/* Stores pointer to current the entry in the provided entry structure * and advances the position of the iterator. Returns 1 when the current * entry is in fact an entry, 0 otherwise. */ int listTypeNext(listTypeIterator *li, listTypeEntry *entry) { /* Protect from converting when iterating */ redisAssert(li->subject->encoding == li->encoding); entry->li = li; if (li->encoding == REDIS_ENCODING_ZIPLIST) { entry->zi = li->zi; if (entry->zi != NULL) { if (li->direction == REDIS_TAIL) li->zi = ziplistNext(li->subject->ptr,li->zi); else li->zi = ziplistPrev(li->subject->ptr,li->zi); return 1; } } else if (li->encoding == REDIS_ENCODING_LINKEDLIST) { entry->ln = li->ln; if (entry->ln != NULL) { if (li->direction == REDIS_TAIL) li->ln = li->ln->next; else li->ln = li->ln->prev; return 1; } } else { redisPanic("Unknown list encoding"); } return 0; }
void listTypeInsert(listTypeEntry *entry, robj *value, int where) { robj *subject = entry->li->subject; if (entry->li->encoding == REDIS_ENCODING_ZIPLIST) { value = getDecodedObject(value); if (where == REDIS_TAIL) { unsigned char *next = ziplistNext(subject->ptr,entry->zi); /* When we insert after the current element, but the current element * is the tail of the list, we need to do a push. */ if (next == NULL) { subject->ptr = ziplistPush(subject->ptr,value->ptr,sdslen(value->ptr),REDIS_TAIL); } else { subject->ptr = ziplistInsert(subject->ptr,next,value->ptr,sdslen(value->ptr)); } } else { subject->ptr = ziplistInsert(subject->ptr,entry->zi,value->ptr,sdslen(value->ptr)); } decrRefCount(value); } else if (entry->li->encoding == REDIS_ENCODING_LINKEDLIST) { if (where == REDIS_TAIL) { listInsertNode(subject->ptr,entry->ln,value,AL_START_TAIL); } else { listInsertNode(subject->ptr,entry->ln,value,AL_START_HEAD); } incrRefCount(value); } else { redisPanic("Unknown list encoding"); } }
/* Get the value from a ziplist encoded hash, identified by field. * Returns -1 when the field cannot be found. */ int hashTypeGetFromZiplist(robj *o, robj *field, unsigned char **vstr, unsigned int *vlen, long long *vll) { unsigned char *zl, *fptr = NULL, *vptr = NULL; int ret; logicErrorExpr(o->encoding == REDIS_ENCODING_ZIPLIST, "Never happend"); field = getDecodedObject(field); zl = o->ptr; fptr = ziplistIndex(zl, ZIPLIST_HEAD); if (fptr != NULL) { fptr = ziplistFind(fptr, field->ptr, sdslen(field->ptr), 1); if (fptr != NULL) { /* Grab pointer to the value (fptr points to the field) */ vptr = ziplistNext(zl, fptr); logicErrorExpr(vptr != NULL, "Never happend"); } } decrRefCount(field); if (vptr != NULL) { ret = ziplistGet(vptr, vstr, vlen, vll); logicErrorExpr(ret, "Never happend"); return 0; } return -1; }
void lrangeCommand(redisClient *c) { robj *o; long start, end, llen, rangelen; if ((getLongFromObjectOrReply(c, c->argv[2], &start, NULL) != REDIS_OK) || (getLongFromObjectOrReply(c, c->argv[3], &end, NULL) != REDIS_OK)) return; if ((o = lookupKeyReadOrReply(c,c->argv[1],shared.emptymultibulk)) == NULL || checkType(c,o,REDIS_LIST)) return; llen = listTypeLength(o); /* convert negative indexes */ if (start < 0) start = llen+start; if (end < 0) end = llen+end; if (start < 0) start = 0; /* Invariant: start >= 0, so this test will be true when end < 0. * The range is empty when start > end or start >= length. */ if (start > end || start >= llen) { addReply(c,shared.emptymultibulk); return; } if (end >= llen) end = llen-1; rangelen = (end-start)+1; /* Return the result in form of a multi-bulk reply */ addReplyMultiBulkLen(c,rangelen); if (o->encoding == REDIS_ENCODING_ZIPLIST) { unsigned char *p = ziplistIndex(o->ptr,start); unsigned char *vstr; unsigned int vlen; long long vlong; while(rangelen--) { ziplistGet(p,&vstr,&vlen,&vlong); if (vstr) { addReplyBulkCBuffer(c,vstr,vlen); } else { addReplyBulkLongLong(c,vlong); } p = ziplistNext(o->ptr,p); } } else if (o->encoding == REDIS_ENCODING_LINKEDLIST) { listNode *ln; /* If we are nearest to the end of the list, reach the element * starting from tail and going backward, as it is faster. */ if (start > llen/2) start -= llen; ln = listIndex(o->ptr,start); while(rangelen--) { addReplyBulk(c,ln->value); ln = ln->next; } } else { redisPanic("List encoding is not LINKEDLIST nor ZIPLIST!"); } }
/* Emit the commands needed to rebuild a list object. * The function returns 0 on error, 1 on success. */ int rewriteListObject(rio *r, robj *key, robj *o) { long long count = 0, items = listTypeLength(o); if (o->encoding == REDIS_ENCODING_ZIPLIST) { unsigned char *zl = o->ptr; unsigned char *p = ziplistIndex(zl,0); unsigned char *vstr; unsigned int vlen; long long vlong; while(ziplistGet(p,&vstr,&vlen,&vlong)) { 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,"RPUSH",5) == 0) return 0; if (rioWriteBulkObject(r,key) == 0) return 0; } if (vstr) { if (rioWriteBulkString(r,(char*)vstr,vlen) == 0) return 0; } else { if (rioWriteBulkLongLong(r,vlong) == 0) return 0; } p = ziplistNext(zl,p); if (++count == REDIS_AOF_REWRITE_ITEMS_PER_CMD) count = 0; items--; } } else if (o->encoding == REDIS_ENCODING_LINKEDLIST) { list *list = o->ptr; listNode *ln; listIter li; listRewind(list,&li); while((ln = listNext(&li))) { robj *eleobj = listNodeValue(ln); 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,"RPUSH",5) == 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--; } } else { redisPanic("Unknown list encoding"); } return 1; }
/* Add an element, discard the old if the key already exists. * Return 0 on insert and 1 on update. * This function will take care of incrementing the reference count of the * retained fields and value objects. */ int hashTypeSet(robj *o, robj *field, robj *value) { int update = 0; if (o->encoding == REDIS_ENCODING_ZIPLIST) { unsigned char *zl, *fptr, *vptr; field = getDecodedObject(field); value = getDecodedObject(value); zl = o->ptr; fptr = ziplistIndex(zl, ZIPLIST_HEAD); if (fptr != NULL) { fptr = ziplistFind(fptr, field->ptr, sdslen(field->ptr), 1); if (fptr != NULL) { /* Grab pointer to the value (fptr points to the field) */ vptr = ziplistNext(zl, fptr); logicErrorExpr(vptr != NULL, "Never happend"); update = 1; /* Delete value */ zl = ziplistDelete(zl, &vptr); /* Insert new value */ zl = ziplistInsert(zl, vptr, value->ptr, sdslen(value->ptr)); } } if (!update) { /* Push new field/value pair onto the tail of the ziplist */ zl = ziplistPush(zl, field->ptr, sdslen(field->ptr), ZIPLIST_TAIL); zl = ziplistPush(zl, value->ptr, sdslen(value->ptr), ZIPLIST_TAIL); } o->ptr = zl; decrRefCount(field); decrRefCount(value); /* Check if the ziplist needs to be converted to a hash table */ if (hashTypeLength(o) > server.hash_max_ziplist_entries) hashTypeConvert(o, REDIS_ENCODING_HT); } else if (o->encoding == REDIS_ENCODING_HT) { if (dictReplace(o->ptr, field, value)) { /* Insert */ incrRefCount(field); } else { /* Update */ update = 1; } incrRefCount(value); } else { logicError("Unknown hash encoding"); } return update; }
