snippet.cc | Explore in Territory

// Copyright 2014 The Chromium Authors
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file.

#ifdef UNSAFE_BUFFERS_BUILD
// TODO(crbug.com/40285824): Remove this and convert code to safer constructs.
#pragma allow_unsafe_buffers
#endif

#include "components/query_parser/snippet.h"

#include <stdint.h>

#include <algorithm>
#include <memory>

#include "base/check.h"
#include "base/strings/string_split.h"
#include "base/strings/string_util.h"
#include "base/strings/utf_string_conversions.h"
#include "third_party/icu/source/common/unicode/brkiter.h"
#include "third_party/icu/source/common/unicode/utext.h"
#include "third_party/icu/source/common/unicode/utf8.h"

namespace query_parser {
namespace {

bool PairFirstLessThan(const Snippet::MatchPosition& a,
                       const Snippet::MatchPosition& b) { … }

// Combines all pairs after offset in match_positions that are contained
// or touch the pair at offset.
void CoalescePositionsFrom(size_t offset,
                           Snippet::MatchPositions* match_positions) { … }

// Makes sure there is a pair in match_positions that contains the specified
// range. This keeps the pairs ordered in match_positions by first, and makes
// sure none of the pairs in match_positions touch each other.
void AddMatch(size_t start,
              size_t end,
              Snippet::MatchPositions* match_positions) { … }

// Converts an index in a utf8 string into the index in the corresponding utf16
// string and returns the utf16 index. This is intended to be called in a loop
// iterating through a utf8 string.
//
// utf8_string: the utf8 string.
// utf8_length: length of the utf8 string.
// offset: the utf8 offset to convert.
// utf8_pos: current offset in the utf8 string. This is modified and on return
//           matches offset.
// wide_pos: current index in the wide string. This is the same as the return
//           value.
size_t AdvanceAndReturnUTF16Pos(const char* utf8_string,
                                int32_t utf8_length,
                                int32_t offset,
                                int32_t* utf8_pos,
                                size_t* utf16_pos) { … }

// Given a character break iterator over a UTF-8 string, set the iterator
// position to |*utf8_pos| and move by |count| characters. |count| can
// be either positive or negative.
void MoveByNGraphemes(icu::BreakIterator* bi, int count, size_t* utf8_pos) { … }

// The amount of context to include for a given hit. Note that it's counted
// in terms of graphemes rather than bytes.
const int kSnippetContext = …;

// Returns true if next match falls within a snippet window
// from the previous match. The window size is counted in terms
// of graphemes rather than bytes in UTF-8.
bool IsNextMatchWithinSnippetWindow(icu::BreakIterator* bi,
                                    size_t previous_match_end,
                                    size_t next_match_start) { … }

}  // namespace

// static
void Snippet::ExtractMatchPositions(const std::string& offsets_str,
                                    const std::string& column_num,
                                    MatchPositions* match_positions) { … }

// static
void Snippet::ConvertMatchPositionsToWide(
    const std::string& utf8_string,
    Snippet::MatchPositions* match_positions) { … }

Snippet::Snippet() { … }

Snippet::Snippet(const Snippet& other) = default;

Snippet::Snippet(Snippet&& other) noexcept = default;

Snippet::~Snippet() { … }

Snippet& Snippet::operator=(const Snippet&) = default;

void Snippet::ComputeSnippet(const MatchPositions& match_positions,
                             const std::string& document) { … }

void Snippet::Swap(Snippet* other) { … }

}  // namespace query_parser
chromium/components/query_parser/snippet.cc
