├── .github
    └── workflows
    │   └── ci.yml
├── .gitignore
├── CHANGELOG.md
├── COPYRIGHT
├── Cargo.toml
├── LICENSE-APACHE
├── LICENSE-MIT
├── README.md
├── benches
    └── xid.rs
├── scripts
    └── unicode.py
├── src
    ├── lib.rs
    ├── tables.rs
    └── tests.rs
└── tests
    └── exhaustive_tests.rs


/.github/workflows/ci.yml:
--------------------------------------------------------------------------------
 1 | name: CI
 2 | 
 3 | on:
 4 |   push:
 5 |   pull_request:
 6 |   schedule: [cron: "40 1 * * *"]
 7 | 
 8 | jobs:
 9 |   test:
10 |     name: Rust ${{matrix.rust}}
11 |     runs-on: ubuntu-latest
12 |     strategy:
13 |       fail-fast: false
14 |       matrix:
15 |         rust: [nightly, beta, stable]
16 |     steps:
17 |       - uses: actions/checkout@v2
18 |       - uses: actions-rs/toolchain@v1
19 |         with:
20 |           toolchain: ${{matrix.rust}}
21 |           profile: minimal
22 |           override: true
23 |       - run: cargo build --features bench
24 |         if: matrix.rust == 'nightly'
25 |       - run: cargo test --features bench
26 |         if: matrix.rust == 'nightly'
27 |       - run: cargo bench --features bench
28 |         if: matrix.rust == 'nightly'
29 |       - run: cargo build
30 |   tables:
31 |     name: Verify tables
32 |     runs-on: ubuntu-latest
33 |     steps:
34 |       - uses: actions/checkout@v2
35 |       - name: Verify regenerated files
36 |         run: ./scripts/unicode.py && rustfmt tables.rs && diff tables.rs src/tables.rs
37 | 


--------------------------------------------------------------------------------
/.gitignore:
--------------------------------------------------------------------------------
1 | target
2 | Cargo.lock
3 | scripts/tmp
4 | 


--------------------------------------------------------------------------------
/CHANGELOG.md:
--------------------------------------------------------------------------------
 1 | # Changelog
 2 | 
 3 | ## 0.2.6
 4 | 
 5 | - Update to Unicode 16.0.0.
 6 | 
 7 | ## 0.2.5
 8 | 
 9 | - Update to Unicode 15.1.0.
10 | 
11 | ## 0.2.4
12 | 
13 | - Update to Unicode 15.0.0.
14 | - Replace `const` tables with `static` tables.
15 | 
16 | ## 0.2.3
17 | 
18 | - Update to Unicode 14.0.0.
19 | 
20 | ## 0.2.2
21 | 
22 | - Add an ASCII fast-path.
23 | 
24 | ## 0.2.1
25 | 
26 | - Update to Unicode 13.0.0.
27 | - Speed up lookup.
28 | 
29 | ## 0.2.0
30 | 
31 | - Update to Unicode 12.1.0.
32 | 
33 | ## 0.1.0
34 | 
35 | - Initial release.
36 | 


--------------------------------------------------------------------------------
/COPYRIGHT:
--------------------------------------------------------------------------------
1 | Licensed under the Apache License, Version 2.0
2 | <LICENSE-APACHE or
3 | http://www.apache.org/licenses/LICENSE-2.0> or the MIT
4 | license <LICENSE-MIT or http://opensource.org/licenses/MIT>,
5 | at your option. All files in the project carrying such
6 | notice may not be copied, modified, or distributed except
7 | according to those terms.
8 | 


--------------------------------------------------------------------------------
/Cargo.toml:
--------------------------------------------------------------------------------
 1 | [package]
 2 | 
 3 | name = "unicode-xid"
 4 | version = "0.2.6"
 5 | authors = ["erick.tryzelaar <erick.tryzelaar@gmail.com>",
 6 |            "kwantam <kwantam@gmail.com>",
 7 |            "Manish Goregaokar <manishsmail@gmail.com>"
 8 |            ]
 9 | 
10 | homepage = "https://github.com/unicode-rs/unicode-xid"
11 | repository = "https://github.com/unicode-rs/unicode-xid"
12 | documentation = "https://unicode-rs.github.io/unicode-xid"
13 | license = "MIT OR Apache-2.0"
14 | keywords = ["text", "unicode", "xid"]
15 | readme = "README.md"
16 | description = """
17 | Determine whether characters have the XID_Start
18 | or XID_Continue properties according to
19 | Unicode Standard Annex #31.
