leveldb-ruby 0.1

data/leveldb/table/merger.h

@@ -0,0 +1,26 @@
+// 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.
+
+#ifndef STORAGE_LEVELDB_TABLE_MERGER_H_
+#define STORAGE_LEVELDB_TABLE_MERGER_H_
+
+namespace leveldb {
+
+class Comparator;
+class Iterator;
+
+// Return an iterator that provided the union of the data in
+// children[0,n-1].  Takes ownership of the child iterators and
+// will delete them when the result iterator is deleted.
+//
+// The result does no duplicate suppression.  I.e., if a particular
+// key is present in K child iterators, it will be yielded K times.
+//
+// REQUIRES: n >= 0
+extern Iterator* NewMergingIterator(
+    const Comparator* comparator, Iterator** children, int n);
+
+}
+
+#endif  // STORAGE_LEVELDB_TABLE_MERGER_H_

data/leveldb/table/table.cc

@@ -0,0 +1,175 @@
+// 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.
+
+#include "leveldb/table.h"
+
+#include "leveldb/cache.h"
+#include "leveldb/env.h"
+#include "table/block.h"
+#include "table/format.h"
+#include "table/two_level_iterator.h"
+#include "util/coding.h"
+
+namespace leveldb {
+
+struct Table::Rep {
+  ~Rep() {
+    delete index_block;
+  }
+
+  Options options;
+  Status status;
+  RandomAccessFile* file;
+  uint64_t cache_id;
+
+  BlockHandle metaindex_handle;  // Handle to metaindex_block: saved from footer
+  Block* index_block;
+};
+
+Status Table::Open(const Options& options,
+                   RandomAccessFile* file,
+                   uint64_t size,
+                   Table** table) {
+  *table = NULL;
+  if (size < Footer::kEncodedLength) {
+    return Status::InvalidArgument("file is too short to be an sstable");
+  }
+
+  char footer_space[Footer::kEncodedLength];
+  Slice footer_input;
+  Status s = file->Read(size - Footer::kEncodedLength, Footer::kEncodedLength,
+                        &footer_input, footer_space);
+  if (!s.ok()) return s;
+
+  Footer footer;
+  s = footer.DecodeFrom(&footer_input);
+  if (!s.ok()) return s;
+
+  // Read the index block
+  Block* index_block = NULL;
+  if (s.ok()) {
+    s = ReadBlock(file, ReadOptions(), footer.index_handle(), &index_block);
+  }
+
+  if (s.ok()) {
+    // We've successfully read the footer and the index block: we're
+    // ready to serve requests.
+    Rep* rep = new Table::Rep;
+    rep->options = options;
+    rep->file = file;
+    rep->metaindex_handle = footer.metaindex_handle();
+    rep->index_block = index_block;
+    rep->cache_id = (options.block_cache ? options.block_cache->NewId() : 0);
+    *table = new Table(rep);
+  } else {
+    if (index_block) delete index_block;
+  }
+
+  return s;
+}
+
+Table::~Table() {
+  delete rep_;
+}
+
+static void DeleteBlock(void* arg, void* ignored) {
+  delete reinterpret_cast<Block*>(arg);
+}
+
+static void DeleteCachedBlock(const Slice& key, void* value) {
+  Block* block = reinterpret_cast<Block*>(value);
+  delete block;
+}
+
+static void ReleaseBlock(void* arg, void* h) {
+  Cache* cache = reinterpret_cast<Cache*>(arg);
+  Cache::Handle* handle = reinterpret_cast<Cache::Handle*>(h);
+  cache->Release(handle);
+}
+
+// Convert an index iterator value (i.e., an encoded BlockHandle)
+// into an iterator over the contents of the corresponding block.
+Iterator* Table::BlockReader(void* arg,
+                             const ReadOptions& options,
+                             const Slice& index_value) {
+  Table* table = reinterpret_cast<Table*>(arg);
+  Cache* block_cache = table->rep_->options.block_cache;
+  Block* block = NULL;
+  Cache::Handle* cache_handle = NULL;
+
+  BlockHandle handle;
+  Slice input = index_value;
+  Status s = handle.DecodeFrom(&input);
+  // We intentionally allow extra stuff in index_value so that we
+  // can add more features in the future.
+
+  if (s.ok()) {
+    if (block_cache != NULL) {
+      char cache_key_buffer[16];
+      EncodeFixed64(cache_key_buffer, table->rep_->cache_id);
+      EncodeFixed64(cache_key_buffer+8, handle.offset());
+      Slice key(cache_key_buffer, sizeof(cache_key_buffer));
+      cache_handle = block_cache->Lookup(key);
+      if (cache_handle != NULL) {
+        block = reinterpret_cast<Block*>(block_cache->Value(cache_handle));
+      } else {
+        s = ReadBlock(table->rep_->file, options, handle, &block);
+        if (s.ok() && options.fill_cache) {
+          cache_handle = block_cache->Insert(
+              key, block, block->size(), &DeleteCachedBlock);
+        }
+      }
+    } else {
+      s = ReadBlock(table->rep_->file, options, handle, &block);
+    }
+  }
+
+  Iterator* iter;
+  if (block != NULL) {
+    iter = block->NewIterator(table->rep_->options.comparator);
+    if (cache_handle == NULL) {
+      iter->RegisterCleanup(&DeleteBlock, block, NULL);
+    } else {
+      iter->RegisterCleanup(&ReleaseBlock, block_cache, cache_handle);
+    }
+  } else {
+    iter = NewErrorIterator(s);
+  }
+  return iter;
+}
+
+Iterator* Table::NewIterator(const ReadOptions& options) const {
+  return NewTwoLevelIterator(
+      rep_->index_block->NewIterator(rep_->options.comparator),
+      &Table::BlockReader, const_cast<Table*>(this), options);
+}
+
+uint64_t Table::ApproximateOffsetOf(const Slice& key) const {
+  Iterator* index_iter =
+      rep_->index_block->NewIterator(rep_->options.comparator);
+  index_iter->Seek(key);
+  uint64_t result;
+  if (index_iter->Valid()) {
+    BlockHandle handle;
+    Slice input = index_iter->value();
+    Status s = handle.DecodeFrom(&input);
+    if (s.ok()) {
+      result = handle.offset();
+    } else {
+      // Strange: we can't decode the block handle in the index block.
