/* Emit the commands needed to rebuild a set object.
* The function returns 0 on error, 1 on success. */
int rewriteSetObject(rio *r, robj *key, robj *o) {
long long count = 0, items = setTypeSize(o);
if (o->encoding == REDIS_ENCODING_INTSET) {
int ii = 0;
int64_t llval;
while(intsetGet(o->ptr,ii++,&llval)) {
if (count == 0) {
int cmd_items = (items > REDIS_AOF_REWRITE_ITEMS_PER_CMD) ?
REDIS_AOF_REWRITE_ITEMS_PER_CMD : items;
if (rioWriteBulkCount(r,'*',2+cmd_items) == 0) return 0;
if (rioWriteBulkString(r,"SADD",4) == 0) return 0;
if (rioWriteBulkObject(r,key) == 0) return 0;
}
if (rioWriteBulkLongLong(r,llval) == 0) return 0;
if (++count == REDIS_AOF_REWRITE_ITEMS_PER_CMD) count = 0;
items--;
}
} else if (o->encoding == REDIS_ENCODING_HT) {
dictIterator *di = dictGetIterator(o->ptr);
dictEntry *de;
while((de = dictNext(di)) != NULL) {
robj *eleobj = dictGetKey(de);
if (count == 0) {
int cmd_items = (items > REDIS_AOF_REWRITE_ITEMS_PER_CMD) ?
REDIS_AOF_REWRITE_ITEMS_PER_CMD : items;
if (rioWriteBulkCount(r,'*',2+cmd_items) == 0) return 0;
if (rioWriteBulkString(r,"SADD",4) == 0) return 0;
if (rioWriteBulkObject(r,key) == 0) return 0;
}
if (rioWriteBulkObject(r,eleobj) == 0) return 0;
if (++count == REDIS_AOF_REWRITE_ITEMS_PER_CMD) count = 0;
items--;
}
dictReleaseIterator(di);
} else {
redisPanic("Unknown set encoding");
}
return 1;
}
/* latencyCommand() helper to produce the reply for the LATEST subcommand,
* listing the last latency sample for every event type registered so far. */
void latencyCommandReplyWithLatestEvents(client *c) {
dictIterator *di;
dictEntry *de;
addReplyMultiBulkLen(c,dictSize(server.latency_events));
di = dictGetIterator(server.latency_events);
while((de = dictNext(di)) != NULL) {
char *event = dictGetKey(de);
struct latencyTimeSeries *ts = dictGetVal(de);
int last = (ts->idx + LATENCY_TS_LEN - 1) % LATENCY_TS_LEN;
addReplyMultiBulkLen(c,4);
addReplyBulkCString(c,event);
addReplyLongLong(c,ts->samples[last].time);
addReplyLongLong(c,ts->samples[last].latency);
addReplyLongLong(c,ts->max);
}
dictReleaseIterator(di);
}
/* Return a random key, in form of a Redis object.
* If there are no keys, NULL is returned.
*随机key从数据库中
* The function makes sure to return keys not already expired. */
robj *dbRandomKey(redisDb *db) {
struct dictEntry *de;
while(1) {
sds key;
robj *keyobj;
de = dictGetRandomKey(db->dict);
if (de == NULL) return NULL;
key = dictGetKey(de);
keyobj = createStringObject(key,sdslen(key));
if (dictFind(db->expires,key)) {//判断键是否过期
if (expireIfNeeded(db,keyobj)) {//如果过期
decrRefCount(keyobj);//减少key的引用计数
continue; /* search for another key. This expired. */
}
}
return keyobj;
}
}
/* Unsubscribe from all the channels. Return the number of channels the
* client was subscribed to. */
int smempubsubUnsubscribeAllChannels(client *c, int notify)
{
dictIterator *di = dictGetSafeIterator(c->smempubsub_channels);
dictEntry *de;
int count = 0;
while((de = dictNext(di)) != NULL) {
robj *channel = dictGetKey(de);
count += smempubsubUnsubscribeChannel(c,channel,notify);
}
/* We were subscribed to nothing? Still reply to the client. */
if (notify && count == 0) {
addReply(c,shared.mbulkhdr[3]);
addReply(c,shared.unsubscribebulk);
addReply(c,shared.nullbulk);
addReplyLongLong(c,dictSize(c->smempubsub_channels));
}
dictReleaseIterator(di);
return count;
}
void keysCommand(client *c) {
dictIterator *di;
dictEntry *de;
sds pattern = c->argv[1]->ptr;
int plen = sdslen(pattern), allkeys;
unsigned long numkeys = 0;
void *replylen = addDeferredMultiBulkLength(c);
di = dictGetSafeIterator(server.counters);
allkeys = (pattern[0] == '*' && pattern[1] == '\0');
while((de = dictNext(di)) != NULL) {
sds key = dictGetKey(de);
robj *keyobj;
if (allkeys || stringmatchlen(pattern,plen,key,sdslen(key),0)) {
keyobj = createStringObject(key,sdslen(key));
addReplyBulk(c,keyobj);
numkeys++;
decrRefCount(keyobj);
}
}
dictReleaseIterator(di);
setDeferredMultiBulkLength(c,replylen,numkeys);
}
/* PUBSUB command for Pub/Sub introspection. */
void pubsubCommand(redisClient *c) {
// PUBSUB CHANNELS [pattern] 子命令
if (!strcasecmp(c->argv[1]->ptr,"channels") &&
(c->argc == 2 || c->argc ==3))
{
/* PUBSUB CHANNELS [<pattern>] */
// 检查命令请求是否给定了 pattern 参数
// 如果没有给定的话，就设为 NULL
sds pat = (c->argc == 2) ? NULL : c->argv[2]->ptr;
// 创建 pubsub_channels 的字典迭代器
// 该字典的键为频道，值为链表
// 链表中保存了所有订阅键所对应的频道的客户端
dictIterator *di = dictGetIterator(server.pubsub_channels);
dictEntry *de;
long mblen = 0;
void *replylen;
replylen = addDeferredMultiBulkLength(c);
// 从迭代器中获取一个客户端
while((de = dictNext(di)) != NULL) {
// 从字典中取出客户端所订阅的频道
robj *cobj = dictGetKey(de);
sds channel = cobj->ptr;
// 顺带一提
// 因为 Redis 的字典实现只能遍历字典的值（客户端）
// 所以这里才会有遍历字典值然后通过字典值取出字典键（频道）的蹩脚用法
// 如果没有给定 pattern 参数，那么打印所有找到的频道
// 如果给定了 pattern 参数，那么只打印和 pattern 相匹配的频道
if (!pat || stringmatchlen(pat, sdslen(pat),
channel, sdslen(channel),0))
{
// 向客户端输出频道
addReplyBulk(c,cobj);
mblen++;
}
}
// 释放字典迭代器
dictReleaseIterator(di);
setDeferredMultiBulkLength(c,replylen,mblen);
// PUBSUB NUMSUB [channel-1 channel-2 ... channel-N] 子命令
} else if (!strcasecmp(c->argv[1]->ptr,"numsub") && c->argc >= 2) {
/* PUBSUB NUMSUB [Channel_1 ... Channel_N] */
int j;
addReplyMultiBulkLen(c,(c->argc-2)*2);
for (j = 2; j < c->argc; j++) {
// c->argv[j] 也即是客户端输入的第 N 个频道名字
// pubsub_channels 的字典为频道名字
// 而值则是保存了 c->argv[j] 频道所有订阅者的链表
// 而调用 dictFetchValue 也就是取出所有订阅给定频道的客户端
list *l = dictFetchValue(server.pubsub_channels,c->argv[j]);
addReplyBulk(c,c->argv[j]);
// 向客户端返回链表的长度属性
// 这个属性就是某个频道的订阅者数量
// 例如：如果一个频道有三个订阅者，那么链表的长度就是 3
// 而返回给客户端的数字也是三
addReplyBulkLongLong(c,l ? listLength(l) : 0);
}
// PUBSUB NUMPAT 子命令
} else if (!strcasecmp(c->argv[1]->ptr,"numpat") && c->argc == 2) {
/* PUBSUB NUMPAT */
// pubsub_patterns 链表保存了服务器中所有被订阅的模式
// pubsub_patterns 的长度就是服务器中被订阅模式的数量
addReplyLongLong(c,listLength(server.pubsub_patterns));
// 错误处理
} else {
addReplyErrorFormat(c,
"Unknown PUBSUB subcommand or wrong number of arguments for '%s'",
(char*)c->argv[1]->ptr);
}
}
int freeMemoryIfNeeded(void) {
size_t mem_reported, mem_used, mem_tofree, mem_freed;
int slaves = listLength(server.slaves);
mstime_t latency, eviction_latency;
long long delta;
/* Check if we are over the memory usage limit. If we are not, no need
* to subtract the slaves output buffers. We can just return ASAP. */
mem_reported = zmalloc_used_memory();
if (mem_reported <= server.maxmemory) return C_OK;
/* Remove the size of slaves output buffers and AOF buffer from the
* count of used memory. */
mem_used = mem_reported;
if (slaves) {
listIter li;
listNode *ln;
listRewind(server.slaves,&li);
while((ln = listNext(&li))) {
client *slave = listNodeValue(ln);
unsigned long obuf_bytes = getClientOutputBufferMemoryUsage(slave);
if (obuf_bytes > mem_used)
mem_used = 0;
else
mem_used -= obuf_bytes;
}
}
if (server.aof_state != AOF_OFF) {
mem_used -= sdslen(server.aof_buf);
mem_used -= aofRewriteBufferSize();
}
/* Check if we are still over the memory limit. */
if (mem_used <= server.maxmemory) return C_OK;
/* Compute how much memory we need to free. */
mem_tofree = mem_used - server.maxmemory;
mem_freed = 0;
if (server.maxmemory_policy == MAXMEMORY_NO_EVICTION)
goto cant_free; /* We need to free memory, but policy forbids. */
latencyStartMonitor(latency);
while (mem_freed < mem_tofree) {
int j, k, i, keys_freed = 0;
static int next_db = 0;
sds bestkey = NULL;
int bestdbid;
redisDb *db;
dict *dict;
dictEntry *de;
if (server.maxmemory_policy == MAXMEMORY_ALLKEYS_LRU ||
server.maxmemory_policy == MAXMEMORY_VOLATILE_LRU)
{
struct evictionPoolEntry *pool = EvictionPoolLRU;
while(bestkey == NULL) {
unsigned long total_keys = 0, keys;
/* We don't want to make local-db choices when expiring keys,
* so to start populate the eviction pool sampling keys from
* every DB. */
for (i = 0; i < server.dbnum; i++) {
db = server.db+i;
dict = (server.maxmemory_policy == MAXMEMORY_ALLKEYS_LRU) ?