/* Get the value from a ziplist encoded hash, identified by field. * Returns -1 when the field cannot be found. */ int hashTypeGetFromZiplist(robj *o, robj *field, unsigned char **vstr, unsigned int *vlen, PORT_LONGLONG *vll) { unsigned char *zl, *fptr = NULL, *vptr = NULL; int ret; redisAssert(o->encoding == REDIS_ENCODING_ZIPLIST); field = getDecodedObject(field); zl = o->ptr; fptr = ziplistIndex(zl, ZIPLIST_HEAD); if (fptr != NULL) { fptr = ziplistFind(fptr, field->ptr, (unsigned int)sdslen(field->ptr), 1); WIN_PORT_FIX /* cast (unsigned int) */ if (fptr != NULL) { /* Grab pointer to the value (fptr points to the field) */ vptr = ziplistNext(zl, fptr); redisAssert(vptr != NULL); } }
static void *loadZsetZiplistObject(unsigned char* zl, unsigned int *rlen) { unsigned int i = 0, len; unsigned char *eptr, *sptr; unsigned char *vstr; unsigned int vlen; int buf_len; long long vlong; double score; char buf[128]; sds ele; len = ziplistLen (zl); eptr = ziplistIndex(zl,0); sptr = ziplistNext(zl,eptr); if(rdb_version < 2) { *rlen = len * 2; } else { *rlen = len; } sds *results = zmalloc(*rlen * sizeof(sds)); while (eptr != NULL) { score = zzlGetScore(sptr); ziplistGet(eptr,&vstr,&vlen,&vlong); if (vstr == NULL) ele = sdsfromlonglong(vlong); else ele = sdsnewlen((char*)vstr,vlen); results[i] = ele; buf_len = snprintf(buf, 128, "%f", score); results[i+1] = sdsnewlen(buf, buf_len); i += 2; zzlNext(zl,&eptr,&sptr); } return results; }
int main(int argc, char *argv[]) { unsigned char *zl, *p; unsigned char *entry; unsigned int elen; long long value; /* If an argument is given, use it as the random seed. */ if (argc == 2) srand(atoi(argv[1])); zl = createIntList(); ziplistRepr(zl); zl = createList(); ziplistRepr(zl); pop(zl,ZIPLIST_TAIL); ziplistRepr(zl); pop(zl,ZIPLIST_HEAD); ziplistRepr(zl); pop(zl,ZIPLIST_TAIL); ziplistRepr(zl); pop(zl,ZIPLIST_TAIL); ziplistRepr(zl); printf("Get element at index 3:\n"); { zl = createList(); p = ziplistIndex(zl, 3); if (!ziplistGet(p, &entry, &elen, &value)) { printf("ERROR: Could not access index 3\n"); return 1; } if (entry) { if (elen && fwrite(entry,elen,1,stdout) == 0) perror("fwrite"); printf("\n"); } else { printf("%lld\n", value); } printf("\n"); } printf("Get element at index 4 (out of range):\n"); { zl = createList(); p = ziplistIndex(zl, 4); if (p == NULL) { printf("No entry\n"); } else { printf("ERROR: Out of range index should return NULL, returned offset: %ld\n", p-zl); return 1; } printf("\n"); } printf("Get element at index -1 (last element):\n"); { zl = createList(); p = ziplistIndex(zl, -1); if (!ziplistGet(p, &entry, &elen, &value)) { printf("ERROR: Could not access index -1\n"); return 1; } if (entry) { if (elen && fwrite(entry,elen,1,stdout) == 0) perror("fwrite"); printf("\n"); } else { printf("%lld\n", value); } printf("\n"); } printf("Get element at index -4 (first element):\n"); { zl = createList(); p = ziplistIndex(zl, -4); if (!ziplistGet(p, &entry, &elen, &value)) { printf("ERROR: Could not access index -4\n"); return 1; } if (entry) { if (elen && fwrite(entry,elen,1,stdout) == 0) perror("fwrite"); printf("\n"); } else { printf("%lld\n", value); } printf("\n"); } printf("Get element at index -5 (reverse out of range):\n"); { zl = createList(); p = ziplistIndex(zl, -5); if (p == NULL) { printf("No entry\n"); } else { printf("ERROR: Out of range index should return NULL, returned offset: %ld\n", p-zl); return 1; } printf("\n"); } printf("Iterate list from 0 to end:\n"); { zl = createList(); p = ziplistIndex(zl, 0); while (ziplistGet(p, &entry, &elen, &value)) { printf("Entry: "); if (entry) { if (elen && fwrite(entry,elen,1,stdout) == 0) perror("fwrite"); } else { printf("%lld", value); } p = ziplistNext(zl,p); printf("\n"); } printf("\n"); } printf("Iterate list from 1 to end:\n"); { zl = createList(); p = ziplistIndex(zl, 1); while (ziplistGet(p, &entry, &elen, &value)) { printf("Entry: "); if (entry) { if (elen && fwrite(entry,elen,1,stdout) == 0) perror("fwrite"); } else { printf("%lld", value); } p = ziplistNext(zl,p); printf("\n"); } printf("\n"); } printf("Iterate list from 2 to end:\n"); { zl = createList(); p = ziplistIndex(zl, 2); while (ziplistGet(p, &entry, &elen, &value)) { printf("Entry: "); if (entry) { if (elen && fwrite(entry,elen,1,stdout) == 0) perror("fwrite"); } else { printf("%lld", value); } p = ziplistNext(zl,p); printf("\n"); } printf("\n"); } printf("Iterate starting out of range:\n"); { zl = createList(); p = ziplistIndex(zl, 4); if (!ziplistGet(p, &entry, &elen, &value)) { printf("No entry\n"); } else { printf("ERROR\n"); } printf("\n"); } printf("Iterate from back to front:\n"); { zl = createList(); p = ziplistIndex(zl, -1); while (ziplistGet(p, &entry, &elen, &value)) { printf("Entry: "); if (entry) { if (elen && fwrite(entry,elen,1,stdout) == 0) perror("fwrite"); } else { printf("%lld", value); } p = ziplistPrev(zl,p); printf("\n"); } printf("\n"); } printf("Iterate from back to front, deleting all items:\n"); { zl = createList(); p = ziplistIndex(zl, -1); while (ziplistGet(p, &entry, &elen, &value)) { printf("Entry: "); if (entry) { if (elen && fwrite(entry,elen,1,stdout) == 0) perror("fwrite"); } else { printf("%lld", value); } zl = ziplistDelete(zl,&p); p = ziplistPrev(zl,p); printf("\n"); } printf("\n"); } printf("Delete inclusive range 0,0:\n"); { zl = createList(); zl = ziplistDeleteRange(zl, 0, 1); ziplistRepr(zl); } printf("Delete inclusive range 0,1:\n"); { zl = createList(); zl = ziplistDeleteRange(zl, 0, 2); ziplistRepr(zl); } printf("Delete inclusive range 1,2:\n"); { zl = createList(); zl = ziplistDeleteRange(zl, 1, 2); ziplistRepr(zl); } printf("Delete with start index out of range:\n"); { zl = createList(); zl = ziplistDeleteRange(zl, 5, 1); ziplistRepr(zl); } printf("Delete with num overflow:\n"); { zl = createList(); zl = ziplistDeleteRange(zl, 1, 5); ziplistRepr(zl); } printf("Delete foo while iterating:\n"); { zl = createList(); p = ziplistIndex(zl,0); while (ziplistGet(p,&entry,&elen,&value)) { if (entry && strncmp("foo",(char*)entry,elen) == 0) { printf("Delete foo\n"); zl = ziplistDelete(zl,&p); } else { printf("Entry: "); if (entry) { if (elen && fwrite(entry,elen,1,stdout) == 0) perror("fwrite"); } else { printf("%lld",value); } p = ziplistNext(zl,p); printf("\n"); } } printf("\n"); ziplistRepr(zl); } printf("Regression test for >255 byte strings:\n"); { char v1[257],v2[257]; memset(v1,'x',256); memset(v2,'y',256); zl = ziplistNew(); zl = ziplistPush(zl,(unsigned char*)v1,strlen(v1),ZIPLIST_TAIL); zl = ziplistPush(zl,(unsigned char*)v2,strlen(v2),ZIPLIST_TAIL); /* Pop values again and compare their value. */ p = ziplistIndex(zl,0); assert(ziplistGet(p,&entry,&elen,&value)); assert(strncmp(v1,(char*)entry,elen) == 0); p = ziplistIndex(zl,1); assert(ziplistGet(p,&entry,&elen,&value)); assert(strncmp(v2,(char*)entry,elen) == 0); printf("SUCCESS\n\n"); } printf("Regression test deleting next to last entries:\n"); { char v[3][257]; zlentry e[3]; int i; for (i = 0; i < (sizeof(v)/sizeof(v[0])); i++) { memset(v[i], 