path: root/tensorflow/core/graph/graph_constructor.cc

/* Copyright 2015 The TensorFlow Authors. All Rights Reserved.

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at

    http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
==============================================================================*/

#include "tensorflow/core/graph/graph_constructor.h"

#include <algorithm>
#include <set>
#include <string>
#include <unordered_map>
#include <vector>

#include "tensorflow/core/common_runtime/shape_refiner.h"
#include "tensorflow/core/framework/function.h"
#include "tensorflow/core/framework/function.pb.h"
#include "tensorflow/core/framework/graph.pb.h"
#include "tensorflow/core/framework/node_def.pb.h"
#include "tensorflow/core/framework/node_def_util.h"
#include "tensorflow/core/framework/tensor_shape.pb.h"
#include "tensorflow/core/framework/types.h"
#include "tensorflow/core/framework/versions.h"
#include "tensorflow/core/framework/versions.pb.h"
#include "tensorflow/core/graph/algorithm.h"
#include "tensorflow/core/graph/graph.h"
#include "tensorflow/core/graph/tensor_id.h"
#include "tensorflow/core/lib/core/errors.h"
#include "tensorflow/core/lib/gtl/inlined_vector.h"
#include "tensorflow/core/lib/strings/scanner.h"
#include "tensorflow/core/platform/logging.h"
#include "tensorflow/core/public/version.h"

namespace tensorflow {

namespace {
inline bool IsMerge(const NodeDef& node_def) {
  return node_def.op() == "Merge" || node_def.op() == "RefMerge";
}

inline bool IsNextIteration(const NodeDef& node_def) {
  return node_def.op() == "NextIteration" ||
         node_def.op() == "RefNextIteration";
}

bool IsValidNodeName(StringPiece s, bool allow_internal_ops) {
  using ::tensorflow::strings::Scanner;
  return Scanner(s)
      .One(allow_internal_ops ? Scanner::LETTER_DIGIT_DOT_UNDERSCORE
                              : Scanner::LETTER_DIGIT_DOT)
      .Any(Scanner::LETTER_DIGIT_DASH_DOT_SLASH_UNDERSCORE)
      .Eos()
      .GetResult();
}

class GraphConstructor {
 public:
  struct Options {
    Options(const GraphConstructorOptions& in)  // NOLINT(runtime/explicit)
        : allow_internal_ops(in.allow_internal_ops),
          expect_device_spec(in.expect_device_spec),
          importing(false),
          validate_colocation_constraints(false) {}
    Options(const ImportGraphDefOptions& in)  // NOLINT(runtime/explicit)
        : allow_internal_ops(false),
          expect_device_spec(false),
          prefix(in.prefix.empty() || StringPiece(in.prefix).ends_with("/")
                     ? in.prefix
                     : in.prefix + "/"),
          uniquify_names(in.uniquify_names),
          input_map(in.input_map),
          skip_mapped_nodes(in.skip_mapped_nodes),
          control_dependencies(in.control_dependencies),
          return_tensors(in.return_tensors),
          return_nodes(in.return_nodes),
          importing(true),
          validate_colocation_constraints(in.validate_colocation_constraints) {}

    bool allow_internal_ops;
    bool expect_device_spec;

    string prefix;
    bool uniquify_names;
    std::map<TensorId, TensorId> input_map;
    bool skip_mapped_nodes;
    std::vector<string> control_dependencies;
    std::vector<TensorId> return_tensors;
    std::vector<string> return_nodes;

    // TODO(ashankar): This bool exists to separate out functionality required
    // to make ImportGraphDef a close equivalent of Python's import_graph_def
    // without affecting the behavior of ConvertGraphDefToGraph at the time
    // ImportGraphDef was added.
    //
    // That said, the functionality here (shape and op validation) seems
    // applicable to ConvertGraphDefToGraph as well, so make an attempt to
    // remove this.
    bool importing;
    bool validate_colocation_constraints;
  };

  typedef gtl::ArraySlice<const NodeDef*> NodeDefSlice;

  // versions and library may be nullptr
  static Status Construct(const Options& opts, NodeDefSlice node_defs,
                          const VersionDef* versions,
                          const FunctionDefLibrary* library, Graph* g,
                          ShapeRefiner* refiner,
                          std::vector<std::pair<Node*, int>>* return_tensors,
                          std::vector<Node*>* return_nodes,
                          std::vector<TensorId>* unused_input_map_keys) {
    if (versions) {
      TF_RETURN_IF_ERROR(CheckVersions(*versions, TF_GRAPH_DEF_VERSION,
                                       TF_GRAPH_DEF_VERSION_MIN_PRODUCER,
                                       "GraphDef", "graph"));
    }
    GraphConstructor c(opts, node_defs, versions, library, g, refiner,
                       return_tensors, return_nodes, unused_input_map_keys);
    const Status s = c.TryImport();
    if (!s.ok()) c.Undo();
    return s;
  }

 private:
  GraphConstructor(const Options& opts, NodeDefSlice node_defs,
                   const VersionDef* versions,
                   const FunctionDefLibrary* library, Graph* g,
                   ShapeRefiner* refiner,
                   std::vector<std::pair<Node*, int>>* return_tensors,
                   std::vector<Node*>* return_nodes,
                   std::vector<TensorId>* unused_input_map_keys)
      : opts_(opts),
        node_defs_(node_defs),
        versions_(versions),
        library_(library),
        g_(g),
        original_versions_(g->versions()),
        refiner_(refiner),
        return_tensors_(return_tensors),
        return_nodes_(return_nodes),
        unused_input_map_keys_(unused_input_map_keys) {}

