metadata_v2: factor out global_entry_data

CMakeLists.txt

@@ -100,6 +100,7 @@ list(
   src/dwarfs/filesystem_v2.cpp
   src/dwarfs/filesystem_writer.cpp
   src/dwarfs/fstypes.cpp
+  src/dwarfs/global_entry_data.cpp
   src/dwarfs/inode_manager.cpp
   src/dwarfs/inode_reader_v2.cpp
   src/dwarfs/logger.cpp

include/dwarfs/entry.h

@@ -38,91 +38,13 @@
 
 namespace dwarfs {
 
-// TODO: clean up
-struct global_entry_data {
-  global_entry_data(bool no_time)
-      : no_time_(no_time) {}
-
-  void add_uid(uint16_t uid) { add(uid, uids, next_uid_index); }
-
-  void add_gid(uint16_t gid) { add(gid, gids, next_gid_index); }
-
-  void add_mode(uint16_t mode) { add(mode, modes, next_mode_index); }
-
-  void add(uint16_t val, std::unordered_map<uint16_t, uint16_t>& map,
-           uint16_t& next_index) {
-    if (map.emplace(val, next_index).second) {
-      ++next_index;
-    }
-  }
-
-  void add_time(uint64_t time) {
-    if (time < timestamp_base) {
-      timestamp_base = time;
-    }
-  }
-
-  void add_name(std::string const& name) { names.emplace(name, 0); }
-
-  void add_link(std::string const& link) { links.emplace(link, 0); }
-
-  void index() {
-    index(names);
-    index(links);
-  }
-
-  void index(std::unordered_map<std::string, uint32_t>& map);
-
-  uint16_t get_uid_index(uint16_t uid) const { return uids.at(uid); }
-
-  uint16_t get_gid_index(uint16_t gid) const { return gids.at(gid); }
-
-  uint16_t get_mode_index(uint16_t mode) const { return modes.at(mode); }
-
-  uint32_t get_name_index(std::string const& name) const {
-    return names.at(name);
-  }
-
-  uint32_t get_link_index(std::string const& link) const {
-    return links.at(link);
-  }
-
-  uint64_t get_time_offset(uint64_t time) const {
-    return no_time_ ? 0 : time - timestamp_base;
-  }
-
-  std::vector<uint16_t> get_uids() const;
-
-  std::vector<uint16_t> get_gids() const;
-
-  std::vector<uint16_t> get_modes() const;
-
-  std::vector<std::string> get_names() const;
-
-  std::vector<std::string> get_links() const;
-
-  // TODO: make private
-  template <typename T, typename U>
-  std::vector<T> get_vector(std::unordered_map<T, U> const& map) const;
-
-  std::unordered_map<uint16_t, uint16_t> uids;
-  std::unordered_map<uint16_t, uint16_t> gids;
-  std::unordered_map<uint16_t, uint16_t> modes;
-  std::unordered_map<std::string, uint32_t> names;
-  std::unordered_map<std::string, uint32_t> links;
-  uint16_t next_uid_index{0};
-  uint16_t next_gid_index{0};
-  uint16_t next_mode_index{0};
-  uint64_t timestamp_base{std::numeric_limits<uint64_t>::max()};
-  bool no_time_;
-};
-
 class file;
 class link;
 class dir;
 class inode;
 class os_access;
 class progress;
+class global_entry_data;
 
 class entry_visitor {
  public:
@@ -143,7 +65,6 @@ class entry : public file_interface {
   bool has_parent() const;
   std::shared_ptr<entry> parent() const;
   void set_name(const std::string& name);
-  void set_name_offset(size_t offset);
   std::string path() const override;
   const std::string& name() const override { return name_; }
   size_t size() const override { return stat_.st_size; }
@@ -165,7 +86,6 @@ class entry : public file_interface {
   std::string name_;
   std::weak_ptr<entry> parent_;
   struct ::stat stat_;
-  uint32_t name_offset_;
 };
 
 class file : public entry {
@@ -204,7 +124,6 @@ class dir : public entry {
   void walk(std::function<void(const entry*)> const& f) const override;
   void accept(entry_visitor& v, bool preorder) override;
   void sort();
-  void set_offset(size_t offset);
   void set_inode(uint32_t inode);
   void
   pack(thrift::metadata::metadata& mv2, global_entry_data const& data) const;
@@ -218,7 +137,6 @@ class dir : public entry {
   using entry_ptr = std::shared_ptr<entry>;
 
   std::vector<std::shared_ptr<entry>> entries_;
-  uint32_t offset_ = 0;
   uint32_t inode_ = 0;
 };
 