db->dict : db->expires;
if ((keys = dictSize(dict)) != 0) {
evictionPoolPopulate(i, dict, db->dict, pool);
total_keys += keys;
}
}
if (!total_keys) break; /* No keys to evict. */
/* Go backward from best to worst element to evict. */
for (k = EVPOOL_SIZE-1; k >= 0; k--) {
if (pool[k].key == NULL) continue;
bestdbid = pool[k].dbid;
if (server.maxmemory_policy == MAXMEMORY_ALLKEYS_LRU) {
de = dictFind(server.db[pool[k].dbid].dict,
pool[k].key);
} else {
de = dictFind(server.db[pool[k].dbid].expires,
pool[k].key);
}
/* Remove the entry from the pool. */
if (pool[k].key != pool[k].cached)
sdsfree(pool[k].key);
pool[k].key = NULL;
pool[k].idle = 0;
/* If the key exists, is our pick. Otherwise it is
* a ghost and we need to try the next element. */
if (de) {
bestkey = dictGetKey(de);
break;
} else {
/* Ghost... Iterate again. */
}
}
}
}
/* volatile-random and allkeys-random policy */
else if (server.maxmemory_policy == MAXMEMORY_ALLKEYS_RANDOM ||
server.maxmemory_policy == MAXMEMORY_VOLATILE_RANDOM)
{
/* When evicting a random key, we try to evict a key for
* each DB, so we use the static 'next_db' variable to
* incrementally visit all DBs. */
for (i = 0; i < server.dbnum; i++) {
j = (++next_db) % server.dbnum;
db = server.db+j;
dict = (server.maxmemory_policy == MAXMEMORY_ALLKEYS_RANDOM) ?
db->dict : db->expires;
if (dictSize(dict) != 0) {
de = dictGetRandomKey(dict);
bestkey = dictGetKey(de);
bestdbid = j;
break;
}
}
}
/* volatile-ttl */
else if (server.maxmemory_policy == MAXMEMORY_VOLATILE_TTL) {
long bestttl = 0; /* Initialized to avoid warning. */
/* In this policy we scan a single DB per iteration (visiting
* a different DB per call), expiring the key with the smallest
* TTL among the few sampled.
*
* Note that this algorithm makes local-DB choices, and should
* use a pool and code more similr to the one used in the
* LRU eviction policies in the future. */
for (i = 0; i < server.dbnum; i++) {
j = (++next_db) % server.dbnum;
db = server.db+j;
dict = db->expires;
if (dictSize(dict) != 0) {
for (k = 0; k < server.maxmemory_samples; k++) {
sds thiskey;
long thisttl;
de = dictGetRandomKey(dict);
thiskey = dictGetKey(de);
thisttl = (long) dictGetVal(de);
/* Keys expiring sooner (smaller unix timestamp) are
* better candidates for deletion */
if (bestkey == NULL || thisttl < bestttl) {
bestkey = thiskey;
bestttl = thisttl;
bestdbid = j;
}
}
}
}
}
/* Finally remove the selected key. */
if (bestkey) {
db = server.db+bestdbid;
robj *keyobj = createStringObject(bestkey,sdslen(bestkey));
propagateExpire(db,keyobj,server.lazyfree_lazy_eviction);
/* We compute the amount of memory freed by db*Delete() alone.
* It is possible that actually the memory needed to propagate
* the DEL in AOF and replication link is greater than the one
* we are freeing removing the key, but we can't account for
* that otherwise we would never exit the loop.
*
* AOF and Output buffer memory will be freed eventually so
* we only care about memory used by the key space. */
delta = (long long) zmalloc_used_memory();
latencyStartMonitor(eviction_latency);
if (server.lazyfree_lazy_eviction)
dbAsyncDelete(db,keyobj);
else
dbSyncDelete(db,keyobj);
latencyEndMonitor(eviction_latency);
latencyAddSampleIfNeeded("eviction-del",eviction_latency);
latencyRemoveNestedEvent(latency,eviction_latency);
delta -= (long long) zmalloc_used_memory();
mem_freed += delta;
server.stat_evictedkeys++;
notifyKeyspaceEvent(NOTIFY_EVICTED, "evicted",
keyobj, db->id);
decrRefCount(keyobj);
keys_freed++;
/* When the memory to free starts to be big enough, we may
* start spending so much time here that is impossible to
* deliver data to the slaves fast enough, so we force the
* transmission here inside the loop. */
if (slaves) flushSlavesOutputBuffers();
}
if (!keys_freed) {
latencyEndMonitor(latency);
latencyAddSampleIfNeeded("eviction-cycle",latency);
goto cant_free; /* nothing to free... */
}
}
latencyEndMonitor(latency);
latencyAddSampleIfNeeded("eviction-cycle",latency);
return C_OK;
cant_free:
/* We are here if we are not able to reclaim memory. There is only one
* last thing we can try: check if the lazyfree thread has jobs in queue
* and wait... */
while(bioPendingJobsOfType(BIO_LAZY_FREE)) {
if (((mem_reported - zmalloc_used_memory()) + mem_freed) >= mem_tofree)
break;
usleep(1000);
}
return C_ERR;
}
void evictionPoolPopulate(int dbid, dict *sampledict, dict *keydict, struct evictionPoolEntry *pool) {
int j, k, count;
dictEntry *samples[server.maxmemory_samples];
count = dictGetSomeKeys(sampledict,samples,server.maxmemory_samples);
for (j = 0; j < count; j++) {
unsigned long long idle;
sds key;
robj *o;
dictEntry *de;
de = samples[j];
key = dictGetKey(de);
/* If the dictionary we are sampling from is not the main
* dictionary (but the expires one) we need to lookup the key
* again in the key dictionary to obtain the value object. */
if (sampledict != keydict) de = dictFind(keydict, key);
o = dictGetVal(de);
idle = estimateObjectIdleTime(o);
/* Insert the element inside the pool.
* First, find the first empty bucket or the first populated
* bucket that has an idle time smaller than our idle time. */
k = 0;
while (k < EVPOOL_SIZE &&
pool[k].key &&
pool[k].idle < idle) k++;
if (k == 0 && pool[EVPOOL_SIZE-1].key != NULL) {
/* Can't insert if the element is < the worst element we have
* and there are no empty buckets. */
continue;
} else if (k < EVPOOL_SIZE && pool[k].key == NULL) {
/* Inserting into empty position. No setup needed before insert. */
} else {
/* Inserting in the middle. Now k points to the first element
* greater than the element to insert. */
if (pool[EVPOOL_SIZE-1].key == NULL) {
/* Free space on the right? Insert at k shifting
* all the elements from k to end to the right. */
/* Save SDS before overwriting. */
sds cached = pool[EVPOOL_SIZE-1].cached;
memmove(pool+k+1,pool+k,
sizeof(pool[0])*(EVPOOL_SIZE-k-1));
pool[k].cached = cached;
} else {
/* No free space on right? Insert at k-1 */
k--;
/* Shift all elements on the left of k (included) to the
* left, so we discard the element with smaller idle time. */
sds cached = pool[0].cached; /* Save SDS before overwriting. */
if (pool[0].key != pool[0].cached) sdsfree(pool[0].key);
memmove(pool,pool+1,sizeof(pool[0])*k);
pool[k].cached = cached;
}
}
/* Try to reuse the cached SDS string allocated in the pool entry,
* because allocating and deallocating this object is costly
* (according to the profiler, not my fantasy. Remember:
* premature optimizbla bla bla bla. */
int klen = sdslen(key);
if (klen > EVPOOL_CACHED_SDS_SIZE) {
pool[k].key = sdsdup(key);
} else {
memcpy(pool[k].cached,key,klen+1);
sdssetlen(pool[k].cached,klen);
pool[k].key = pool[k].cached;
}
pool[k].idle = idle;
pool[k].dbid = dbid;
}
}
static void
slotsScanSdsKeyCallback(void *l, const dictEntry *de) {
sds skey = dictGetKey(de);
robj *key = createStringObject(skey, sdslen(skey));
listAddNodeTail((list *)l, key);
}
/* The SORT command is the most complex command in Redis. Warning: this code
* is optimized for speed and a bit less for readability */
void sortCommand(redisClient *c) {
list *operations;
unsigned int outputlen = 0;
int desc = 0, alpha = 0;
long limit_start = 0, limit_count = -1, start, end;
int j, dontsort = 0, vectorlen;
int getop = 0; /* GET operation counter */
int int_convertion_error = 0;
robj *sortval, *sortby = NULL, *storekey = NULL;
redisSortObject *vector; /* Resulting vector to sort */
/* Lookup the key to sort. It must be of the right types */
sortval = lookupKeyRead(c->db,c->argv[1]);
if (sortval && sortval->type != REDIS_SET &&
sortval->type != REDIS_LIST &&
sortval->type != REDIS_ZSET)
{
addReply(c,shared.wrongtypeerr);
return;
}
/* Create a list of operations to perform for every sorted element.