  Status TryImport() {
    TF_RETURN_IF_ERROR(EnsureNoNameCollisions());
    TF_RETURN_IF_ERROR(ValidateInputMapAndControlDependencies());
    TF_RETURN_IF_ERROR(BuildNodeIndex());
    TF_RETURN_IF_ERROR(InitFromEdges());
    TF_RETURN_IF_ERROR(Convert());
    TF_RETURN_IF_ERROR(AddBackEdges());
    TF_RETURN_IF_ERROR(UpdateVersionDef());
    TF_RETURN_IF_ERROR(PopulateReturnTensors());
    TF_RETURN_IF_ERROR(PopulateReturnNodes());
    FixupSourceAndSinkEdges(g_);
    return Status::OK();
  }

  Status EnsureNoNameCollisions();
  Status ValidateInputMapAndControlDependencies();
  Status BuildNodeIndex();
  Status InitFromEdges();
  Status Convert();
  Status AddBackEdges();
  Status UpdateVersionDef();
  Status PopulateReturnTensors();
  Status PopulateReturnNodes();

  void Undo();

  Status IsNodeFullyMapped(const NodeDef& node_def, bool* is_node_mapped);
  Status ValidateColocationConstraints(const NodeDef& node_def);
  Status MakeNode(const NodeDef& node_def, Node** node);
  Status MakeEdge(Node* src, int output_index, Node* dst, int input_index);
  Status ValidateShape(Node* node);
  Status ModifyNodeDefForImport(NodeDef* node_def);
  // Modifies node_def's inputs according to opts_.input_map.
  // input_already_exists is a pre-initialized vector of length
  // node_def->input_size(). This function will mark inputs that are remapped to
  // true.
  void RemapNodeDefInputs(NodeDef* node_def,
                          std::vector<bool>* input_already_exists);
  // input_already_exists is a pre-initialized vector of length
  // node_def->input_size(). This function will add and mark control inputs as
  // true.
  void AddControlDependencies(NodeDef* node_def,
                              std::vector<bool>* input_already_exists);
  void AddPrefixToNodeDef(const std::vector<bool>& input_already_exists,
                          NodeDef* node_def);

  // Modifies `node_def` if its name isn't unique, or if any of its inputs'
  // names have been uniquified. This must be called in topological order on all
  // nodes.
  void UniquifyNames(const std::vector<bool>& input_already_exists,
                     NodeDef* node_def);

  // Returns true if `name` already exists in `g_` (either as a node name or
  // prefix).
  bool NameExists(StringPiece name);

  // Returns a unique version of `original_name`, or `original_name` if it's
  // already unique in the graph.
  string FindUniqueName(StringPiece original_name);

  // From constructor
  const Options opts_;
  const NodeDefSlice node_defs_;
  const VersionDef* versions_;
  const FunctionDefLibrary* library_;
  Graph* g_;
  const VersionDef original_versions_;

  ShapeRefiner* refiner_;

  // May be null. Not owned.
  std::vector<std::pair<Node*, int>>* return_tensors_;

  // May be null. Not owned.
  std::vector<Node*>* return_nodes_;

  // May be null. Not owned.
  std::vector<TensorId>* unused_input_map_keys_;

  // Intermediate datastructure used to populate `unused_input_map_keys_`.
  std::set<TensorId> used_input_map_keys_;

  // Mapping from node name to the index within node_defs_.
  struct NodeInfo {
    explicit NodeInfo(int i) : gdef_index(i), node(nullptr) {}
    // std::unordered_map<> requires that we have a default constructor.
    NodeInfo() : NodeInfo(-1) {}
    int gdef_index;
    Node* node;  // nullptr until the NodeDef is converted to a Node.
  };
  // TODO(vrv): Profile this data structure to see if we should use an
  // alternative implementation of std::unordered_map.
  std::unordered_map<StringPiece, NodeInfo, StringPiece::Hasher> gdef_nodes_;

  // Mapping from node name to the existing node in g_.
  std::unordered_map<StringPiece, Node*, StringPiece::Hasher> existing_nodes_;

  // Prefixes already used in the graph.
  std::unordered_set<StringPiece, StringPiece::Hasher> existing_prefixes_;

  // Imported node names that have been uniquified. The key is the original
  // name, the value is the new unique name.
  std::unordered_map<string, string> uniquified_names_;

  // Index of NodeDefs in node_defs_ with all inputs already converted.
  std::vector<int> ready_;

  // Mapping between index within node_defs_ and the number of inputs that
  // still need to be converted.
  std::vector<int> pending_count_;

  // Mapping between index within node_defs_ and the index within node_defs_ of
  // all nodes it outputs to.
  std::vector<gtl::InlinedVector<int, 4>> outputs_;

  // Used in the conversion from node_defs_ to g_ to represent the ith input
  // of a node.
  struct InputInfo {
    explicit InputInfo(const string& node_name, Node* n, int i)
        : name(node_name), node(n), index(i) {}
    // Use string instead of StringPiece so we don't have to manage lifetime
    string name;
    Node* node;
    int index;
  };

  // Used in the conversion from node_defs_ to g_ to represent an edge from
  // the node named 'name' to node 'n'.
  struct EdgeInfo {
    explicit EdgeInfo(const string& name, int i1, Node* n, int i2)
        : src_name(name), src_index(i1), dst_node(n), dst_index(i2) {}
    // Use string instead of StringPiece so we don't have to manage lifetime
    string src_name;
    int src_index;
    Node* dst_node;
    int dst_index;
  };
  std::vector<EdgeInfo> back_edges_;
};

// This could be expensive but we don't expect to call it often, if at all (only
// if there are multiple nodes in g_ with the same name)
bool NodeNameInValues(const std::map<TensorId, TensorId>& input_map,
                      const StringPiece& node_name) {
  for (auto iter = input_map.begin(); iter != input_map.end(); ++iter) {
    if (iter->second.first == node_name) return true;
  }
  return false;
}

bool NodeNameInValues(const std::vector<string>& control_dependencies,
                      const StringPiece& node_name) {
  return std::find(control_dependencies.begin(), control_dependencies.end(),
                   node_name) != control_dependencies.end();
}

