summaryrefslogtreecommitdiff
path: root/src/journal/sd-journal.c
diff options
context:
space:
mode:
authorMichal Schmidt <mschmidt@redhat.com>2014-08-19 13:38:53 +0200
committerMichal Schmidt <mschmidt@redhat.com>2014-10-23 17:38:02 +0200
commitc1f906bd91c388fd84a006a56e1e6692e23f8ae3 (patch)
treea7a587ada295261a4b92ea69ca80adf0b7e5a7b7 /src/journal/sd-journal.c
parent43cf8388ea4ffed1801468d4b650d6e48eefce9e (diff)
journal: make sd_journal::files a OrderedHashmap
Anything that uses hashmap_next() almost certainly cares about the order and needs to be an OrderedHashmap.
Diffstat (limited to 'src/journal/sd-journal.c')
-rw-r--r--src/journal/sd-journal.c38
1 files changed, 19 insertions, 19 deletions
diff --git a/src/journal/sd-journal.c b/src/journal/sd-journal.c
index ac57f4f6f8..cf21c4d899 100644
--- a/src/journal/sd-journal.c
+++ b/src/journal/sd-journal.c
@@ -86,7 +86,7 @@ static void detach_location(sd_journal *j) {
j->current_file = NULL;
j->current_field = 0;
- HASHMAP_FOREACH(f, j->files, i)
+ ORDERED_HASHMAP_FOREACH(f, j->files, i)
f->current_offset = 0;
}
@@ -879,7 +879,7 @@ static int real_journal_next(sd_journal *j, direction_t direction) {
assert_return(j, -EINVAL);
assert_return(!journal_pid_changed(j), -ECHILD);
- HASHMAP_FOREACH(f, j->files, i) {
+ ORDERED_HASHMAP_FOREACH(f, j->files, i) {
bool found;
r = next_beyond_location(j, f, direction, &o, &p);
@@ -1288,10 +1288,10 @@ static int add_any_file(sd_journal *j, const char *path) {
assert(j);
assert(path);
- if (hashmap_get(j->files, path))
+ if (ordered_hashmap_get(j->files, path))
return 0;
- if (hashmap_size(j->files) >= JOURNAL_FILES_MAX) {
+ if (ordered_hashmap_size(j->files) >= JOURNAL_FILES_MAX) {
log_warning("Too many open journal files, not adding %s.", path);
return set_put_error(j, -ETOOMANYREFS);
}
@@ -1302,7 +1302,7 @@ static int add_any_file(sd_journal *j, const char *path) {
/* journal_file_dump(f); */
- r = hashmap_put(j->files, f->path, f);
+ r = ordered_hashmap_put(j->files, f->path, f);
if (r < 0) {
journal_file_close(f);
return r;
@@ -1351,7 +1351,7 @@ static int remove_file(sd_journal *j, const char *prefix, const char *filename)
if (!path)
return -ENOMEM;
- f = hashmap_get(j->files, path);
+ f = ordered_hashmap_get(j->files, path);
if (!f)
return 0;
@@ -1363,7 +1363,7 @@ static void remove_file_real(sd_journal *j, JournalFile *f) {
assert(j);
assert(f);
- hashmap_remove(j->files, f->path);
+ ordered_hashmap_remove(j->files, f->path);
log_debug("File %s removed.", f->path);
@@ -1374,7 +1374,7 @@ static void remove_file_real(sd_journal *j, JournalFile *f) {
if (j->unique_file == f) {
/* Jump to the next unique_file or NULL if that one was last */
- j->unique_file = hashmap_next(j->files, j->unique_file->path);
+ j->unique_file = ordered_hashmap_next(j->files, j->unique_file->path);
j->unique_offset = 0;
if (!j->unique_file)
j->unique_file_lost = true;
@@ -1634,7 +1634,7 @@ static int add_current_paths(sd_journal *j) {