@@ -228,7 +146,6 @@ class link : public entry {
 
   type_t type() const override;
   const std::string& linkname() const;
-  void set_offset(size_t offset);
   void set_inode(uint32_t inode);
   void accept(entry_visitor& v, bool preorder) override;
   uint32_t inode_num() const override { return inode_; }
@@ -238,7 +155,6 @@ class link : public entry {
 
  private:
   std::string link_;
-  uint32_t offset_ = 0;
   uint32_t inode_ = 0;
 };
 

include/dwarfs/global_entry_data.h

@@ -0,0 +1,111 @@
+/* vim:set ts=2 sw=2 sts=2 et: */
+/**
+ * \author     Marcus Holland-Moritz (github@mhxnet.de)
+ * \copyright  Copyright (c) Marcus Holland-Moritz
+ *
+ * This file is part of dwarfs.
+ *
+ * dwarfs is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * dwarfs is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with dwarfs.  If not, see <https://www.gnu.org/licenses/>.
+ */
+
+#pragma once
+
+#include <cstdint>
+#include <limits>
+#include <unordered_map>
+#include <vector>
+
+namespace dwarfs {
+
+// TODO: clean up
+class global_entry_data {
+ public:
+  global_entry_data(bool no_time)
+      : no_time_(no_time) {}
+
+  void add_uid(uint16_t uid) { add(uid, uids, next_uid_index); }
+
+  void add_gid(uint16_t gid) { add(gid, gids, next_gid_index); }
+
+  void add_mode(uint16_t mode) { add(mode, modes, next_mode_index); }
+
+  void add(uint16_t val, std::unordered_map<uint16_t, uint16_t>& map,
+           uint16_t& next_index) {
+    if (map.emplace(val, next_index).second) {
+      ++next_index;
+    }
+  }
+
+  void add_time(uint64_t time) {
+    if (time < timestamp_base) {
+      timestamp_base = time;
+    }
+  }
+
+  void add_name(std::string const& name) { names.emplace(name, 0); }
+
+  void add_link(std::string const& link) { links.emplace(link, 0); }
+
+  void index() {
+    index(names);
+    index(links);
+  }
+
+  void index(std::unordered_map<std::string, uint32_t>& map);
+
+  uint16_t get_uid_index(uint16_t uid) const { return uids.at(uid); }
+
+  uint16_t get_gid_index(uint16_t gid) const { return gids.at(gid); }
+
+  uint16_t get_mode_index(uint16_t mode) const { return modes.at(mode); }
+
+  uint32_t get_name_index(std::string const& name) const {
+    return names.at(name);
+  }
+
+  uint32_t get_link_index(std::string const& link) const {
+    return links.at(link);
+  }
+
+  uint64_t get_time_offset(uint64_t time) const {
+    return no_time_ ? 0 : time - timestamp_base;
+  }
+
+  std::vector<uint16_t> get_uids() const;
+
+  std::vector<uint16_t> get_gids() const;
+
+  std::vector<uint16_t> get_modes() const;
+
+  std::vector<std::string> get_names() const;
+
+  std::vector<std::string> get_links() const;
+
+  // TODO: make private
+  template <typename T, typename U>
+  std::vector<T> get_vector(std::unordered_map<T, U> const& map) const;
+
+  std::unordered_map<uint16_t, uint16_t> uids;
+  std::unordered_map<uint16_t, uint16_t> gids;
+  std::unordered_map<uint16_t, uint16_t> modes;
+  std::unordered_map<std::string, uint32_t> names;
+  std::unordered_map<std::string, uint32_t> links;
+  uint16_t next_uid_index{0};
+  uint16_t next_gid_index{0};
+  uint16_t next_mode_index{0};
+  uint64_t timestamp_base{std::numeric_limits<uint64_t>::max()};
+  bool no_time_;
+};
+
+} // namespace dwarfs

src/dwarfs/entry.cpp

@@ -27,12 +27,10 @@
 #include <sys/types.h>
 #include <unistd.h>
 
-#include <folly/Conv.h>
-#include <folly/gen/Base.h>
-
 #include <openssl/sha.h>
 