Status GraphConstructor::EnsureNoNameCollisions() {
  existing_nodes_.reserve(g_->num_nodes());
  // Populate existing_nodes_ and existing_prefixes_.
  for (Node* n : g_->nodes()) {
    bool already_exists = !existing_nodes_.insert({n->name(), n}).second;
    if (already_exists) {
      if (NodeNameInValues(opts_.input_map, n->name())) {
        return errors::InvalidArgument(
            "cannot resolve input_map because multiple nodes exist with name '",
            n->name(), "'");
      }
      if (NodeNameInValues(opts_.control_dependencies, n->name())) {
        return errors::InvalidArgument(
            "cannot resolve control_dependencies because multiple nodes exist "
            "with name '",
            n->name(), "'");
      }
    }
    // Add all of node's prefixes to existing_prefixes_ (if it has any).
    size_t idx = -1;
    while ((idx = n->name().find('/', idx + 1)) != string::npos) {
      StringPiece name(n->name());
      existing_prefixes_.insert(name.substr(0, idx));
    }
  }
  if (opts_.prefix.empty() && opts_.importing && !opts_.uniquify_names) {
    for (const NodeDef* n : node_defs_) {
      const string& name = n->name();
      if (NameExists(name)) {
        return errors::InvalidArgument("Node name '", name,
                                       "' already exists in the Graph");
      }
    }
  } else if (!opts_.prefix.empty()) {
    StringPiece prefix_no_slash(opts_.prefix);
    prefix_no_slash.remove_suffix(1);
    if (!IsValidNodeName(prefix_no_slash, false)) {
      return errors::InvalidArgument("Imported node name prefix '",
                                     opts_.prefix,
                                     "' would lead to invalid node names");
    }
    if (NameExists(prefix_no_slash)) {
      return errors::InvalidArgument("Import node name prefix '",
                                     prefix_no_slash,
                                     "' conflicts with "
                                     "name already used in the graph");
    }
  }
  return Status::OK();
}

Status GraphConstructor::ValidateInputMapAndControlDependencies() {
  for (const auto& mapping : opts_.input_map) {
    TensorId src = mapping.first;
    TensorId dst = mapping.second;
    if (existing_nodes_.count(dst.first) == 0) {
      return errors::InvalidArgument(
          "node '", dst.first, "' in input_map does not exist in graph ",
          "(input_map entry: ", src.ToString(), "->", dst.ToString(), ")");
    }
    if ((src.second == Graph::kControlSlot) !=
        (dst.second == Graph::kControlSlot)) {
      return errors::InvalidArgument("input_map entry ", src.ToString(), "->",
                                     dst.ToString(), " between ",
                                     "control edge and non-control edge");
    }
  }
  for (const string& node : opts_.control_dependencies) {
    if (existing_nodes_.count(node) == 0) {
      return errors::InvalidArgument(
          "node '", node,
          "' in control_dependencies does not exist in "
          "graph");
    }
  }
  return Status::OK();
}

Status GraphConstructor::BuildNodeIndex() {
  // Validate the node names and add them to gdef_nodes_.
  for (int n = 0; n < node_defs_.size(); ++n) {
    const NodeDef& node_def = *node_defs_[n];
    if (!IsValidNodeName(node_def.name(), opts_.allow_internal_ops)) {
      return errors::InvalidArgument(
          "Node '", node_def.name(),
          "': Node name contains invalid characters");
    }
    if (!gdef_nodes_
             .insert(std::make_pair(StringPiece(node_def.name()), NodeInfo(n)))
             .second) {
      return errors::InvalidArgument("Node '", node_def.name(),
                                     "' is not unique");
    }
    // Validate the operation's type.
    if (node_def.op().empty()) {
      return errors::InvalidArgument("Node '", node_def.name(),
                                     "' does not specify an operation");
    }
    if (opts_.expect_device_spec && node_def.device().empty()) {
      return errors::InvalidArgument("Node '", node_def.name(),
                                     "' is missing a device specification");
    }
    // Validate control edges at end
    bool in_control_dependence = false;
    for (int i = 0; i < node_def.input_size(); ++i) {
      StringPiece input_name = node_def.input(i);
      if (!input_name.empty() && input_name.starts_with("^")) {
        in_control_dependence = true;
      } else if (in_control_dependence) {
        return errors::InvalidArgument(
            "Node '", node_def.name(),
            "': Control dependencies must come after regular dependencies");
      }
    }
  }
  return Status::OK();
}

std::unordered_set<string> GetNextIterationNodes(
    const GraphConstructor::NodeDefSlice& node_defs) {
  std::unordered_set<string> next_iteration_nodes;

  for (int n = 0; n < node_defs.size(); ++n) {
    const NodeDef& node_def = *node_defs[n];
    if (IsNextIteration(node_def)) {
      next_iteration_nodes.insert(node_def.name());
    }
  }

  return next_iteration_nodes;
}

Status GraphConstructor::InitFromEdges() {
  const int num_nodes = node_defs_.size();
  pending_count_.reserve(num_nodes);
  outputs_.resize(num_nodes);
  std::unordered_set<string> next_iteration_nodes_ =
      GetNextIterationNodes(node_defs_);

