| // Copyright (c) 2011 The Chromium Authors. All rights reserved. |
| // Use of this source code is governed by a BSD-style license that can be |
| // found in the LICENSE file. |
| |
| #ifndef MEDIA_BASE_DJB2_H_ |
| #define MEDIA_BASE_DJB2_H_ |
| |
| #include <stddef.h> |
| #include <stdint.h> |
| |
| #include "media/base/media_export.h" |
| |
| // DJB2 is a hash algorithm with excellent distribution and speed |
| // on many different sets. |
| // It has marginally more collisions than FNV1, but makes up for it in |
| // performance. |
| // The return value is suitable for table lookups. |
| // For small fixed sizes (ie a pixel), it has low overhead and inlines well. |
| // For large data sets, it optimizes into assembly/simd and is appropriate |
| // for realtime applications. |
| // See Also: |
| // http://www.cse.yorku.ca/~oz/hash.html |
| |
| static const uint32_t kDJB2HashSeed = 5381u; |
| |
| // These functions perform DJB2 hash. The simplest call is DJB2Hash() to |
| // generate the DJB2 hash of the given data: |
| // uint32_t hash = DJB2Hash(data1, length1, kDJB2HashSeed); |
| // |
| // You can also compute the DJB2 hash of data incrementally by making multiple |
| // calls to DJB2Hash(): |
| // uint32_t hash_value = kDJB2HashSeed; // Initial seed for DJB2. |
| // for (size_t i = 0; i < copy_lines; ++i) { |
| // hash_value = DJB2Hash(source, bytes_per_line, hash_value); |
| // source += source_stride; |
| // } |
| |
| // For the given buffer of data, compute the DJB2 hash of |
| // the data. You can call this any number of times during the computation. |
| MEDIA_EXPORT uint32_t DJB2Hash(const void* buf, size_t len, uint32_t seed); |
| |
| #endif // MEDIA_BASE_DJB2_H_ |
| |