xxhash 0.3.0 → 0.4.0

data/ext/xxhash/libxxhash.h

@@ -1,7 +1,8 @@
 /*
    xxHash - Extremely Fast Hash algorithm
    Header File
-   Copyright (C) 2012-2014, Yann Collet.
+   Copyright (C) 2012-2016, Yann Collet.
+
    BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
 
    Redistribution and use in source and binary forms, with or without
@@ -28,7 +29,7 @@
    OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 
    You can contact the author at :
-   - xxHash source repository : http://code.google.com/p/xxhash/
+   - xxHash source repository : https://github.com/Cyan4973/xxHash
 */
 
 /* Notice extracted from xxHash homepage :
@@ -55,102 +56,246 @@ SHA1-32         0.28 GB/s    10
 Q.Score is a measure of quality of the hash function.
 It depends on successfully passing SMHasher test set.
 10 is a perfect score.
+
+A 64-bits version, named XXH64, is available since r35.
+It offers much better speed, but for 64-bits applications only.
+Name     Speed on 64 bits    Speed on 32 bits
+XXH64       13.8 GB/s            1.9 GB/s
+XXH32        6.8 GB/s            6.0 GB/s
 */
 
-#pragma once
+#ifndef XXHASH_H_5627135585666179
+#define XXHASH_H_5627135585666179 1
 
 #if defined (__cplusplus)
 extern "C" {
 #endif
 
 
-/*****************************
-   Includes
-*****************************/
-#include <stddef.h>   /* size_t */
+/* ****************************
+*  Compiler specifics
+******************************/
+#if !(defined(__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L))   /* ! C99 */
+#  define restrict   /* disable restrict */
+#endif
 
 
-/*****************************
-   Type
-*****************************/
+/* ****************************
+*  Definitions
+******************************/
+#include <stddef.h>   /* size_t */
 typedef enum { XXH_OK=0, XXH_ERROR } XXH_errorcode;
 
 
+/* ****************************
+*  API modifier
+******************************/
+/** XXH_PRIVATE_API
+*   This is useful to include xxhash functions in `static` mode
+*   in order to inline them, and remove their symbol from the public list.
+*   Methodology :
+*     #define XXH_PRIVATE_API
+*     #include "xxhash.h"
+*   `xxhash.c` is automatically included.
+*   It's not useful to compile and link it as a separate module.
+*/
+#ifdef XXH_PRIVATE_API
+#  ifndef XXH_STATIC_LINKING_ONLY
+#    define XXH_STATIC_LINKING_ONLY
+#  endif
+#  if defined(__GNUC__)
+#    define XXH_PUBLIC_API static __inline __attribute__((unused))
+#  elif defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */)
+#    define XXH_PUBLIC_API static inline
+#  elif defined(_MSC_VER)
+#    define XXH_PUBLIC_API static __inline
+#  else
+#    define XXH_PUBLIC_API static   /* this version may generate warnings for unused static functions; disable the relevant warning */
+#  endif
+#else
+#  define XXH_PUBLIC_API   /* do nothing */
+#endif /* XXH_PRIVATE_API */
+
+/*!XXH_NAMESPACE, aka Namespace Emulation :
+
+If you want to include _and expose_ xxHash functions from within your own library,
+but also want to avoid symbol collisions with other libraries which may also include xxHash,
+
+you can use XXH_NAMESPACE, to automatically prefix any public symbol from xxhash library
+with the value of XXH_NAMESPACE (therefore, avoid NULL and numeric values).
+
+Note that no change is required within the calling program as long as it includes `xxhash.h` :
+regular symbol name will be automatically translated by this header.
+*/
+#ifdef XXH_NAMESPACE
+#  define XXH_CAT(A,B) A##B
+#  define XXH_NAME2(A,B) XXH_CAT(A,B)
+#  define XXH_versionNumber XXH_NAME2(XXH_NAMESPACE, XXH_versionNumber)
+#  define XXH32 XXH_NAME2(XXH_NAMESPACE, XXH32)
+#  define XXH32_createState XXH_NAME2(XXH_NAMESPACE, XXH32_createState)
+#  define XXH32_freeState XXH_NAME2(XXH_NAMESPACE, XXH32_freeState)
+#  define XXH32_reset XXH_NAME2(XXH_NAMESPACE, XXH32_reset)
+#  define XXH32_update XXH_NAME2(XXH_NAMESPACE, XXH32_update)
+#  define XXH32_digest XXH_NAME2(XXH_NAMESPACE, XXH32_digest)
+#  define XXH32_copyState XXH_NAME2(XXH_NAMESPACE, XXH32_copyState)
+#  define XXH32_canonicalFromHash XXH_NAME2(XXH_NAMESPACE, XXH32_canonicalFromHash)
+#  define XXH32_hashFromCanonical XXH_NAME2(XXH_NAMESPACE, XXH32_hashFromCanonical)
+#  define XXH64 XXH_NAME2(XXH_NAMESPACE, XXH64)
+#  define XXH64_createState XXH_NAME2(XXH_NAMESPACE, XXH64_createState)
+#  define XXH64_freeState XXH_NAME2(XXH_NAMESPACE, XXH64_freeState)
+#  define XXH64_reset XXH_NAME2(XXH_NAMESPACE, XXH64_reset)
+#  define XXH64_update XXH_NAME2(XXH_NAMESPACE, XXH64_update)
+#  define XXH64_digest XXH_NAME2(XXH_NAMESPACE, XXH64_digest)
+#  define XXH64_copyState XXH_NAME2(XXH_NAMESPACE, XXH64_copyState)
+#  define XXH64_canonicalFromHash XXH_NAME2(XXH_NAMESPACE, XXH64_canonicalFromHash)
+#  define XXH64_hashFromCanonical XXH_NAME2(XXH_NAMESPACE, XXH64_hashFromCanonical)
+#endif
 
