-/* Memory layout of a ziplist, containing "foo", "bar", "quux":
- * <zlbytes><zllen><len>"foo"<len>"bar"<len>"quux"
+/* The ziplist is a specially encoded dually linked list that is designed
+ * to be very memory efficient. It stores both strings and integer values,
+ * where integers are encoded as actual integers instead of a series of
+ * characters. It allows push and pop operations on either side of the list
+ * in O(1) time. However, because every operation requires a reallocation of
+ * the memory used by the ziplist, the actual complexity is related to the
+ * amount of memory used by the ziplist.
*
- * <zlbytes> is an unsigned integer to hold the number of bytes that
- * the ziplist occupies. This is stored to not have to traverse the ziplist
- * to know the new length when pushing.
+ * ----------------------------------------------------------------------------
*
- * <zllen> is the number of items in the ziplist. When this value is
- * greater than 254, we need to traverse the entire list to know
- * how many items it holds.
+ * ZIPLIST OVERALL LAYOUT:
+ * The general layout of the ziplist is as follows:
+ * <zlbytes><zltail><zllen><entry><entry><zlend>
*
- * <len> is the number of bytes occupied by a single entry. When this
- * number is greater than 253, the length will occupy 5 bytes, where
- * the extra bytes contain an unsigned integer to hold the length.
+ * <zlbytes> is an unsigned integer to hold the number of bytes that the
+ * ziplist occupies. This value needs to be stored to be able to resize the
+ * entire structure without the need to traverse it first.
+ *
+ * <zltail> is the offset to the last entry in the list. This allows a pop
+ * operation on the far side of the list without the need for full traversal.
+ *
+ * <zllen> is the number of entries.When this value is larger than 2**16-2,
+ * we need to traverse the entire list to know how many items it holds.
+ *
+ * <zlend> is a single byte special value, equal to 255, which indicates the
+ * end of the list.
+ *
+ * ZIPLIST ENTRIES:
+ * Every entry in the ziplist is prefixed by a header that contains two pieces
+ * of information. First, the length of the previous entry is stored to be
+ * able to traverse the list from back to front. Second, the encoding with an
+ * optional string length of the entry itself is stored.
+ *
+ * The length of the previous entry is encoded in the following way:
+ * If this length is smaller than 254 bytes, it will only consume a single
+ * byte that takes the length as value. When the length is greater than or
+ * equal to 254, it will consume 5 bytes. The first byte is set to 254 to
+ * indicate a larger value is following. The remaining 4 bytes take the
+ * length of the previous entry as value.
+ *
+ * The other header field of the entry itself depends on the contents of the
+ * entry. When the entry is a string, the first 2 bits of this header will hold
+ * the type of encoding used to store the length of the string, followed by the
+ * actual length of the string. When the entry is an integer the first 2 bits
+ * are both set to 1. The following 2 bits are used to specify what kind of
+ * integer will be stored after this header. An overview of the different
+ * types and encodings is as follows:
+ *
+ * |00pppppp| - 1 byte
+ * String value with length less than or equal to 63 bytes (6 bits).
+ * |01pppppp|qqqqqqqq| - 2 bytes
+ * String value with length less than or equal to 16383 bytes (14 bits).
+ * |10______|qqqqqqqq|rrrrrrrr|ssssssss|tttttttt| - 5 bytes
+ * String value with length greater than or equal to 16384 bytes.
+ * |1100____| - 1 byte
+ * Integer encoded as int16_t (2 bytes).
+ * |1101____| - 1 byte
+ * Integer encoded as int32_t (4 bytes).
+ * |1110____| - 1 byte
+ * Integer encoded as int64_t (8 bytes).