+      // We'll just return the offset of the metaindex block, which is
+      // close to the whole file size for this case.
+      result = rep_->metaindex_handle.offset();
+    }
+  } else {
+    // key is past the last key in the file.  Approximate the offset
+    // by returning the offset of the metaindex block (which is
+    // right near the end of the file).
+    result = rep_->metaindex_handle.offset();
+  }
+  delete index_iter;
+  return result;
+}
+
+}

data/leveldb/table/table_builder.cc

@@ -0,0 +1,227 @@
+// 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.
+
+#include "leveldb/table_builder.h"
+
+#include <assert.h>
+#include <stdio.h>
+#include "leveldb/comparator.h"
+#include "leveldb/env.h"
+#include "table/block_builder.h"
+#include "table/format.h"
+#include "util/coding.h"
+#include "util/crc32c.h"
+#include "util/logging.h"
+
+namespace leveldb {
+
+struct TableBuilder::Rep {
+  Options options;
+  Options index_block_options;
+  WritableFile* file;
+  uint64_t offset;
+  Status status;
+  BlockBuilder data_block;
+  BlockBuilder index_block;
+  std::string last_key;
+  int64_t num_entries;
+  bool closed;          // Either Finish() or Abandon() has been called.
+
+  // We do not emit the index entry for a block until we have seen the
+  // first key for the next data block.  This allows us to use shorter
+  // keys in the index block.  For example, consider a block boundary
+  // between the keys "the quick brown fox" and "the who".  We can use
+  // "the r" as the key for the index block entry since it is >= all
+  // entries in the first block and < all entries in subsequent
+  // blocks.
+  //
+  // Invariant: r->pending_index_entry is true only if data_block is empty.
+  bool pending_index_entry;
+  BlockHandle pending_handle;  // Handle to add to index block
+
+  std::string compressed_output;
+
+  Rep(const Options& opt, WritableFile* f)
+      : options(opt),
+        index_block_options(opt),
+        file(f),
+        offset(0),
+        data_block(&options),
+        index_block(&index_block_options),
+        num_entries(0),
+        closed(false),
+        pending_index_entry(false) {
+    index_block_options.block_restart_interval = 1;
+  }
+};
+
+TableBuilder::TableBuilder(const Options& options, WritableFile* file)
+    : rep_(new Rep(options, file)) {
+}
+
+TableBuilder::~TableBuilder() {
+  assert(rep_->closed);  // Catch errors where caller forgot to call Finish()
+  delete rep_;
+}
+
+Status TableBuilder::ChangeOptions(const Options& options) {
+  // Note: if more fields are added to Options, update
+  // this function to catch changes that should not be allowed to
+  // change in the middle of building a Table.
+  if (options.comparator != rep_->options.comparator) {
+    return Status::InvalidArgument("changing comparator while building table");
+  }
+
+  // Note that any live BlockBuilders point to rep_->options and therefore
+  // will automatically pick up the updated options.
+  rep_->options = options;
+  rep_->index_block_options = options;
+  rep_->index_block_options.block_restart_interval = 1;
+  return Status::OK();
+}
+
+void TableBuilder::Add(const Slice& key, const Slice& value) {
+  Rep* r = rep_;
+  assert(!r->closed);
+  if (!ok()) return;
+  if (r->num_entries > 0) {
+    assert(r->options.comparator->Compare(key, Slice(r->last_key)) > 0);
+  }
+
+  if (r->pending_index_entry) {
+    assert(r->data_block.empty());
+    r->options.comparator->FindShortestSeparator(&r->last_key, key);
+    std::string handle_encoding;
+    r->pending_handle.EncodeTo(&handle_encoding);
+    r->index_block.Add(r->last_key, Slice(handle_encoding));
+    r->pending_index_entry = false;
+  }
+
+  r->last_key.assign(key.data(), key.size());
+  r->num_entries++;
+  r->data_block.Add(key, value);
+
+  const size_t estimated_block_size = r->data_block.CurrentSizeEstimate();
+  if (estimated_block_size >= r->options.block_size) {
+    Flush();
+  }
+}
+
+void TableBuilder::Flush() {
+  Rep* r = rep_;
+  assert(!r->closed);
+  if (!ok()) return;
+  if (r->data_block.empty()) return;
+  assert(!r->pending_index_entry);
+  WriteBlock(&r->data_block, &r->pending_handle);
+  if (ok()) {
+    r->pending_index_entry = true;
+    r->status = r->file->Flush();
+  }
+}
+
+void TableBuilder::WriteBlock(BlockBuilder* block, BlockHandle* handle) {
+  // File format contains a sequence of blocks where each block has:
+  //    block_data: uint8[n]
+  //    type: uint8
+  //    crc: uint32
+  assert(ok());
+  Rep* r = rep_;
+  Slice raw = block->Finish();
+
+  Slice block_contents;
+  CompressionType type = r->options.compression;
+  // TODO(postrelease): Support more compression options: zlib?