-/*****************************
-   Simple Hash Functions
-*****************************/
 
-unsigned int       XXH32 (const void* input, size_t length, unsigned seed);
-unsigned long long XXH64 (const void* input, size_t length, unsigned long long seed);
+/* *************************************
+*  Version
+***************************************/
+#define XXH_VERSION_MAJOR    0
+#define XXH_VERSION_MINOR    6
+#define XXH_VERSION_RELEASE  2
+#define XXH_VERSION_NUMBER  (XXH_VERSION_MAJOR *100*100 + XXH_VERSION_MINOR *100 + XXH_VERSION_RELEASE)
+XXH_PUBLIC_API unsigned XXH_versionNumber (void);
 
-/*
-XXH32() :
+
+/*-**********************************************************************
+*  32-bits hash
+************************************************************************/
+typedef unsigned int       XXH32_hash_t;
+
+/*! XXH32() :
     Calculate the 32-bits hash of sequence "length" bytes stored at memory address "input".
     The memory between input & input+length must be valid (allocated and read-accessible).
     "seed" can be used to alter the result predictably.
-    This function successfully passes all SMHasher tests.
-    Speed on Core 2 Duo @ 3 GHz (single thread, SMHasher benchmark) : 5.4 GB/s
-XXH64() :
-    Calculate the 64-bits hash of sequence of length "len" stored at memory address "input".
-*/
-
+    Speed on Core 2 Duo @ 3 GHz (single thread, SMHasher benchmark) : 5.4 GB/s */
+XXH_PUBLIC_API XXH32_hash_t XXH32 (const void* input, size_t length, unsigned int seed);
 
+/*======   Streaming   ======*/
+typedef struct XXH32_state_s XXH32_state_t;   /* incomplete type */
+XXH_PUBLIC_API XXH32_state_t* XXH32_createState(void);
+XXH_PUBLIC_API XXH_errorcode  XXH32_freeState(XXH32_state_t* statePtr);
+XXH_PUBLIC_API void XXH32_copyState(XXH32_state_t* restrict dst_state, const XXH32_state_t* restrict src_state);
 