'a' + i, sizeof(v[0])); } v[0][256] = '\0'; v[1][ 1] = '\0'; v[2][256] = '\0'; zl = ziplistNew(); for (i = 0; i < (sizeof(v)/sizeof(v[0])); i++) { zl = ziplistPush(zl, (unsigned char *) v[i], strlen(v[i]), ZIPLIST_TAIL); } verify(zl, e); assert(e[0].prevrawlensize == 1); assert(e[1].prevrawlensize == 5); assert(e[2].prevrawlensize == 1); /* Deleting entry 1 will increase `prevrawlensize` for entry 2 */ unsigned char *p = e[1].p; zl = ziplistDelete(zl, &p); verify(zl, e); assert(e[0].prevrawlensize == 1); assert(e[1].prevrawlensize == 5); printf("SUCCESS\n\n"); } printf("Create long list and check indices:\n"); { zl = ziplistNew(); char buf[32]; int i,len; for (i = 0; i < 1000; i++) { len = sprintf(buf,"%d",i); zl = ziplistPush(zl,(unsigned char*)buf,len,ZIPLIST_TAIL); } for (i = 0; i < 1000; i++) { p = ziplistIndex(zl,i); assert(ziplistGet(p,NULL,NULL,&value)); assert(i == value); p = ziplistIndex(zl,-i-1); assert(ziplistGet(p,NULL,NULL,&value)); assert(999-i == value); } printf("SUCCESS\n\n"); } printf("Compare strings with ziplist entries:\n"); { zl = createList(); p = ziplistIndex(zl,0); if (!ziplistCompare(p,(unsigned char*)"hello",5)) { printf("ERROR: not \"hello\"\n"); return 1; } if (ziplistCompare(p,(unsigned char*)"hella",5)) { printf("ERROR: \"hella\"\n"); return 1; } p = ziplistIndex(zl,3); if (!ziplistCompare(p,(unsigned char*)"1024",4)) { printf("ERROR: not \"1024\"\n"); return 1; } if (ziplistCompare(p,(unsigned char*)"1025",4)) { printf("ERROR: \"1025\"\n"); return 1; } printf("SUCCESS\n\n"); } printf("Stress with random payloads of different encoding:\n"); { int i,j,len,where; unsigned char *p; char buf[1024]; int buflen; list *ref; listNode *refnode; /* Hold temp vars from ziplist */ unsigned char *sstr; unsigned int slen; long long sval; for (i = 0; i < 20000; i++) { zl = ziplistNew(); ref = listCreate(); listSetFreeMethod(ref, sdsfree); len = rand() % 256; /* Create lists */ for (j = 0; j < len; j++) { where = (rand() & 1) ? ZIPLIST_HEAD : ZIPLIST_TAIL; if (rand() % 2) { buflen = randstring(buf,1,sizeof(buf)-1); } else { switch(rand() % 3) { case 0: buflen = sprintf(buf,"%lld",(0LL + rand()) >> 20); break; case 1: buflen = sprintf(buf,"%lld",(0LL + rand())); break; case 2: buflen = sprintf(buf,"%lld",(0LL + rand()) << 20); break; default: assert(NULL); } } /* Add to ziplist */ zl = ziplistPush(zl, (unsigned char*)buf, buflen, where); /* Add to reference list */ if (where == ZIPLIST_HEAD) { listAddNodeHead(ref,sdsnewlen(buf, buflen)); } else if (where == ZIPLIST_TAIL) { listAddNodeTail(ref,sdsnewlen(buf, buflen)); } else { assert(NULL); } } assert(listLength(ref) == ziplistLen(zl)); for (j = 0; j < len; j++) { /* Naive way to get elements, but similar to the stresser * executed from the Tcl test suite. */ p = ziplistIndex(zl,j); refnode = listIndex(ref,j); assert(ziplistGet(p,&sstr,&slen,&sval)); if (sstr == NULL) { buflen = sprintf(buf,"%lld",sval); } else { buflen = slen; memcpy(buf,sstr,buflen); buf[buflen] = '\0'; } assert(memcmp(buf,listNodeValue(refnode),buflen) == 0); } zfree(zl); listRelease(ref); } printf("SUCCESS\n\n"); } printf("Stress with variable ziplist size:\n"); { stress(ZIPLIST_HEAD,100000,16384,256); stress(ZIPLIST_TAIL,100000,16384,256); } return 0; }