*/
#include <stdio.h>
int ll2string(char *s, size_t len, long long value);
-/* Important note: the ZIP_END value is used to depict the end of the
- * ziplist structure. When a pointer contains an entry, the first couple
- * of bytes contain the encoded length of the previous entry. This length
- * is encoded as ZIP_ENC_RAW length, so the first two bits will contain 00
- * and the byte will therefore never have a value of 255. */
#define ZIP_END 255
#define ZIP_BIGLEN 254
-/* Entry encoding */
-#define ZIP_ENC_RAW 0
-#define ZIP_ENC_INT16 1
-#define ZIP_ENC_INT32 2
-#define ZIP_ENC_INT64 3
-#define ZIP_ENCODING(p) ((p)[0] >> 6)
+/* Different encoding/length possibilities */
+#define ZIP_STR_06B (0 << 6)
+#define ZIP_STR_14B (1 << 6)
+#define ZIP_STR_32B (2 << 6)
+#define ZIP_INT_16B (0xc0 | 0<<4)
+#define ZIP_INT_32B (0xc0 | 1<<4)
+#define ZIP_INT_64B (0xc0 | 2<<4)
-/* Length encoding for raw entries */
-#define ZIP_LEN_INLINE 0
-#define ZIP_LEN_UINT16 1
-#define ZIP_LEN_UINT32 2
+/* Macro's to determine type */
+#define ZIP_IS_STR(enc) (((enc) & 0xc0) < 0xc0)
+#define ZIP_IS_INT(enc) (!ZIP_IS_STR(enc) && ((enc) & 0x30) < 0x30)
/* Utility macros */
#define ZIPLIST_BYTES(zl) (*((uint32_t*)(zl)))
unsigned char *p;
} zlentry;
+/* Return the encoding pointer to by 'p'. */
+static unsigned int zipEntryEncoding(unsigned char *p) {
+ /* String encoding: 2 MSBs */
+ unsigned char b = p[0] & 0xc0;
+ if (b < 0xc0) {
+ return b;
+ } else {
+ /* Integer encoding: 4 MSBs */
+ return p[0] & 0xf0;
+ }
+ assert(NULL);
+}
+
/* Return bytes needed to store integer encoded by 'encoding' */
-static unsigned int zipEncodingSize(unsigned char encoding) {
- if (encoding == ZIP_ENC_INT16) {
- return sizeof(int16_t);
- } else if (encoding == ZIP_ENC_INT32) {
- return sizeof(int32_t);
- } else if (encoding == ZIP_ENC_INT64) {
- return sizeof(int64_t);
+static unsigned int zipIntSize(unsigned char encoding) {
+ switch(encoding) {
+ case ZIP_INT_16B: return sizeof(int16_t);
+ case ZIP_INT_32B: return sizeof(int32_t);
+ case ZIP_INT_64B: return sizeof(int64_t);
}
assert(NULL);
}
/* Decode the encoded length pointed by 'p'. If a pointer to 'lensize' is
* provided, it is set to the number of bytes required to encode the length. */
static unsigned int zipDecodeLength(unsigned char *p, unsigned int *lensize) {
- unsigned char encoding = ZIP_ENCODING(p), lenenc;
+ unsigned char encoding = zipEntryEncoding(p);
unsigned int len;
- if (encoding == ZIP_ENC_RAW) {
- lenenc = (p[0] >> 4) & 0x3;
- if (lenenc == ZIP_LEN_INLINE) {
- len = p[0] & 0xf;
+ if (ZIP_IS_STR(encoding)) {
+ switch(encoding) {
+ case ZIP_STR_06B:
+ len = p[0] & 0x3f;
if (lensize) *lensize = 1;
- } else if (lenenc == ZIP_LEN_UINT16) {
- len = p[1] | (p[2] << 8);
- if (lensize) *lensize = 3;
- } else {
- len = p[1] | (p[2] << 8) | (p[3] << 16) | (p[4] << 24);
+ break;
+ case ZIP_STR_14B:
+ len = ((p[0] & 0x3f) << 6) | p[1];
+ if (lensize) *lensize = 2;
+ break;
+ case ZIP_STR_32B:
+ len = (p[1] << 24) | (p[2] << 16) | (p[3] << 8) | p[4];
if (lensize) *lensize = 5;
+ break;
+ default:
+ assert(NULL);
}
} else {
- len = zipEncodingSize(encoding);
+ len = zipIntSize(encoding);
if (lensize) *lensize = 1;
}
return len;
/* Encode the length 'l' writing it in 'p'. If p is NULL it just returns
* the amount of bytes required to encode such a length. */
-static unsigned int zipEncodeLength(unsigned char *p, char encoding, unsigned int rawlen) {
- unsigned char len = 1, lenenc, buf[5];
- if (encoding == ZIP_ENC_RAW) {
- if (rawlen <= 0xf) {
+static unsigned int zipEncodeLength(unsigned char *p, unsigned char encoding, unsigned int rawlen) {
+ unsigned char len = 1, buf[5];
+
+ if (ZIP_IS_STR(encoding)) {
+ /* Although encoding is given it may not be set for strings,
+ * so we determine it here using the raw length. */
+ if (rawlen <= 0x3f) {
if (!p) return len;
- lenenc = ZIP_LEN_INLINE;
- buf[0] = rawlen;
- } else if (rawlen <= 0xffff) {
- len += 2;
+ buf[0] = ZIP_STR_06B | rawlen;
+ } else if (rawlen <= 0x3fff) {
+ len += 1;
if (!p) return len;
- lenenc = ZIP_LEN_UINT16;
- buf[1] = (rawlen ) & 0xff;