-/*****************************
-   Advanced Hash Functions
-*****************************/
-typedef struct { long long ll[ 6]; } XXH32_state_t;
-typedef struct { long long ll[11]; } XXH64_state_t;
+XXH_PUBLIC_API XXH_errorcode XXH32_reset  (XXH32_state_t* statePtr, unsigned int seed);
+XXH_PUBLIC_API XXH_errorcode XXH32_update (XXH32_state_t* statePtr, const void* input, size_t length);
+XXH_PUBLIC_API XXH32_hash_t  XXH32_digest (const XXH32_state_t* statePtr);
 
 /*
-These structures allow static allocation of XXH states.
-States must then be initialized using XXHnn_reset() before first use.
-
-If you prefer dynamic allocation, please refer to functions below.
-*/
+These functions generate the xxHash of an input provided in multiple segments.
+Note that, for small input, they are slower than single-call functions, due to state management.
+For small input, prefer `XXH32()` and `XXH64()` .
 
-XXH32_state_t* XXH32_createState(void);
-XXH_errorcode  XXH32_freeState(XXH32_state_t* statePtr);
+XXH state must first be allocated, using XXH*_createState() .
 
-XXH64_state_t* XXH64_createState(void);
-XXH_errorcode  XXH64_freeState(XXH64_state_t* statePtr);
+Start a new hash by initializing state with a seed, using XXH*_reset().
 
-/*
-These functions create and release memory for XXH state.
-States must then be initialized using XXHnn_reset() before first use.
-*/
+Then, feed the hash state by calling XXH*_update() as many times as necessary.
+Obviously, input must be allocated and read accessible.
+The function returns an error code, with 0 meaning OK, and any other value meaning there is an error.
 
+Finally, a hash value can be produced anytime, by using XXH*_digest().
+This function returns the nn-bits hash as an int or long long.
 
-XXH_errorcode XXH32_reset  (XXH32_state_t* statePtr, unsigned seed);
-XXH_errorcode XXH32_update (XXH32_state_t* statePtr, const void* input, size_t length);
-unsigned int  XXH32_digest (const XXH32_state_t* statePtr);
+It's still possible to continue inserting input into the hash state after a digest,
+and generate some new hashes later on, by calling again XXH*_digest().
 
-XXH_errorcode      XXH64_reset  (XXH64_state_t* statePtr, unsigned long long seed);
-XXH_errorcode      XXH64_update (XXH64_state_t* statePtr, const void* input, size_t length);
-unsigned long long XXH64_digest (const XXH64_state_t* statePtr);
+When done, free XXH state space if it was allocated dynamically.
+*/
 
-/*
-These functions calculate the xxHash of an input provided in multiple smaller packets,
-as opposed to an input provided as a single block.
+/*======   Canonical representation   ======*/
 
-XXH state space must first be allocated, using either static or dynamic method provided above.
+typedef struct { unsigned char digest[4]; } XXH32_canonical_t;
+XXH_PUBLIC_API void XXH32_canonicalFromHash(XXH32_canonical_t* dst, XXH32_hash_t hash);
+XXH_PUBLIC_API XXH32_hash_t XXH32_hashFromCanonical(const XXH32_canonical_t* src);
 
-Start a new hash by initializing state with a seed, using XXHnn_reset().
+/* Default result type for XXH functions are primitive unsigned 32 and 64 bits.
+*  The canonical representation uses human-readable write convention, aka big-endian (large digits first).
+*  These functions allow transformation of hash result into and from its canonical format.
+*  This way, hash values can be written into a file / memory, and remain comparable on different systems and programs.
+*/
 
-Then, feed the hash state by calling XXHnn_update() as many times as necessary.
-Obviously, input must be valid, meaning allocated and read accessible.
-The function returns an error code, with 0 meaning OK, and any other value meaning there is an error.
 
