@nxtedition/rocksdb 1.0.0

package/deps/rocksdb/rocksdb/include/rocksdb/slice.h

@@ -0,0 +1,269 @@
+// Copyright (c) 2011-present, Facebook, Inc.  All rights reserved.
+//  This source code is licensed under both the GPLv2 (found in the
+//  COPYING file in the root directory) and Apache 2.0 License
+//  (found in the LICENSE.Apache file in the root directory).
+// Copyright (c) 2011 The LevelDB Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style license that can be
+// found in the LICENSE file. See the AUTHORS file for names of contributors.
+//
+// Slice is a simple structure containing a pointer into some external
+// storage and a size.  The user of a Slice must ensure that the slice
+// is not used after the corresponding external storage has been
+// deallocated.
+//
+// Multiple threads can invoke const methods on a Slice without
+// external synchronization, but if any of the threads may call a
+// non-const method, all threads accessing the same Slice must use
+// external synchronization.
+
+#pragma once
+
+#include <assert.h>
+#include <stddef.h>
+#include <string.h>
+#include <cstdio>
+#include <string>
+
+#ifdef __cpp_lib_string_view
+#include <string_view>
+#endif
+
+#include "rocksdb/cleanable.h"
+
+namespace ROCKSDB_NAMESPACE {
+
+class Slice {
+ public:
+  // Create an empty slice.
+  Slice() : data_(""), size_(0) {}
+
+  // Create a slice that refers to d[0,n-1].
+  Slice(const char* d, size_t n) : data_(d), size_(n) {}
+
+  // Create a slice that refers to the contents of "s"
+  /* implicit */
+  Slice(const std::string& s) : data_(s.data()), size_(s.size()) {}
+
+#ifdef __cpp_lib_string_view
+  // Create a slice that refers to the same contents as "sv"
+  /* implicit */
+  Slice(std::string_view sv) : data_(sv.data()), size_(sv.size()) {}
+#endif
+
+  // Create a slice that refers to s[0,strlen(s)-1]
+  /* implicit */
+  Slice(const char* s) : data_(s) { size_ = (s == nullptr) ? 0 : strlen(s); }
+
+  // Create a single slice from SliceParts using buf as storage.
+  // buf must exist as long as the returned Slice exists.
+  Slice(const struct SliceParts& parts, std::string* buf);
+
+  // Return a pointer to the beginning of the referenced data
+  const char* data() const { return data_; }
+
+  // Return the length (in bytes) of the referenced data
+  size_t size() const { return size_; }
+
+  // Return true iff the length of the referenced data is zero
+  bool empty() const { return size_ == 0; }
+
+  // Return the ith byte in the referenced data.
+  // REQUIRES: n < size()
+  char operator[](size_t n) const {
+    assert(n < size());
+    return data_[n];
+  }
+
+  // Change this slice to refer to an empty array
+  void clear() {
+    data_ = "";
+    size_ = 0;
+  }
+
+  // Drop the first "n" bytes from this slice.
+  void remove_prefix(size_t n) {
+    assert(n <= size());
+    data_ += n;
+    size_ -= n;
+  }
+
+  void remove_suffix(size_t n) {
+    assert(n <= size());
+    size_ -= n;
+  }
+
+  // Return a string that contains the copy of the referenced data.
+  // when hex is true, returns a string of twice the length hex encoded (0-9A-F)
+  std::string ToString(bool hex = false) const;
+
+#ifdef __cpp_lib_string_view
+  // Return a string_view that references the same data as this slice.
+  std::string_view ToStringView() const {
+    return std::string_view(data_, size_);
+  }
+#endif
+
+  // Decodes the current slice interpreted as an hexadecimal string into result,
+  // if successful returns true, if this isn't a valid hex string
+  // (e.g not coming from Slice::ToString(true)) DecodeHex returns false.
+  // This slice is expected to have an even number of 0-9A-F characters
+  // also accepts lowercase (a-f)
+  bool DecodeHex(std::string* result) const;
+
+  // Three-way comparison.  Returns value:
+  //   <  0 iff "*this" <  "b",
+  //   == 0 iff "*this" == "b",
+  //   >  0 iff "*this" >  "b"
+  int compare(const Slice& b) const;
+
+  // Return true iff "x" is a prefix of "*this"