  // Parse the inputs for each node.
  for (int n = 0; n < num_nodes; ++n) {
    const NodeDef& node_def = *node_defs_[n];
    if (IsMerge(node_def)) {
      // Cycles in the graph are only allowed for while loops. A while loop is
      // identified by an edge from a NextIteration node to a Merge node. For
      // such Merge nodes, only wait for one non-control input before
      // considering the node ready to process in Convert().
      int32 num_control_edges = 0;
      bool has_loop_back_edge = false;
      for (int i = 0; i < node_def.input_size(); ++i) {
        StringPiece input_name(node_def.input(i));
        if (input_name.starts_with("^")) {
          num_control_edges++;
        } else {
          TensorId id(ParseTensorName(input_name));
          if (next_iteration_nodes_.find(id.first.ToString()) !=
              next_iteration_nodes_.end()) {
            has_loop_back_edge = true;
          }
        }
      }
      if (has_loop_back_edge) {
        pending_count_.push_back(num_control_edges + 1);
      } else {
        pending_count_.push_back(node_def.input_size());
      }
    } else {
      pending_count_.push_back(node_def.input_size());
    }
    if (node_def.input_size() == 0) {
      ready_.push_back(n);
      continue;
    }
    for (int i = 0; i < node_def.input_size(); ++i) {
      StringPiece input_name = node_def.input(i);
      TensorId id(ParseTensorName(input_name));
      auto iter = gdef_nodes_.find(id.first);
      if (iter == gdef_nodes_.end()) {
        return errors::InvalidArgument("Node '", node_def.name(),
                                       "': Unknown input node '",
                                       node_def.input(i), "'");
      }
      outputs_[iter->second.gdef_index].push_back(n);
    }
  }
  return Status::OK();
}

Status GraphConstructor::ValidateColocationConstraints(
    const NodeDef& node_def) {
  if (!opts_.validate_colocation_constraints || !opts_.importing)
    return Status::OK();
  const auto iter = node_def.attr().find(kColocationAttrName);
  if (iter == node_def.attr().end()) return Status::OK();
  for (const string& c : iter->second.list().s()) {
    StringPiece s(c);
    if (s.Consume(kColocationGroupPrefix) &&
        gdef_nodes_.find(s) == gdef_nodes_.end()) {
      return errors::InvalidArgument(
          "Node '", node_def.name(),
          "' expects to be colocated with unknown node '", s, "'");
    }
  }
  return Status::OK();
}

Status GraphConstructor::MakeNode(const NodeDef& node_def, Node** node) {
  // Add the node to the graph.
  Status status;
  *node = g_->AddNode(node_def, &status);
  if (!status.ok()) return status;
  if (opts_.expect_device_spec) {
    (*node)->set_assigned_device_name(node_def.device());
  }
  return Status::OK();
}

Status GraphConstructor::ValidateShape(Node* node) {
  if (!opts_.importing) return Status::OK();
  TF_RETURN_IF_ERROR(refiner_->AddNode(node));
  // For nodes with the _output_shapes attribute, override the shape.
  std::vector<TensorShapeProto> shape_attrs;
  const char* kAttrName = "_output_shapes";
  if (!GetNodeAttr(node->attrs(), kAttrName, &shape_attrs).ok()) {
    // No _output_shapes attribute, the AddNode call above was sufficient.
    return Status::OK();
  }
  auto* ic = refiner_->GetContext(node);
  DCHECK(ic != nullptr)
      << "ShapeRefiner::AddNode() should have created the InferenceContext";
  if (shape_attrs.size() != node->num_outputs()) {
    return errors::InvalidArgument(
        "Node '", node->name(), "' has ", node->num_outputs(),
        " outputs but the ", kAttrName, " attribute specifies shapes for ",
        shape_attrs.size(), " outputs");
  }
  for (int i = 0; i < shape_attrs.size(); ++i) {
    const TensorShapeProto& p = shape_attrs[i];
    shape_inference::ShapeHandle h;
    Status s = ic->MakeShapeFromShapeProto(p, &h);
    if (!s.ok()) {
      return errors::InvalidArgument("Node '", node->name(), " has an invalid ",
                                     kAttrName, " attribute (shape #", i,
                                     " error:'", s.error_message(), "'");
    }
    s = refiner_->SetShape(node, i, h);
    if (!s.ok()) {
      // If the output shape is incompatible with what is inferred
      // by the graph for a very specific whitelist of ops, then we
      // ignore this output shape.  This can happen if there is a
      // bug in the shape function for some operation, and the
      // serialized graph def has the incorrect shape set when
      // running on a newer binary with the fixed shape function.
      // This is an escape hatch that allows us to correct shape
      // functions that are not critical to correct execution but
      // would cause graphs to fail if imported after correcting.
      //
      const string& op = node->type_string();
      const std::vector<string> whitelist = {
          // To be removed after 2017/03/08.
          "RandomShuffleQueue", "PaddingFIFOQueue", "FIFOQueue",
          "PriorityQueue", "QueueSize", "Stack", "Barrier", "BarrierReadySize",
          "BarrierIncompleteSize", "HashTable", "MutableHashTable",
          "MutableHashTableOfTensors", "Mutex", "CuckooTable", "IndexTable",
          "WholeFileReader", "TextLineReader", "FixedLengthRecordReader",
          "TFRecordReader", "IdentityReader", "RefSwitch", "RefEnter",
          "RefNextIteration", "RefMerge", "RefIdentity", "LMDBReader",
          // To be removed after 2017/04/24.
          "ConditionalAccumulator", "SparseConditionalAccumulator", "Table",
      };
      if (std::find(whitelist.begin(), whitelist.end(), op) ==
          whitelist.end()) {
        return errors::InvalidArgument(
            "Node '", node->name(), "' has an ", kAttrName,
            " attribute inconsistent with the GraphDef for output #", i, ": ",
            s.error_message());
      }
    }
  }
  node->ClearAttr(kAttrName);
  return Status::OK();
}