-Finally, you can produce a hash anytime, by using XXHnn_digest().
-This function returns the final nn-bits hash.
-You can nonetheless continue feeding the hash state with more input,
-and therefore get some new hashes, by calling again XXHnn_digest().
+#ifndef XXH_NO_LONG_LONG
+/*-**********************************************************************
+*  64-bits hash
+************************************************************************/
+typedef unsigned long long XXH64_hash_t;
 
-When you are done, don't forget to free XXH state space, using typically XXHnn_freeState().
+/*! XXH64() :
+    Calculate the 64-bits hash of sequence of length "len" stored at memory address "input".
+    "seed" can be used to alter the result predictably.
+    This function runs faster on 64-bits systems, but slower on 32-bits systems (see benchmark).
 */
+XXH_PUBLIC_API XXH64_hash_t XXH64 (const void* input, size_t length, unsigned long long seed);
+
+/*======   Streaming   ======*/
+typedef struct XXH64_state_s XXH64_state_t;   /* incomplete type */
+XXH_PUBLIC_API XXH64_state_t* XXH64_createState(void);
+XXH_PUBLIC_API XXH_errorcode  XXH64_freeState(XXH64_state_t* statePtr);
+XXH_PUBLIC_API void XXH64_copyState(XXH64_state_t* restrict dst_state, const XXH64_state_t* restrict src_state);
+
+XXH_PUBLIC_API XXH_errorcode XXH64_reset  (XXH64_state_t* statePtr, unsigned long long seed);
+XXH_PUBLIC_API XXH_errorcode XXH64_update (XXH64_state_t* statePtr, const void* input, size_t length);
+XXH_PUBLIC_API XXH64_hash_t  XXH64_digest (const XXH64_state_t* statePtr);
+
+/*======   Canonical representation   ======*/
+typedef struct { unsigned char digest[8]; } XXH64_canonical_t;
+XXH_PUBLIC_API void XXH64_canonicalFromHash(XXH64_canonical_t* dst, XXH64_hash_t hash);
+XXH_PUBLIC_API XXH64_hash_t XXH64_hashFromCanonical(const XXH64_canonical_t* src);
+#endif  /* XXH_NO_LONG_LONG */
+
+
+#ifdef XXH_STATIC_LINKING_ONLY
+
+/* ================================================================================================
+   This section contains definitions which are not guaranteed to remain stable.
+   They may change in future versions, becoming incompatible with a different version of the library.
+   They shall only be used with static linking.
+   Never use these definitions in association with dynamic linking !
+=================================================================================================== */
+
+/* These definitions are only meant to allow allocation of XXH state
+   statically, on stack, or in a struct for example.
+   Do not use members directly. */
+
+   struct XXH32_state_s {
+       unsigned total_len_32;
+       unsigned large_len;
+       unsigned v1;
+       unsigned v2;
+       unsigned v3;
+       unsigned v4;
+       unsigned mem32[4];   /* buffer defined as U32 for alignment */
+       unsigned memsize;
+       unsigned reserved;   /* never read nor write, will be removed in a future version */
+   };   /* typedef'd to XXH32_state_t */
+
+#ifndef XXH_NO_LONG_LONG
+   struct XXH64_state_s {
+       unsigned long long total_len;
+       unsigned long long v1;
+       unsigned long long v2;
+       unsigned long long v3;
+       unsigned long long v4;
+       unsigned long long mem64[4];   /* buffer defined as U64 for alignment */
+       unsigned memsize;
+       unsigned reserved[2];          /* never read nor write, will be removed in a future version */
+   };   /* typedef'd to XXH64_state_t */
+#endif
+
+#  ifdef XXH_PRIVATE_API
+#    include "xxhash.c"   /* include xxhash function bodies as `static`, for inlining */
+#  endif
+
+#endif /* XXH_STATIC_LINKING_ONLY */
 
 
 #if defined (__cplusplus)
 }
-#endif
+#endif
+
+#endif /* XXHASH_H_5627135585666179 */

data/ext/xxhash/xxhash.c

@@ -5,29 +5,49 @@ VALUE xxhash_xxh32(VALUE mod, VALUE input, VALUE seed)
   return ULL2NUM(XXH32(StringValuePtr(input), (size_t)RSTRING_LEN(input), (unsigned int)NUM2ULL(seed)));
 }
 
-void xxhash32_streaming_hash_free(XXH32_state_t* state)
+void xxhash32_streaming_hash_free(xxhash_xxh32_t* storage)
 {
   // Digest frees the memory.
-  XXH32_freeState(state);
+  XXH32_freeState(storage->state);
+  xfree(storage);
 }
 
