1 | /* This file is part of the KDE libraries |
2 | SPDX-FileCopyrightText: 2006 Jacob R Rideout <kde@jacobrideout.net> |
3 | |
4 | SPDX-License-Identifier: LGPL-2.0-or-later |
5 | */ |
6 | |
7 | #ifndef TEXTBREAKS_H |
8 | #define TEXTBREAKS_H |
9 | |
10 | class QString; |
11 | |
12 | #include "sonnetcore_export.h" |
13 | |
14 | #include <QList> |
15 | |
16 | #include <memory> |
17 | |
18 | namespace Sonnet |
19 | { |
20 | class TextBreaksPrivate; |
21 | |
22 | /*! |
23 | * \internal |
24 | * |
25 | * \brief TextBreaks determines the barriers between linguistic structures in any given text. |
26 | * |
27 | * TextBreaks is a class that determines the boundaries between graphemes |
28 | * (characters as per the unicode definition,) words and sentences. The |
29 | * default implementation conforms to Unicode Standard Annex #29 https://unicode.org/reports/tr29/. |
30 | * You can subclass TextBreaks to create the correct behaviour for languages that require it. |
31 | * |
32 | * \since 4.3 |
33 | */ |
34 | class SONNETCORE_EXPORT TextBreaks |
35 | { |
36 | public: |
37 | struct Position { |
38 | int start, length; |
39 | }; |
40 | |
41 | /*! |
42 | * \brief This structure abstracts the positions of breaks in the test. As per the |
43 | * unicode annex, both the start and end of the text are returned. |
44 | */ |
45 | typedef QList<Position> Positions; |
46 | |
47 | /*! Constructor |
48 | * Creates a new TextBreaks instance. If \a text is specified, |
49 | * it sets the text to be checked. |
50 | * \a text the text that is to be checked |
51 | */ |
52 | explicit TextBreaks(const QString &text = QString()); |
53 | |
54 | /*! Virtual Destructor |
55 | */ |
56 | virtual ~TextBreaks(); |
57 | |
58 | /*! |
59 | * Returns the text to be checked |
60 | * Returns text |
61 | */ |
62 | QString text() const; |
63 | |
64 | /*! |
65 | * Sets the text to \a text |
66 | * \a text to be set |
67 | * Returns true if the word is misspelled. false otherwise |
68 | */ |
69 | void setText(const QString &text); |
70 | |
71 | /*! |
72 | * Return the Positions of each word for the given \a text. |
73 | * \a text to be checked |
74 | * Returns positions of breaks |
75 | */ |
76 | static Positions wordBreaks(const QString &text); |
77 | |
78 | /*! |
79 | * Return the Positions of each sentence for the given \a text. |
80 | * \a text to be checked |
81 | * Returns positions of breaks |
82 | */ |
83 | static Positions sentenceBreaks(const QString &text); |
84 | |
85 | /*! |
86 | * Return the Positions of each word for the text previously set. |
87 | * Returns positions of breaks |
88 | */ |
89 | virtual Positions wordBreaks() const; |
90 | |
91 | /*! |
92 | * Return the Positions of each sentence for the text previously set. |
93 | * Returns positions of breaks |
94 | */ |
95 | virtual Positions sentenceBreaks() const; |
96 | |
97 | private: |
98 | std::unique_ptr<TextBreaksPrivate> const d; |
99 | }; |
100 | } |
101 | |
102 | Q_DECLARE_TYPEINFO(Sonnet::TextBreaks::Position, Q_PRIMITIVE_TYPE); |
103 | |
104 | #endif |
105 | |