 #include "dwarfs/entry.h"
+#include "dwarfs/global_entry_data.h"
 #include "dwarfs/inode.h"
 #include "dwarfs/os_access.h"
 #include "dwarfs/progress.h"
@@ -40,47 +38,11 @@
 
 namespace dwarfs {
 
-template <typename T, typename U>
-std::vector<T>
-global_entry_data::get_vector(std::unordered_map<T, U> const& map) const {
-  using namespace folly::gen;
-  std::vector<std::pair<T, U>> pairs(map.begin(), map.end());
-  return from(pairs) | orderBy([](auto const& p) { return p.second; }) |
-         get<0>() | as<std::vector>();
-}
-
-std::vector<uint16_t> global_entry_data::get_uids() const {
-  return get_vector(uids);
-}
-
-std::vector<uint16_t> global_entry_data::get_gids() const {
-  return get_vector(gids);
-}
-
-std::vector<uint16_t> global_entry_data::get_modes() const {
-  return get_vector(modes);
-}
-
-std::vector<std::string> global_entry_data::get_names() const {
-  return get_vector(names);
-}
-
-std::vector<std::string> global_entry_data::get_links() const {
-  return get_vector(links);
-}
-
-void global_entry_data::index(std::unordered_map<std::string, uint32_t>& map) {
-  using namespace folly::gen;
-  uint32_t ix = 0;
-  from(map) | get<0>() | order | [&](std::string const& s) { map[s] = ix++; };
-}
-
 entry::entry(const std::string& name, std::shared_ptr<entry> parent,
              const struct ::stat& st)
     : name_(name)
     , parent_(std::move(parent))
-    , stat_(st)
+    , stat_(st) {}
-    , name_offset_(0) {}
 
 void entry::scan(os_access& os, progress& prog) {
   const std::string& p = path();
@@ -100,10 +62,6 @@ std::shared_ptr<entry> entry::parent() const { return parent_.lock(); }
 
 void entry::set_name(const std::string& name) { name_ = name; }
 
-void entry::set_name_offset(size_t offset) {
-  name_offset_ = folly::to<uint32_t>(offset);
-}
-
 std::string entry::path() const {
   if (auto parent = parent_.lock()) {
     return parent->path() + "/" + name_;
@@ -243,8 +201,6 @@ void dir::sort() {
             });
 }
 
-void dir::set_offset(size_t offset) { offset_ = folly::to<uint32_t>(offset); }
-
 void dir::set_inode(uint32_t inode) { inode_ = inode; }
 
 void dir::scan(os_access&, const std::string&, progress&) {}
@@ -275,8 +231,6 @@ entry::type_t link::type() const { return E_LINK; }
 
 const std::string& link::linkname() const { return link_; }
 
-void link::set_offset(size_t offset) { offset_ = folly::to<uint32_t>(offset); }
-
 void link::set_inode(uint32_t inode) { inode_ = inode; }
 
 void link::accept(entry_visitor& v, bool) { v.visit(this); }

src/dwarfs/global_entry_data.cpp

@@ -0,0 +1,63 @@
+/* vim:set ts=2 sw=2 sts=2 et: */
+/**
+ * \author     Marcus Holland-Moritz (github@mhxnet.de)
+ * \copyright  Copyright (c) Marcus Holland-Moritz
+ *
+ * This file is part of dwarfs.
+ *
+ * dwarfs is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * dwarfs is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with dwarfs.  If not, see <https://www.gnu.org/licenses/>.
+ */
+
+#include <folly/gen/Base.h>
+
+#include "dwarfs/global_entry_data.h"
+
+namespace dwarfs {
+
+template <typename T, typename U>
+std::vector<T>
+global_entry_data::get_vector(std::unordered_map<T, U> const& map) const {
+  using namespace folly::gen;
+  std::vector<std::pair<T, U>> pairs(map.begin(), map.end());
+  return from(pairs) | orderBy([](auto const& p) { return p.second; }) |
+         get<0>() | as<std::vector>();
+}
+
+std::vector<uint16_t> global_entry_data::get_uids() const {
+  return get_vector(uids);
+}
+
+std::vector<uint16_t> global_entry_data::get_gids() const {
+  return get_vector(gids);
+}
+
+std::vector<uint16_t> global_entry_data::get_modes() const {
+  return get_vector(modes);
+}
+
+std::vector<std::string> global_entry_data::get_names() const {
+  return get_vector(names);
+}
+
+std::vector<std::string> global_entry_data::get_links() const {
+  return get_vector(links);
+}
+
+void global_entry_data::index(std::unordered_map<std::string, uint32_t>& map) {
+  using namespace folly::gen;
+  uint32_t ix = 0;
+  from(map) | get<0>() | order | [&](std::string const& s) { map[s] = ix++; };
+}
+
+} // namespace dwarfs

src/dwarfs/scanner.cpp

@@ -32,6 +32,7 @@
 #include "dwarfs/entry.h"
 #include "dwarfs/filesystem_writer.h"
 #include "dwarfs/fstypes.h"
+#include "dwarfs/global_entry_data.h"
 #include "dwarfs/hash_util.h"
 #include "dwarfs/inode_manager.h"
 #include "dwarfs/logger.h"