20 | """
21 | exclude = ["/.github/**", "/scripts/*"]
22 | rust-version = "1.17"
23 | 
24 | [features]
25 | default = []
26 | no_std = []
27 | bench = []
28 | 
29 | [dev-dependencies]
30 | criterion = "0.3"
31 | 
32 | [[bench]]
33 | name = "xid"
34 | harness = false
35 | 


--------------------------------------------------------------------------------
/LICENSE-APACHE:
--------------------------------------------------------------------------------
  1 |                               Apache License
  2 |                         Version 2.0, January 2004
  3 |                      http://www.apache.org/licenses/
  4 | 
  5 | TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
  6 | 
  7 | 1. Definitions.
  8 | 
  9 |    "License" shall mean the terms and conditions for use, reproduction,
 10 |    and distribution as defined by Sections 1 through 9 of this document.
 11 | 
 12 |    "Licensor" shall mean the copyright owner or entity authorized by
 13 |    the copyright owner that is granting the License.
 14 | 
 15 |    "Legal Entity" shall mean the union of the acting entity and all
 16 |    other entities that control, are controlled by, or are under common
 17 |    control with that entity. For the purposes of this definition,
 18 |    "control" means (i) the power, direct or indirect, to cause the
 19 |    direction or management of such entity, whether by contract or
 20 |    otherwise, or (ii) ownership of fifty percent (50%) or more of the
 21 |    outstanding shares, or (iii) beneficial ownership of such entity.
 22 | 
 23 |    "You" (or "Your") shall mean an individual or Legal Entity
 24 |    exercising permissions granted by this License.
 25 | 
 26 |    "Source" form shall mean the preferred form for making modifications,
 27 |    including but not limited to software source code, documentation
 28 |    source, and configuration files.
 29 | 
 30 |    "Object" form shall mean any form resulting from mechanical
 31 |    transformation or translation of a Source form, including but
 32 |    not limited to compiled object code, generated documentation,
 33 |    and conversions to other media types.
 34 | 
 35 |    "Work" shall mean the work of authorship, whether in Source or
 36 |    Object form, made available under the License, as indicated by a
 37 |    copyright notice that is included in or attached to the work
 38 |    (an example is provided in the Appendix below).
 39 | 
 40 |    "Derivative Works" shall mean any work, whether in Source or Object
 41 |    form, that is based on (or derived from) the Work and for which the
 42 |    editorial revisions, annotations, elaborations, or other modifications
 43 |    represent, as a whole, an original work of authorship. For the purposes
 44 |    of this License, Derivative Works shall not include works that remain
 45 |    separable from, or merely link (or bind by name) to the interfaces of,
 46 |    the Work and Derivative Works thereof.
 47 | 
 48 |    "Contribution" shall mean any work of authorship, including
 49 |    the original version of the Work and any modifications or additions
 50 |    to that Work or Derivative Works thereof, that is intentionally
 51 |    submitted to Licensor for inclusion in the Work by the copyright owner
 52 |    or by an individual or Legal Entity authorized to submit on behalf of
 53 |    the copyright owner. For the purposes of this definition, "submitted"
 54 |    means any form of electronic, verbal, or written communication sent
 55 |    to the Licensor or its representatives, including but not limited to
 56 |    communication on electronic mailing lists, source code control systems,
 57 |    and issue tracking systems that are managed by, or on behalf of, the
 58 |    Licensor for the purpose of discussing and improving the Work, but
 59 |    excluding communication that is conspicuously marked or otherwise
 60 |    designated in writing by the copyright owner as "Not a Contribution."
 61 | 
 62 |    "Contributor" shall mean Licensor and any individual or Legal Entity
 63 |    on behalf of whom a Contribution has been received by Licensor and
 64 |    subsequently incorporated within the Work.
 65 | 
 66 | 2. Grant of Copyright License. Subject to the terms and conditions of
 67 |    this License, each Contributor hereby grants to You a perpetual,
 68 |    worldwide, non-exclusive, no-charge, royalty-free, irrevocable
 69 |    copyright license to reproduce, prepare Derivative Works of,
 70 |    publicly display, publicly perform, sublicense, and distribute the
 71 |    Work and such Derivative Works in Source or Object form.