+  switch (type) {
+    case kNoCompression:
+      block_contents = raw;
+      break;
+
+    case kSnappyCompression: {
+      std::string* compressed = &r->compressed_output;
+      if (port::Snappy_Compress(raw.data(), raw.size(), compressed) &&
+          compressed->size() < raw.size() - (raw.size() / 8u)) {
+        block_contents = *compressed;
+      } else {
+        // Snappy not supported, or compressed less than 12.5%, so just
+        // store uncompressed form
+        block_contents = raw;
+        type = kNoCompression;
+      }
+      break;
+    }
+  }
+  handle->set_offset(r->offset);
+  handle->set_size(block_contents.size());
+  r->status = r->file->Append(block_contents);
+  if (r->status.ok()) {
+    char trailer[kBlockTrailerSize];
+    trailer[0] = type;
+    uint32_t crc = crc32c::Value(block_contents.data(), block_contents.size());
+    crc = crc32c::Extend(crc, trailer, 1);  // Extend crc to cover block type
+    EncodeFixed32(trailer+1, crc32c::Mask(crc));
+    r->status = r->file->Append(Slice(trailer, kBlockTrailerSize));
+    if (r->status.ok()) {
+      r->offset += block_contents.size() + kBlockTrailerSize;
+    }
+  }
+  r->compressed_output.clear();
+  block->Reset();
+}
+
+Status TableBuilder::status() const {
+  return rep_->status;
+}
+
+Status TableBuilder::Finish() {
+  Rep* r = rep_;
+  Flush();
+  assert(!r->closed);
+  r->closed = true;
+  BlockHandle metaindex_block_handle;
+  BlockHandle index_block_handle;
+  if (ok()) {
+    BlockBuilder meta_index_block(&r->options);
+    // TODO(postrelease): Add stats and other meta blocks
+    WriteBlock(&meta_index_block, &metaindex_block_handle);
+  }
+  if (ok()) {
+    if (r->pending_index_entry) {
+      r->options.comparator->FindShortSuccessor(&r->last_key);
+      std::string handle_encoding;
+      r->pending_handle.EncodeTo(&handle_encoding);
+      r->index_block.Add(r->last_key, Slice(handle_encoding));
+      r->pending_index_entry = false;
+    }
+    WriteBlock(&r->index_block, &index_block_handle);
+  }
+  if (ok()) {
+    Footer footer;
+    footer.set_metaindex_handle(metaindex_block_handle);
+    footer.set_index_handle(index_block_handle);
+    std::string footer_encoding;
+    footer.EncodeTo(&footer_encoding);
+    r->status = r->file->Append(footer_encoding);
+    if (r->status.ok()) {
+      r->offset += footer_encoding.size();
+    }
+  }
+  return r->status;
+}
+
+void TableBuilder::Abandon() {
+  Rep* r = rep_;
+  assert(!r->closed);
+  r->closed = true;
+}
+
+uint64_t TableBuilder::NumEntries() const {
+  return rep_->num_entries;
+}
+
+uint64_t TableBuilder::FileSize() const {
+  return rep_->offset;
+}
+
+}

data/leveldb/table/table_test.cc

@@ -0,0 +1,845 @@
+// 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.
+
+#include "leveldb/table.h"
+
+#include <map>
+#include "db/dbformat.h"
+#include "db/memtable.h"
+#include "db/write_batch_internal.h"
+#include "leveldb/db.h"
+#include "leveldb/env.h"
+#include "leveldb/iterator.h"
+#include "leveldb/table_builder.h"
+#include "table/block.h"
+#include "table/block_builder.h"
+#include "table/format.h"
+#include "util/random.h"
+#include "util/testharness.h"
+#include "util/testutil.h"
+
+namespace leveldb {
+
+// Return reverse of "key".
+// Used to test non-lexicographic comparators.