* "root" directories. We don't expect errors here, so we
* treat them as fatal. */
- HASHMAP_FOREACH(f, j->files, i) {
+ ORDERED_HASHMAP_FOREACH(f, j->files, i) {
_cleanup_free_ char *dir;
int r;
@@ -1689,7 +1689,7 @@ static sd_journal *journal_new(int flags, const char *path) {
goto fail;
}
- j->files = hashmap_new(&string_hash_ops);
+ j->files = ordered_hashmap_new(&string_hash_ops);
j->directories_by_path = hashmap_new(&string_hash_ops);
j->mmap = mmap_cache_new();
if (!j->files || !j->directories_by_path || !j->mmap)
@@ -1835,10 +1835,10 @@ _public_ void sd_journal_close(sd_journal *j) {
sd_journal_flush_matches(j);
- while ((f = hashmap_steal_first(j->files)))
+ while ((f = ordered_hashmap_steal_first(j->files)))
journal_file_close(f);
- hashmap_free(j->files);
+ ordered_hashmap_free(j->files);
while ((d = hashmap_first(j->directories_by_path)))
remove_directory(j, d);
@@ -2368,7 +2368,7 @@ _public_ int sd_journal_get_cutoff_realtime_usec(sd_journal *j, uint64_t *from,
assert_return(from || to, -EINVAL);
assert_return(from != to, -EINVAL);
- HASHMAP_FOREACH(f, j->files, i) {
+ ORDERED_HASHMAP_FOREACH(f, j->files, i) {
usec_t fr, t;
r = journal_file_get_cutoff_realtime_usec(f, &fr, &t);
@@ -2408,7 +2408,7 @@ _public_ int sd_journal_get_cutoff_monotonic_usec(sd_journal *j, sd_id128_t boot
assert_return(from || to, -EINVAL);
assert_return(from != to, -EINVAL);
- HASHMAP_FOREACH(f, j->files, i) {
+ ORDERED_HASHMAP_FOREACH(f, j->files, i) {
usec_t fr, t;
r = journal_file_get_cutoff_monotonic_usec(f, boot_id, &fr, &t);
@@ -2443,7 +2443,7 @@ void journal_print_header(sd_journal *j) {
assert(j);
- HASHMAP_FOREACH(f, j->files, i) {
+ ORDERED_HASHMAP_FOREACH(f, j->files, i) {
if (newline)
putchar('\n');
else
@@ -2462,7 +2462,7 @@ _public_ int sd_journal_get_usage(sd_journal *j, uint64_t *bytes) {
assert_return(!journal_pid_changed(j), -ECHILD);
assert_return(bytes, -EINVAL);
- HASHMAP_FOREACH(f, j->files, i) {
+ ORDERED_HASHMAP_FOREACH(f, j->files, i) {
struct stat st;
if (fstat(f->fd, &st) < 0)
@@ -2511,7 +2511,7 @@ _public_ int sd_journal_enumerate_unique(sd_journal *j, const void **data, size_
if (j->unique_file_lost)
return 0;
- j->unique_file = hashmap_first(j->files);
+ j->unique_file = ordered_hashmap_first(j->files);
if (!j->unique_file)
return 0;
@@ -2545,7 +2545,7 @@ _public_ int sd_journal_enumerate_unique(sd_journal *j, const void **data, size_
/* We reached the end of the list? Then start again, with the next file */
if (j->unique_offset == 0) {
- j->unique_file = hashmap_next(j->files, j->unique_file->path);
+ j->unique_file = ordered_hashmap_next(j->files, j->unique_file->path);
if (!j->unique_file)
return 0;
@@ -2594,7 +2594,7 @@ _public_ int sd_journal_enumerate_unique(sd_journal *j, const void **data, size_
* object by checking if it exists in the earlier
* traversed files. */
found = false;
- HASHMAP_FOREACH(of, j->files, i) {
+ ORDERED_HASHMAP_FOREACH(of, j->files, i) {
Object *oo;
uint64_t op;