blob: 1fa54af9d97f527557e694131b3836e9894e686f [file] [log] [blame]
/* -*- Mode:C++; c-file-style:"gnu"; indent-tabs-mode:nil; -*- */
/*
* Copyright (c) 2014-2017, Regents of the University of California,
* Arizona Board of Regents,
* Colorado State University,
* University Pierre & Marie Curie, Sorbonne University,
* Washington University in St. Louis,
* Beijing Institute of Technology,
* The University of Memphis.
*
* This file is part of NFD (Named Data Networking Forwarding Daemon).
* See AUTHORS.md for complete list of NFD authors and contributors.
*
* NFD 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.
*
* NFD 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
* NFD, e.g., in COPYING.md file. If not, see <http://www.gnu.org/licenses/>.
*/
#include "name-tree.hpp"
#include "core/logger.hpp"
#include <boost/concept/assert.hpp>
#include <boost/concept_check.hpp>
#include <type_traits>
namespace nfd {
namespace name_tree {
NFD_LOG_INIT("NameTree");
NameTree::NameTree(size_t nBuckets)
: m_ht(HashtableOptions(nBuckets))
{
}
Entry&
NameTree::lookup(const Name& name, bool enforceMaxDepth)
{
NFD_LOG_TRACE("lookup " << name);
size_t depth = enforceMaxDepth ? std::min(name.size(), getMaxDepth()) : name.size();
HashSequence hashes = computeHashes(name, depth);
const Node* node = nullptr;
Entry* parent = nullptr;
for (size_t prefixLen = 0; prefixLen <= depth; ++prefixLen) {
bool isNew = false;
std::tie(node, isNew) = m_ht.insert(name, prefixLen, hashes);
if (isNew && parent != nullptr) {
node->entry.setParent(*parent);
}
parent = &node->entry;
}
return node->entry;
}
Entry&
NameTree::lookup(const fib::Entry& fibEntry)
{
Entry* nte = this->getEntry(fibEntry);
if (nte == nullptr) {
// special case: Fib::s_emptyEntry is unattached
BOOST_ASSERT(fibEntry.getPrefix().empty());
return this->lookup(fibEntry.getPrefix());
}
BOOST_ASSERT(nte->getFibEntry() == &fibEntry);
return *nte;
}
Entry&
NameTree::lookup(const pit::Entry& pitEntry)
{
Entry* nte = this->getEntry(pitEntry);
BOOST_ASSERT(nte != nullptr);
BOOST_ASSERT(std::count_if(nte->getPitEntries().begin(), nte->getPitEntries().end(),
[&pitEntry] (const shared_ptr<pit::Entry>& pitEntry1) {
return pitEntry1.get() == &pitEntry;
}) == 1);
if (nte->getName().size() == pitEntry.getName().size()) {
return *nte;
}
// special case: PIT entry whose Interest name ends with an implicit digest
// are attached to the name tree entry with one-shorter-prefix.
BOOST_ASSERT(pitEntry.getName().at(-1).isImplicitSha256Digest());
BOOST_ASSERT(nte->getName() == pitEntry.getName().getPrefix(-1));
return this->lookup(pitEntry.getName());
}
Entry&
NameTree::lookup(const measurements::Entry& measurementsEntry)
{
Entry* nte = this->getEntry(measurementsEntry);
BOOST_ASSERT(nte != nullptr);
BOOST_ASSERT(nte->getMeasurementsEntry() == &measurementsEntry);
return *nte;
}
Entry&
NameTree::lookup(const strategy_choice::Entry& strategyChoiceEntry)
{
Entry* nte = this->getEntry(strategyChoiceEntry);
BOOST_ASSERT(nte != nullptr);
BOOST_ASSERT(nte->getStrategyChoiceEntry() == &strategyChoiceEntry);
return *nte;
}
size_t
NameTree::eraseIfEmpty(Entry* entry, bool canEraseAncestors)
{
BOOST_ASSERT(entry != nullptr);
size_t nErased = 0;
for (Entry* parent = nullptr; entry != nullptr && entry->isEmpty(); entry = parent) {
parent = entry->getParent();
if (parent != nullptr) {
entry->unsetParent();
}
m_ht.erase(getNode(*entry));
++nErased;
if (!canEraseAncestors) {
break;
}
}
if (nErased == 0) {
NFD_LOG_TRACE("not-erase " << entry->getName());
}
return nErased;
}
Entry*
NameTree::findExactMatch(const Name& name, size_t prefixLen) const
{
const Node* node = m_ht.find(name, std::min(name.size(), prefixLen));
return node == nullptr ? nullptr : &node->entry;
}
Entry*
NameTree::findLongestPrefixMatch(const Name& name, const EntrySelector& entrySelector) const
{
HashSequence hashes = computeHashes(name);
for (ssize_t prefixLen = name.size(); prefixLen >= 0; --prefixLen) {
const Node* node = m_ht.find(name, prefixLen, hashes);
if (node != nullptr && entrySelector(node->entry)) {
return &node->entry;
}
}
return nullptr;
}
Entry*
NameTree::findLongestPrefixMatch(const Entry& entry1, const EntrySelector& entrySelector) const
{
Entry* entry = const_cast<Entry*>(&entry1);
while (entry != nullptr) {
if (entrySelector(*entry)) {
return entry;
}
entry = entry->getParent();
}
return nullptr;
}
Entry*
NameTree::findLongestPrefixMatch(const pit::Entry& pitEntry, const EntrySelector& entrySelector) const
{
const Entry* nte = this->getEntry(pitEntry);
BOOST_ASSERT(nte != nullptr);
// PIT entry Interest name either exceeds depth limit or ends with an implicit digest: go deeper
if (nte->getName().size() < pitEntry.getName().size()) {
for (size_t prefixLen = nte->getName().size() + 1; prefixLen <= pitEntry.getName().size(); ++prefixLen) {
const Entry* exact = this->findExactMatch(pitEntry.getName(), prefixLen);
if (exact == nullptr) {
break;
}
nte = exact;
}
}
return this->findLongestPrefixMatch(*nte, entrySelector);
}
boost::iterator_range<NameTree::const_iterator>
NameTree::findAllMatches(const Name& name, const EntrySelector& entrySelector) const
{
// As we are using Name Prefix Hash Table, and the current LPM() is
// implemented as starting from full name, and reduce the number of
// components by 1 each time, we could use it here.
// For trie-like design, it could be more efficient by walking down the
// trie from the root node.
Entry* entry = this->findLongestPrefixMatch(name, entrySelector);
return {Iterator(make_shared<PrefixMatchImpl>(*this, entrySelector), entry), end()};
}
boost::iterator_range<NameTree::const_iterator>
NameTree::fullEnumerate(const EntrySelector& entrySelector) const
{
return {Iterator(make_shared<FullEnumerationImpl>(*this, entrySelector), nullptr), end()};
}
boost::iterator_range<NameTree::const_iterator>
NameTree::partialEnumerate(const Name& prefix,
const EntrySubTreeSelector& entrySubTreeSelector) const
{
Entry* entry = this->findExactMatch(prefix);
return {Iterator(make_shared<PartialEnumerationImpl>(*this, entrySubTreeSelector), entry), end()};
}
} // namespace name_tree
} // namespace nfd