+static std::string Reverse(const Slice& key) {
+  std::string str(key.ToString());
+  std::string rev(str.rbegin(), str.rend());
+  return rev;
+}
+
+namespace {
+class ReverseKeyComparator : public Comparator {
+ public:
+  virtual const char* Name() const {
+    return "leveldb.ReverseBytewiseComparator";
+  }
+
+  virtual int Compare(const Slice& a, const Slice& b) const {
+    return BytewiseComparator()->Compare(Reverse(a), Reverse(b));
+  }
+
+  virtual void FindShortestSeparator(
+      std::string* start,
+      const Slice& limit) const {
+    std::string s = Reverse(*start);
+    std::string l = Reverse(limit);
+    BytewiseComparator()->FindShortestSeparator(&s, l);
+    *start = Reverse(s);
+  }
+
+  virtual void FindShortSuccessor(std::string* key) const {
+    std::string s = Reverse(*key);
+    BytewiseComparator()->FindShortSuccessor(&s);
+    *key = Reverse(s);
+  }
+};
+}
+static ReverseKeyComparator reverse_key_comparator;
+
+static void Increment(const Comparator* cmp, std::string* key) {
+  if (cmp == BytewiseComparator()) {
+    key->push_back('\0');
+  } else {
+    assert(cmp == &reverse_key_comparator);
+    std::string rev = Reverse(*key);
+    rev.push_back('\0');
+    *key = Reverse(rev);
+  }
+}
+
+// An STL comparator that uses a Comparator
+namespace {
+struct STLLessThan {
+  const Comparator* cmp;
+
+  STLLessThan() : cmp(BytewiseComparator()) { }
+  STLLessThan(const Comparator* c) : cmp(c) { }
+  bool operator()(const std::string& a, const std::string& b) const {
+    return cmp->Compare(Slice(a), Slice(b)) < 0;
+  }
+};
+}
+
+class StringSink: public WritableFile {
+ public:
+  ~StringSink() { }
+
+  const std::string& contents() const { return contents_; }
+
+  virtual Status Close() { return Status::OK(); }
+  virtual Status Flush() { return Status::OK(); }
+  virtual Status Sync() { return Status::OK(); }
+
+  virtual Status Append(const Slice& data) {
+    contents_.append(data.data(), data.size());
+    return Status::OK();
+  }
+
+ private:
+  std::string contents_;
+};
+
+
+class StringSource: public RandomAccessFile {
+ public:
+  StringSource(const Slice& contents)
+      : contents_(contents.data(), contents.size()) {
+  }
+
+  virtual ~StringSource() { }
+
+  uint64_t Size() const { return contents_.size(); }
+
+  virtual Status Read(uint64_t offset, size_t n, Slice* result,
+                       char* scratch) const {
+    if (offset > contents_.size()) {
+      return Status::InvalidArgument("invalid Read offset");
+    }
+    if (offset + n > contents_.size()) {
+      n = contents_.size() - offset;
+    }
+    memcpy(scratch, &contents_[offset], n);
+    *result = Slice(scratch, n);
+    return Status::OK();
+  }
+
+ private:
+  std::string contents_;
+};
+
+typedef std::map<std::string, std::string, STLLessThan> KVMap;
+
+// Helper class for tests to unify the interface between
+// BlockBuilder/TableBuilder and Block/Table.
+class Constructor {
+ public:
+  explicit Constructor(const Comparator* cmp) : data_(STLLessThan(cmp)) { }
+  virtual ~Constructor() { }
+
+  void Add(const std::string& key, const Slice& value) {
+    data_[key] = value.ToString();
+  }
+
+  // Finish constructing the data structure with all the keys that have
+  // been added so far.  Returns the keys in sorted order in "*keys"
+  // and stores the key/value pairs in "*kvmap"
+  void Finish(const Options& options,
+              std::vector<std::string>* keys,
+              KVMap* kvmap) {
+    *kvmap = data_;
+    keys->clear();
+    for (KVMap::const_iterator it = data_.begin();
+         it != data_.end();
+         ++it) {
+      keys->push_back(it->first);
+    }
+    data_.clear();
+    Status s = FinishImpl(options, *kvmap);
+    ASSERT_TRUE(s.ok()) << s.ToString();
+  }
+
+  // Construct the data structure from the data in "data"
+  virtual Status FinishImpl(const Options& options, const KVMap& data) = 0;
+
+  virtual size_t NumBytes() const = 0;
+
+  virtual Iterator* NewIterator() const = 0;
+
+  virtual const KVMap& data() { return data_; }
+
+  virtual DB* db() const { return NULL; }  // Overridden in DBConstructor
+
+ private:
+  KVMap data_;
+};
+
+class BlockConstructor: public Constructor {
+ public:
+  explicit BlockConstructor(const Comparator* cmp)
+      : Constructor(cmp),
+        comparator_(cmp),
+        block_size_(-1),
+        block_(NULL) { }
+  ~BlockConstructor() {
+    delete block_;
+  }
+  virtual Status FinishImpl(const Options& options, const KVMap& data) {
+    delete block_;
+    block_ = NULL;
+    BlockBuilder builder(&options);
+
+    for (KVMap::const_iterator it = data.begin();
+         it != data.end();
+         ++it) {
+      builder.Add(it->first, it->second);
+    }