* Operations can be GET/DEL/INCR/DECR */
operations = listCreate();
listSetFreeMethod(operations,zfree);
j = 2; /* options start at argv[2] */
/* Now we need to protect sortval incrementing its count, in the future
* SORT may have options able to overwrite/delete keys during the sorting
* and the sorted key itself may get destroyed */
if (sortval)
incrRefCount(sortval);
else
sortval = createListObject();
/* The SORT command has an SQL-alike syntax, parse it */
while(j < c->argc) {
int leftargs = c->argc-j-1;
if (!strcasecmp(c->argv[j]->ptr,"asc")) {
desc = 0;
} else if (!strcasecmp(c->argv[j]->ptr,"desc")) {
desc = 1;
} else if (!strcasecmp(c->argv[j]->ptr,"alpha")) {
alpha = 1;
} else if (!strcasecmp(c->argv[j]->ptr,"limit") && leftargs >= 2) {
if ((getLongFromObjectOrReply(c, c->argv[j+1], &limit_start, NULL) != REDIS_OK) ||
(getLongFromObjectOrReply(c, c->argv[j+2], &limit_count, NULL) != REDIS_OK)) return;
j+=2;
} else if (!strcasecmp(c->argv[j]->ptr,"store") && leftargs >= 1) {
storekey = c->argv[j+1];
j++;
} else if (!strcasecmp(c->argv[j]->ptr,"by") && leftargs >= 1) {
sortby = c->argv[j+1];
/* If the BY pattern does not contain '*', i.e. it is constant,
* we don't need to sort nor to lookup the weight keys. */
if (strchr(c->argv[j+1]->ptr,'*') == NULL) dontsort = 1;
j++;
} else if (!strcasecmp(c->argv[j]->ptr,"get") && leftargs >= 1) {
listAddNodeTail(operations,createSortOperation(
REDIS_SORT_GET,c->argv[j+1]));
getop++;
j++;
} else {
decrRefCount(sortval);
listRelease(operations);
addReply(c,shared.syntaxerr);
return;
}
j++;
}
/* For the STORE option, or when SORT is called from a Lua script,
* we want to force a specific ordering even when no explicit ordering
* was asked (SORT BY nosort). This guarantees that replication / AOF
* is deterministic.
*
* However in the case 'dontsort' is true, but the type to sort is a
* sorted set, we don't need to do anything as ordering is guaranteed
* in this special case. */
if ((storekey || c->flags & REDIS_LUA_CLIENT) &&
(dontsort && sortval->type != REDIS_ZSET))
{
/* Force ALPHA sorting */
dontsort = 0;
alpha = 1;
sortby = NULL;
}
/* Destructively convert encoded sorted sets for SORT. */
if (sortval->type == REDIS_ZSET)
zsetConvert(sortval, REDIS_ENCODING_SKIPLIST);
/* Objtain the length of the object to sort. */
switch(sortval->type) {
case REDIS_LIST: vectorlen = listTypeLength(sortval); break;
case REDIS_SET: vectorlen = setTypeSize(sortval); break;
case REDIS_ZSET: vectorlen = dictSize(((zset*)sortval->ptr)->dict); break;
default: vectorlen = 0; redisPanic("Bad SORT type"); /* Avoid GCC warning */
}
/* Perform LIMIT start,count sanity checking. */
start = (limit_start < 0) ? 0 : limit_start;
end = (limit_count < 0) ? vectorlen-1 : start+limit_count-1;
if (start >= vectorlen) {
start = vectorlen-1;
end = vectorlen-2;
}
if (end >= vectorlen) end = vectorlen-1;
/* Optimization:
*
* 1) if the object to sort is a sorted set.
* 2) There is nothing to sort as dontsort is true (BY <constant string>).
* 3) We have a LIMIT option that actually reduces the number of elements
* to fetch.
*
* In this case to load all the objects in the vector is a huge waste of
* resources. We just allocate a vector that is big enough for the selected
* range length, and make sure to load just this part in the vector. */
if (sortval->type == REDIS_ZSET &&
dontsort &&
(start != 0 || end != vectorlen-1))
{
vectorlen = end-start+1;
}
/* Load the sorting vector with all the objects to sort */
vector = zmalloc(sizeof(redisSortObject)*vectorlen);
j = 0;
if (sortval->type == REDIS_LIST) {
listTypeIterator *li = listTypeInitIterator(sortval,0,REDIS_TAIL);
listTypeEntry entry;
while(listTypeNext(li,&entry)) {
vector[j].obj = listTypeGet(&entry);
vector[j].u.score = 0;
vector[j].u.cmpobj = NULL;
j++;
}
listTypeReleaseIterator(li);
} else if (sortval->type == REDIS_SET) {
setTypeIterator *si = setTypeInitIterator(sortval);
robj *ele;
while((ele = setTypeNextObject(si)) != NULL) {
vector[j].obj = ele;
vector[j].u.score = 0;
vector[j].u.cmpobj = NULL;
j++;
}
setTypeReleaseIterator(si);
} else if (sortval->type == REDIS_ZSET && dontsort) {
/* Special handling for a sorted set, if 'dontsort' is true.
* This makes sure we return elements in the sorted set original
* ordering, accordingly to DESC / ASC options.
*
* Note that in this case we also handle LIMIT here in a direct
* way, just getting the required range, as an optimization. */
zset *zs = sortval->ptr;
zskiplist *zsl = zs->zsl;
zskiplistNode *ln;
robj *ele;
int rangelen = vectorlen;
/* Check if starting point is trivial, before doing log(N) lookup. */
if (desc) {
long zsetlen = dictSize(((zset*)sortval->ptr)->dict);
ln = zsl->tail;
if (start > 0)
ln = zslGetElementByRank(zsl,zsetlen-start);
} else {
ln = zsl->header->level[0].forward;
if (start > 0)
ln = zslGetElementByRank(zsl,start+1);
}
while(rangelen--) {
redisAssertWithInfo(c,sortval,ln != NULL);
ele = ln->obj;
vector[j].obj = ele;
vector[j].u.score = 0;
vector[j].u.cmpobj = NULL;
j++;
ln = desc ? ln->backward : ln->level[0].forward;
}
/* The code producing the output does not know that in the case of
* sorted set, 'dontsort', and LIMIT, we are able to get just the
* range, already sorted, so we need to adjust "start" and "end"
* to make sure start is set to 0. */
end -= start;
start = 0;
} else if (sortval->type == REDIS_ZSET) {
dict *set = ((zset*)sortval->ptr)->dict;
dictIterator *di;
dictEntry *setele;
di = dictGetIterator(set);
while((setele = dictNext(di)) != NULL) {
vector[j].obj = dictGetKey(setele);
vector[j].u.score = 0;
vector[j].u.cmpobj = NULL;
j++;
}
dictReleaseIterator(di);
} else {
redisPanic("Unknown type");
}
redisAssertWithInfo(c,sortval,j == vectorlen);
/* Now it's time to load the right scores in the sorting vector */
if (dontsort == 0) {
for (j = 0; j < vectorlen; j++) {
robj *byval;
if (sortby) {
/* lookup value to sort by */
byval = lookupKeyByPattern(c->db,sortby,vector[j].obj);
if (!byval) continue;
} else {
/* use object itself to sort by */
byval = vector[j].obj;
}
if (alpha) {
if (sortby) vector[j].u.cmpobj = getDecodedObject(byval);
} else {
if (byval->encoding == REDIS_ENCODING_RAW) {
char *eptr;
vector[j].u.score = strtod(byval->ptr,&eptr);
if (eptr[0] != '\0' || errno == ERANGE ||
isnan(vector[j].u.score))
{
int_convertion_error = 1;
}
} else if (byval->encoding == REDIS_ENCODING_INT) {
/* Don't need to decode the object if it's
* integer-encoded (the only encoding supported) so
* far. We can just cast it */
vector[j].u.score = (long)byval->ptr;
} else {
redisAssertWithInfo(c,sortval,1 != 1);
}
}
/* when the object was retrieved using lookupKeyByPattern,
* its refcount needs to be decreased. */
if (sortby) {
decrRefCount(byval);
}
}
}
if (dontsort == 0) {
server.sort_desc = desc;
server.sort_alpha = alpha;
server.sort_bypattern = sortby ? 1 : 0;
server.sort_store = storekey ? 1 : 0;
if (sortby && (start != 0 || end != vectorlen-1))
pqsort(vector,vectorlen,sizeof(redisSortObject),sortCompare, start,end);
else
qsort(vector,vectorlen,sizeof(redisSortObject),sortCompare);
}
/* Send command output to the output buffer, performing the specified
* GET/DEL/INCR/DECR operations if any. */
outputlen = getop ? getop*(end-start+1) : end-start+1;
if (int_convertion_error) {
addReplyError(c,"One or more scores can't be converted into double");
} else if (storekey == NULL) {
/* STORE option not specified, sent the sorting result to client */
addReplyMultiBulkLen(c,outputlen);
for (j = start; j <= end; j++) {
listNode *ln;
listIter li;
if (!getop) addReplyBulk(c,vector[j].obj);
listRewind(operations,&li);
while((ln = listNext(&li))) {
redisSortOperation *sop = ln->value;
robj *val = lookupKeyByPattern(c->db,sop->pattern,
vector[j].obj);
if (sop->type == REDIS_SORT_GET) {
if (!val) {
addReply(c,shared.nullbulk);
} else {
addReplyBulk(c,val);
decrRefCount(val);
}
} else {
/* Always fails */
redisAssertWithInfo(c,sortval,sop->type == REDIS_SORT_GET);
}
}
}
} else {
robj *sobj = createZiplistObject();
/* STORE option specified, set the sorting result as a List object */
for (j = start; j <= end; j++) {
listNode *ln;
listIter li;
if (!getop) {
listTypePush(sobj,vector[j].obj,REDIS_TAIL);
} else {
listRewind(operations,&li);
while((ln = listNext(&li))) {
redisSortOperation *sop = ln->value;
robj *val = lookupKeyByPattern(c->db,sop->pattern,
vector[j].obj);
if (sop->type == REDIS_SORT_GET) {
if (!val) val = createStringObject("",0);
/* listTypePush does an incrRefCount, so we should take care
* care of the incremented refcount caused by either
* lookupKeyByPattern or createStringObject("",0) */
listTypePush(sobj,val,REDIS_TAIL);
decrRefCount(val);
} else {
/* Always fails */
redisAssertWithInfo(c,sortval,sop->type == REDIS_SORT_GET);
}
}
}
}
if (outputlen) {
setKey(c->db,storekey,sobj);
notifyKeyspaceEvent(REDIS_NOTIFY_LIST,"sortstore",storekey,
c->db->id);
server.dirty += outputlen;
} else if (dbDelete(c->db,storekey)) {
signalModifiedKey(c->db,storekey);
notifyKeyspaceEvent(REDIS_NOTIFY_GENERIC,"del",storekey,c->db->id);
server.dirty++;
}
decrRefCount(sobj);
addReplyLongLong(c,outputlen);
}
/* Cleanup */
if (sortval->type == REDIS_LIST || sortval->type == REDIS_SET)
for (j = 0; j < vectorlen; j++)
decrRefCount(vector[j].obj);
decrRefCount(sortval);
listRelease(operations);
for (j = 0; j < vectorlen; j++) {
if (alpha && vector[j].u.cmpobj)
decrRefCount(vector[j].u.cmpobj);
}
zfree(vector);
}
void evictionPoolPopulate(int dbid, dict *sampledict, dict *keydict, struct evictionPoolEntry *pool) {
int j, k, count;
dictEntry *samples[server.maxmemory_samples];
count = dictGetSomeKeys(sampledict,samples,server.maxmemory_samples);
for (j = 0; j < count; j++) {
unsigned long long idle;
sds key;
robj *o;
dictEntry *de;
de = samples[j];
key = dictGetKey(de);
/* If the dictionary we are sampling from is not the main
* dictionary (but the expires one) we need to lookup the key
* again in the key dictionary to obtain the value object. */
if (server.maxmemory_policy != MAXMEMORY_VOLATILE_TTL) {
if (sampledict != keydict) de = dictFind(keydict, key);
o = dictGetVal(de);
}
/* Calculate the idle time according to the policy. This is called
* idle just because the code initially handled LRU, but is in fact
* just a score where an higher score means better candidate. */
if (server.maxmemory_policy & MAXMEMORY_FLAG_LRU) {
idle = estimateObjectIdleTime(o);
} else if (server.maxmemory_policy & MAXMEMORY_FLAG_LFU) {
/* When we use an LRU policy, we sort the keys by idle time
* so that we expire keys starting from greater idle time.