Status GraphConstructor::ModifyNodeDefForImport(NodeDef* node_def) {
  const OpDef* op_def;
  TF_RETURN_IF_ERROR(g_->op_registry()->LookUpOpDef(node_def->op(), &op_def));
  AddDefaultsToNodeDef(*op_def, node_def);
  TF_RETURN_IF_ERROR(ValidateNodeDef(*node_def, *op_def));
  if (versions_) {
    TF_RETURN_IF_ERROR(CheckOpDeprecation(*op_def, versions_->producer()));
  }
  return Status::OK();
}

void RemoveInputs(const std::vector<int>& inputs_to_remove, NodeDef* node_def,
                  std::vector<bool>* input_already_exists) {
  // Remove 'inputs_to_remove' from 'node_def'
  // TODO(skyewm): is there a better way to do this?
  std::vector<string> inputs;
  inputs.reserve(node_def->input_size());
  for (int i = 0; i < node_def->input_size(); ++i) {
    inputs.push_back(node_def->input(i));
  }
  node_def->clear_input();
  for (int i = 0, j = 0; i < inputs.size(); ++i) {
    if (j < inputs_to_remove.size() && i == inputs_to_remove[j]) {
      ++j;
    } else {
      node_def->add_input(inputs[i]);
    }
  }
  // Remove 'inputs_to_remove' from 'input_already_exists'
  for (int idx : inputs_to_remove) {
    input_already_exists->erase(input_already_exists->begin() + idx);
  }
  DCHECK_EQ(input_already_exists->size(), node_def->input_size());
}

void GraphConstructor::RemapNodeDefInputs(
    NodeDef* node_def, std::vector<bool>* input_already_exists) {
  DCHECK_EQ(input_already_exists->size(), node_def->input_size());
  std::set<TensorId> control_inputs;
  std::vector<int> inputs_to_remove;

  for (int i = 0; i < node_def->input_size(); ++i) {
    auto iter = opts_.input_map.find(ParseTensorName(node_def->input(i)));
    if (iter == opts_.input_map.end()) continue;
    used_input_map_keys_.insert(iter->first);

    TensorId new_input = iter->second;
    if (new_input.second == Graph::kControlSlot) {
      // Check if we've already remapped a different input to new_input, and if
      // so remove this input.
      if (control_inputs.count(new_input) > 0) {
        inputs_to_remove.push_back(i);
        continue;
      }
      control_inputs.insert(new_input);
    }
    node_def->set_input(i, new_input.ToString());
    (*input_already_exists)[i] = true;
  }
  if (!inputs_to_remove.empty()) {
    RemoveInputs(inputs_to_remove, node_def, input_already_exists);
  }
}

void GraphConstructor::AddControlDependencies(
    NodeDef* node_def, std::vector<bool>* input_already_exists) {
  // To avoid adding redundant control dependencies to every imported node, skip
  // nodes that will inherit the dependencies from another imported node.
  bool inherits_deps = false;
  for (int i = 0; i < node_def->input_size(); ++i) {
    // Assume we won't inherit dependencies from remapped inputs that already
    // exist in the graph. Even if we're wrong, we'll only add redundant
    // dependencies.
    if ((*input_already_exists)[i]) continue;

    // If this input is a backedge, assume we won't inherit the dependencies.
    // TODO(skyewm): we have many redundant ParseTensorName calls. It could be
    // worth optimizing these.
    TensorId id(ParseTensorName(node_def->input(i)));
    auto iter = gdef_nodes_.find(id.first);
    DCHECK(iter != gdef_nodes_.end()) << id.first;
    if (iter->second.node == nullptr) {
      // Input hasn't been created yet, indicating it's a backedge.
      continue;
    }
    inherits_deps = true;
  }
  if (inherits_deps) return;

  // node_def either has no inputs or all remapped inputs, add the control
  // dependencies
  for (const string& control_dep : opts_.control_dependencies) {
    string input = TensorId(control_dep, Graph::kControlSlot).ToString();
    const protobuf::RepeatedPtrField<string>& inputs = node_def->input();
    if (std::find(inputs.begin(), inputs.end(), input) != inputs.end()) {
      // Control dependency already exists
      continue;
    }
    node_def->add_input(input);
    input_already_exists->push_back(true);
  }
}

void GraphConstructor::AddPrefixToNodeDef(
    const std::vector<bool>& input_already_exists, NodeDef* node_def) {
  if (opts_.prefix.empty()) return;
  node_def->set_name(strings::StrCat(opts_.prefix, node_def->name()));
  // Update names of input nodes
  for (int i = 0; i < node_def->input_size(); ++i) {
    StringPiece input(node_def->input(i));
    // Skip remapped inputs (which already exist in g_ and are not being
    // imported).
    if (input_already_exists[i]) continue;
    if (input.Consume("^")) {
      node_def->set_input(i, strings::StrCat("^", opts_.prefix, input));
    } else {
      node_def->set_input(i, strings::StrCat(opts_.prefix, input));
    }
  }
  // Update names of colocation groups
  if (node_def->attr().find(kColocationAttrName) != node_def->attr().end()) {
    auto* list =
        node_def->mutable_attr()->at(kColocationAttrName).mutable_list();
    for (int i = 0; i < list->s_size(); ++i) {
      StringPiece v(list->s(i));
      if (v.Consume(kColocationGroupPrefix)) {
        list->set_s(i,
                    strings::StrCat(kColocationGroupPrefix, opts_.prefix, v));
      }
    }
  }
}