+    // Open the block
+    Slice block_data = builder.Finish();
+    block_size_ = block_data.size();
+    char* block_data_copy = new char[block_size_];
+    memcpy(block_data_copy, block_data.data(), block_size_);
+    block_ = new Block(block_data_copy, block_size_);
+    return Status::OK();
+  }
+  virtual size_t NumBytes() const { return block_size_; }
+
+  virtual Iterator* NewIterator() const {
+    return block_->NewIterator(comparator_);
+  }
+
+ private:
+  const Comparator* comparator_;
+  int block_size_;
+  Block* block_;
+
+  BlockConstructor();
+};
+
+class TableConstructor: public Constructor {
+ public:
+  TableConstructor(const Comparator* cmp)
+      : Constructor(cmp),
+        source_(NULL), table_(NULL) {
+  }
+  ~TableConstructor() {
+    Reset();
+  }
+  virtual Status FinishImpl(const Options& options, const KVMap& data) {
+    Reset();
+    StringSink sink;
+    TableBuilder builder(options, &sink);
+
+    for (KVMap::const_iterator it = data.begin();
+         it != data.end();
+         ++it) {
+      builder.Add(it->first, it->second);
+      ASSERT_TRUE(builder.status().ok());
+    }
+    Status s = builder.Finish();
+    ASSERT_TRUE(s.ok()) << s.ToString();
+
+    ASSERT_EQ(sink.contents().size(), builder.FileSize());
+
+    // Open the table
+    source_ = new StringSource(sink.contents());
+    Options table_options;
+    table_options.comparator = options.comparator;
+    return Table::Open(table_options, source_, sink.contents().size(), &table_);
+  }
+  virtual size_t NumBytes() const { return source_->Size(); }
+
+  virtual Iterator* NewIterator() const {
+    return table_->NewIterator(ReadOptions());
+  }
+
+  uint64_t ApproximateOffsetOf(const Slice& key) const {
+    return table_->ApproximateOffsetOf(key);
+  }
+
+ private:
+  void Reset() {
+    delete table_;
+    delete source_;
+    table_ = NULL;
+    source_ = NULL;
+  }
+
+  StringSource* source_;
+  Table* table_;
+
+  TableConstructor();
+};
+
+// A helper class that converts internal format keys into user keys
+class KeyConvertingIterator: public Iterator {
+ public:
+  explicit KeyConvertingIterator(Iterator* iter) : iter_(iter) { }
+  virtual ~KeyConvertingIterator() { delete iter_; }
+  virtual bool Valid() const { return iter_->Valid(); }
+  virtual void Seek(const Slice& target) {
+    ParsedInternalKey ikey(target, kMaxSequenceNumber, kTypeValue);
+    std::string encoded;
+    AppendInternalKey(&encoded, ikey);
+    iter_->Seek(encoded);
+  }
+  virtual void SeekToFirst() { iter_->SeekToFirst(); }
+  virtual void SeekToLast() { iter_->SeekToLast(); }
+  virtual void Next() { iter_->Next(); }
+  virtual void Prev() { iter_->Prev(); }
+
+  virtual Slice key() const {
+    assert(Valid());
+    ParsedInternalKey key;
+    if (!ParseInternalKey(iter_->key(), &key)) {
+      status_ = Status::Corruption("malformed internal key");
+      return Slice("corrupted key");
+    }
+    return key.user_key;
+  }
+
+  virtual Slice value() const { return iter_->value(); }
+  virtual Status status() const {
+    return status_.ok() ? iter_->status() : status_;
+  }
+
+ private:
+  mutable Status status_;
+  Iterator* iter_;
+
+  // No copying allowed
+  KeyConvertingIterator(const KeyConvertingIterator&);
+  void operator=(const KeyConvertingIterator&);
+};
+
+class MemTableConstructor: public Constructor {
+ public:
+  explicit MemTableConstructor(const Comparator* cmp)
+      : Constructor(cmp),
+        internal_comparator_(cmp) {
+    memtable_ = new MemTable(internal_comparator_);
+    memtable_->Ref();
+  }
+  ~MemTableConstructor() {
+    memtable_->Unref();
+  }
+  virtual Status FinishImpl(const Options& options, const KVMap& data) {
+    memtable_->Unref();
+    memtable_ = new MemTable(internal_comparator_);
+    memtable_->Ref();
+    int seq = 1;
+    for (KVMap::const_iterator it = data.begin();
+         it != data.end();
+         ++it) {
+      memtable_->Add(seq, kTypeValue, it->first, it->second);
+      seq++;
+    }
+    return Status::OK();
+  }
+  virtual size_t NumBytes() const {
+    return memtable_->ApproximateMemoryUsage();
+  }
+
+  virtual Iterator* NewIterator() const {
+    return new KeyConvertingIterator(memtable_->NewIterator());
+  }
+
+ private:
+  InternalKeyComparator internal_comparator_;
+  MemTable* memtable_;
+};
+
+class DBConstructor: public Constructor {
+ public:
+  explicit DBConstructor(const Comparator* cmp)
+      : Constructor(cmp),
+        comparator_(cmp) {
+    db_ = NULL;
+    NewDB();
+  }
+  ~DBConstructor() {
+    delete db_;
+  }
+  virtual Status FinishImpl(const Options& options, const KVMap& data) {
+    delete db_;
+    db_ = NULL;
+    NewDB();
+    for (KVMap::const_iterator it = data.begin();