 72 | 
 73 | 3. Grant of Patent License. Subject to the terms and conditions of
 74 |    this License, each Contributor hereby grants to You a perpetual,
 75 |    worldwide, non-exclusive, no-charge, royalty-free, irrevocable
 76 |    (except as stated in this section) patent license to make, have made,
 77 |    use, offer to sell, sell, import, and otherwise transfer the Work,
 78 |    where such license applies only to those patent claims licensable
 79 |    by such Contributor that are necessarily infringed by their
 80 |    Contribution(s) alone or by combination of their Contribution(s)
 81 |    with the Work to which such Contribution(s) was submitted. If You
 82 |    institute patent litigation against any entity (including a
 83 |    cross-claim or counterclaim in a lawsuit) alleging that the Work
 84 |    or a Contribution incorporated within the Work constitutes direct
 85 |    or contributory patent infringement, then any patent licenses
 86 |    granted to You under this License for that Work shall terminate
 87 |    as of the date such litigation is filed.
 88 | 
 89 | 4. Redistribution. You may reproduce and distribute copies of the
 90 |    Work or Derivative Works thereof in any medium, with or without
 91 |    modifications, and in Source or Object form, provided that You
 92 |    meet the following conditions:
 93 | 
 94 |    (a) You must give any other recipients of the Work or
 95 |        Derivative Works a copy of this License; and
 96 | 
 97 |    (b) You must cause any modified files to carry prominent notices
 98 |        stating that You changed the files; and
 99 | 
100 |    (c) You must retain, in the Source form of any Derivative Works
101 |        that You distribute, all copyright, patent, trademark, and
102 |        attribution notices from the Source form of the Work,
103 |        excluding those notices that do not pertain to any part of
104 |        the Derivative Works; and
105 | 
106 |    (d) If the Work includes a "NOTICE" text file as part of its
107 |        distribution, then any Derivative Works that You distribute must
108 |        include a readable copy of the attribution notices contained
109 |        within such NOTICE file, excluding those notices that do not
110 |        pertain to any part of the Derivative Works, in at least one
111 |        of the following places: within a NOTICE text file distributed
112 |        as part of the Derivative Works; within the Source form or
113 |        documentation, if provided along with the Derivative Works; or,
114 |        within a display generated by the Derivative Works, if and
115 |        wherever such third-party notices normally appear. The contents
116 |        of the NOTICE file are for informational purposes only and
117 |        do not modify the License. You may add Your own attribution
118 |        notices within Derivative Works that You distribute, alongside
119 |        or as an addendum to the NOTICE text from the Work, provided
120 |        that such additional attribution notices cannot be construed
121 |        as modifying the License.
122 | 
123 |    You may add Your own copyright statement to Your modifications and
124 |    may provide additional or different license terms and conditions
125 |    for use, reproduction, or distribution of Your modifications, or
126 |    for any such Derivative Works as a whole, provided Your use,
127 |    reproduction, and distribution of the Work otherwise complies with
128 |    the conditions stated in this License.
129 | 
130 | 5. Submission of Contributions. Unless You explicitly state otherwise,
131 |    any Contribution intentionally submitted for inclusion in the Work
132 |    by You to the Licensor shall be under the terms and conditions of
133 |    this License, without any additional terms or conditions.
134 |    Notwithstanding the above, nothing herein shall supersede or modify
135 |    the terms of any separate license agreement you may have executed
136 |    with Licensor regarding such Contributions.
137 | 
138 | 6. Trademarks. This License does not grant permission to use the trade
139 |    names, trademarks, service marks, or product names of the Licensor,
140 |    except as required for reasonable and customary use in describing the
141 |    origin of the Work and reproducing the content of the NOTICE file.
142 | 
143 | 7. Disclaimer of Warranty. Unless required by applicable law or
144 |    agreed to in writing, Licensor provides the Work (and each
145 |    Contributor provides its Contributions) on an "AS IS" BASIS,
146 |    WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
147 |    implied, including, without limitation, any warranties or conditions
148 |    of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
149 |    PARTICULAR PURPOSE. You are solely responsible for determining the
150 |    appropriateness of using or redistributing the Work and assume any
151 |    risks associated with Your exercise of permissions under this License.