+  bool starts_with(const Slice& x) const {
+    return ((size_ >= x.size_) && (memcmp(data_, x.data_, x.size_) == 0));
+  }
+
+  bool ends_with(const Slice& x) const {
+    return ((size_ >= x.size_) &&
+            (memcmp(data_ + size_ - x.size_, x.data_, x.size_) == 0));
+  }
+
+  // Compare two slices and returns the first byte where they differ
+  size_t difference_offset(const Slice& b) const;
+
+  // private: make these public for rocksdbjni access
+  const char* data_;
+  size_t size_;
+
+  // Intentionally copyable
+};
+
+/**
+ * A Slice that can be pinned with some cleanup tasks, which will be run upon
+ * ::Reset() or object destruction, whichever is invoked first. This can be used
+ * to avoid memcpy by having the PinnableSlice object referring to the data
+ * that is locked in the memory and release them after the data is consumed.
+ */
+class PinnableSlice : public Slice, public Cleanable {
+ public:
+  PinnableSlice() { buf_ = &self_space_; }
+  explicit PinnableSlice(std::string* buf) { buf_ = buf; }
+
+  PinnableSlice(PinnableSlice&& other);
+  PinnableSlice& operator=(PinnableSlice&& other);
+
+  // No copy constructor and copy assignment allowed.
+  PinnableSlice(PinnableSlice&) = delete;
+  PinnableSlice& operator=(PinnableSlice&) = delete;
+
+  inline void PinSlice(const Slice& s, CleanupFunction f, void* arg1,
+                       void* arg2) {
+    assert(!pinned_);
+    pinned_ = true;
+    data_ = s.data();
+    size_ = s.size();
+    RegisterCleanup(f, arg1, arg2);
+    assert(pinned_);
+  }
+
+  inline void PinSlice(const Slice& s, Cleanable* cleanable) {
+    assert(!pinned_);
+    pinned_ = true;
+    data_ = s.data();
+    size_ = s.size();
+    cleanable->DelegateCleanupsTo(this);
+    assert(pinned_);
+  }
+
+  inline void PinSelf(const Slice& slice) {
+    assert(!pinned_);
+    buf_->assign(slice.data(), slice.size());
+    data_ = buf_->data();
+    size_ = buf_->size();
+    assert(!pinned_);
+  }
+
+  inline void PinSelf() {
+    assert(!pinned_);
+    data_ = buf_->data();
+    size_ = buf_->size();
+    assert(!pinned_);
+  }
+
+  void remove_suffix(size_t n) {
+    assert(n <= size());
+    if (pinned_) {
+      size_ -= n;
+    } else {
+      buf_->erase(size() - n, n);
+      PinSelf();
+    }
+  }
+
+  void remove_prefix(size_t n) {
+    assert(n <= size());
+    if (pinned_) {
+      data_ += n;
+      size_ -= n;
+    } else {
+      buf_->erase(0, n);
+      PinSelf();
+    }
+  }
+
+  void Reset() {
+    Cleanable::Reset();
+    pinned_ = false;
+    size_ = 0;
+  }
+
+  inline std::string* GetSelf() { return buf_; }
+
+  inline bool IsPinned() const { return pinned_; }
+
+ private:
+  friend class PinnableSlice4Test;
+  std::string self_space_;
+  std::string* buf_;
+  bool pinned_ = false;
+};
+
+// A set of Slices that are virtually concatenated together.  'parts' points
+// to an array of Slices.  The number of elements in the array is 'num_parts'.
+struct SliceParts {
+  SliceParts(const Slice* _parts, int _num_parts)
+      : parts(_parts), num_parts(_num_parts) {}
+  SliceParts() : parts(nullptr), num_parts(0) {}
+
+  const Slice* parts;
+  int num_parts;
+};
+
+inline bool operator==(const Slice& x, const Slice& y) {
+  return ((x.size() == y.size()) &&
+          (memcmp(x.data(), y.data(), x.size()) == 0));
+}
+
+inline bool operator!=(const Slice& x, const Slice& y) { return !(x == y); }
+
+inline int Slice::compare(const Slice& b) const {
+  assert(data_ != nullptr && b.data_ != nullptr);
+  const size_t min_len = (size_ < b.size_) ? size_ : b.size_;
+  int r = memcmp(data_, b.data_, min_len);
+  if (r == 0) {
+    if (size_ < b.size_)
+      r = -1;
+    else if (size_ > b.size_)
+      r = +1;
+  }
+  return r;
+}
+
+inline size_t Slice::difference_offset(const Slice& b) const {
+  size_t off = 0;
+  const size_t len = (size_ < b.size_) ? size_ : b.size_;
+  for (; off < len; off++) {
+    if (data_[off] != b.data_[off]) break;
+  }
+  return off;
+}
+
+}  // namespace ROCKSDB_NAMESPACE