/* This command implements SCAN, HSCAN and SSCAN commands. * If object 'o' is passed, then it must be a Hash or Set object, otherwise * if 'o' is NULL the command will operate on the dictionary associated with * the current database. * * When 'o' is not NULL the function assumes that the first argument in * the client arguments vector is a key so it skips it before iterating * in order to parse options. * * In the case of a Hash object the function returns both the field and value * of every element on the Hash. */ void scanGenericCommand(client *c, robj *o, unsigned long cursor) { int i, j; list *keys = listCreate(); listNode *node, *nextnode; long count = 10; sds pat = NULL; int patlen = 0, use_pattern = 0; dict *ht; /* Object must be NULL (to iterate keys names), or the type of the object * must be Set, Sorted Set, or Hash. */ serverAssert(o == NULL || o->type == OBJ_SET || o->type == OBJ_HASH || o->type == OBJ_ZSET); /* Set i to the first option argument. The previous one is the cursor. */ i = (o == NULL) ? 2 : 3; /* Skip the key argument if needed. */ /* Step 1: Parse options. */ while (i < c->argc) { j = c->argc - i; if (!strcasecmp(c->argv[i]->ptr, "count") && j >= 2) { if (getLongFromObjectOrReply(c, c->argv[i+1], &count, NULL) != C_OK) { goto cleanup; } if (count < 1) { addReply(c,shared.syntaxerr); goto cleanup; } i += 2; } else if (!strcasecmp(c->argv[i]->ptr, "match") && j >= 2) { pat = c->argv[i+1]->ptr; patlen = sdslen(pat); /* The pattern always matches if it is exactly "*", so it is * equivalent to disabling it. */ use_pattern = !(pat[0] == '*' && patlen == 1); i += 2; } else { addReply(c,shared.syntaxerr); goto cleanup; } } /* Step 2: Iterate the collection. * * Note that if the object is encoded with a ziplist, intset, or any other * representation that is not a hash table, we are sure that it is also * composed of a small number of elements. So to avoid taking state we * just return everything inside the object in a single call, setting the * cursor to zero to signal the end of the iteration. */ /* Handle the case of a hash table. */ ht = NULL; if (o == NULL) { ht = c->db->dict; } else if (o->type == OBJ_SET && o->encoding == OBJ_ENCODING_HT) { ht = o->ptr; } else if (o->type == OBJ_HASH && o->encoding == OBJ_ENCODING_HT) { ht = o->ptr; count *= 2; /* We return key / value for this type. */ } else if (o->type == OBJ_ZSET && o->encoding == OBJ_ENCODING_SKIPLIST) { zset *zs = o->ptr; ht = zs->dict; count *= 2; /* We return key / value for this type. */ } if (ht) { void *privdata[2]; /* We set the max number of iterations to ten times the specified * COUNT, so if the hash table is in a pathological state (very * sparsely populated) we avoid to block too much time at the cost * of returning no or very few elements. */ long maxiterations = count*10; /* We pass two pointers to the callback: the list to which it will * add new elements, and the object containing the dictionary so that * it is possible to fetch more data in a type-dependent way. */ privdata[0] = keys; privdata[1] = o; do { cursor = dictScan(ht, cursor, scanCallback, privdata); } while (cursor && maxiterations-- && listLength(keys) < (unsigned long)count); } else if (o->type == OBJ_SET) { int pos = 0; int64_t ll; while(intsetGet(o->ptr,pos++,&ll)) listAddNodeTail(keys,createStringObjectFromLongLong(ll)); cursor = 0; } else if (o->type == OBJ_HASH || o->type == OBJ_ZSET) { unsigned char *p = ziplistIndex(o->ptr,0); unsigned char *vstr; unsigned int vlen; long long vll; while(p) { ziplistGet(p,&vstr,&vlen,&vll); listAddNodeTail(keys, (vstr != NULL) ? createStringObject((char*)vstr,vlen) : createStringObjectFromLongLong(vll)); p = ziplistNext(o->ptr,p); } cursor = 0; } else { serverPanic("Not handled