152 | 
153 | 8. Limitation of Liability. In no event and under no legal theory,
154 |    whether in tort (including negligence), contract, or otherwise,
155 |    unless required by applicable law (such as deliberate and grossly
156 |    negligent acts) or agreed to in writing, shall any Contributor be
157 |    liable to You for damages, including any direct, indirect, special,
158 |    incidental, or consequential damages of any character arising as a
159 |    result of this License or out of the use or inability to use the
160 |    Work (including but not limited to damages for loss of goodwill,
161 |    work stoppage, computer failure or malfunction, or any and all
162 |    other commercial damages or losses), even if such Contributor
163 |    has been advised of the possibility of such damages.
164 | 
165 | 9. Accepting Warranty or Additional Liability. While redistributing
166 |    the Work or Derivative Works thereof, You may choose to offer,
167 |    and charge a fee for, acceptance of support, warranty, indemnity,
168 |    or other liability obligations and/or rights consistent with this
169 |    License. However, in accepting such obligations, You may act only
170 |    on Your own behalf and on Your sole responsibility, not on behalf
171 |    of any other Contributor, and only if You agree to indemnify,
172 |    defend, and hold each Contributor harmless for any liability
173 |    incurred by, or claims asserted against, such Contributor by reason
174 |    of your accepting any such warranty or additional liability.
175 | 
176 | END OF TERMS AND CONDITIONS
177 | 
178 | APPENDIX: How to apply the Apache License to your work.
179 | 
180 |    To apply the Apache License to your work, attach the following
181 |    boilerplate notice, with the fields enclosed by brackets "[]"
182 |    replaced with your own identifying information. (Don't include
183 |    the brackets!)  The text should be enclosed in the appropriate
184 |    comment syntax for the file format. We also recommend that a
185 |    file or class name and description of purpose be included on the
186 |    same "printed page" as the copyright notice for easier
187 |    identification within third-party archives.
188 | 
189 | Copyright [yyyy] [name of copyright owner]
190 | 
191 | Licensed under the Apache License, Version 2.0 (the "License");
192 | you may not use this file except in compliance with the License.
193 | You may obtain a copy of the License at
194 | 
195 | 	http://www.apache.org/licenses/LICENSE-2.0
196 | 
197 | Unless required by applicable law or agreed to in writing, software
198 | distributed under the License is distributed on an "AS IS" BASIS,
199 | WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
200 | See the License for the specific language governing permissions and
201 | limitations under the License.
202 | 


--------------------------------------------------------------------------------
/LICENSE-MIT:
--------------------------------------------------------------------------------
 1 | Copyright (c) 2015 The Rust Project Developers
 2 | 
 3 | Permission is hereby granted, free of charge, to any
 4 | person obtaining a copy of this software and associated
 5 | documentation files (the "Software"), to deal in the
 6 | Software without restriction, including without
 7 | limitation the rights to use, copy, modify, merge,
 8 | publish, distribute, sublicense, and/or sell copies of
 9 | the Software, and to permit persons to whom the Software
10 | is furnished to do so, subject to the following
11 | conditions:
12 | 
13 | The above copyright notice and this permission notice
14 | shall be included in all copies or substantial portions
15 | of the Software.
16 | 
17 | THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF
18 | ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED
19 | TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A
20 | PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT
21 | SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
22 | CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
23 | OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR
24 | IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
25 | DEALINGS IN THE SOFTWARE.
26 | 


--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
 1 | # unicode-xid
 2 | 
 3 | Determine if a `char` is a valid identifier for a parser and/or lexer according to
 4 | [Unicode Standard Annex #31](https://www.unicode.org/reports/tr31/) rules.
 5 | 
 6 | [![Build Status](https://img.shields.io/github/actions/workflow/status/unicode-rs/unicode-xid/ci.yml?branch=master)](https://github.com/unicode-rs/unicode-xid/actions?query=branch%3Amaster)
 7 | 
 8 | [Documentation](https://unicode-rs.github.io/unicode-xid/unicode_xid/index.html)
 9 | 
10 | ```rust
11 | extern crate unicode_xid;
12 | 
13 | use unicode_xid::UnicodeXID;
14 | 
15 | fn main() {
16 |     let ch = 'a';
17 |     println!("Is {} a valid start of an identifier? {}", ch, UnicodeXID::is_xid_start(ch));
18 | }
19 | ```
20 | 
21 | # features
22 | 
23 | unicode-xid supports a `no_std` feature. This eliminates dependence
24 | on std, and instead uses equivalent functions from core.
