ttf_parser/tables/cmap/
format2.rs

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
// This table has a pretty complex parsing algorithm.
// A detailed explanation can be found here:
// https://docs.microsoft.com/en-us/typography/opentype/spec/cmap#format-2-high-byte-mapping-through-table
// https://developer.apple.com/fonts/TrueType-Reference-Manual/RM06/Chap6cmap.html
// https://github.com/fonttools/fonttools/blob/a360252709a3d65f899915db0a5bd753007fdbb7/Lib/fontTools/ttLib/tables/_c_m_a_p.py#L360

use core::convert::TryFrom;

use crate::parser::{FromData, LazyArray16, Stream};
use crate::GlyphId;

#[derive(Clone, Copy)]
struct SubHeaderRecord {
    first_code: u16,
    entry_count: u16,
    id_delta: i16,
    id_range_offset: u16,
}

impl FromData for SubHeaderRecord {
    const SIZE: usize = 8;

    #[inline]
    fn parse(data: &[u8]) -> Option<Self> {
        let mut s = Stream::new(data);
        Some(SubHeaderRecord {
            first_code: s.read::<u16>()?,
            entry_count: s.read::<u16>()?,
            id_delta: s.read::<i16>()?,
            id_range_offset: s.read::<u16>()?,
        })
    }
}

/// A [format 2](https://docs.microsoft.com/en-us/typography/opentype/spec/cmap#format-2-high-byte-mapping-through-table)
/// subtable.
#[derive(Clone, Copy)]
pub struct Subtable2<'a> {
    sub_header_keys: LazyArray16<'a, u16>,
    sub_headers_offset: usize,
    sub_headers: LazyArray16<'a, SubHeaderRecord>,
    // The whole subtable data.
    data: &'a [u8],
}

impl<'a> Subtable2<'a> {
    /// Parses a subtable from raw data.
    pub fn parse(data: &'a [u8]) -> Option<Self> {
        let mut s = Stream::new(data);
        s.skip::<u16>(); // format
        s.skip::<u16>(); // length
        s.skip::<u16>(); // language
        let sub_header_keys = s.read_array16::<u16>(256)?;
        // The maximum index in a sub_header_keys is a sub_headers count.
        let sub_headers_count = sub_header_keys.into_iter().map(|n| n / 8).max()? + 1;

        // Remember sub_headers offset before reading. Will be used later.
        let sub_headers_offset = s.offset();
        let sub_headers = s.read_array16::<SubHeaderRecord>(sub_headers_count)?;

        Some(Self {
            sub_header_keys,
            sub_headers_offset,
            sub_headers,
            data,
        })
    }

    /// Returns a glyph index for a code point.
    ///
    /// Returns `None` when `code_point` is larger than `u16`.
    pub fn glyph_index(&self, code_point: u32) -> Option<GlyphId> {
        // This subtable supports code points only in a u16 range.
        let code_point = u16::try_from(code_point).ok()?;

        let code_point = code_point;
        let high_byte = code_point >> 8;
        let low_byte = code_point & 0x00FF;

        let i = if code_point < 0xff {
            // 'SubHeader 0 is special: it is used for single-byte character codes.'
            0
        } else {
            // 'Array that maps high bytes to subHeaders: value is subHeader index × 8.'
            self.sub_header_keys.get(high_byte)? / 8
        };

        let sub_header = self.sub_headers.get(i)?;

        let first_code = sub_header.first_code;
        let range_end = first_code.checked_add(sub_header.entry_count)?;
        if low_byte < first_code || low_byte >= range_end {
            return None;
        }

        // SubHeaderRecord::id_range_offset points to SubHeaderRecord::first_code
        // in the glyphIndexArray. So we have to advance to our code point.
        let index_offset = usize::from(low_byte.checked_sub(first_code)?) * u16::SIZE;

        // 'The value of the idRangeOffset is the number of bytes
        // past the actual location of the idRangeOffset'.
        let offset = self.sub_headers_offset
                // Advance to required subheader.
                + SubHeaderRecord::SIZE * usize::from(i + 1)
                // Move back to idRangeOffset start.
                - u16::SIZE
                // Use defined offset.
                + usize::from(sub_header.id_range_offset)
                // Advance to required index in the glyphIndexArray.
                + index_offset;

        let glyph: u16 = Stream::read_at(self.data, offset)?;
        if glyph == 0 {
            return None;
        }

        u16::try_from((i32::from(glyph) + i32::from(sub_header.id_delta)) % 65536)
            .ok()
            .map(GlyphId)
    }

    /// Calls `f` for each codepoint defined in this table.
    pub fn codepoints(&self, f: impl FnMut(u32)) {
        let _ = self.codepoints_inner(f);
    }

    #[inline]
    fn codepoints_inner(&self, mut f: impl FnMut(u32)) -> Option<()> {
        for first_byte in 0u16..256 {
            let i = self.sub_header_keys.get(first_byte)? / 8;
            let sub_header = self.sub_headers.get(i)?;
            let first_code = sub_header.first_code;

            if i == 0 {
                // This is a single byte code.
                let range_end = first_code.checked_add(sub_header.entry_count)?;
                if first_byte >= first_code && first_byte < range_end {
                    f(u32::from(first_byte));
                }
            } else {
                // This is a two byte code.
                let base = first_code.checked_add(first_byte << 8)?;
                for k in 0..sub_header.entry_count {
                    let code_point = base.checked_add(k)?;
                    f(u32::from(code_point));
                }
            }
        }

        Some(())
    }
}

impl core::fmt::Debug for Subtable2<'_> {
    fn fmt(&self, f: &mut core::fmt::Formatter) -> core::fmt::Result {
        write!(f, "Subtable2 {{ ... }}")
    }
}