+         it != data.end();
+         ++it) {
+      WriteBatch batch;
+      batch.Put(it->first, it->second);
+      ASSERT_TRUE(db_->Write(WriteOptions(), &batch).ok());
+    }
+    return Status::OK();
+  }
+  virtual size_t NumBytes() const {
+    Range r("", "\xff\xff");
+    uint64_t size;
+    db_->GetApproximateSizes(&r, 1, &size);
+    return size;
+  }
+
+  virtual Iterator* NewIterator() const {
+    return db_->NewIterator(ReadOptions());
+  }
+
+  virtual DB* db() const { return db_; }
+
+ private:
+  void NewDB() {
+    std::string name = test::TmpDir() + "/table_testdb";
+
+    Options options;
+    options.comparator = comparator_;
+    Status status = DestroyDB(name, options);
+    ASSERT_TRUE(status.ok()) << status.ToString();
+
+    options.create_if_missing = true;
+    options.error_if_exists = true;
+    options.write_buffer_size = 10000;  // Something small to force merging
+    status = DB::Open(options, name, &db_);
+    ASSERT_TRUE(status.ok()) << status.ToString();
+  }
+
+  const Comparator* comparator_;
+  DB* db_;
+};
+
+enum TestType {
+  TABLE_TEST,
+  BLOCK_TEST,
+  MEMTABLE_TEST,
+  DB_TEST,
+};
+
+struct TestArgs {
+  TestType type;
+  bool reverse_compare;
+  int restart_interval;
+};
+
+static const TestArgs kTestArgList[] = {
+  { TABLE_TEST, false, 16 },
+  { TABLE_TEST, false, 1 },
+  { TABLE_TEST, false, 1024 },
+  { TABLE_TEST, true, 16 },
+  { TABLE_TEST, true, 1 },
+  { TABLE_TEST, true, 1024 },
+
+  { BLOCK_TEST, false, 16 },
+  { BLOCK_TEST, false, 1 },
+  { BLOCK_TEST, false, 1024 },
+  { BLOCK_TEST, true, 16 },
+  { BLOCK_TEST, true, 1 },
+  { BLOCK_TEST, true, 1024 },
+
+  // Restart interval does not matter for memtables
+  { MEMTABLE_TEST, false, 16 },
+  { MEMTABLE_TEST, true, 16 },
+
+  // Do not bother with restart interval variations for DB
+  { DB_TEST, false, 16 },
+  { DB_TEST, true, 16 },
+};
+static const int kNumTestArgs = sizeof(kTestArgList) / sizeof(kTestArgList[0]);
+
+class Harness {
+ public:
+  Harness() : constructor_(NULL) { }
+
+  void Init(const TestArgs& args) {
+    delete constructor_;
+    constructor_ = NULL;
+    options_ = Options();
+
+    options_.block_restart_interval = args.restart_interval;
+    // Use shorter block size for tests to exercise block boundary
+    // conditions more.
+    options_.block_size = 256;
+    if (args.reverse_compare) {
+      options_.comparator = &reverse_key_comparator;
+    }
+    switch (args.type) {
+      case TABLE_TEST:
+        constructor_ = new TableConstructor(options_.comparator);
+        break;
+      case BLOCK_TEST:
+        constructor_ = new BlockConstructor(options_.comparator);
+        break;
+      case MEMTABLE_TEST:
+        constructor_ = new MemTableConstructor(options_.comparator);
+        break;
+      case DB_TEST:
+        constructor_ = new DBConstructor(options_.comparator);
+        break;
+    }
+  }
+
+  ~Harness() {
+    delete constructor_;
+  }
+
+  void Add(const std::string& key, const std::string& value) {
+    constructor_->Add(key, value);
+  }
+
+  void Test(Random* rnd) {
+    std::vector<std::string> keys;
+    KVMap data;
+    constructor_->Finish(options_, &keys, &data);
+
+    TestForwardScan(keys, data);
+    TestBackwardScan(keys, data);
+    TestRandomAccess(rnd, keys, data);
+  }
+
+  void TestForwardScan(const std::vector<std::string>& keys,
+                       const KVMap& data) {
+    Iterator* iter = constructor_->NewIterator();
+    ASSERT_TRUE(!iter->Valid());
+    iter->SeekToFirst();
+    for (KVMap::const_iterator model_iter = data.begin();
+         model_iter != data.end();
+         ++model_iter) {
+      ASSERT_EQ(ToString(data, model_iter), ToString(iter));
+      iter->Next();
+    }
+    ASSERT_TRUE(!iter->Valid());
+    delete iter;
+  }
+
+  void TestBackwardScan(const std::vector<std::string>& keys,
+                        const KVMap& data) {
+    Iterator* iter = constructor_->NewIterator();
+    ASSERT_TRUE(!iter->Valid());
+    iter->SeekToLast();
+    for (KVMap::const_reverse_iterator model_iter = data.rbegin();
+         model_iter != data.rend();
+         ++model_iter) {
+      ASSERT_EQ(ToString(data, model_iter), ToString(iter));
+      iter->Prev();
+    }
+    ASSERT_TRUE(!iter->Valid());
+    delete iter;
+  }
+
+  void TestRandomAccess(Random* rnd,
+                        const std::vector<std::string>& keys,
+                        const KVMap& data) {
+    static const bool kVerbose = false;
+    Iterator* iter = constructor_->NewIterator();
+    ASSERT_TRUE(!iter->Valid());
+    KVMap::const_iterator model_iter = data.begin();
+    if (kVerbose) fprintf(stderr, "---\n");
+    for (int i = 0; i < 200; i++) {
+      const int toss = rnd->Uniform(5);