* However when the policy is an LFU one, we have a frequency
* estimation, and we want to evict keys with lower frequency
* first. So inside the pool we put objects using the inverted
* frequency subtracting the actual frequency to the maximum
* frequency of 255. */
idle = 255-LFUDecrAndReturn(o);
} else if (server.maxmemory_policy == MAXMEMORY_VOLATILE_TTL) {
/* In this case the sooner the expire the better. */
idle = ULLONG_MAX - (long)dictGetVal(de);
} else {
serverPanic("Unknown eviction policy in evictionPoolPopulate()");
}
/* Insert the element inside the pool.
* First, find the first empty bucket or the first populated
* bucket that has an idle time smaller than our idle time. */
k = 0;
while (k < EVPOOL_SIZE &&
pool[k].key &&
pool[k].idle < idle) k++;
if (k == 0 && pool[EVPOOL_SIZE-1].key != NULL) {
/* Can't insert if the element is < the worst element we have
* and there are no empty buckets. */
continue;
} else if (k < EVPOOL_SIZE && pool[k].key == NULL) {
/* Inserting into empty position. No setup needed before insert. */
} else {
/* Inserting in the middle. Now k points to the first element
* greater than the element to insert. */
if (pool[EVPOOL_SIZE-1].key == NULL) {
/* Free space on the right? Insert at k shifting
* all the elements from k to end to the right. */
/* Save SDS before overwriting. */
sds cached = pool[EVPOOL_SIZE-1].cached;
memmove(pool+k+1,pool+k,
sizeof(pool[0])*(EVPOOL_SIZE-k-1));
pool[k].cached = cached;
} else {
/* No free space on right? Insert at k-1 */
k--;
/* Shift all elements on the left of k (included) to the
* left, so we discard the element with smaller idle time. */
sds cached = pool[0].cached; /* Save SDS before overwriting. */
if (pool[0].key != pool[0].cached) sdsfree(pool[0].key);
memmove(pool,pool+1,sizeof(pool[0])*k);
pool[k].cached = cached;
}
}
/* Try to reuse the cached SDS string allocated in the pool entry,
* because allocating and deallocating this object is costly
* (according to the profiler, not my fantasy. Remember:
* premature optimizbla bla bla bla. */
int klen = sdslen(key);
if (klen > EVPOOL_CACHED_SDS_SIZE) {
pool[k].key = sdsdup(key);
} else {
memcpy(pool[k].cached,key,klen+1);
sdssetlen(pool[k].cached,klen);
pool[k].key = pool[k].cached;
}
pool[k].idle = idle;
pool[k].dbid = dbid;
}
}
static rstatus_t
gossip_forward_state(struct server_pool *sp)
{
//assume each record needs maximum 256 bytes
struct ring_msg *msg = create_ring_msg_with_data(256 * node_count);
uint8_t *data = msg->data; //dn_zalloc(sizeof(uint8_t) * 256 * node_count);//msg->data;
uint8_t *pos = data;
dictIterator *dc_it;
dictEntry *dc_de;
dc_it = dictGetIterator(gn_pool.dict_dc);
while ((dc_de = dictNext(dc_it)) != NULL) {
struct gossip_dc *g_dc = dictGetVal(dc_de);
//log_debug(LOG_VERB, "\tDC name : '%.*s'", g_dc->name.len, g_dc->name.data);
dictIterator *rack_it = dictGetIterator(g_dc->dict_rack);
dictEntry *rack_de;
while ((rack_de = dictNext(rack_it)) != NULL) {
struct gossip_rack *g_rack = dictGetVal(rack_de);
//log_debug(LOG_VERB, "\tRack name : '%.*s'", g_rack->name.len, g_rack->name.data);
dictIterator *node_it = dictGetIterator(g_rack->dict_token_nodes);
dictEntry *node_de;
int i = 0;
while ((node_de = dictNext(node_it)) != NULL) {
struct node *gnode = dictGetVal(node_de);
//log_debug(LOG_VERB, "\tNode name : '%.*s'", gnode->name.len, gnode->name.data);
if (i++ > 0) {
//pipe separator
*pos = '|';
pos += 1;
}
//write dc name
dn_memcpy(pos, g_dc->name.data, g_dc->name.len);
pos += g_dc->name.len;
//$ separator
*pos = '$';
pos += 1;
//write rack name
dn_memcpy(pos, g_rack->name.data, g_rack->name.len);
pos += g_rack->name.len;
//$ separator
*pos = '$';
pos += 1;
//write node token
struct string *token_str = dictGetKey(node_de);
//log_debug(LOG_VERB, "\tToken string : '%.*s'", token_str->len, token_str->data);
int k;
for(k=0; k<token_str->len;k++, pos++) {
*pos = *(token_str->data + k);
}
//comma separator
*pos = ',';
pos += 1;
//write ts
int count = 0;
uint64_t ts;
if (gnode->is_local) //only update my own timestamp
ts = (uint64_t) time(NULL);
else ts = gnode->ts;
count = 0;
write_number(pos, ts, &count);
pos += count;
//comma separator
*pos = ',';
pos += 1;
//write state
uint8_t new_state = gossip_failure_detector(gnode);
gnode->state = new_state;
count = 0;
write_number(pos, gnode->state, &count);
pos += count;
//comma separator
*pos = ',';
pos += 1;
//write addresss
for(k=0; k<gnode->name.len; k++, pos++) {
*pos = *(gnode->name.data + k);
}
}
dictReleaseIterator(node_it);
}
dictReleaseIterator(rack_it);
}
msg->len = pos-data;
log_debug(LOG_VERB, "\tForwarding my current gossip states : '%.*s'", (pos-data), data);
dictReleaseIterator(dc_it);
return gossip_ring_msg_to_core(sp, msg, dnode_peer_forward_state);
}
void Object::print(FILE *f) {
Object obj;
int i;
switch (type) {
case objBool:
fprintf(f, "%s", booln ? "true" : "false");
break;
case objInt:
fprintf(f, "%d", intg);
break;
case objReal:
fprintf(f, "%g", real);
break;
case objString:
fprintf(f, "(");
fwrite(string->getCString(), 1, string->getLength(), f);
fprintf(f, ")");
break;
case objName:
fprintf(f, "/%s", name);
break;
case objNull:
fprintf(f, "null");
break;
case objArray:
fprintf(f, "[");
for (i = 0; i < arrayGetLength(); ++i) {
if (i > 0)
fprintf(f, " ");
arrayGetNF(i, &obj);
obj.print(f);
obj.free();
}
fprintf(f, "]");
break;
case objDict:
fprintf(f, "<<");
for (i = 0; i < dictGetLength(); ++i) {
fprintf(f, " /%s ", dictGetKey(i));
dictGetValNF(i, &obj);
obj.print(f);
obj.free();
}
fprintf(f, " >>");
break;
case objStream:
fprintf(f, "<stream>");
break;
case objRef:
fprintf(f, "%d %d R", ref.num, ref.gen);
break;
case objCmd:
fprintf(f, "%s", cmd);
break;
case objError:
fprintf(f, "<error>");
break;
case objEOF:
fprintf(f, "<EOF>");
break;
case objNone:
fprintf(f, "<none>");
break;
}
}
static long
_term_version(dictEntry *entry) {
Term *term = dictGetKey(entry);
return term->offset;
}
/* Emit the commands needed to rebuild a sorted set object.