package/deps/rocksdb/rocksdb/include/rocksdb/slice_transform.h

@@ -0,0 +1,103 @@
+// Copyright (c) 2011-present, Facebook, Inc.  All rights reserved.
+//  This source code is licensed under both the GPLv2 (found in the
+//  COPYING file in the root directory) and Apache 2.0 License
+//  (found in the LICENSE.Apache file in the root directory).
+// Copyright (c) 2012 The LevelDB Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style license that can be
+// found in the LICENSE file. See the AUTHORS file for names of contributors.
+//
+// Class for specifying user-defined functions which perform a
+// transformation on a slice.  It is not required that every slice
+// belong to the domain and/or range of a function.  Subclasses should
+// define InDomain and InRange to determine which slices are in either
+// of these sets respectively.
+
+#pragma once
+
+#include <string>
+
+#include "rocksdb/rocksdb_namespace.h"
+
+namespace ROCKSDB_NAMESPACE {
+
+class Slice;
+
+/*
+ * A SliceTransform is a generic pluggable way of transforming one string
+ * to another. Its primary use-case is in configuring rocksdb
+ * to store prefix blooms by setting prefix_extractor in
+ * ColumnFamilyOptions.
+ */
+class SliceTransform {
+ public:
+  virtual ~SliceTransform(){};
+
+  // Return the name of this transformation.
+  virtual const char* Name() const = 0;
+
+  // Extract a prefix from a specified key. This method is called when
+  // a key is inserted into the db, and the returned slice is used to
+  // create a bloom filter.
+  virtual Slice Transform(const Slice& key) const = 0;
+
+  // Determine whether the specified key is compatible with the logic
+  // specified in the Transform method. This method is invoked for every
+  // key that is inserted into the db. If this method returns true,
+  // then Transform is called to translate the key to its prefix and
+  // that returned prefix is inserted into the bloom filter. If this
+  // method returns false, then the call to Transform is skipped and
+  // no prefix is inserted into the bloom filters.
+  //
+  // For example, if the Transform method operates on a fixed length
+  // prefix of size 4, then an invocation to InDomain("abc") returns
+  // false because the specified key length(3) is shorter than the
+  // prefix size of 4.
+  //
+  // Wiki documentation here:
+  // https://github.com/facebook/rocksdb/wiki/Prefix-Seek-API-Changes
+  //
+  virtual bool InDomain(const Slice& key) const = 0;
+
+  // This is currently not used and remains here for backward compatibility.
+  virtual bool InRange(const Slice& /*dst*/) const { return false; }
+
+  // Some SliceTransform will have a full length which can be used to
+  // determine if two keys are consecuitive. Can be disabled by always
+  // returning 0
+  virtual bool FullLengthEnabled(size_t* /*len*/) const { return false; }
+
+  // Transform(s)=Transform(`prefix`) for any s with `prefix` as a prefix.
+  //
+  // This function is not used by RocksDB, but for users. If users pass
+  // Options by string to RocksDB, they might not know what prefix extractor
+  // they are using. This function is to help users can determine:
+  //   if they want to iterate all keys prefixing `prefix`, whether it is
+  //   safe to use prefix bloom filter and seek to key `prefix`.
+  // If this function returns true, this means a user can Seek() to a prefix
+  // using the bloom filter. Otherwise, user needs to skip the bloom filter
+  // by setting ReadOptions.total_order_seek = true.
+  //
+  // Here is an example: Suppose we implement a slice transform that returns
+  // the first part of the string after splitting it using delimiter ",":
+  // 1. SameResultWhenAppended("abc,") should return true. If applying prefix
+  //    bloom filter using it, all slices matching "abc:.*" will be extracted
+  //    to "abc,", so any SST file or memtable containing any of those key
+  //    will not be filtered out.
+  // 2. SameResultWhenAppended("abc") should return false. A user will not
+  //    guaranteed to see all the keys matching "abc.*" if a user seek to "abc"
+  //    against a DB with the same setting. If one SST file only contains
+  //    "abcd,e", the file can be filtered out and the key will be invisible.
+  //
+  // i.e., an implementation always returning false is safe.
+  virtual bool SameResultWhenAppended(const Slice& /*prefix*/) const {
+    return false;
+  }
+};
+
+extern const SliceTransform* NewFixedPrefixTransform(size_t prefix_len);
+
+extern const SliceTransform* NewCappedPrefixTransform(size_t cap_len);
+
+extern const SliceTransform* NewNoopTransform();
+
+}  // namespace ROCKSDB_NAMESPACE

