diff options
author | Daniel Baumann <daniel.baumann@progress-linux.org> | 2024-04-07 09:06:44 +0000 |
---|---|---|
committer | Daniel Baumann <daniel.baumann@progress-linux.org> | 2024-04-07 09:06:44 +0000 |
commit | ed5640d8b587fbcfed7dd7967f3de04b37a76f26 (patch) | |
tree | 7a5f7c6c9d02226d7471cb3cc8fbbf631b415303 /sc/inc/rangecache.hxx | |
parent | Initial commit. (diff) | |
download | libreoffice-ed5640d8b587fbcfed7dd7967f3de04b37a76f26.tar.xz libreoffice-ed5640d8b587fbcfed7dd7967f3de04b37a76f26.zip |
Adding upstream version 4:7.4.7.upstream/4%7.4.7upstream
Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>
Diffstat (limited to 'sc/inc/rangecache.hxx')
-rw-r--r-- | sc/inc/rangecache.hxx | 126 |
1 files changed, 126 insertions, 0 deletions
diff --git a/sc/inc/rangecache.hxx b/sc/inc/rangecache.hxx new file mode 100644 index 000000000..c65e8653a --- /dev/null +++ b/sc/inc/rangecache.hxx @@ -0,0 +1,126 @@ +/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */ +/* + * This file is part of the LibreOffice project. + * + * This Source Code Form is subject to the terms of the Mozilla Public + * License, v. 2.0. If a copy of the MPL was not distributed with this + * file, You can obtain one at http://mozilla.org/MPL/2.0/. + * + * This file incorporates work covered by the following license notice: + * + * Licensed to the Apache Software Foundation (ASF) under one or more + * contributor license agreements. See the NOTICE file distributed + * with this work for additional information regarding copyright + * ownership. The ASF licenses this file to you 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 . + */ + +#pragma once + +#include "address.hxx" +#include "queryentry.hxx" +#include <o3tl/hash_combine.hxx> +#include <svl/listener.hxx> + +#include <memory> +#include <unordered_map> + +class ScDocument; +struct ScInterpreterContext; +struct ScQueryParam; +struct ScSortedRangeCacheMap; + +/** Sorted cache for one range used with interpreter functions such as VLOOKUP + and MATCH. Caches sorted order for cells in the given range, which must + be one column. This allows faster lookups when cells are not sorted. + + The class has a vector of SCROW items, which is sorted according to values + of those cells. Therefore e.g. binary search of those cells can be done + by doing binary search of the vector while mapping the indexes to rows. + */ + +class ScSortedRangeCache final : public SvtListener +{ +public: + /// MUST be new'd because Notify() deletes. + ScSortedRangeCache(ScDocument* pDoc, const ScRange& rRange, const ScQueryParam& param, + ScInterpreterContext* context, bool invalid = false); + + /// Returns if the cache is usable. + bool isValid() const { return mValid; } + + /// Remove from document structure and delete (!) cache on modify hint. + virtual void Notify(const SfxHint& rHint) override; + + const ScRange& getRange() const { return maRange; } + + enum class ValueType + { + Values, + StringsCaseSensitive, + StringsCaseInsensitive + }; + struct HashKey + { + ScRange range; + ValueType valueType; + ScQueryOp queryOp; + ScQueryEntry::QueryType queryType; + bool operator==(const HashKey& other) const + { + return range == other.range && valueType == other.valueType && queryOp == other.queryOp + && queryType == other.queryType; + } + }; + HashKey getHashKey() const { return { maRange, mValueType, mQueryOp, mQueryType }; } + static HashKey makeHashKey(const ScRange& range, const ScQueryParam& param); + + struct Hash + { + size_t operator()(const HashKey& key) const + { + // Range should be just one column. + size_t hash = key.range.hashStartColumn(); + o3tl::hash_combine(hash, key.valueType); + o3tl::hash_combine(hash, key.queryOp); + o3tl::hash_combine(hash, key.queryType); + return hash; + } + }; + + const std::vector<SCROW>& sortedRows() const { return mSortedRows; } + size_t size() const { return mSortedRows.size(); } + size_t indexForRow(SCROW row) const + { + assert(row >= maRange.aStart.Row() && row <= maRange.aEnd.Row()); + assert(mRowToIndex[row - maRange.aStart.Row()] != mSortedRows.max_size()); + return mRowToIndex[row - maRange.aStart.Row()]; + } + SCROW rowForIndex(size_t index) const { return mSortedRows[index]; } + +private: + // Rows sorted by their value. + std::vector<SCROW> mSortedRows; + std::vector<size_t> mRowToIndex; // indexed by 'SCROW - maRange.aStart.Row()' + ScRange maRange; + ScDocument* mpDoc; + bool mValid; + ValueType mValueType; + ScQueryOp mQueryOp; + ScQueryEntry::QueryType mQueryType; + + ScSortedRangeCache(const ScSortedRangeCache&) = delete; + ScSortedRangeCache& operator=(const ScSortedRangeCache&) = delete; +}; + +// Struct because including lookupcache.hxx in document.hxx isn't wanted. +struct ScSortedRangeCacheMap +{ + std::unordered_map<ScSortedRangeCache::HashKey, std::unique_ptr<ScSortedRangeCache>, + ScSortedRangeCache::Hash> + aCacheMap; +}; + +/* vim:set shiftwidth=4 softtabstop=4 expandtab: */ |