void GraphConstructor::UniquifyNames(
    const std::vector<bool>& input_already_exists, NodeDef* node_def) {
  if (NameExists(node_def->name())) {
    string old_name = node_def->name();
    node_def->set_name(FindUniqueName(node_def->name()));
    uniquified_names_[old_name] = node_def->name();
  }
  for (int i = 0; i < node_def->input_size(); ++i) {
    // Skip remapped inputs (which already exist in g_ and are not being
    // imported).
    if (input_already_exists[i]) continue;
    TensorId id = ParseTensorName(node_def->input(i));
    // We require that UniquifyNames() is called on all NodeDefs in topological
    // order. This guarantees that node_def's inputs will already be uniquified
    // if necessary.
    auto iter = uniquified_names_.find(id.first.ToString());
    if (iter == uniquified_names_.end()) continue;
    id.first = iter->second;
    node_def->set_input(i, id.ToString());
  }
  // Update names of colocation groups
  if (node_def->attr().find(kColocationAttrName) != node_def->attr().end()) {
    auto* list =
        node_def->mutable_attr()->at(kColocationAttrName).mutable_list();
    for (int i = 0; i < list->s_size(); ++i) {
      StringPiece v(list->s(i));
      if (v.Consume(kColocationGroupPrefix)) {
        auto iter = uniquified_names_.find(v.ToString());
        if (iter == uniquified_names_.end()) continue;
        list->set_s(i, strings::StrCat(kColocationGroupPrefix, iter->second));
      }
    }
  }
}

bool GraphConstructor::NameExists(StringPiece name) {
  if (existing_nodes_.find(name) != existing_nodes_.end()) return true;
  return existing_prefixes_.find(name) != existing_prefixes_.end();
}

string GraphConstructor::FindUniqueName(StringPiece original_name) {
  string name = original_name.ToString();
  int count = 1;
  while (NameExists(name)) {
    name = strings::StrCat(original_name, "_", count++);
  }
  return name;
}

Status GraphConstructor::IsNodeFullyMapped(const NodeDef& node_def,
                                           bool* is_node_mapped) {
  const OpDef* op_def;
  TF_RETURN_IF_ERROR(g_->op_registry()->LookUpOpDef(node_def.op(), &op_def));
  for (int i = 0; i < op_def->output_arg_size(); ++i) {
    if (opts_.input_map.find({node_def.name(), i}) == opts_.input_map.end()) {
      *is_node_mapped = false;
      return Status::OK();
    }
  }
  *is_node_mapped = true;
  return Status::OK();
}

namespace {

void UpdatePendingCountAndReady(
    const std::vector<gtl::InlinedVector<int, 4>>& outputs, int o,
    std::vector<int>* pending_count, std::vector<int>* ready) {
  for (size_t i = 0; i < outputs[o].size(); ++i) {
    const int output = outputs[o][i];
    (*pending_count)[output]--;
    if ((*pending_count)[output] == 0) {
      ready->push_back(output);
    }
  }
}

}  // anonymous namespace

Status GraphConstructor::Convert() {
  // Import functions before adding nodes, since imported nodes may refer to
  // functions
  if (library_) {
    TF_RETURN_IF_ERROR(g_->AddFunctionLibrary(*library_));
  }

  std::vector<InputInfo> inputs;
  int processed = 0;

  std::vector<bool> input_already_exists;

  // Process the NodeDefs in topological order.
  // (InitFromEdges() sets this up by filling in ready_ with nodes that have no
  // inputs, pending_counts_ with the number of inputs for each node and
  // outputs_ with the outputs of each node).
  while (!ready_.empty()) {
    int o = ready_.back();
    ready_.pop_back();
    ++processed;
    inputs.clear();
    bool has_data_back_edge = false;

    const NodeDef& original_node_def = *node_defs_[o];
    NodeDef imported_node_def;
    const NodeDef* node_def;

    // input_already_exists[i] is true iff the i-th input of the node we're
    // importing refers to a preexisting node in g_ (i.e. input[i] existed prior
    // to importing node_defs_).  Conversely, input_already_exists[i] is false
    // iff the input refers to a node in node_defs_.
    input_already_exists.clear();
    input_already_exists.resize(original_node_def.input_size(), false);

    if (opts_.importing) {
      if (opts_.skip_mapped_nodes) {
        bool is_node_mapped = false;
        TF_RETURN_IF_ERROR(
            IsNodeFullyMapped(original_node_def, &is_node_mapped));
        if (is_node_mapped) {
          // Skip this node after updating pending_count_ for outputs
          UpdatePendingCountAndReady(outputs_, o, &pending_count_, &ready_);
          continue;
        }
      }

      // TODO(ashankar): The line below means an additional copy of the NodeDef,
      // which can be expensive if the NodeDef contains large tensors in it.
      // Might make sense to change the API for ImportGraphDef to take a mutable
      // GraphDef* and avoid the copying.
      imported_node_def = original_node_def;
      if (!opts_.input_map.empty()) {
        // Note that input_already_exists can shrink here
        RemapNodeDefInputs(&imported_node_def, &input_already_exists);
      }
      if (!opts_.control_dependencies.empty()) {
        // Note that input_already_exists can grow here
        AddControlDependencies(&imported_node_def, &input_already_exists);
      }
      node_def = &imported_node_def;
    } else {
      node_def = &original_node_def;
    }

    DCHECK_EQ(node_def->input_size(), input_already_exists.size());
    TF_RETURN_IF_ERROR(ValidateColocationConstraints(*node_def));
    for (int i = 0; i < node_def->input_size(); ++i) {
      TensorId id(ParseTensorName(node_def->input(i)));
      Node* src_node;
      int src_index;

      if (!input_already_exists[i]) {
        // Locate input in newly-imported nodes
        auto iter = gdef_nodes_.find(id.first);
        DCHECK(iter != gdef_nodes_.end()) << id.first;
        src_node = iter->second.node;
        src_index = id.second;
        if (src_node == nullptr) has_data_back_edge = true;
      } else {
        // Input refers to preexistng node in graph
        auto iter = existing_nodes_.find(id.first);
        DCHECK(iter != existing_nodes_.end()) << id.first;
        src_node = iter->second;
        src_index = id.second;
      }

      if (src_node != nullptr && src_index >= src_node->num_outputs()) {
        return errors::InvalidArgument(
            "Node '", node_def->name(), "': Connecting to invalid output ",
            id.second, " of source node ", id.first, " which has ",
            src_node->num_outputs(), " outputs");
      }

      inputs.push_back(InputInfo(id.first.ToString(), src_node, src_index));
    }

    if (has_data_back_edge && !IsMerge(*node_def)) {
      return errors::InvalidArgument(
          "Node '", node_def->name(),
          "' had a back edge, but only Merge nodes can have back edges.");
    }