 VALUE xxhash32_streaming_hash_new(VALUE klass, VALUE seed)
 {
-  XXH32_state_t* state = XXH32_createState();
-  XXH_errorcode code = XXH32_reset(state, (unsigned int)NUM2ULL(seed));
+  XXH_errorcode code;
+  xxhash_xxh32_t* storage;
+  storage = ALLOC(xxhash_xxh32_t);
+  storage->state = XXH32_createState();
+  storage->seed = (unsigned int)NUM2ULL(seed);
+  code = XXH32_reset(storage->state, storage->seed);
   if(code != XXH_OK) {
     rb_raise(rb_eRuntimeError, "Error during reset.");
     return Qnil;
   }
-  return Data_Wrap_Struct(klass, 0, xxhash32_streaming_hash_free, state);
+  return Data_Wrap_Struct(klass, 0, xxhash32_streaming_hash_free, storage);
 }
 
+VALUE xxhash32_streaming_hash_reset(VALUE self)
+{
+  XXH_errorcode code;
+  xxhash_xxh32_t* storage;
+  Data_Get_Struct(self, xxhash_xxh32_t, storage);
+
+  code = XXH32_reset(storage->state, storage->seed);
+  if(code != XXH_OK) {
+    rb_raise(rb_eRuntimeError, "Error during reset.");
+  }
+  return Qnil;
+}
+
+
 VALUE xxhash32_streaming_hash_update(VALUE self, VALUE data)
 {
-  XXH32_state_t* state;
-  Data_Get_Struct(self, XXH32_state_t, state);
+  XXH_errorcode code;
+  xxhash_xxh32_t* storage;
+  Data_Get_Struct(self, xxhash_xxh32_t, storage);
 
-  XXH_errorcode code = XXH32_update(state, StringValuePtr(data), (size_t)RSTRING_LEN(data));
+  code = XXH32_update(storage->state, StringValuePtr(data), (size_t)RSTRING_LEN(data));
   if(code != XXH_OK) {
     rb_raise(rb_eRuntimeError, "Error during update.");
   }
@@ -36,11 +56,11 @@ VALUE xxhash32_streaming_hash_update(VALUE self, VALUE data)
 
 VALUE xxhash32_streaming_hash_digest(VALUE self)
 {
-  XXH32_state_t* state;
-  Data_Get_Struct(self, XXH32_state_t, state);
+  xxhash_xxh32_t* storage;
+  Data_Get_Struct(self, xxhash_xxh32_t, storage);
 
   // Do not free memory now.
-  return ULL2NUM(XXH32_digest(state));
+  return ULL2NUM(XXH32_digest(storage->state));
 }
 
 VALUE xxhash_xxh64(VALUE mod, VALUE input, VALUE seed)
@@ -48,29 +68,49 @@ VALUE xxhash_xxh64(VALUE mod, VALUE input, VALUE seed)
   return ULL2NUM(XXH64(StringValuePtr(input), (size_t)RSTRING_LEN(input), (unsigned int)NUM2ULL(seed)));
 }
 
-void xxhash64_streaming_hash_free(XXH64_state_t* state)
+void xxhash64_streaming_hash_free(xxhash_xxh64_t* storage)
 {
   // Digest frees the memory.
-  XXH64_freeState(state);
+  XXH64_freeState(storage->state);
+  xfree(storage);
 }
 
 VALUE xxhash64_streaming_hash_new(VALUE klass, VALUE seed)
 {
-  XXH64_state_t* state = XXH64_createState();
-  XXH_errorcode code = XXH64_reset(state, NUM2ULL(seed));
+  XXH_errorcode code;
+  xxhash_xxh64_t* storage;
+  storage = ALLOC(xxhash_xxh64_t);
+  storage->state = XXH64_createState();
+  storage->seed = (unsigned int)NUM2ULL(seed);
+
+  code = XXH64_reset(storage->state, storage->seed);
   if(code != XXH_OK) {
     rb_raise(rb_eRuntimeError, "Error during reset.");
     return Qnil;
   }
-  return Data_Wrap_Struct(klass, 0, xxhash64_streaming_hash_free, state);
+  return Data_Wrap_Struct(klass, 0, xxhash64_streaming_hash_free, storage);
+}
+
+VALUE xxhash64_streaming_hash_reset(VALUE self)
+{
+  XXH_errorcode code;
+  xxhash_xxh64_t* storage;
+  Data_Get_Struct(self, xxhash_xxh64_t, storage);
+
+  code = XXH64_reset(storage->state, storage->seed);
+  if(code != XXH_OK) {
+    rb_raise(rb_eRuntimeError, "Error during reset.");
+  }
+  return Qnil;
 }
 