25 | 


--------------------------------------------------------------------------------
/benches/xid.rs:
--------------------------------------------------------------------------------
 1 | extern crate criterion;
 2 | extern crate unicode_xid;
 3 | 
 4 | use criterion::{criterion_group, criterion_main, BenchmarkId, Criterion, Throughput};
 5 | use unicode_xid::UnicodeXID;
 6 | 
 7 | fn bench_unicode_xid(c: &mut Criterion) {
 8 |     let unicode_chars = chars(1..0x3000);
 9 |     let ascii_chars = chars(1..0x80);
10 | 
11 |     let mut group = c.benchmark_group("UnicodeXID");
12 |     group.throughput(Throughput::Bytes(unicode_chars.len() as u64));
13 |     group.bench_with_input(
14 |         BenchmarkId::new("is_xid_start", "unicode"),
15 |         &unicode_chars,
16 |         |b, chars| b.iter(|| chars.iter().copied().map(UnicodeXID::is_xid_start).last()),
17 |     );
18 |     group.throughput(Throughput::Bytes(ascii_chars.len() as u64));
19 |     group.bench_with_input(
20 |         BenchmarkId::new("is_xid_start", "ascii"),
21 |         &ascii_chars,
22 |         |b, chars| b.iter(|| chars.iter().copied().map(UnicodeXID::is_xid_start).last()),
23 |     );
24 |     group.throughput(Throughput::Bytes(unicode_chars.len() as u64));
25 |     group.bench_with_input(
26 |         BenchmarkId::new("is_xid_continue", "unicode"),
27 |         &unicode_chars,
28 |         |b, chars| {
29 |             b.iter(|| {
30 |                 chars
31 |                     .iter()
32 |                     .copied()
33 |                     .map(UnicodeXID::is_xid_continue)
34 |                     .last()
35 |             })
36 |         },
37 |     );
38 |     group.throughput(Throughput::Bytes(ascii_chars.len() as u64));
39 |     group.bench_with_input(
40 |         BenchmarkId::new("is_xid_continue", "ascii"),
41 |         &ascii_chars,
42 |         |b, chars| {
43 |             b.iter(|| {
44 |                 chars
45 |                     .iter()
46 |                     .copied()
47 |                     .map(UnicodeXID::is_xid_continue)
48 |                     .last()
49 |             })
50 |         },
51 |     );
52 |     group.finish();
53 | }
54 | 
55 | fn chars(range: std::ops::Range<u32>) -> Vec<char> {
56 |     range.filter_map(|i| std::char::from_u32(i)).collect()
57 | }
58 | 
59 | criterion_group!(benches, bench_unicode_xid);
60 | criterion_main!(benches);
61 | 


--------------------------------------------------------------------------------
/scripts/unicode.py:
--------------------------------------------------------------------------------
  1 | #!/usr/bin/env python
  2 | #
  3 | # Copyright 2011-2015 The Rust Project Developers. See the COPYRIGHT
  4 | # file at the top-level directory of this distribution and at
  5 | # http://rust-lang.org/COPYRIGHT.
  6 | #
  7 | # Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or
  8 | # http://www.apache.org/licenses/LICENSE-2.0> or the MIT license
  9 | # <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your
 10 | # option. This file may not be copied, modified, or distributed
 11 | # except according to those terms.
 12 | 
 13 | # This script uses the following Unicode tables:
 14 | # - DerivedCoreProperties.txt
 15 | # - ReadMe.txt
 16 | #
 17 | # Since this should not require frequent updates, we just store this
 18 | # out-of-line and check the unicode.rs file into git.
 19 | 
 20 | import fileinput, re, os, sys
 21 | 
 22 | preamble = '''// Copyright 2012-2025 The Rust Project Developers. See the COPYRIGHT
 23 | // file at the top-level directory of this distribution and at
 24 | // https://www.rust-lang.org/policies/licenses.
 25 | //
 26 | // Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or
 27 | // https://www.apache.org/licenses/LICENSE-2.0> or the MIT license
 28 | // <LICENSE-MIT or https://opensource.org/licenses/MIT>, at your
 29 | // option. This file may not be copied, modified, or distributed
 30 | // except according to those terms.
