Merge branch 'main' into wesley

benchmarks/db_bench.cc

@@ -74,7 +74,7 @@ static int FLAGS_reads = -1;
 static int FLAGS_threads = 1;
 
 // Size of each value
-static int FLAGS_value_size = 100;
+static int FLAGS_value_size = 1000;
 
 // Arrange to generate values that shrink to this fraction of
 // their original size after compression
@@ -134,6 +134,15 @@ namespace leveldb {
 namespace {
 leveldb::Env* g_env = nullptr;
 
+void EncodeNonIndexValue(const Slice& value, std::string* res) {
+  enum Type : unsigned char {
+    kNonIndexValue = 2,
+  };
+  res->push_back(kNonIndexValue);
+  res->append(value.ToString());
+}
+
+
 class CountComparator : public Comparator {
  public:
   CountComparator(const Comparator* wrapped) : wrapped_(wrapped) {}
@@ -852,8 +861,15 @@ class Benchmark {
       for (int j = 0; j < entries_per_batch_; j++) {
         const int k = seq ? i + j : thread->rand.Uniform(FLAGS_num);
         key.Set(k);
-        batch.Put(key.slice(), gen.Generate(value_size_));
-        bytes += value_size_ + key.slice().size();
+        auto value = gen.Generate(value_size_);
+        FieldArray field_array = {
+          {"1", value.ToString()},
+        };
+        auto field_str = Fields(field_array).Serialize();
+        std::string encoded_value;
+        EncodeNonIndexValue(field_str, &encoded_value);
+        batch.Put(key.slice(), Slice(encoded_value));
+        bytes += encoded_value.size() + key.slice().size();
         thread->stats.FinishedSingleOp();
       }
       s = db_->Write(write_options_, &batch);

db/db_impl.cc

@@ -260,6 +260,9 @@ void DBImpl::RemoveObsoleteFiles() {
         case kTableFile:
           keep = (live.find(number) != live.end());
           break;
+        case kVTableFile:
+          keep = (live.find(number) != live.end());
+          break;
         case kTempFile:
           // Any temp files that are currently being written to must
           // be recorded in pending_outputs_, which is inserted into "live"
@@ -1215,8 +1218,12 @@ Status DBImpl::Get(const ReadOptions& options, const Slice& key,
     if (s.ok()) {
       s = DecodeValue(value);
     }
-    auto fields = Fields(Slice(*value));
-    *value = fields["1"];
+    if (s.ok()) {
+      auto fields = Fields(Slice(*value));
+      *value = fields["1"];
+    } else {
+      *value = "";
+    }
     mutex_.Lock();
   }
 
@@ -1268,7 +1275,11 @@ Status DBImpl::Get(const ReadOptions& options, const Slice& key,
     if (s.ok()) {
       s = DecodeValue(value);
     }
-    *fields = Fields(Slice(*value));
+    if (s.ok()) {
+      *fields = Fields(Slice(*value));
+    } else {
+      *fields = Fields();
+    }
     mutex_.Lock();
   }
 

db/filename.cc

@@ -117,6 +117,8 @@ bool ParseFileName(const std::string& filename, uint64_t* number,
       *type = kTableFile;
     } else if (suffix == Slice(".dbtmp")) {
       *type = kTempFile;
+    } else if (suffix == Slice(".vrb")) {
+      *type = kVTableFile;
     } else {
       return false;
     }

db/filename.h

@@ -25,7 +25,8 @@ enum FileType {
   kDescriptorFile,
   kCurrentFile,
   kTempFile,
-  kInfoLogFile  // Either the current one, or an old one
+  kInfoLogFile,
+  kVTableFile// Either the current one, or an old one
 };
 
 // Return the name of the log file with the specified number

include/leveldb/options.h

@@ -101,7 +101,7 @@ struct LEVELDB_EXPORT Options {
   size_t block_size = 4 * 1024;
 
   // Threshold of value size that decide whether to separate the key and value
-  size_t kv_sep_size = 1;
+  size_t kv_sep_size = 100;
 
   // Number of keys between restart points for delta encoding of keys.
   // This parameter can be changed dynamically.  Most clients should

test/test_basicio.cc

@@ -3,7 +3,7 @@
 #include "leveldb/db.h"
 using namespace leveldb;
 
-constexpr int value_size = 2000;
+constexpr int value_size = 2048;
 constexpr int data_size = 128 << 20;
 
 std::map<std::string, std::string> data;