 VALUE xxhash64_streaming_hash_update(VALUE self, VALUE data)
 {
-  XXH64_state_t* state;
-  Data_Get_Struct(self, XXH64_state_t, state);
+  XXH_errorcode code;
+  xxhash_xxh64_t* storage;
+  Data_Get_Struct(self, xxhash_xxh64_t, storage);
 
-  XXH_errorcode code = XXH64_update(state, StringValuePtr(data), (size_t)RSTRING_LEN(data));
+  code = XXH64_update(storage->state, StringValuePtr(data), (size_t)RSTRING_LEN(data));
   if(code != XXH_OK) {
     rb_raise(rb_eRuntimeError, "Error during update.");
   }
@@ -79,31 +119,38 @@ VALUE xxhash64_streaming_hash_update(VALUE self, VALUE data)
 
 VALUE xxhash64_streaming_hash_digest(VALUE self)
 {
-  XXH64_state_t* state;
-  Data_Get_Struct(self, XXH64_state_t, state);
+  xxhash_xxh64_t* storage;
+  Data_Get_Struct(self, xxhash_xxh64_t, storage);
 
   // Do not free memory now.
-  return ULL2NUM(XXH64_digest(state));
+  return ULL2NUM(XXH64_digest(storage->state));
 }
 
 
 void Init_xxhash(void)
 {
-  VALUE mXXhash = rb_define_module("XXhash");
-  VALUE mInternal = rb_define_module_under(mXXhash, "XXhashInternal");
+  VALUE cStreamingHash;
+  VALUE cStreamingHash64;
+  VALUE mXXhash;
+  VALUE mInternal;
+
+  mXXhash = rb_define_module("XXhash");
+  mInternal = rb_define_module_under(mXXhash, "XXhashInternal");
 
   rb_define_singleton_method(mInternal, "xxh32", (ruby_method*) &xxhash_xxh32, 2);
   rb_define_singleton_method(mInternal, "xxh64", (ruby_method*) &xxhash_xxh64, 2);
 
-  VALUE cStreamingHash = rb_define_class_under(mInternal, "StreamingHash32", rb_cObject);
+  cStreamingHash = rb_define_class_under(mInternal, "StreamingHash32", rb_cObject);
 
   rb_define_singleton_method(cStreamingHash, "new", (ruby_method*) &xxhash32_streaming_hash_new, 1);
   rb_define_method(cStreamingHash, "update", (ruby_method*) &xxhash32_streaming_hash_update, 1);
   rb_define_method(cStreamingHash, "digest", (ruby_method*) &xxhash32_streaming_hash_digest, 0);
+  rb_define_method(cStreamingHash, "reset", (ruby_method*) &xxhash32_streaming_hash_reset, 0);
 
-  VALUE cStreamingHash64 = rb_define_class_under(mInternal, "StreamingHash64", rb_cObject);
+  cStreamingHash64 = rb_define_class_under(mInternal, "StreamingHash64", rb_cObject);
 
   rb_define_singleton_method(cStreamingHash64, "new", (ruby_method*) &xxhash64_streaming_hash_new, 1);
   rb_define_method(cStreamingHash64, "update", (ruby_method*) &xxhash64_streaming_hash_update, 1);
   rb_define_method(cStreamingHash64, "digest", (ruby_method*) &xxhash64_streaming_hash_digest, 0);
+  rb_define_method(cStreamingHash64, "reset", (ruby_method*) &xxhash64_streaming_hash_reset, 0);
 }