From 98d0f6ec08290b3c0848ac2fdc9a2e319dc91186 Mon Sep 17 00:00:00 2001
From: huangmengbin <44433515+huangmengbin@users.noreply.github.com>
Date: Wed, 19 Jul 2023 10:44:10 -0700
Subject: [PATCH] fix: VersionSet::DumpManifest (#11605)

Summary:
Fixes https://github.com/facebook/rocksdb/issues/11604

Pull Request resolved: https://github.com/facebook/rocksdb/pull/11605

Reviewed By: jowlyzhang

Differential Revision: D47459254

Pulled By: ajkr

fbshipit-source-id: 4420e443fbf4bd01ddaa2b47285fc4445bf36246
---
 db/version_set.cc | 22 ++++++++++++++++------
 db/version_set.h  |  4 ++--
 tools/ldb_cmd.cc  | 13 ++++++++-----
 3 files changed, 26 insertions(+), 13 deletions(-)

diff --git a/db/version_set.cc b/db/version_set.cc
index 0bdc0304f..e95e98f79 100644
--- a/db/version_set.cc
+++ b/db/version_set.cc
@@ -6302,8 +6302,9 @@ Status VersionSet::GetLiveFilesChecksumInfo(FileChecksumList* checksum_list) {
   return s;
 }
 
-Status VersionSet::DumpManifest(Options& options, std::string& dscname,
-                                bool verbose, bool hex, bool json) {
+Status VersionSet::DumpManifest(
+    Options& options, std::string& dscname, bool verbose, bool hex, bool json,
+    const std::vector<ColumnFamilyDescriptor>& cf_descs) {
   assert(options.env);
   // TODO: plumb Env::IOActivity
   const ReadOptions read_options;
@@ -6329,13 +6330,22 @@ Status VersionSet::DumpManifest(Options& options, std::string& dscname,
         std::move(file), dscname, db_options_->log_readahead_size, io_tracer_);
   }
 
-  std::vector<ColumnFamilyDescriptor> cf_descs;
+  std::map<std::string, const ColumnFamilyDescriptor*> cf_name_to_desc;
+  for (const auto& cf_desc : cf_descs) {
+    cf_name_to_desc[cf_desc.name] = &cf_desc;
+  }
+  std::vector<ColumnFamilyDescriptor> final_cf_descs;
   for (const auto& cf : column_families) {
-    cf_descs.emplace_back(cf, options);
+    const auto iter = cf_name_to_desc.find(cf);
+    if (iter != cf_name_to_desc.cend()) {
+      final_cf_descs.push_back(*iter->second);
+    } else {
+      final_cf_descs.emplace_back(cf, options);
+    }
   }
 
-  DumpManifestHandler handler(cf_descs, this, io_tracer_, read_options, verbose,
-                              hex, json);
+  DumpManifestHandler handler(final_cf_descs, this, io_tracer_, read_options,
+                              verbose, hex, json);
   {
     VersionSet::LogReporter reporter;
     reporter.status = &s;
diff --git a/db/version_set.h b/db/version_set.h
index c3e48bb4e..e32d0ff11 100644
--- a/db/version_set.h
+++ b/db/version_set.h
@@ -1273,8 +1273,8 @@ class VersionSet {
 
   // printf contents (for debugging)
   Status DumpManifest(Options& options, std::string& manifestFileName,
-                      bool verbose, bool hex = false, bool json = false);
-
+                      bool verbose, bool hex = false, bool json = false,
+                      const std::vector<ColumnFamilyDescriptor>& cf_descs = {});
 
   const std::string& DbSessionId() const { return db_session_id_; }
 
diff --git a/tools/ldb_cmd.cc b/tools/ldb_cmd.cc
index f0119b31e..46dd36ccc 100644
--- a/tools/ldb_cmd.cc
+++ b/tools/ldb_cmd.cc
@@ -1310,7 +1310,8 @@ void DBLoaderCommand::DoCommand() {
 namespace {
 
 void DumpManifestFile(Options options, std::string file, bool verbose, bool hex,
-                      bool json) {
+                      bool json,
+                      const std::vector<ColumnFamilyDescriptor>& cf_descs) {
   EnvOptions sopt;
   std::string dbname("dummy");
   std::shared_ptr<Cache> tc(NewLRUCache(options.max_open_files - 10,
@@ -1326,7 +1327,7 @@ void DumpManifestFile(Options options, std::string file, bool verbose, bool hex,
   VersionSet versions(dbname, &immutable_db_options, sopt, tc.get(), &wb, &wc,
                       /*block_cache_tracer=*/nullptr, /*io_tracer=*/nullptr,
                       /*db_id*/ "", /*db_session_id*/ "");
-  Status s = versions.DumpManifest(options, file, verbose, hex, json);
+  Status s = versions.DumpManifest(options, file, verbose, hex, json, cf_descs);
   if (!s.ok()) {
     fprintf(stderr, "Error in processing file %s %s\n", file.c_str(),
             s.ToString().c_str());
@@ -1438,7 +1439,8 @@ void ManifestDumpCommand::DoCommand() {
     fprintf(stdout, "Processing Manifest file %s\n", manifestfile.c_str());
   }
 
-  DumpManifestFile(options_, manifestfile, verbose_, is_key_hex_, json_);
+  DumpManifestFile(options_, manifestfile, verbose_, is_key_hex_, json_,
+                   column_families_);
 
   if (verbose_) {
     fprintf(stdout, "Processing Manifest file %s done\n", manifestfile.c_str());
@@ -2046,7 +2048,7 @@ void DBDumperCommand::DoCommand() {
         break;
       case kDescriptorFile:
         DumpManifestFile(options_, path_, /* verbose_ */ false, is_key_hex_,
-                         /*  json_ */ false);
+                         /*  json_ */ false, column_families_);
         break;
       case kBlobFile:
         DumpBlobFile(path_, is_key_hex_, is_value_hex_,
@@ -3706,7 +3708,8 @@ void DBFileDumperCommand::DoCommand() {
   manifest_filepath = NormalizePath(manifest_filepath);
 
   std::cout << manifest_filepath << std::endl;
-  DumpManifestFile(options_, manifest_filepath, false, false, false);
+  DumpManifestFile(options_, manifest_filepath, false, false, false,
+                   column_families_);
   std::cout << std::endl;
 
   std::vector<ColumnFamilyMetaData> column_families;