encoding in SCAN."); } /* Step 3: Filter elements. */ node = listFirst(keys); while (node) { robj *kobj = listNodeValue(node); nextnode = listNextNode(node); int filter = 0; /* Filter element if it does not match the pattern. */ if (!filter && use_pattern) { if (sdsEncodedObject(kobj)) { if (!stringmatchlen(pat, patlen, kobj->ptr, sdslen(kobj->ptr), 0)) filter = 1; } else { char buf[LONG_STR_SIZE]; int len; serverAssert(kobj->encoding == OBJ_ENCODING_INT); len = ll2string(buf,sizeof(buf),(long)kobj->ptr); if (!stringmatchlen(pat, patlen, buf, len, 0)) filter = 1; } } /* Filter element if it is an expired key. */ if (!filter && o == NULL && expireIfNeeded(c->db, kobj)) filter = 1; /* Remove the element and its associted value if needed. */ if (filter) { decrRefCount(kobj); listDelNode(keys, node); } /* If this is a hash or a sorted set, we have a flat list of * key-value elements, so if this element was filtered, remove the * value, or skip it if it was not filtered: we only match keys. */ if (o && (o->type == OBJ_ZSET || o->type == OBJ_HASH)) { node = nextnode; nextnode = listNextNode(node); if (filter) { kobj = listNodeValue(node); decrRefCount(kobj); listDelNode(keys, node); } } node = nextnode; } /* Step 4: Reply to the client. */ addReplyMultiBulkLen(c, 2); addReplyBulkLongLong(c,cursor); addReplyMultiBulkLen(c, listLength(keys)); while ((node = listFirst(keys)) != NULL) { robj *kobj = listNodeValue(node); addReplyBulk(c, kobj); decrRefCount(kobj); listDelNode(keys, node); } cleanup: listSetFreeMethod(keys,decrRefCountVoid); listRelease(keys); }
/* 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. */ int rewriteAppendOnlyFile(char *filename) { dictIterator *di = NULL; dictEntry *de; FILE *fp; char tmpfile[256]; int j; time_t now = time(NULL); /* 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, "Failed rewriting the append only file: %s", strerror(errno)); return REDIS_ERR; } 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 (fwrite(selectcmd,sizeof(selectcmd)-1,1,fp) == 0) goto werr; if (fwriteBulkLongLong(fp,j) == 0) goto werr; /* Iterate this DB writing every entry */ while((de = dictNext(di)) != NULL) { sds keystr = dictGetEntryKey(de); robj key, *o; time_t expiretime; int swapped; keystr = dictGetEntryKey(de); o = dictGetEntryVal(de); initStaticStringObject(key,keystr); /* If the value for this key is swapped, load a preview in memory. * We use a "swapped" flag to remember if we need to free the * value object instead to just increment the ref count anyway * in order to avoid copy-on-write of pages if we are forked() */ if (!server.vm_enabled || o->storage == REDIS_VM_MEMORY || o->storage == REDIS_VM_SWAPPING) { swapped = 0; } else { o = vmPreviewObject(o); swapped = 1; } 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 (fwrite(cmd,sizeof(cmd)-1,1,fp) == 0) goto werr; /* Key and value */ if (fwriteBulkObject(fp,&key) == 0) goto werr; if (fwriteBulkObject(fp,o) == 0) goto werr; } else if (o->type == REDIS_LIST) { /* Emit the RPUSHes needed to rebuild the list */ char cmd[]="*3\r\n$5\r\nRPUSH\r\n"; if (o->encoding == REDIS_ENCODING_ZIPLIST) { unsigned char *zl = o->ptr; unsigned char *p = ziplistIndex(zl,0); unsigned char *vstr; unsigned int vlen; long long vlong; while(ziplistGet(p,&vstr,&vlen,&vlong)) { if (fwrite(cmd,sizeof(cmd)-1,1,fp) == 0) goto werr; if (fwriteBulkObject(fp,&key) == 0) goto werr; if (vstr) { if (fwriteBulkString(fp,(char*)vstr,vlen) == 0) goto werr; } else { if (fwriteBulkLongLong(fp,vlong) == 