package/deps/rocksdb/rocksdb/include/rocksdb/snapshot.h

@@ -0,0 +1,48 @@
+//  Copyright (c) 2011-present, Facebook, Inc.  All rights reserved.
+//  This source code is licensed under both the GPLv2 (found in the
+//  COPYING file in the root directory) and Apache 2.0 License
+//  (found in the LICENSE.Apache file in the root directory).
+
+#pragma once
+
+#include "rocksdb/types.h"
+
+namespace ROCKSDB_NAMESPACE {
+
+class DB;
+
+// Abstract handle to particular state of a DB.
+// A Snapshot is an immutable object and can therefore be safely
+// accessed from multiple threads without any external synchronization.
+//
+// To Create a Snapshot, call DB::GetSnapshot().
+// To Destroy a Snapshot, call DB::ReleaseSnapshot(snapshot).
+class Snapshot {
+ public:
+  // returns Snapshot's sequence number
+  virtual SequenceNumber GetSequenceNumber() const = 0;
+
+ protected:
+  virtual ~Snapshot();
+};
+
+// Simple RAII wrapper class for Snapshot.
+// Constructing this object will create a snapshot.  Destructing will
+// release the snapshot.
+class ManagedSnapshot {
+ public:
+  explicit ManagedSnapshot(DB* db);
+
+  // Instead of creating a snapshot, take ownership of the input snapshot.
+  ManagedSnapshot(DB* db, const Snapshot* _snapshot);
+
+  ~ManagedSnapshot();
+
+  const Snapshot* snapshot();
+
+ private:
+  DB* db_;
+  const Snapshot* snapshot_;
+};
+
+}  // namespace ROCKSDB_NAMESPACE

package/deps/rocksdb/rocksdb/include/rocksdb/sst_dump_tool.h

@@ -0,0 +1,19 @@
+// Copyright (c) 2011-present, Facebook, Inc.  All rights reserved.
+//  This source code is licensed under both the GPLv2 (found in the
+//  COPYING file in the root directory) and Apache 2.0 License
+//  (found in the LICENSE.Apache file in the root directory).
+#ifndef ROCKSDB_LITE
+#pragma once
+
+#include "rocksdb/options.h"
+
+namespace ROCKSDB_NAMESPACE {
+
+class SSTDumpTool {
+ public:
+  int Run(int argc, char const* const* argv, Options options = Options());
+};
+
+}  // namespace ROCKSDB_NAMESPACE
+
+#endif  // ROCKSDB_LITE