* The function returns 0 on error, 1 on success. */
int rewriteSortedSetObject(rio *r, robj *key, robj *o) {
long long count = 0, items = zsetLength(o);
if (o->encoding == REDIS_ENCODING_ZIPLIST) {
unsigned char *zl = o->ptr;
unsigned char *eptr, *sptr;
unsigned char *vstr;
unsigned int vlen;
long long vll;
double score;
eptr = ziplistIndex(zl,0);
redisAssert(eptr != NULL);
sptr = ziplistNext(zl,eptr);
redisAssert(sptr != NULL);
while (eptr != NULL) {
redisAssert(ziplistGet(eptr,&vstr,&vlen,&vll));
score = zzlGetScore(sptr);
if (count == 0) {
int cmd_items = (items > REDIS_AOF_REWRITE_ITEMS_PER_CMD) ?
REDIS_AOF_REWRITE_ITEMS_PER_CMD : items;
if (rioWriteBulkCount(r,'*',2+cmd_items*2) == 0) return 0;
if (rioWriteBulkString(r,"ZADD",4) == 0) return 0;
if (rioWriteBulkObject(r,key) == 0) return 0;
}
if (rioWriteBulkDouble(r,score) == 0) return 0;
if (vstr != NULL) {
if (rioWriteBulkString(r,(char*)vstr,vlen) == 0) return 0;
} else {
if (rioWriteBulkLongLong(r,vll) == 0) return 0;
}
zzlNext(zl,&eptr,&sptr);
if (++count == REDIS_AOF_REWRITE_ITEMS_PER_CMD) count = 0;
items--;
}
} else if (o->encoding == REDIS_ENCODING_SKIPLIST) {
zset *zs = o->ptr;
dictIterator *di = dictGetIterator(zs->dict);
dictEntry *de;
while((de = dictNext(di)) != NULL) {
robj *eleobj = dictGetKey(de);
double *score = dictGetVal(de);
if (count == 0) {
int cmd_items = (items > REDIS_AOF_REWRITE_ITEMS_PER_CMD) ?
REDIS_AOF_REWRITE_ITEMS_PER_CMD : items;
if (rioWriteBulkCount(r,'*',2+cmd_items*2) == 0) return 0;
if (rioWriteBulkString(r,"ZADD",4) == 0) return 0;
if (rioWriteBulkObject(r,key) == 0) return 0;
}
if (rioWriteBulkDouble(r,*score) == 0) return 0;
if (rioWriteBulkObject(r,eleobj) == 0) return 0;
if (++count == REDIS_AOF_REWRITE_ITEMS_PER_CMD) count = 0;
items--;
}
dictReleaseIterator(di);
} else {
redisPanic("Unknown sorted zset encoding");
}
return 1;
}
/* Write a sequence of commands able to fully rebuild the dataset into
* "filename". Used both by REWRITEAOF and BGREWRITEAOF.
*
* In order to minimize the number of commands needed in the rewritten
* log Redis uses variadic commands when possible, such as RPUSH, SADD
* and ZADD. However at max REDIS_AOF_REWRITE_ITEMS_PER_CMD items per time
* are inserted using a single command. */
int rewriteAppendOnlyFile(char *filename) {
dictIterator *di = NULL;
dictEntry *de;
rio aof;
FILE *fp;
char tmpfile[256];
int j;
long long now = mstime();
/* Note that we have to use a different temp name here compared to the
* one used by rewriteAppendOnlyFileBackground() function. */
snprintf(tmpfile,256,"temp-rewriteaof-%d.aof", (int) getpid());
fp = fopen(tmpfile,"w");
if (!fp) {
redisLog(REDIS_WARNING, "Opening the temp file for AOF rewrite in rewriteAppendOnlyFile(): %s", strerror(errno));
return REDIS_ERR;
}
rioInitWithFile(&aof,fp);
for (j = 0; j < server.dbnum; j++) {
char selectcmd[] = "*2\r\n$6\r\nSELECT\r\n";
redisDb *db = server.db+j;
dict *d = db->dict;
if (dictSize(d) == 0) continue;
di = dictGetSafeIterator(d);
if (!di) {
fclose(fp);
return REDIS_ERR;
}
/* SELECT the new DB */
if (rioWrite(&aof,selectcmd,sizeof(selectcmd)-1) == 0) goto werr;
if (rioWriteBulkLongLong(&aof,j) == 0) goto werr;
/* Iterate this DB writing every entry */
while((de = dictNext(di)) != NULL) {
sds keystr;
robj key, *o;
long long expiretime;
keystr = dictGetKey(de);
o = dictGetVal(de);
initStaticStringObject(key,keystr);
expiretime = getExpire(db,&key);
/* Save the key and associated value */
if (o->type == REDIS_STRING) {
/* Emit a SET command */
char cmd[]="*3\r\n$3\r\nSET\r\n";
if (rioWrite(&aof,cmd,sizeof(cmd)-1) == 0) goto werr;
/* Key and value */
if (rioWriteBulkObject(&aof,&key) == 0) goto werr;
if (rioWriteBulkObject(&aof,o) == 0) goto werr;
} else if (o->type == REDIS_LIST) {
if (rewriteListObject(&aof,&key,o) == 0) goto werr;
} else if (o->type == REDIS_SET) {
if (rewriteSetObject(&aof,&key,o) == 0) goto werr;
} else if (o->type == REDIS_ZSET) {
if (rewriteSortedSetObject(&aof,&key,o) == 0) goto werr;
} else if (o->type == REDIS_HASH) {
if (rewriteHashObject(&aof,&key,o) == 0) goto werr;
} else {
redisPanic("Unknown object type");
}
/* Save the expire time */
if (expiretime != -1) {
char cmd[]="*3\r\n$9\r\nPEXPIREAT\r\n";
/* If this key is already expired skip it */
if (expiretime < now) continue;
if (rioWrite(&aof,cmd,sizeof(cmd)-1) == 0) goto werr;
if (rioWriteBulkObject(&aof,&key) == 0) goto werr;
if (rioWriteBulkLongLong(&aof,expiretime) == 0) goto werr;
}
}
dictReleaseIterator(di);
}
/* Make sure data will not remain on the OS's output buffers */
fflush(fp);
aof_fsync(fileno(fp));
fclose(fp);
/* Use RENAME to make sure the DB file is changed atomically only
* if the generate DB file is ok. */
if (rename(tmpfile,filename) == -1) {
redisLog(REDIS_WARNING,"Error moving temp append only file on the final destination: %s", strerror(errno));
unlink(tmpfile);
return REDIS_ERR;
}
redisLog(REDIS_NOTICE,"SYNC append only file rewrite performed");
return REDIS_OK;
werr:
fclose(fp);
unlink(tmpfile);
redisLog(REDIS_WARNING,"Write error writing append only file on disk: %s", strerror(errno));
if (di) dictReleaseIterator(di);
return REDIS_ERR;
}
int freeMemoryIfNeeded(void) {
size_t mem_reported, mem_used, mem_tofree, mem_freed;
mstime_t latency, eviction_latency;
long long delta;
int slaves = listLength(server.slaves);
/* When clients are paused the dataset should be static not just from the
* POV of clients not being able to write, but also from the POV of
* expires and evictions of keys not being performed. */
if (clientsArePaused()) return C_OK;
/* Check if we are over the memory usage limit. If we are not, no need
* to subtract the slaves output buffers. We can just return ASAP. */
mem_reported = zmalloc_used_memory();
if (mem_reported <= server.maxmemory) return C_OK;
/* Remove the size of slaves output buffers and AOF buffer from the
* count of used memory. */
mem_used = mem_reported;
size_t overhead = freeMemoryGetNotCountedMemory();
mem_used = (mem_used > overhead) ? mem_used-overhead : 0;
/* Check if we are still over the memory limit. */
if (mem_used <= server.maxmemory) return C_OK;
/* Compute how much memory we need to free. */
mem_tofree = mem_used - server.maxmemory;
mem_freed = 0;
if (server.maxmemory_policy == MAXMEMORY_NO_EVICTION)
goto cant_free; /* We need to free memory, but policy forbids. */
latencyStartMonitor(latency);
while (mem_freed < mem_tofree) {
int j, k, i, keys_freed = 0;
static int next_db = 0;
sds bestkey = NULL;
int bestdbid;
redisDb *db;
dict *dict;
dictEntry *de;
if (server.maxmemory_policy & (MAXMEMORY_FLAG_LRU|MAXMEMORY_FLAG_LFU) ||
server.maxmemory_policy == MAXMEMORY_VOLATILE_TTL)
{
struct evictionPoolEntry *pool = EvictionPoolLRU;
while(bestkey == NULL) {
unsigned long total_keys = 0, keys;
/* We don't want to make local-db choices when expiring keys,
* so to start populate the eviction pool sampling keys from
* every DB. */
for (i = 0; i < server.dbnum; i++) {
db = server.db+i;
dict = (server.maxmemory_policy & MAXMEMORY_FLAG_ALLKEYS) ?
db->dict : db->expires;
if ((keys = dictSize(dict)) != 0) {
evictionPoolPopulate(i, dict, db->dict, pool);
total_keys += keys;
}
}
if (!total_keys) break; /* No keys to evict. */
/* Go backward from best to worst element to evict. */
for (k = EVPOOL_SIZE-1; k >= 0; k--) {
if (pool[k].key == NULL) continue;
bestdbid = pool[k].dbid;
if (server.maxmemory_policy & MAXMEMORY_FLAG_ALLKEYS) {
de = dictFind(server.db[pool[k].dbid].dict,
pool[k].key);
} else {
de = dictFind(server.db[pool[k].dbid].expires,
pool[k].key);
}
/* Remove the entry from the pool. */
if (pool[k].key != pool[k].cached)
sdsfree(pool[k].key);
pool[k].key = NULL;
pool[k].idle = 0;
/* If the key exists, is our pick. Otherwise it is
* a ghost and we need to try the next element. */
if (de) {
bestkey = dictGetKey(de);
break;
} else {
/* Ghost... Iterate again. */
}
}
}
}
/* volatile-random and allkeys-random policy */
else if (server.maxmemory_policy == MAXMEMORY_ALLKEYS_RANDOM ||
server.maxmemory_policy == MAXMEMORY_VOLATILE_RANDOM)
{
/* When evicting a random key, we try to evict a key for
* each DB, so we use the static 'next_db' variable to
* incrementally visit all DBs. */
for (i = 0; i < server.dbnum; i++) {
j = (++next_db) % server.dbnum;
db = server.db+j;
dict = (server.maxmemory_policy == MAXMEMORY_ALLKEYS_RANDOM) ?
db->dict : db->expires;
if (dictSize(dict) != 0) {
de = dictGetRandomKey(dict);
bestkey = dictGetKey(de);
bestdbid = j;
break;
}
}
}
/* Finally remove the selected key. */
if (bestkey) {
db = server.db+bestdbid;
robj *keyobj = createStringObject(bestkey,sdslen(bestkey));
propagateExpire(db,keyobj,server.lazyfree_lazy_eviction);
/* We compute the amount of memory freed by db*Delete() alone.