0) goto werr; } p = ziplistNext(zl,p); } } else if (o->encoding == REDIS_ENCODING_LINKEDLIST) { list *list = o->ptr; listNode *ln; listIter li; listRewind(list,&li); while((ln = listNext(&li))) { robj *eleobj = listNodeValue(ln); if (fwrite(cmd,sizeof(cmd)-1,1,fp) == 0) goto werr; if (fwriteBulkObject(fp,&key) == 0) goto werr; if (fwriteBulkObject(fp,eleobj) == 0) goto werr; } } else { redisPanic("Unknown list encoding"); } } else if (o->type == REDIS_SET) { char cmd[]="*3\r\n$4\r\nSADD\r\n"; /* Emit the SADDs needed to rebuild the set */ if (o->encoding == REDIS_ENCODING_INTSET) { int ii = 0; int64_t llval; while(intsetGet(o->ptr,ii++,&llval)) { if (fwrite(cmd,sizeof(cmd)-1,1,fp) == 0) goto werr; if (fwriteBulkObject(fp,&key) == 0) goto werr; if (fwriteBulkLongLong(fp,llval) == 0) goto werr; } } else if (o->encoding == REDIS_ENCODING_HT) { dictIterator *di = dictGetIterator(o->ptr); dictEntry *de; while((de = dictNext(di)) != NULL) { robj *eleobj = dictGetEntryKey(de); if (fwrite(cmd,sizeof(cmd)-1,1,fp) == 0) goto werr; if (fwriteBulkObject(fp,&key) == 0) goto werr; if (fwriteBulkObject(fp,eleobj) == 0) goto werr; } dictReleaseIterator(di); } else { redisPanic("Unknown set encoding"); } } else if (o->type == REDIS_ZSET) { /* Emit the ZADDs needed to rebuild the sorted set */ char cmd[]="*4\r\n$4\r\nZADD\r\n"; 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 (fwrite(cmd,sizeof(cmd)-1,1,fp) == 0) goto werr; if (fwriteBulkObject(fp,&key) == 0) goto werr; if (fwriteBulkDouble(fp,score) == 0) goto werr; if (vstr != NULL) { if (fwriteBulkString(fp,(char*)vstr,vlen) == 0) goto werr; } else { if (fwriteBulkLongLong(fp,vll) == 0) goto werr; } zzlNext(zl,&eptr,&sptr); } } 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 = dictGetEntryKey(de); double *score = dictGetEntryVal(de); if (fwrite(cmd,sizeof(cmd)-1,1,fp) == 0) goto werr; if (fwriteBulkObject(fp,&key) == 0) goto werr; if (fwriteBulkDouble(fp,*score) == 0) goto werr; if (fwriteBulkObject(fp,eleobj) == 0) goto werr; } dictReleaseIterator(di); } else { redisPanic("Unknown sorted set encoding"); } } else if (o->type == REDIS_HASH) { char cmd[]="*4\r\n$4\r\nHSET\r\n"; /* Emit the HSETs needed to rebuild the hash */ if (o->encoding == REDIS_ENCODING_ZIPMAP) { unsigned char *p = zipmapRewind(o->ptr); unsigned char *field, *val; unsigned int flen, vlen; while((p = zipmapNext(p,&field,&flen,&val,&vlen)) != NULL) { if (fwrite(cmd,sizeof(cmd)-1,1,fp) == 0) goto werr; if (fwriteBulkObject(fp,&key) == 0) goto werr; if (fwriteBulkString(fp,(char*)field,flen) == 0) goto werr; if (fwriteBulkString(fp,(char*)val,vlen) == 0) goto werr; } } else { dictIterator *di = dictGetIterator(o->ptr); dictEntry *de; while((de = dictNext(di)) != NULL) { robj *field = dictGetEntryKey(de); robj *val = dictGetEntryVal(de); if (fwrite(cmd,sizeof(cmd)-1,1,fp) == 0) goto werr; if (fwriteBulkObject(fp,&key) == 0) goto werr; if (fwriteBulkObject(fp,field) == 0) goto werr; if (fwriteBulkObject(fp,val) == 0) goto werr; } dictReleaseIterator(di); } } else { redisPanic("Unknown object type"); } /* Save the expire time */ if (expiretime != -1) { char cmd[]="*3\r\n$8\r\nEXPIREAT\r\n"; /* If this key is already expired skip it */ if (expiretime < now) continue; if (fwrite(cmd,sizeof(cmd)-1,1,fp) == 0) goto werr; if (fwriteBulkObject(fp,&key) == 0) goto werr; if (fwriteBulkLongLong(fp,expiretime) == 0) goto werr; } if (swapped) decrRefCount(o); } 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; }