    Node* node;
    if (opts_.importing) {
      if (!opts_.prefix.empty()) {
        AddPrefixToNodeDef(input_already_exists, &imported_node_def);
      } else if (opts_.uniquify_names) {
        UniquifyNames(input_already_exists, &imported_node_def);
      }
      TF_RETURN_IF_ERROR(ModifyNodeDefForImport(&imported_node_def));
    }
    TF_RETURN_IF_ERROR(MakeNode(*node_def, &node));
    // Use original_node_def so name StringPiece remains valid
    gdef_nodes_[original_node_def.name()].node = node;

    // Add edges from inputs to *node to the graph.
    for (size_t i = 0; i < inputs.size(); ++i) {
      if (inputs[i].node == nullptr) {
        // Record this back edge, which will be added after all nodes
        // are created.
        back_edges_.push_back(
            EdgeInfo(inputs[i].name, inputs[i].index, node, i));
      } else if (inputs[i].index == Graph::kControlSlot) {
        g_->AddControlEdge(inputs[i].node, node);
      } else {
        TF_RETURN_IF_ERROR(MakeEdge(inputs[i].node, inputs[i].index, node, i));
      }
    }

    // Function shape inference is supported on an opt-in basis per
    // ShapeRefiner.
    if (refiner_->function_shape_inference_supported() ||
        g_->flib_def().Find(node_def->name()) == nullptr) {
      TF_RETURN_IF_ERROR(ValidateShape(node));
    }

    // Update pending_count_ for outputs.
    UpdatePendingCountAndReady(outputs_, o, &pending_count_, &ready_);
  }

  if (processed < node_defs_.size()) {
    return errors::InvalidArgument(node_defs_.size() - processed,
                                   " nodes in a cycle");
  }

  // Update unused_input_map_keys_
  if (unused_input_map_keys_ != nullptr) {
    for (const auto& pair : opts_.input_map) {
      if (used_input_map_keys_.find(pair.first) == used_input_map_keys_.end()) {
        unused_input_map_keys_->push_back(pair.first);
      }
    }
  }

  return Status::OK();
}

Status GraphConstructor::AddBackEdges() {
  // Add the back edges after all nodes are created.
  for (auto e : back_edges_) {
    Node* src_node = gdef_nodes_[e.src_name].node;
    if (e.src_index == Graph::kControlSlot) {
      g_->AddControlEdge(src_node, e.dst_node);
    } else {
      TF_RETURN_IF_ERROR(
          MakeEdge(src_node, e.src_index, e.dst_node, e.dst_index));
    }

    VLOG(2) << "Add back edge: " << src_node->name() << " -> "
            << e.dst_node->name();
  }
  return Status::OK();
}

Status GraphConstructor::UpdateVersionDef() {
  if (versions_ == nullptr) return Status::OK();

  if (!opts_.importing) {
    g_->set_versions(*versions_);
    return Status::OK();
  }
  VersionDef versions = g_->versions();
  versions.set_producer(std::min(versions.producer(), versions_->producer()));
  versions.set_min_consumer(
      std::max(versions.min_consumer(), versions_->min_consumer()));
  if (versions_->bad_consumers_size() > 0) {
    std::set<int> bad(versions.bad_consumers().begin(),
                      versions.bad_consumers().end());
    bad.insert(versions_->bad_consumers().begin(),
               versions_->bad_consumers().end());
    versions.clear_bad_consumers();
    for (int v : bad) {
      versions.add_bad_consumers(v);
    }
  }
  g_->set_versions(versions);
  return Status::OK();
}

Status GraphConstructor::PopulateReturnTensors() {
  if (opts_.return_tensors.empty()) return Status::OK();
  for (const TensorId& id : opts_.return_tensors) {
    auto iter = opts_.input_map.find(id);
    if (iter == opts_.input_map.end()) {
      // Locate id in imported nodes
      auto iter = gdef_nodes_.find(id.first);
      if (iter == gdef_nodes_.end()) {
        return errors::InvalidArgument("Requested return tensor '",
                                       id.ToString(),
                                       "' not found in graph def");
      }
      int num_outputs = iter->second.node->num_outputs();
      if ((id.second < 0 || id.second >= num_outputs) &&
          id.second != Graph::kControlSlot) {
        return errors::InvalidArgument("Invalid return output ", id.second,
                                       " of node '", id.first, "', which has ",
                                       num_outputs, " output(s)");
      }
      return_tensors_->push_back({iter->second.node, id.second});
    } else {
      // id was remapped to existing node
      TensorId remapped_id = iter->second;
      DCHECK_GT(existing_nodes_.count(remapped_id.first), 0);
      Node* node = existing_nodes_[remapped_id.first];
      return_tensors_->push_back({node, remapped_id.second});
    }
  }
  return Status::OK();
}

