unicode_normalization/
lookups.rs

Help
1// Copyright 2019 The Rust Project Developers. See the COPYRIGHT
2// file at the top-level directory of this distribution and at
3// http://rust-lang.org/COPYRIGHT.
4//
5// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or
6// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license
7// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your
8// option. This file may not be copied, modified, or distributed
9// except according to those terms.
10
11//! Lookups of unicode properties using minimal perfect hashing.
12
13use crate::perfect_hash::mph_lookup;
14use crate::tables::*;
15
16/// Look up the canonical combining class for a codepoint.
17///
18/// The value returned is as defined in the Unicode Character Database.
19pub fn canonical_combining_class(c: char) -> u8 {
20    mph_lookup(
21        c.into(),
22        CANONICAL_COMBINING_CLASS_SALT,
23        CANONICAL_COMBINING_CLASS_KV,
24        u8_lookup_fk,
25        u8_lookup_fv,
26        0,
27    )
28}
29
30pub(crate) fn composition_table(c1: char, c2: char) -> Option<char> {
31    if c1 < '\u{10000}' && c2 < '\u{10000}' {
32        mph_lookup(
33            (c1 as u32) << 16 | (c2 as u32),
34            COMPOSITION_TABLE_SALT,
35            COMPOSITION_TABLE_KV,
36            pair_lookup_fk,
37            pair_lookup_fv_opt,
38            None,
39        )
40    } else {
41        composition_table_astral(c1, c2)
42    }
43}
44
45pub(crate) fn canonical_fully_decomposed(c: char) -> Option<&'static [char]> {
46    mph_lookup(
47        c.into(),
48        CANONICAL_DECOMPOSED_SALT,
49        CANONICAL_DECOMPOSED_KV,
50        pair_lookup_fk,
51        pair_lookup_fv_opt,
52        None,
53    )
54    .map(|(start, len)| &CANONICAL_DECOMPOSED_CHARS[start as usize..][..len as usize])
55}
56
57pub(crate) fn compatibility_fully_decomposed(c: char) -> Option<&'static [char]> {
58    mph_lookup(
59        c.into(),
60        COMPATIBILITY_DECOMPOSED_SALT,
61        COMPATIBILITY_DECOMPOSED_KV,
62        pair_lookup_fk,
63        pair_lookup_fv_opt,
64        None,
65    )
66    .map(|(start, len)| &COMPATIBILITY_DECOMPOSED_CHARS[start as usize..][..len as usize])
67}
68
69pub(crate) fn cjk_compat_variants_fully_decomposed(c: char) -> Option<&'static [char]> {
70    mph_lookup(
71        c.into(),
72        CJK_COMPAT_VARIANTS_DECOMPOSED_SALT,
73        CJK_COMPAT_VARIANTS_DECOMPOSED_KV,
74        pair_lookup_fk,
75        pair_lookup_fv_opt,
76        None,
77    )
78    .map(|(start, len)| &CJK_COMPAT_VARIANTS_DECOMPOSED_CHARS[start as usize..][..len as usize])
79}
80
81/// Return whether the given character is a combining mark (`General_Category=Mark`)
82pub fn is_combining_mark(c: char) -> bool {
83    mph_lookup(
84        c.into(),
85        COMBINING_MARK_SALT,
86        COMBINING_MARK_KV,
87        bool_lookup_fk,
88        bool_lookup_fv,
89        false,
90    )
91}
92
93pub fn stream_safe_trailing_nonstarters(c: char) -> usize {
94    mph_lookup(
95        c.into(),
96        TRAILING_NONSTARTERS_SALT,
97        TRAILING_NONSTARTERS_KV,
98        u8_lookup_fk,
99        u8_lookup_fv,
100        0,
101    ) as usize
102}
103
104/// Extract the key in a 24 bit key and 8 bit value packed in a u32.
105#[inline]
106fn u8_lookup_fk(kv: u32) -> u32 {
107    kv >> 8
108}
109
110/// Extract the value in a 24 bit key and 8 bit value packed in a u32.
111#[inline]
112fn u8_lookup_fv(kv: u32) -> u8 {
113    (kv & 0xff) as u8
114}
115
116/// Extract the key for a boolean lookup.
117#[inline]
118fn bool_lookup_fk(kv: u32) -> u32 {
119    kv
120}
121
122/// Extract the value for a boolean lookup.
123#[inline]
124fn bool_lookup_fv(_kv: u32) -> bool {
125    true
126}
127
128/// Extract the key in a pair.
129#[inline]
130fn pair_lookup_fk<T>(kv: (u32, T)) -> u32 {
131    kv.0
132}
133
134/// Extract the value in a pair, returning an option.
135#[inline]
136fn pair_lookup_fv_opt<T>(kv: (u32, T)) -> Option<T> {
137    Some(kv.1)
138}
unicode_normalization/lookups.rs

unicode_normalization/
lookups.rs