* It is possible that actually the memory needed to propagate
* the DEL in AOF and replication link is greater than the one
* we are freeing removing the key, but we can't account for
* that otherwise we would never exit the loop.
*
* AOF and Output buffer memory will be freed eventually so
* we only care about memory used by the key space. */
delta = (long long) zmalloc_used_memory();
latencyStartMonitor(eviction_latency);
if (server.lazyfree_lazy_eviction)
dbAsyncDelete(db,keyobj);
else
dbSyncDelete(db,keyobj);
latencyEndMonitor(eviction_latency);
latencyAddSampleIfNeeded("eviction-del",eviction_latency);
latencyRemoveNestedEvent(latency,eviction_latency);
delta -= (long long) zmalloc_used_memory();
mem_freed += delta;
server.stat_evictedkeys++;
notifyKeyspaceEvent(NOTIFY_EVICTED, "evicted",
keyobj, db->id);
decrRefCount(keyobj);
keys_freed++;
/* When the memory to free starts to be big enough, we may
* start spending so much time here that is impossible to
* deliver data to the slaves fast enough, so we force the
* transmission here inside the loop. */
if (slaves) flushSlavesOutputBuffers();
/* Normally our stop condition is the ability to release
* a fixed, pre-computed amount of memory. However when we
* are deleting objects in another thread, it's better to
* check, from time to time, if we already reached our target
* memory, since the "mem_freed" amount is computed only
* across the dbAsyncDelete() call, while the thread can
* release the memory all the time. */
if (server.lazyfree_lazy_eviction && !(keys_freed % 16)) {
overhead = freeMemoryGetNotCountedMemory();
mem_used = zmalloc_used_memory();
mem_used = (mem_used > overhead) ? mem_used-overhead : 0;
if (mem_used <= server.maxmemory) {
mem_freed = mem_tofree;
}
}
}
if (!keys_freed) {
latencyEndMonitor(latency);
latencyAddSampleIfNeeded("eviction-cycle",latency);
goto cant_free; /* nothing to free... */
}
}
latencyEndMonitor(latency);
latencyAddSampleIfNeeded("eviction-cycle",latency);
return C_OK;
cant_free:
/* We are here if we are not able to reclaim memory. There is only one
* last thing we can try: check if the lazyfree thread has jobs in queue
* and wait... */
while(bioPendingJobsOfType(BIO_LAZY_FREE)) {
if (((mem_reported - zmalloc_used_memory()) + mem_freed) >= mem_tofree)
break;
usleep(1000);
}
return C_ERR;
}
void srandmemberWithCountCommand(redisClient *c) {
long l;
unsigned long count, size;
int uniq = 1;
robj *set, *ele;
int64_t llele;
int encoding;
dict *d;
if (getLongFromObjectOrReply(c,c->argv[2],&l,NULL) != REDIS_OK) return;
if (l >= 0) {
count = (unsigned) l;
} else {
/* A negative count means: return the same elements multiple times
* (i.e. don't remove the extracted element after every extraction). */
count = -l;
uniq = 0;
}
if ((set = lookupKeyReadOrReply(c,c->argv[1],shared.emptymultibulk))
== NULL || checkType(c,set,REDIS_SET)) return;
size = setTypeSize(set);
/* If count is zero, serve it ASAP to avoid special cases later. */
if (count == 0) {
addReply(c,shared.emptymultibulk);
return;
}
/* CASE 1: The count was negative, so the extraction method is just:
* "return N random elements" sampling the whole set every time.
* This case is trivial and can be served without auxiliary data
* structures. */
if (!uniq) {
addReplyMultiBulkLen(c,count);
while(count--) {
encoding = setTypeRandomElement(set,&ele,&llele);
if (encoding == REDIS_ENCODING_INTSET) {
addReplyBulkLongLong(c,llele);
} else {
addReplyBulk(c,ele);
}
}
return;
}
/* CASE 2:
* The number of requested elements is greater than the number of
* elements inside the set: simply return the whole set. */
if (count >= size) {
sunionDiffGenericCommand(c,c->argv+1,1,NULL,REDIS_OP_UNION);
return;
}
/* For CASE 3 and CASE 4 we need an auxiliary dictionary. */
d = dictCreate(&setDictType,NULL);
/* CASE 3:
* The number of elements inside the set is not greater than
* SRANDMEMBER_SUB_STRATEGY_MUL times the number of requested elements.
* In this case we create a set from scratch with all the elements, and
* subtract random elements to reach the requested number of elements.
*
* This is done because if the number of requsted elements is just
* a bit less than the number of elements in the set, the natural approach
* used into CASE 3 is highly inefficient. */
if (count*SRANDMEMBER_SUB_STRATEGY_MUL > size) {
setTypeIterator *si;
/* Add all the elements into the temporary dictionary. */
si = setTypeInitIterator(set);
while((encoding = setTypeNext(si,&ele,&llele)) != -1) {
int retval = DICT_ERR;
if (encoding == REDIS_ENCODING_INTSET) {
retval = dictAdd(d,createStringObjectFromLongLong(llele),NULL);
} else if (ele->encoding == REDIS_ENCODING_RAW) {
retval = dictAdd(d,dupStringObject(ele),NULL);
} else if (ele->encoding == REDIS_ENCODING_INT) {
retval = dictAdd(d,
createStringObjectFromLongLong((long)ele->ptr),NULL);
}
redisAssert(retval == DICT_OK);
}
setTypeReleaseIterator(si);
redisAssert(dictSize(d) == size);
/* Remove random elements to reach the right count. */
while(size > count) {
dictEntry *de;
de = dictGetRandomKey(d);
dictDelete(d,dictGetKey(de));
size--;
}
}
/* CASE 4: We have a big set compared to the requested number of elements.
* In this case we can simply get random elements from the set and add
* to the temporary set, trying to eventually get enough unique elements
* to reach the specified count. */
else {
unsigned long added = 0;
while(added < count) {
encoding = setTypeRandomElement(set,&ele,&llele);
if (encoding == REDIS_ENCODING_INTSET) {
ele = createStringObjectFromLongLong(llele);
} else if (ele->encoding == REDIS_ENCODING_RAW) {
ele = dupStringObject(ele);
} else if (ele->encoding == REDIS_ENCODING_INT) {
ele = createStringObjectFromLongLong((long)ele->ptr);
}
/* Try to add the object to the dictionary. If it already exists
* free it, otherwise increment the number of objects we have
* in the result dictionary. */
if (dictAdd(d,ele,NULL) == DICT_OK)
added++;
else
decrRefCount(ele);
}
}
/* CASE 3 & 4: send the result to the user. */
{
dictIterator *di;
dictEntry *de;
addReplyMultiBulkLen(c,count);
di = dictGetIterator(d);
while((de = dictNext(di)) != NULL)
addReplyBulk(c,dictGetKey(de));
dictReleaseIterator(di);
dictRelease(d);
}
}
/* Create a human readable report of latency events for this Disque instance. */
sds createLatencyReport(void) {
sds report = sdsempty();
int advise_better_vm = 0; /* Better virtual machines. */
int advise_slowlog_enabled = 0; /* Enable slowlog. */
int advise_slowlog_tuning = 0; /* Reconfigure slowlog. */
int advise_slowlog_inspect = 0; /* Check your slowlog. */
int advise_disk_contention = 0; /* Try to lower disk contention. */
int advise_scheduler = 0; /* Intrinsic latency. */
int advise_data_writeback = 0; /* data=writeback. */
int advise_no_appendfsync = 0; /* don't fsync during rewrites. */
int advise_local_disk = 0; /* Avoid remote disks. */
int advise_ssd = 0; /* Use an SSD drive. */
int advise_write_load_info = 0; /* Print info about AOF and write load. */
int advise_hz = 0; /* Use higher HZ. */
int advise_large_objects = 0; /* Deletion of large objects. */
int advise_relax_fsync_policy = 0; /* appendfsync always is slow. */
int advices = 0;
/* Return ASAP if the latency engine is disabled and it looks like it
* was never enabled so far. */
if (dictSize(server.latency_events) == 0 &&
server.latency_monitor_threshold == 0)
{
report = sdscat(report,"I'm sorry, Dave, I can't do that. Latency monitoring is disabled in this Disque instance. You may use \"CONFIG SET latency-monitor-threshold <milliseconds>.\" in order to enable it. If we weren't in a deep space mission I'd suggest to take a look at http://disque.io/topics/latency-monitor.\n");
return report;
}
/* Show all the events stats and add for each event some event-related
* comment depending on the values. */
dictIterator *di;
dictEntry *de;
int eventnum = 0;
di = dictGetSafeIterator(server.latency_events);
while((de = dictNext(di)) != NULL) {
char *event = dictGetKey(de);
struct latencyTimeSeries *ts = dictGetVal(de);
struct latencyStats ls;
if (ts == NULL) continue;
eventnum++;
if (eventnum == 1) {
report = sdscat(report,"Dave, I have observed latency spikes in this Disque instance. You don't mind talking about it, do you Dave?\n\n");
}
analyzeLatencyForEvent(event,&ls);
report = sdscatprintf(report,
"%d. %s: %d latency spikes (average %lums, mean deviation %lums, period %.2f sec). Worst all time event %lums.",
eventnum, event,
ls.samples,
(unsigned long) ls.avg,