Status GraphConstructor::PopulateReturnNodes() {
  if (opts_.return_nodes.empty()) return Status::OK();
  for (StringPiece name : opts_.return_nodes) {
    auto iter = gdef_nodes_.find(name);
    if (iter == gdef_nodes_.end()) {
      return errors::InvalidArgument("Requested return node '", name,
                                     "' not found in graph def");
    }
    return_nodes_->push_back(iter->second.node);
  }
  return Status::OK();
}

void GraphConstructor::Undo() {
  for (const auto& iter : gdef_nodes_) {
    if (iter.second.node != nullptr) {
      g_->RemoveNode(iter.second.node);
    }
  }
  g_->set_versions(original_versions_);
}

Status GraphConstructor::MakeEdge(Node* src, int output_index, Node* dst,
                                  int input_index) {
  DataType src_out = src->output_type(output_index);
  DataType dst_in = dst->input_type(input_index);
  if (!TypesCompatible(dst_in, src_out)) {
    return errors::InvalidArgument(
        "Input ", input_index, " of node ", dst->name(), " was passed ",
        DataTypeString(src_out), " from ", src->name(), ":", output_index,
        " incompatible with expected ", DataTypeString(dst_in), ".");
  }
  g_->AddEdge(src, output_index, dst, input_index);
  return Status::OK();
}

}  // namespace

Status ConvertGraphDefToGraph(const GraphConstructorOptions& opts,
                              const GraphDef& gdef, Graph* g) {
  ShapeRefiner refiner(gdef.versions().producer(), g->op_registry());
  return GraphConstructor::Construct(
      opts, gdef.node(), &gdef.versions(), &gdef.library(), g, &refiner,
      /*return_tensors=*/nullptr, /*return_nodes=*/nullptr,
      /*unused_input_map_keys=*/nullptr);
}

Status ConvertNodeDefsToGraph(const GraphConstructorOptions& opts,
                              gtl::ArraySlice<NodeDef> nodes, Graph* g) {
  ShapeRefiner refiner(TF_GRAPH_DEF_VERSION, g->op_registry());
  // TODO(irving): Copy will go away once NodeInfo exists
  std::vector<const NodeDef*> node_defs;
  for (const auto& n : nodes) {
    node_defs.push_back(&n);
  }
  return GraphConstructor::Construct(opts, node_defs, nullptr, nullptr, g,
                                     &refiner, /*return_tensors=*/nullptr,
                                     /*return_nodes=*/nullptr,
                                     /*unused_input_map_keys=*/nullptr);
}

Status ImportGraphDef(const ImportGraphDefOptions& opts, const GraphDef& gdef,
                      Graph* g, ShapeRefiner* refiner,
                      ImportGraphDefResults* results) {
  if (!opts.return_tensors.empty()) {
    if (results == nullptr) {
      return errors::InvalidArgument(
          "results argument to ImportGraphDef() must be non-null if "
          "opts.return_tensors is non-empty");
    }
  }

  if (!opts.return_nodes.empty()) {
    if (opts.skip_mapped_nodes) {
      return errors::InvalidArgument(
          "Requesting return_nodes with skip_mapped_nodes set is not currently "
          "supported");
    }
    if (results == nullptr) {
      return errors::InvalidArgument(
          "results argument to ImportGraphDef() must be non-null if "
          "opts.return_nodes is non-empty");
    }
  }

  if (results != nullptr) {
    if (!results->return_tensors.empty() || !results->return_nodes.empty() ||
        !results->unused_input_map_keys.empty()) {
      return errors::InvalidArgument(
          "All fields in results argument to ImportGraphDef() must be empty.");
    }
  }

  ShapeRefiner default_refiner(gdef.versions().producer(), g->op_registry());
  if (refiner == nullptr) {
    refiner = &default_refiner;
  } else {
    // Log a warning if we are importing a GraphDef at an older
    // producer version after already having added non-source/sink
    // nodes to the graph in the past.
    if (gdef.versions().producer() > 0 &&
        gdef.versions().producer() < refiner->graph_def_version() &&
        g->num_nodes() > 2) {
      LOG(WARNING) << "Importing a graph with a lower producer version "
                   << gdef.versions().producer()
                   << " into an existing graph with producer version "
                   << refiner->graph_def_version() << ". Shape inference will "
                   << "have run different parts of the graph with different "
                   << "producer versions.";
    }
  }

  // Set the graph def version of the refiner as the min of the
  // current value and the version from the graph we are about to
  // import.
  //
  // Note: to match Run() semantics, we should re-run shape inference
  // on the entire graph if the producer version has changed.  For now
  // we log the warning above.
  refiner->set_graph_def_version(
      std::min(refiner->graph_def_version(), gdef.versions().producer()));

  if (results == nullptr) {
    return GraphConstructor::Construct(opts, gdef.node(), &gdef.versions(),
                                       &gdef.library(), g, refiner, nullptr,
                                       nullptr, nullptr);
  } else {
    return GraphConstructor::Construct(
        opts, gdef.node(), &gdef.versions(), &gdef.library(), g, refiner,
        &results->return_tensors, &results->return_nodes,
        &results->unused_input_map_keys);
  }
}

void CopyGraph(const Graph& src, Graph* dest) {
  for (Node* n : dest->nodes()) {
    CHECK(n->IsSource() || n->IsSink()) << "*dest must be empty";
  }

  // Copy GraphDef versions
  dest->set_versions(src.versions());

  // Copy the nodes
  std::unordered_map<Node*, Node*>
      node_map;  // "Node in src" -> "Node in *dest"
  node_map[src.source_node()] = dest->source_node();
  node_map[src.sink_node()] = dest->sink_node();
  for (Node* n : src.op_nodes()) {
    node_map[n] = dest->CopyNode(n);
  }

  // Copy the edges
  for (const Edge* e : src.edges()) {
    Node* src_copy = node_map[e->src()];
    Node* dst_copy = node_map[e->dst()];
    dest->AddEdge(src_copy, e->src_output(), dst_copy, e->dst_input());
  }
}

}  // namespace tensorflow