+      switch (toss) {
+        case 0: {
+          if (iter->Valid()) {
+            if (kVerbose) fprintf(stderr, "Next\n");
+            iter->Next();
+            ++model_iter;
+            ASSERT_EQ(ToString(data, model_iter), ToString(iter));
+          }
+          break;
+        }
+
+        case 1: {
+          if (kVerbose) fprintf(stderr, "SeekToFirst\n");
+          iter->SeekToFirst();
+          model_iter = data.begin();
+          ASSERT_EQ(ToString(data, model_iter), ToString(iter));
+          break;
+        }
+
+        case 2: {
+          std::string key = PickRandomKey(rnd, keys);
+          model_iter = data.lower_bound(key);
+          if (kVerbose) fprintf(stderr, "Seek '%s'\n",
+                                EscapeString(key).c_str());
+          iter->Seek(Slice(key));
+          ASSERT_EQ(ToString(data, model_iter), ToString(iter));
+          break;
+        }
+
+        case 3: {
+          if (iter->Valid()) {
+            if (kVerbose) fprintf(stderr, "Prev\n");
+            iter->Prev();
+            if (model_iter == data.begin()) {
+              model_iter = data.end();   // Wrap around to invalid value
+            } else {
+              --model_iter;
+            }
+            ASSERT_EQ(ToString(data, model_iter), ToString(iter));
+          }
+          break;
+        }
+
+        case 4: {
+          if (kVerbose) fprintf(stderr, "SeekToLast\n");
+          iter->SeekToLast();
+          if (keys.empty()) {
+            model_iter = data.end();
+          } else {
+            std::string last = data.rbegin()->first;
+            model_iter = data.lower_bound(last);
+          }
+          ASSERT_EQ(ToString(data, model_iter), ToString(iter));
+          break;
+        }
+      }
+    }
+    delete iter;
+  }
+
+  std::string ToString(const KVMap& data, const KVMap::const_iterator& it) {
+    if (it == data.end()) {
+      return "END";
+    } else {
+      return "'" + it->first + "->" + it->second + "'";
+    }
+  }
+
+  std::string ToString(const KVMap& data,
+                       const KVMap::const_reverse_iterator& it) {
+    if (it == data.rend()) {
+      return "END";
+    } else {
+      return "'" + it->first + "->" + it->second + "'";
+    }
+  }
+
+  std::string ToString(const Iterator* it) {
+    if (!it->Valid()) {
+      return "END";
+    } else {
+      return "'" + it->key().ToString() + "->" + it->value().ToString() + "'";
+    }
+  }
+
+  std::string PickRandomKey(Random* rnd, const std::vector<std::string>& keys) {
+    if (keys.empty()) {
+      return "foo";
+    } else {
+      const int index = rnd->Uniform(keys.size());
+      std::string result = keys[index];
+      switch (rnd->Uniform(3)) {
+        case 0:
+          // Return an existing key
+          break;
+        case 1: {
+          // Attempt to return something smaller than an existing key
+          if (result.size() > 0 && result[result.size()-1] > '\0') {
+            result[result.size()-1]--;
+          }
+          break;
+        }
+        case 2: {
+          // Return something larger than an existing key
+          Increment(options_.comparator, &result);
+          break;
+        }
+      }
+      return result;
+    }
+  }
+
+  // Returns NULL if not running against a DB
+  DB* db() const { return constructor_->db(); }
+
+ private:
+  Options options_;
+  Constructor* constructor_;
+};
+
+// Test the empty key
+TEST(Harness, SimpleEmptyKey) {
+  for (int i = 0; i < kNumTestArgs; i++) {
+    Init(kTestArgList[i]);
+    Random rnd(test::RandomSeed() + 1);
+    Add("", "v");
+    Test(&rnd);
+  }
+}
+
+TEST(Harness, SimpleSingle) {
+  for (int i = 0; i < kNumTestArgs; i++) {
+    Init(kTestArgList[i]);
+    Random rnd(test::RandomSeed() + 2);
+    Add("abc", "v");
+    Test(&rnd);
+  }
+}
+
+TEST(Harness, SimpleMulti) {
+  for (int i = 0; i < kNumTestArgs; i++) {
+    Init(kTestArgList[i]);
+    Random rnd(test::RandomSeed() + 3);
+    Add("abc", "v");
+    Add("abcd", "v");
+    Add("ac", "v2");
+    Test(&rnd);
+  }
+}
+
+TEST(Harness, SimpleSpecialKey) {
+  for (int i = 0; i < kNumTestArgs; i++) {
+    Init(kTestArgList[i]);
+    Random rnd(test::RandomSeed() + 4);
+    Add("\xff\xff", "v3");
+    Test(&rnd);
+  }
+}
+
+TEST(Harness, Randomized) {
+  for (int i = 0; i < kNumTestArgs; i++) {
+    Init(kTestArgList[i]);
+    Random rnd(test::RandomSeed() + 5);
+    for (int num_entries = 0; num_entries < 2000;
+         num_entries += (num_entries < 50 ? 1 : 200)) {
+      if ((num_entries % 10) == 0) {
+        fprintf(stderr, "case %d of %d: num_entries = %d\n",
+                (i + 1), int(kNumTestArgs), num_entries);
+      }
+      for (int e = 0; e < num_entries; e++) {