(unsigned long) ls.mad,
(double) ls.period/ls.samples,
(unsigned long) ts->max);
/* Fork */
if (!strcasecmp(event,"fork")) {
char *fork_quality;
if (server.stat_fork_rate < 10) {
fork_quality = "terrible";
advise_better_vm = 1;
advices++;
} else if (server.stat_fork_rate < 25) {
fork_quality = "poor";
advise_better_vm = 1;
advices++;
} else if (server.stat_fork_rate < 100) {
fork_quality = "good";
} else {
fork_quality = "excellent";
}
report = sdscatprintf(report,
" Fork rate is %.2f GB/sec (%s).", server.stat_fork_rate,
fork_quality);
}
/* Potentially commands. */
if (!strcasecmp(event,"command")) {
if (server.slowlog_log_slower_than == 0) {
advise_slowlog_enabled = 1;
advices++;
} else if (server.slowlog_log_slower_than/1000 >
server.latency_monitor_threshold)
{
advise_slowlog_tuning = 1;
advices++;
}
advise_slowlog_inspect = 1;
advise_large_objects = 1;
advices += 2;
}
/* fast-command. */
if (!strcasecmp(event,"fast-command")) {
advise_scheduler = 1;
advices++;
}
/* AOF and I/O. */
if (!strcasecmp(event,"aof-write-pending-fsync")) {
advise_local_disk = 1;
advise_disk_contention = 1;
advise_ssd = 1;
advise_data_writeback = 1;
advices += 4;
}
if (!strcasecmp(event,"aof-write-active-child")) {
advise_no_appendfsync = 1;
advise_data_writeback = 1;
advise_ssd = 1;
advices += 3;
}
if (!strcasecmp(event,"aof-write-alone")) {
advise_local_disk = 1;
advise_data_writeback = 1;
advise_ssd = 1;
advices += 3;
}
if (!strcasecmp(event,"aof-fsync-always")) {
advise_relax_fsync_policy = 1;
advices++;
}
if (!strcasecmp(event,"aof-fstat") ||
!strcasecmp(event,"rdb-unlik-temp-file")) {
advise_disk_contention = 1;
advise_local_disk = 1;
advices += 2;
}
if (!strcasecmp(event,"aof-rewrite-diff-write") ||
!strcasecmp(event,"aof-rename")) {
advise_write_load_info = 1;
advise_data_writeback = 1;
advise_ssd = 1;
advise_local_disk = 1;
advices += 4;
}
/* Expire cycle. */
if (!strcasecmp(event,"expire-cycle")) {
advise_hz = 1;
advise_large_objects = 1;
advices += 2;
}
/* Eviction cycle. */
if (!strcasecmp(event,"eviction-cycle")) {
advise_large_objects = 1;
advices++;
}
report = sdscatlen(report,"\n",1);
}
dictReleaseIterator(di);
if (eventnum == 0) {
report = sdscat(report,"Dave, no latency spike was observed during the lifetime of this Disque instance, not in the slightest bit. I honestly think you ought to sit down calmly, take a stress pill, and think things over.\n");
} else if (advices == 0) {
report = sdscat(report,"\nWhile there are latency events logged, I'm not able to suggest any easy fix. Please use the Disque community to get some help, providing this report in your help request.\n");
} else {
/* Add all the suggestions accumulated so far. */
/* Better VM. */
report = sdscat(report,"\nI have a few advices for you:\n\n");
if (advise_better_vm) {
report = sdscat(report,"- If you are using a virtual machine, consider upgrading it with a faster one using an hypervisior that provides less latency during fork() calls. Xen is known to have poor fork() performance. Even in the context of the same VM provider, certain kinds of instances can execute fork faster than others.\n");
}
/* Slow log. */
if (advise_slowlog_enabled) {
report = sdscatprintf(report,"- There are latency issues with potentially slow commands you are using. Try to enable the Slow Log Disque feature using the command 'CONFIG SET slowlog-log-slower-than %llu'. If the Slow log is disabled Disque is not able to log slow commands execution for you.\n", (unsigned long long)server.latency_monitor_threshold*1000);
}
if (advise_slowlog_tuning) {
report = sdscatprintf(report,"- Your current Slow Log configuration only logs events that are slower than your configured latency monitor threshold. Please use 'CONFIG SET slowlog-log-slower-than %llu'.\n", (unsigned long long)server.latency_monitor_threshold*1000);
}
if (advise_slowlog_inspect) {
report = sdscat(report,"- Check your Slow Log to understand what are the commands you are running which are too slow to execute. Please check http://disque.io/commands/slowlog for more information.\n");
}
/* Intrinsic latency. */
if (advise_scheduler) {
report = sdscat(report,"- The system is slow to execute Disque code paths not containing system calls. This usually means the system does not provide Disque CPU time to run for long periods. You should try to:\n"
" 1) Lower the system load.\n"
" 2) Use a computer / VM just for Disque if you are running other softawre in the same system.\n"
" 3) Check if you have a \"noisy neighbour\" problem.\n"
" 4) Check with 'disque-cli --intrinsic-latency 100' what is the intrinsic latency in your system.\n"
" 5) Check if the problem is allocator-related by recompiling Disque with MALLOC=libc, if you are using Jemalloc. However this may create fragmentation problems.\n");
}
/* AOF / Disk latency. */
if (advise_local_disk) {
report = sdscat(report,"- It is strongly advised to use local disks for persistence, especially if you are using AOF. Remote disks provided by platform-as-a-service providers are known to be slow.\n");
}
if (advise_ssd) {
report = sdscat(report,"- SSD disks are able to reduce fsync latency, and total time needed for snapshotting and AOF log rewriting (resulting in smaller memory usage and smaller final AOF rewrite buffer flushes). With extremely high write load SSD disks can be a good option. However Disque should perform reasonably with high load using normal disks. Use this advice as a last resort.\n");
}
if (advise_data_writeback) {
report = sdscat(report,"- Mounting ext3/4 filesystems with data=writeback can provide a performance boost compared to data=ordered, however this mode of operation provides less guarantees, and sometimes it can happen that after a hard crash the AOF file will have an half-written command at the end and will require to be repaired before Disque restarts.\n");
}
if (advise_disk_contention) {
report = sdscat(report,"- Try to lower the disk contention. This is often caused by other disk intensive processes running in the same computer (including other Disque instances).\n");
}
if (advise_no_appendfsync) {
report = sdscat(report,"- Assuming from the point of view of data safety this is viable in your environment, you could try to enable the 'no-appendfsync-on-rewrite' option, so that fsync will not be performed while there is a child rewriting the AOF file or producing an RDB file (the moment where there is high disk contention).\n");
}
if (advise_relax_fsync_policy && server.aof_fsync == AOF_FSYNC_ALWAYS) {
report = sdscat(report,"- Your fsync policy is set to 'always'. It is very hard to get good performances with such a setup, if possible try to relax the fsync policy to 'onesec'.\n");
}
if (advise_write_load_info) {
report = sdscat(report,"- Latency during the AOF atomic rename operation or when the final difference is flushed to the AOF file at the end of the rewrite, sometimes is caused by very high write load, causing the AOF buffer to get very large. If possible try to send less commands to accomplish the same work, or use Lua scripts to group multiple operations into a single EVALSHA call.\n");
}
if (advise_hz && server.hz < 100) {
report = sdscat(report,"- In order to make the Disque keys expiring process more incremental, try to set the 'hz' configuration parameter to 100 using 'CONFIG SET hz 100'.\n");
}
if (advise_large_objects) {
report = sdscat(report,"- Deleting, expiring or evicting (because of maxmemory policy) large objects is a blocking operation. If you have very large objects that are often deleted, expired, or evicted, try to fragment those objects into multiple smaller objects.\n");
}
}
return report;
}
/* The SORT command is the most complex command in Redis. Warning: this code
* is optimized for speed and a bit less for readability */
void sortCommand(redisClient *c) {
list *operations;
unsigned int outputlen = 0;
int desc = 0, alpha = 0;
long limit_start = 0, limit_count = -1, start, end;
int j, dontsort = 0, vectorlen;
int getop = 0; /* GET operation counter */
int int_convertion_error = 0;
robj *sortval, *sortby = NULL, *storekey = NULL;
redisSortObject *vector; /* Resulting vector to sort */
/* Lookup the key to sort. It must be of the right types */
sortval = lookupKeyRead(c->db,c->argv[1]);
if (sortval && sortval->type != REDIS_SET && sortval->type != REDIS_LIST &&
sortval->type != REDIS_ZSET)
{
addReply(c,shared.wrongtypeerr);
return;
}
/* Create a list of operations to perform for every sorted element.