- buf[2] = (rawlen >> 8) & 0xff;
+ buf[0] = ZIP_STR_14B | ((rawlen >> 8) & 0x3f);
+ buf[1] = rawlen & 0xff;
} else {
len += 4;
if (!p) return len;
- lenenc = ZIP_LEN_UINT32;
- buf[1] = (rawlen ) & 0xff;
- buf[2] = (rawlen >> 8) & 0xff;
- buf[3] = (rawlen >> 16) & 0xff;
- buf[4] = (rawlen >> 24) & 0xff;
+ buf[0] = ZIP_STR_32B;
+ buf[1] = (rawlen >> 24) & 0xff;
+ buf[2] = (rawlen >> 16) & 0xff;
+ buf[3] = (rawlen >> 8) & 0xff;
+ buf[4] = rawlen & 0xff;
}
- buf[0] = (lenenc << 4) | (buf[0] & 0xf);
+ } else {
+ /* Implies integer encoding, so length is always 1. */
+ if (!p) return len;
+ buf[0] = encoding;
}
- if (!p) return len;
- /* Apparently we need to store the length in 'p' */
- buf[0] = (encoding << 6) | (buf[0] & 0x3f);
+ /* Store this length at p */
memcpy(p,buf,len);
return len;
}
/* Great, the string can be encoded. Check what's the smallest
* of our encoding types that can hold this value. */
if (value >= INT16_MIN && value <= INT16_MAX) {
- *encoding = ZIP_ENC_INT16;
+ *encoding = ZIP_INT_16B;
} else if (value >= INT32_MIN && value <= INT32_MAX) {
- *encoding = ZIP_ENC_INT32;
+ *encoding = ZIP_INT_32B;
} else {
- *encoding = ZIP_ENC_INT64;
+ *encoding = ZIP_INT_64B;
}
*v = value;
return 1;
int16_t i16;
int32_t i32;
int64_t i64;
- if (encoding == ZIP_ENC_INT16) {
+ if (encoding == ZIP_INT_16B) {
i16 = value;
memcpy(p,&i16,sizeof(i16));
- } else if (encoding == ZIP_ENC_INT32) {
+ } else if (encoding == ZIP_INT_32B) {
i32 = value;
memcpy(p,&i32,sizeof(i32));
- } else if (encoding == ZIP_ENC_INT64) {
+ } else if (encoding == ZIP_INT_64B) {
i64 = value;
memcpy(p,&i64,sizeof(i64));
} else {
int16_t i16;
int32_t i32;
int64_t i64, ret;
- if (encoding == ZIP_ENC_INT16) {
+ if (encoding == ZIP_INT_16B) {
memcpy(&i16,p,sizeof(i16));
ret = i16;
- } else if (encoding == ZIP_ENC_INT32) {
+ } else if (encoding == ZIP_INT_32B) {
memcpy(&i32,p,sizeof(i32));
ret = i32;
- } else if (encoding == ZIP_ENC_INT64) {
+ } else if (encoding == ZIP_INT_64B) {
memcpy(&i64,p,sizeof(i64));
ret = i64;
} else {
e.prevrawlen = zipPrevDecodeLength(p,&e.prevrawlensize);
e.len = zipDecodeLength(p+e.prevrawlensize,&e.lensize);
e.headersize = e.prevrawlensize+e.lensize;
- e.encoding = ZIP_ENCODING(p+e.prevrawlensize);
+ e.encoding = zipEntryEncoding(p+e.prevrawlensize);
e.p = p;
return e;
}
unsigned int curlen = ZIPLIST_BYTES(zl), reqlen, prevlen = 0;
unsigned int offset, nextdiff = 0;
unsigned char *tail;
- unsigned char encoding = ZIP_ENC_RAW;
+ unsigned char encoding = 0;
long long value;
zlentry entry;
/* See if the entry can be encoded */
if (zipTryEncoding(s,slen,&value,&encoding)) {
- reqlen = zipEncodingSize(encoding);
+ /* 'encoding' is set to the appropriate integer encoding */
+ reqlen = zipIntSize(encoding);
} else {
+ /* 'encoding' is untouched, however zipEncodeLength will use the
+ * string length to figure out how to encode it. */
reqlen = slen;
}
-
/* We need space for both the length of the previous entry and
* the length of the payload. */
reqlen += zipPrevEncodeLength(NULL,prevlen);
/* Write the entry */
p += zipPrevEncodeLength(p,prevlen);
p += zipEncodeLength(p,encoding,slen);
- if (encoding != ZIP_ENC_RAW) {
- zipSaveInteger(p,value,encoding);
- } else {
+ if (ZIP_IS_STR(encoding)) {
memcpy(p,s,slen);
+ } else {
+ zipSaveInteger(p,value,encoding);
}
ZIPLIST_INCR_LENGTH(zl,1);
return zl;
if (sstr) *sstr = NULL;
entry = zipEntry(p);
- if (entry.encoding == ZIP_ENC_RAW) {
+ if (ZIP_IS_STR(entry.encoding)) {
if (sstr) {
*slen = entry.len;
*sstr = p+entry.headersize;
if (p[0] == ZIP_END) return 0;
entry = zipEntry(p);
- if (entry.encoding == ZIP_ENC_RAW) {
+ if (ZIP_IS_STR(entry.encoding)) {
/* Raw compare */
if (entry.len == slen) {
return memcmp(p+entry.headersize,sstr,slen) == 0;
entry = zipEntry(p);
printf("{offset %ld, header %u, payload %u} ",p-zl,entry.headersize,entry.len);
p += entry.headersize;
- if (entry.encoding == ZIP_ENC_RAW) {
+ if (ZIP_IS_STR(entry.encoding)) {
fwrite(p,entry.len,1,stdout);
} else {
printf("%lld", (long long) zipLoadInteger(p,entry.encoding));
projects / redis.git / commitdiff