package/deps/rocksdb/rocksdb/include/rocksdb/sst_file_manager.h

@@ -0,0 +1,136 @@
+//  Copyright (c) 2011-present, Facebook, Inc.  All rights reserved.
+//  This source code is licensed under both the GPLv2 (found in the
+//  COPYING file in the root directory) and Apache 2.0 License
+//  (found in the LICENSE.Apache file in the root directory).
+
+#pragma once
+
+#include <memory>
+#include <string>
+#include <unordered_map>
+#include <vector>
+
+#include "rocksdb/file_system.h"
+#include "rocksdb/statistics.h"
+#include "rocksdb/status.h"
+
+namespace ROCKSDB_NAMESPACE {
+
+class Env;
+class Logger;
+
+// SstFileManager is used to track SST files in the DB and control their
+// deletion rate.
+// All SstFileManager public functions are thread-safe.
+// SstFileManager is not extensible.
+class SstFileManager {
+ public:
+  virtual ~SstFileManager() {}
+
+  // Update the maximum allowed space that should be used by RocksDB, if
+  // the total size of the SST files exceeds max_allowed_space, writes to
+  // RocksDB will fail.
+  //
+  // Setting max_allowed_space to 0 will disable this feature; maximum allowed
+  // space will be infinite (Default value).
+  //
+  // thread-safe.
+  virtual void SetMaxAllowedSpaceUsage(uint64_t max_allowed_space) = 0;
+
+  // Set the amount of buffer room each compaction should be able to leave.
+  // In other words, at its maximum disk space consumption, the compaction
+  // should still leave compaction_buffer_size available on the disk so that
+  // other background functions may continue, such as logging and flushing.
+  virtual void SetCompactionBufferSize(uint64_t compaction_buffer_size) = 0;
+
+  // Return true if the total size of SST files exceeded the maximum allowed
+  // space usage.
+  //
+  // thread-safe.
+  virtual bool IsMaxAllowedSpaceReached() = 0;
+
+  // Returns true if the total size of SST files as well as estimated size
+  // of ongoing compactions exceeds the maximums allowed space usage.
+  virtual bool IsMaxAllowedSpaceReachedIncludingCompactions() = 0;
+
+  // Return the total size of all tracked files.
+  // thread-safe
+  virtual uint64_t GetTotalSize() = 0;
+
+  // Return a map containing all tracked files and their corresponding sizes.
+  // thread-safe
+  virtual std::unordered_map<std::string, uint64_t> GetTrackedFiles() = 0;
+
+  // Return delete rate limit in bytes per second.
+  // thread-safe
+  virtual int64_t GetDeleteRateBytesPerSecond() = 0;
+
+  // Update the delete rate limit in bytes per second.
+  // zero means disable delete rate limiting and delete files immediately
+  // thread-safe
+  virtual void SetDeleteRateBytesPerSecond(int64_t delete_rate) = 0;
+
+  // Return trash/DB size ratio where new files will be deleted immediately
+  // thread-safe
+  virtual double GetMaxTrashDBRatio() = 0;
+
+  // Update trash/DB size ratio where new files will be deleted immediately
+  // thread-safe
+  virtual void SetMaxTrashDBRatio(double ratio) = 0;
+
+  // Return the total size of trash files
+  // thread-safe
+  virtual uint64_t GetTotalTrashSize() = 0;
+
+  // Set the statistics ptr to dump the stat information
+  virtual void SetStatisticsPtr(const std::shared_ptr<Statistics>& stats) = 0;
+};
+
+// Create a new SstFileManager that can be shared among multiple RocksDB
+// instances to track SST file and control there deletion rate.
+// Even though SstFileManager don't track WAL files but it still control
+// there deletion rate.
+//
+// @param env: Pointer to Env object, please see "rocksdb/env.h".
+// @param fs: Pointer to FileSystem object (rocksdb/file_system.h"
+// @param info_log: If not nullptr, info_log will be used to log errors.
+//
+// == Deletion rate limiting specific arguments ==
+// @param trash_dir: Deprecated, this argument have no effect
+// @param rate_bytes_per_sec: How many bytes should be deleted per second, If
+//    this value is set to 1024 (1 Kb / sec) and we deleted a file of size 4 Kb
+//    in 1 second, we will wait for another 3 seconds before we delete other
+//    files, Set to 0 to disable deletion rate limiting.
+//    This option also affect the delete rate of WAL files in the DB.
+// @param delete_existing_trash: Deprecated, this argument have no effect, but
+//    if user provide trash_dir we will schedule deletes for files in the dir
+// @param status: If not nullptr, status will contain any errors that happened
+//    during creating the missing trash_dir or deleting existing files in trash.
+// @param max_trash_db_ratio: If the trash size constitutes for more than this
+//    fraction of the total DB size we will start deleting new files passed to
+//    DeleteScheduler immediately
+// @param bytes_max_delete_chunk: if a file to delete is larger than delete
+//    chunk, ftruncate the file by this size each time, rather than dropping the
+//    whole file. 0 means to always delete the whole file. If the file has more
+//    than one linked names, the file will be deleted as a whole. Either way,
+//    `rate_bytes_per_sec` will be appreciated. NOTE that with this option,
+//    files already renamed as a trash may be partial, so users should not
+//    directly recover them without checking.
+extern SstFileManager* NewSstFileManager(
+    Env* env, std::shared_ptr<FileSystem> fs,
+    std::shared_ptr<Logger> info_log = nullptr,
+    const std::string& trash_dir = "", int64_t rate_bytes_per_sec = 0,
+    bool delete_existing_trash = true, Status* status = nullptr,
+    double max_trash_db_ratio = 0.25,
+    uint64_t bytes_max_delete_chunk = 64 * 1024 * 1024);
+
+// Same as above, but takes a pointer to a legacy Env object, instead of
+// Env and FileSystem objects
+extern SstFileManager* NewSstFileManager(
+    Env* env, std::shared_ptr<Logger> info_log = nullptr,
+    std::string trash_dir = "", int64_t rate_bytes_per_sec = 0,
+    bool delete_existing_trash = true, Status* status = nullptr,
+    double max_trash_db_ratio = 0.25,
+    uint64_t bytes_max_delete_chunk = 64 * 1024 * 1024);
+
+}  // namespace ROCKSDB_NAMESPACE