* Operations can be GET/DEL/INCR/DECR */
operations = listCreate();
listSetFreeMethod(operations,zfree);
j = 2;
/* Now we need to protect sortval incrementing its count, in the future
* SORT may have options able to overwrite/delete keys during the sorting
* and the sorted key itself may get destroied */
if (sortval)
incrRefCount(sortval);
else
sortval = createListObject();
/* The SORT command has an SQL-alike syntax, parse it */
while(j < c->argc) {
int leftargs = c->argc-j-1;
if (!strcasecmp(c->argv[j]->ptr,"asc")) {
desc = 0;
} else if (!strcasecmp(c->argv[j]->ptr,"desc")) {
desc = 1;
} else if (!strcasecmp(c->argv[j]->ptr,"alpha")) {
alpha = 1;
} else if (!strcasecmp(c->argv[j]->ptr,"limit") && leftargs >= 2) {
if ((getLongFromObjectOrReply(c, c->argv[j+1], &limit_start, NULL) != REDIS_OK) ||
(getLongFromObjectOrReply(c, c->argv[j+2], &limit_count, NULL) != REDIS_OK)) return;
j+=2;
} else if (!strcasecmp(c->argv[j]->ptr,"store") && leftargs >= 1) {
storekey = c->argv[j+1];
j++;
} else if (!strcasecmp(c->argv[j]->ptr,"by") && leftargs >= 1) {
sortby = c->argv[j+1];
/* If the BY pattern does not contain '*', i.e. it is constant,
* we don't need to sort nor to lookup the weight keys. */
if (strchr(c->argv[j+1]->ptr,'*') == NULL) dontsort = 1;
j++;
} else if (!strcasecmp(c->argv[j]->ptr,"get") && leftargs >= 1) {
listAddNodeTail(operations,createSortOperation(
REDIS_SORT_GET,c->argv[j+1]));
getop++;
j++;
} else {
decrRefCount(sortval);
listRelease(operations);
addReply(c,shared.syntaxerr);
return;
}
j++;
}
/* If we have STORE we need to force sorting for deterministic output
* and replication. We use alpha sorting since this is guaranteed to
* work with any input. */
if (storekey && dontsort) {
dontsort = 0;
alpha = 1;
sortby = NULL;
}
/* Destructively convert encoded sorted sets for SORT. */
if (sortval->type == REDIS_ZSET)
zsetConvert(sortval, REDIS_ENCODING_SKIPLIST);
/* Load the sorting vector with all the objects to sort */
switch(sortval->type) {
case REDIS_LIST: vectorlen = listTypeLength(sortval); break;
case REDIS_SET: vectorlen = setTypeSize(sortval); break;
case REDIS_ZSET: vectorlen = dictSize(((zset*)sortval->ptr)->dict); break;
default: vectorlen = 0; redisPanic("Bad SORT type"); /* Avoid GCC warning */
}
vector = zmalloc(sizeof(redisSortObject)*vectorlen);
j = 0;
if (sortval->type == REDIS_LIST) {
listTypeIterator *li = listTypeInitIterator(sortval,0,REDIS_TAIL);
listTypeEntry entry;
while(listTypeNext(li,&entry)) {
vector[j].obj = listTypeGet(&entry);
vector[j].u.score = 0;
vector[j].u.cmpobj = NULL;
j++;
}
listTypeReleaseIterator(li);
} else if (sortval->type == REDIS_SET) {
setTypeIterator *si = setTypeInitIterator(sortval);
robj *ele;
while((ele = setTypeNextObject(si)) != NULL) {
vector[j].obj = ele;
vector[j].u.score = 0;
vector[j].u.cmpobj = NULL;
j++;
}
setTypeReleaseIterator(si);
} else if (sortval->type == REDIS_ZSET) {
dict *set = ((zset*)sortval->ptr)->dict;
dictIterator *di;
dictEntry *setele;
di = dictGetIterator(set);
while((setele = dictNext(di)) != NULL) {
vector[j].obj = dictGetKey(setele);
vector[j].u.score = 0;
vector[j].u.cmpobj = NULL;
j++;
}
dictReleaseIterator(di);
} else {
redisPanic("Unknown type");
}
redisAssertWithInfo(c,sortval,j == vectorlen);
/* Now it's time to load the right scores in the sorting vector */
if (dontsort == 0) {
for (j = 0; j < vectorlen; j++) {
robj *byval;
if (sortby) {
/* lookup value to sort by */
byval = lookupKeyByPattern(c->db,sortby,vector[j].obj);
if (!byval) continue;
} else {
/* use object itself to sort by */
byval = vector[j].obj;
}
if (alpha) {
if (sortby) vector[j].u.cmpobj = getDecodedObject(byval);
} else {
if (byval->encoding == REDIS_ENCODING_RAW) {
char *eptr;
vector[j].u.score = strtod(byval->ptr,&eptr);
if (eptr[0] != '\0' || errno == ERANGE ||
isnan(vector[j].u.score))
{
int_convertion_error = 1;
}
} else if (byval->encoding == REDIS_ENCODING_INT) {
/* Don't need to decode the object if it's
* integer-encoded (the only encoding supported) so
* far. We can just cast it */
vector[j].u.score = (long)byval->ptr;
} else {
redisAssertWithInfo(c,sortval,1 != 1);
}
}
/* when the object was retrieved using lookupKeyByPattern,
* its refcount needs to be decreased. */
if (sortby) {
decrRefCount(byval);
}
}
}
/* We are ready to sort the vector... perform a bit of sanity check
* on the LIMIT option too. We'll use a partial version of quicksort. */
start = (limit_start < 0) ? 0 : limit_start;
end = (limit_count < 0) ? vectorlen-1 : start+limit_count-1;
if (start >= vectorlen) {
start = vectorlen-1;
end = vectorlen-2;
}
if (end >= vectorlen) end = vectorlen-1;
server.sort_dontsort = dontsort;
if (dontsort == 0) {
server.sort_desc = desc;
server.sort_alpha = alpha;
server.sort_bypattern = sortby ? 1 : 0;
if (sortby && (start != 0 || end != vectorlen-1))
pqsort(vector,vectorlen,sizeof(redisSortObject),sortCompare, start,end);
else
qsort(vector,vectorlen,sizeof(redisSortObject),sortCompare);
}
/* Send command output to the output buffer, performing the specified
* GET/DEL/INCR/DECR operations if any. */
outputlen = getop ? getop*(end-start+1) : end-start+1;
if (int_convertion_error) {
addReplyError(c,"One or more scores can't be converted into double");
} else if (storekey == NULL) {
/* STORE option not specified, sent the sorting result to client */
addReplyMultiBulkLen(c,outputlen);
for (j = start; j <= end; j++) {
listNode *ln;
listIter li;
if (!getop) addReplyBulk(c,vector[j].obj);
listRewind(operations,&li);
while((ln = listNext(&li))) {
redisSortOperation *sop = ln->value;
robj *val = lookupKeyByPattern(c->db,sop->pattern,
vector[j].obj);
if (sop->type == REDIS_SORT_GET) {
if (!val) {
addReply(c,shared.nullbulk);
} else {
addReplyBulk(c,val);
decrRefCount(val);
}
} else {
/* Always fails */
redisAssertWithInfo(c,sortval,sop->type == REDIS_SORT_GET);
}
}
}
} else {
robj *sobj = createZiplistObject();
/* STORE option specified, set the sorting result as a List object */
for (j = start; j <= end; j++) {
listNode *ln;
listIter li;
if (!getop) {
listTypePush(sobj,vector[j].obj,REDIS_TAIL);
} else {
listRewind(operations,&li);
while((ln = listNext(&li))) {
redisSortOperation *sop = ln->value;
robj *val = lookupKeyByPattern(c->db,sop->pattern,
vector[j].obj);
if (sop->type == REDIS_SORT_GET) {
if (!val) val = createStringObject("",0);
/* listTypePush does an incrRefCount, so we should take care
* care of the incremented refcount caused by either
* lookupKeyByPattern or createStringObject("",0) */
listTypePush(sobj,val,REDIS_TAIL);
decrRefCount(val);
} else {
/* Always fails */
redisAssertWithInfo(c,sortval,sop->type == REDIS_SORT_GET);
}
}
}
}
if (outputlen) {
setKey(c->db,storekey,sobj);
server.dirty += outputlen;
} else if (dbDelete(c->db,storekey)) {
signalModifiedKey(c->db,storekey);
server.dirty++;
}
decrRefCount(sobj);
addReplyLongLong(c,outputlen);
}
/* Cleanup */
if (sortval->type == REDIS_LIST || sortval->type == REDIS_SET)
for (j = 0; j < vectorlen; j++)
decrRefCount(vector[j].obj);
decrRefCount(sortval);
listRelease(operations);
for (j = 0; j < vectorlen; j++) {
if (alpha && vector[j].u.cmpobj)
decrRefCount(vector[j].u.cmpobj);
}
zfree(vector);
}
static int
conf_parse(rmt_conf *cf)
{
int ret;
conf_pool *cp;
dict *orgs, *org;
dictEntry *de;
sds key;
if(cf == NULL){
return RMT_ERROR;
}
orgs = cf->organizations;
if(orgs == NULL){
return RMT_ERROR;
}
//source pool
key = sdsnew(CONF_ORGANIZATION_NAME_SOURCE);
de = dictFind(orgs, key);
if(de == NULL){
log_error("ERROR: Can not find %s organization in conf file %s",
CONF_ORGANIZATION_NAME_SOURCE, cf->fname);
sdsfree(key);
return RMT_ERROR;
}
org = dictGetVal(de);
if(org == NULL){
log_error("ERROR: Dict %s entry value is NULL", dictGetKey(de));
sdsfree(key);
return RMT_ERROR;
}
cp = &cf->source_pool;
ret = conf_parse_conf_pool(cp, org);
if(ret != RMT_OK){
log_error("ERROR: source pool conf parse error");
sdsfree(key);
return RMT_ERROR;
}
//target pool
key = sdscpy(key, CONF_ORGANIZATION_NAME_TARGET);
de = dictFind(orgs, key);
if(de == NULL){
log_error("ERROR: Can not find %s organization in conf file %s",
CONF_ORGANIZATION_NAME_TARGET, cf->fname);
sdsfree(key);
return RMT_ERROR;
}
org = dictGetVal(de);
if(org == NULL){
log_error("ERROR: Dict %s entry value is NULL", dictGetKey(de));
sdsfree(key);
return RMT_ERROR;
}
cp = &cf->target_pool;
ret = conf_parse_conf_pool(cp, org);
if(ret != RMT_OK){
log_error("ERROR: target pool conf parse error");
sdsfree(key);
return RMT_ERROR;
}
//common
key = sdscpy(key, CONF_ORGANIZATION_NAME_COMMAN);
de = dictFind(orgs, key);
if(de == NULL){
log_error("ERROR: Can not find %s organization in conf file %s",
CONF_ORGANIZATION_NAME_COMMAN, cf->fname);
sdsfree(key);
return RMT_ERROR;
}
org = dictGetVal(de);
if(org == NULL){
log_error("ERROR: Dict %s entry value is NULL", dictGetKey(de));
sdsfree(key);
return RMT_ERROR;
}
ret = conf_parse_conf_common(cf, org);
if(ret != RMT_OK){
log_error("ERROR: common conf parse error");
sdsfree(key);
return RMT_ERROR;
}
sdsfree(key);
return RMT_OK;
}
