_1(opt, name, buf, size, 0);
+ return grep_source_1(opt, gs, 0);
+}
+
+int grep_buffer(struct grep_opt *opt, char *buf, unsigned long size)
+{
+ struct grep_source gs;
+ int r;
+
+ grep_source_init(&gs, GREP_SOURCE_BUF, NULL, NULL);
+ gs.buf = buf;
+ gs.size = size;
+
+ r = grep_source(opt, &gs);
+
+ grep_source_clear(&gs);
+ return r;
+}
+
+void grep_source_init(struct grep_source *gs, enum grep_source_type type,
+ const char *name, const void *identifier)
+{
+ gs->type = type;
+ gs->name = name ? xstrdup(name) : NULL;
+ gs->buf = NULL;
+ gs->size = 0;
+ gs->driver = NULL;
+
+ switch (type) {
+ case GREP_SOURCE_FILE:
+ gs->identifier = xstrdup(identifier);
+ break;
+ case GREP_SOURCE_SHA1:
+ gs->identifier = xmalloc(20);
+ memcpy(gs->identifier, identifier, 20);
+ break;
+ case GREP_SOURCE_BUF:
+ gs->identifier = NULL;
+ }
+}
+
+void grep_source_clear(struct grep_source *gs)
+{
+ free(gs->name);
+ gs->name = NULL;
+ free(gs->identifier);
+ gs->identifier = NULL;
+ grep_source_clear_data(gs);
+}
+
+void grep_source_clear_data(struct grep_source *gs)
+{
+ switch (gs->type) {
+ case GREP_SOURCE_FILE:
+ case GREP_SOURCE_SHA1:
+ free(gs->buf);
+ gs->buf = NULL;
+ gs->size = 0;
+ break;
+ case GREP_SOURCE_BUF:
+ /* leave user-provided buf intact */
+ break;
+ }
+}
+
+static int grep_source_load_sha1(struct grep_source *gs)
+{
+ enum object_type type;
+
+ grep_read_lock();
+ gs->buf = read_sha1_file(gs->identifier, &type, &gs->size);
+ grep_read_unlock();
+
+ if (!gs->buf)
+ return error(_("'%s': unable to read %s"),
+ gs->name,
+ sha1_to_hex(gs->identifier));
+ return 0;
+}
+
+static int grep_source_load_file(struct grep_source *gs)
+{
+ const char *filename = gs->identifier;
+ struct stat st;
+ char *data;
+ size_t size;
+ int i;
+
+ if (lstat(filename, &st) < 0) {
+ err_ret:
+ if (errno != ENOENT)
+ error(_("'%s': %s"), filename, strerror(errno));
+ return -1;
+ }
+ if (!S_ISREG(st.st_mode))
+ return -1;
+ size = xsize_t(st.st_size);
+ i = open(filename, O_RDONLY);
+ if (i < 0)
+ goto err_ret;
+ data = xmalloc(size + 1);
+ if (st.st_size != read_in_full(i, data, size)) {
+ error(_("'%s': short read %s"), filename, strerror(errno));
+ close(i);
+ free(data);
+ return -1;
+ }
+ close(i);
+ data[size] = 0;
+
+ gs->buf = data;
+ gs->size = size;
+ return 0;
+}
+
+int grep_source_load(struct grep_source *gs)
+{
+ if (gs->buf)
+ return 0;
+
+ switch (gs->type) {
+ case GREP_SOURCE_FILE:
+ return grep_source_load_file(gs);
+ case GREP_SOURCE_SHA1:
+ return grep_source_load_sha1(gs);
+ case GREP_SOURCE_BUF:
+ return gs->buf ? 0 : -1;
+ }
+ die("BUG: invalid grep_source type");
+}
+
+void grep_source_load_driver(struct grep_source *gs)
+{
+ if (gs->driver)
+ return;
+
+ grep_attr_lock();
+ gs->driver = userdiff_find_by_path(gs->name);
+ if (!gs->driver)
+ gs->driver = userdiff_find_by_name("default");
+ grep_attr_unlock();
+}
+
+int grep_source_is_binary(struct grep_source *gs)
+{
+ grep_source_load_driver(gs);
+ if (gs->driver->binary != -1)
+ return gs->driver->binary;
+
+ if (!grep_source_load(gs))
+ return buffer_is_binary(gs->buf, gs->size);
+
+ return 0;
}