package/deps/rocksdb/rocksdb/include/rocksdb/sst_file_reader.h

@@ -0,0 +1,47 @@
+//  Copyright (c) 2011-present, Facebook, Inc.  All rights reserved.
+//  This source code is licensed under both the GPLv2 (found in the
+//  COPYING file in the root directory) and Apache 2.0 License
+//  (found in the LICENSE.Apache file in the root directory).
+
+#pragma once
+
+#ifndef ROCKSDB_LITE
+
+#include "rocksdb/iterator.h"
+#include "rocksdb/options.h"
+#include "rocksdb/slice.h"
+#include "rocksdb/table_properties.h"
+
+namespace ROCKSDB_NAMESPACE {
+
+// SstFileReader is used to read sst files that are generated by DB or
+// SstFileWriter.
+class SstFileReader {
+ public:
+  SstFileReader(const Options& options);
+
+  ~SstFileReader();
+
+  // Prepares to read from the file located at "file_path".
+  Status Open(const std::string& file_path);
+
+  // Returns a new iterator over the table contents.
+  // Most read options provide the same control as we read from DB.
+  // If "snapshot" is nullptr, the iterator returns only the latest keys.
+  Iterator* NewIterator(const ReadOptions& options);
+
+  std::shared_ptr<const TableProperties> GetTableProperties() const;
+
+  // Verifies whether there is corruption in this table.
+  Status VerifyChecksum(const ReadOptions& /*read_options*/);
+
+  Status VerifyChecksum() { return VerifyChecksum(ReadOptions()); }
+
+ private:
+  struct Rep;
+  std::unique_ptr<Rep> rep_;
+};
+
+}  // namespace ROCKSDB_NAMESPACE
+
+#endif  // !ROCKSDB_LITE