From 64219276f063fd01e95a5b8f79d40a5a826fe60a Mon Sep 17 00:00:00 2001
From: heyongqiang <heyongqiang@fb.com>
Date: Fri, 24 Aug 2012 11:28:59 -0700
Subject: [PATCH] sst_dump tool

Summary:
as subject

add a tool to read sst file

as subject.

./sst_reader --command=check --file=
./sst_reader --command=scan --file=

Test Plan:
db_test

run this command

Reviewers: dhruba

Differential Revision: https://reviews.facebook.net/D4881
---
 db/version_set.cc |   2 +-
 tools/sst_dump.cc | 148 ++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 149 insertions(+), 1 deletion(-)
 create mode 100644 tools/sst_dump.cc

diff --git a/db/version_set.cc b/db/version_set.cc
index 966f4d182..7e16c1fd6 100644
--- a/db/version_set.cc
+++ b/db/version_set.cc
@@ -939,7 +939,7 @@ Status VersionSet::Recover() {
     log_number_ = log_number;
     prev_log_number_ = prev_log_number;
 
-    Log(options_->info_log, "Recovering from manifest file:%s succeeded,"
+    Log(options_->info_log, "Recovered from manifest file:%s succeeded,"
         "manifest_file_number is %lld, next_file_number is %lld, "
         "last_sequence is %lld, log_number is %lld,"
         "prev_log_number is %lld\n",
diff --git a/tools/sst_dump.cc b/tools/sst_dump.cc
new file mode 100644
index 000000000..873fb4f24
--- /dev/null
+++ b/tools/sst_dump.cc
@@ -0,0 +1,148 @@
+#include "leveldb/table.h"
+
+#include <map>
+#include <string>
+#include <vector>
+
+#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 {
+
+class SstFileReader {
+public:
+  SstFileReader(std::string file_name);
+  Status ReadSequential(bool print_kv, uint64_t read_num = -1);
+
+  uint64_t GetReadNumber() { return read_num_; }
+
+private:
+  std::string file_name_;
+  uint64_t read_num_;
+};
+
+SstFileReader::SstFileReader(std::string file_path)
+:file_name_(file_path), read_num_(0) {
+}
+
+Status SstFileReader::ReadSequential(bool print_kv, uint64_t read_num)
+{
+  Table* table;
+  Options table_options;
+  RandomAccessFile* file = NULL;
+  Status s = table_options.env->NewRandomAccessFile(file_name_, &file);
+  if(!s.ok()) {
+   return s;
+  }
+  uint64_t file_size;
+  table_options.env->GetFileSize(file_name_, &file_size);
+  s = Table::Open(table_options, file, file_size, &table);
+  if(!s.ok()) {
+   return s;
+  }
+
+  Iterator* iter = table->NewIterator(ReadOptions());
+  long i = 0;
+  int64_t bytes = 0;
+  for (iter->SeekToFirst(); iter->Valid(); iter->Next()) {
+    Slice key = iter->key();
+    Slice value = iter->value();
+    ++i;
+    if (i > read_num)
+      break;
+    if (print_kv) {
+      fprintf(stdout, "%s : %s\n",
+          key.ToString().c_str(), value.ToString().c_str());
+    }
+   }
+
+   read_num_ += i;
+
+   Status ret = iter->status();
+   delete iter;
+   return ret;
+}
+
+} // namespace leveldb
+
+static void print_help() {
+  fprintf(stderr,
+      "sst_dump [--command=check|scan] --file=data_dir_OR_sst_file"
+      " [--read_num=NUM]\n");
+}
+
+int main(int argc, char** argv) {
+
+  const char* dir_or_file;
+  uint64_t read_num = -1;
+  std::string command;
+
+  char junk;
+  uint64_t n;
+  for (int i = 1; i < argc; i++)
+  {
+    if (strncmp(argv[i], "--file=", 7) == 0) {
+      dir_or_file = argv[i] + 7;
+    } else if (sscanf(argv[i], "--read_num=%ld%c", &n, &junk) == 1) {
+      read_num = n;
+    } else if (strncmp(argv[i], "--command=", 10) == 0) {
+      command = argv[i] + 10;
+    } else {
+      print_help();
+      exit(1);
+    }
+  }
+
+  if(dir_or_file == NULL) {
+    print_help();
+    exit(1);
+  }
+
+  std::vector<std::string> filenames;
+  leveldb::Env* env = leveldb::Env::Default();
+  leveldb::Status st = env->GetChildren(dir_or_file, &filenames);
+  bool dir = true;
+  if (!st.ok()) {
+    filenames.clear();
+    filenames.push_back(dir_or_file);
+    dir = false;
+  }
+
+  int total_read = 0;
+  for (int i = 0; i < filenames.size(); i++) {
+    std::string filename = filenames.at(i);
+    if (filename.length() <= 4 ||
+        filename.rfind(".sst") != filename.length() - 4) {
+      //ignore
+      continue;
+    }
+    if(dir) {
+      filename = dir_or_file + filename;
+    }
+    leveldb::SstFileReader reader(filename);
+    leveldb::Status st;
+    // scan all files in give file path.
+    if (command == "" || command == "scan" || command == "check") {
+      st = reader.ReadSequential(command != "check");
+      if (!st.ok()) {
+        fprintf(stderr, "%s: %s\n", filename.c_str(),
+            st.ToString().c_str());
+      }
+      total_read += reader.GetReadNumber();
+      if (read_num >= 0 && total_read > read_num) {
+        break;
+      }
+    }
+  }
+}