+        std::string v;
+        Add(test::RandomKey(&rnd, rnd.Skewed(4)),
+            test::RandomString(&rnd, rnd.Skewed(5), &v).ToString());
+      }
+      Test(&rnd);
+    }
+  }
+}
+
+TEST(Harness, RandomizedLongDB) {
+  Random rnd(test::RandomSeed());
+  TestArgs args = { DB_TEST, false, 16 };
+  Init(args);
+  int num_entries = 100000;
+  for (int e = 0; e < num_entries; e++) {
+    std::string v;
+    Add(test::RandomKey(&rnd, rnd.Skewed(4)),
+        test::RandomString(&rnd, rnd.Skewed(5), &v).ToString());
+  }
+  Test(&rnd);
+
+  // We must have created enough data to force merging
+  std::string l0_files, l1_files;
+  ASSERT_TRUE(db()->GetProperty("leveldb.num-files-at-level0", &l0_files));
+  ASSERT_TRUE(db()->GetProperty("leveldb.num-files-at-level1", &l1_files));
+  ASSERT_GT(atoi(l0_files.c_str()) + atoi(l1_files.c_str()), 0);
+
+}
+
+class MemTableTest { };
+
+TEST(MemTableTest, Simple) {
+  InternalKeyComparator cmp(BytewiseComparator());
+  MemTable* memtable = new MemTable(cmp);
+  memtable->Ref();
+  WriteBatch batch;
+  WriteBatchInternal::SetSequence(&batch, 100);
+  batch.Put(std::string("k1"), std::string("v1"));
+  batch.Put(std::string("k2"), std::string("v2"));
+  batch.Put(std::string("k3"), std::string("v3"));
+  batch.Put(std::string("largekey"), std::string("vlarge"));
+  ASSERT_TRUE(WriteBatchInternal::InsertInto(&batch, memtable).ok());
+
+  Iterator* iter = memtable->NewIterator();
+  iter->SeekToFirst();
+  while (iter->Valid()) {
+    fprintf(stderr, "key: '%s' -> '%s'\n",
+            iter->key().ToString().c_str(),
+            iter->value().ToString().c_str());
+    iter->Next();
+  }
+
+  delete iter;
+  memtable->Unref();
+}
+
+static bool Between(uint64_t val, uint64_t low, uint64_t high) {
+  bool result = (val >= low) && (val <= high);
+  if (!result) {
+    fprintf(stderr, "Value %llu is not in range [%llu, %llu]\n",
+            (unsigned long long)(val),
+            (unsigned long long)(low),
+            (unsigned long long)(high));
+  }
+  return result;
+}
+
+class TableTest { };
+
+TEST(TableTest, ApproximateOffsetOfPlain) {
+  TableConstructor c(BytewiseComparator());
+  c.Add("k01", "hello");
+  c.Add("k02", "hello2");
+  c.Add("k03", std::string(10000, 'x'));
+  c.Add("k04", std::string(200000, 'x'));
+  c.Add("k05", std::string(300000, 'x'));
+  c.Add("k06", "hello3");
+  c.Add("k07", std::string(100000, 'x'));
+  std::vector<std::string> keys;
+  KVMap kvmap;
+  Options options;
+  options.block_size = 1024;
+  options.compression = kNoCompression;
+  c.Finish(options, &keys, &kvmap);
+
+  ASSERT_TRUE(Between(c.ApproximateOffsetOf("abc"),       0,      0));
+  ASSERT_TRUE(Between(c.ApproximateOffsetOf("k01"),       0,      0));
+  ASSERT_TRUE(Between(c.ApproximateOffsetOf("k01a"),      0,      0));
+  ASSERT_TRUE(Between(c.ApproximateOffsetOf("k02"),       0,      0));
+  ASSERT_TRUE(Between(c.ApproximateOffsetOf("k03"),       0,      0));
+  ASSERT_TRUE(Between(c.ApproximateOffsetOf("k04"),   10000,  11000));
+  ASSERT_TRUE(Between(c.ApproximateOffsetOf("k04a"), 210000, 211000));
+  ASSERT_TRUE(Between(c.ApproximateOffsetOf("k05"),  210000, 211000));
+  ASSERT_TRUE(Between(c.ApproximateOffsetOf("k06"),  510000, 511000));
+  ASSERT_TRUE(Between(c.ApproximateOffsetOf("k07"),  510000, 511000));
+  ASSERT_TRUE(Between(c.ApproximateOffsetOf("xyz"),  610000, 611000));
+
+}
+
+static bool SnappyCompressionSupported() {
+  std::string out;
+  Slice in = "aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa";
+  return port::Snappy_Compress(in.data(), in.size(), &out);
+}
+
+TEST(TableTest, ApproximateOffsetOfCompressed) {
+  if (!SnappyCompressionSupported()) {
+    fprintf(stderr, "skipping compression tests\n");
+    return;
+  }
+
+  Random rnd(301);
+  TableConstructor c(BytewiseComparator());
+  std::string tmp;
+  c.Add("k01", "hello");
+  c.Add("k02", test::CompressibleString(&rnd, 0.25, 10000, &tmp));
+  c.Add("k03", "hello3");
+  c.Add("k04", test::CompressibleString(&rnd, 0.25, 10000, &tmp));
+  std::vector<std::string> keys;
+  KVMap kvmap;
+  Options options;
+  options.block_size = 1024;
+  options.compression = kSnappyCompression;
+  c.Finish(options, &keys, &kvmap);
+
+  ASSERT_TRUE(Between(c.ApproximateOffsetOf("abc"),       0,      0));
+  ASSERT_TRUE(Between(c.ApproximateOffsetOf("k01"),       0,      0));
+  ASSERT_TRUE(Between(c.ApproximateOffsetOf("k02"),       0,      0));
+  ASSERT_TRUE(Between(c.ApproximateOffsetOf("k03"),    2000,   3000));
+  ASSERT_TRUE(Between(c.ApproximateOffsetOf("k04"),    2000,   3000));
+  ASSERT_TRUE(Between(c.ApproximateOffsetOf("xyz"),    4000,   6000));
+}
+
+}
+
+int main(int argc, char** argv) {
+  return leveldb::test::RunAllTests();
+}