 31 | 
 32 | // NOTE: The following code was generated by "scripts/unicode.py", do not edit directly
 33 | 
 34 | #![allow(missing_docs, non_upper_case_globals, non_snake_case)]
 35 | '''
 36 | 
 37 | def fetch(f):
 38 |     if not os.path.exists(os.path.basename(f)):
 39 |         os.system("curl -O http://www.unicode.org/Public/UNIDATA/%s"
 40 |                   % f)
 41 | 
 42 |     if not os.path.exists(os.path.basename(f)):
 43 |         sys.stderr.write("cannot load %s" % f)
 44 |         exit(1)
 45 | 
 46 | def group_cat(cat):
 47 |     cat_out = []
 48 |     letters = sorted(set(cat))
 49 |     cur_start = letters.pop(0)
 50 |     cur_end = cur_start
 51 |     for letter in letters:
 52 |         assert letter > cur_end, \
 53 |             "cur_end: %s, letter: %s" % (hex(cur_end), hex(letter))
 54 |         if letter == cur_end + 1:
 55 |             cur_end = letter
 56 |         else:
 57 |             cat_out.append((cur_start, cur_end))
 58 |             cur_start = cur_end = letter
 59 |     cat_out.append((cur_start, cur_end))
 60 |     return cat_out
 61 | 
 62 | def ungroup_cat(cat):
 63 |     cat_out = []
 64 |     for (lo, hi) in cat:
 65 |         while lo <= hi:
 66 |             cat_out.append(lo)
 67 |             lo += 1
 68 |     return cat_out
 69 | 
 70 | def format_table_content(f, content, indent):
 71 |     line = " "*indent
 72 |     first = True
 73 |     for chunk in content.split(","):
 74 |         if len(line) + len(chunk) < 98:
 75 |             if first:
 76 |                 line += chunk
 77 |             else:
 78 |                 line += ", " + chunk
 79 |             first = False
 80 |         else:
 81 |             f.write(line + ",\n")
 82 |             line = " "*indent + chunk
 83 |     f.write(line)
 84 | 
 85 | def load_properties(f, interestingprops):
 86 |     fetch(f)
 87 |     props = {}
 88 |     re1 = re.compile("^ *([0-9A-F]+) *; *(\w+)")
 89 |     re2 = re.compile("^ *([0-9A-F]+)\.\.([0-9A-F]+) *; *(\w+)")
 90 | 
 91 |     for line in fileinput.input(os.path.basename(f)):
 92 |         prop = None
 93 |         d_lo = 0
 94 |         d_hi = 0
 95 |         m = re1.match(line)
 96 |         if m:
 97 |             d_lo = m.group(1)
 98 |             d_hi = m.group(1)
 99 |             prop = m.group(2)
100 |         else:
101 |             m = re2.match(line)
102 |             if m:
103 |                 d_lo = m.group(1)
104 |                 d_hi = m.group(2)
105 |                 prop = m.group(3)
106 |             else:
107 |                 continue
108 |         if interestingprops and prop not in interestingprops:
109 |             continue
110 |         d_lo = int(d_lo, 16)
111 |         d_hi = int(d_hi, 16)
112 |         if prop not in props:
113 |             props[prop] = []
114 |         props[prop].append((d_lo, d_hi))
115 | 
116 |     # optimize if possible
117 |     for prop in props:
118 |         props[prop] = group_cat(ungroup_cat(props[prop]))
119 | 
120 |     return props
121 | 
122 | def escape_char(c):
123 |     return "'\\u{%x}'" % c
124 | 
125 | def emit_bsearch_range_table(f):
126 |     f.write("""
127 | fn bsearch_range_table(c: char, r: &[(char,char)]) -> bool {
128 |     use core::cmp::Ordering::{Equal, Less, Greater};
129 | 
130 |     r.binary_search_by(|&(lo,hi)| {
131 |         // Because ASCII ranges are at the start of the tables, a search for an
132 |         // ASCII char will involve more `Greater` results (i.e. the `(lo,hi)`
133 |         // table entry is greater than `c`) than `Less` results. And given that
134 |         // ASCII chars are so common, it makes sense to favor them. Therefore,
135 |         // the `Greater` case is tested for before the `Less` case.
136 |         if lo > c { Greater }
137 |         else if hi < c { Less }
138 |         else { Equal }
139 |     }).is_ok()
140 | }\n
141 | """)
142 | 
143 | def emit_table(f, name, t_data, t_type = "&[(char, char)]", is_pub=True,
144 |         pfun=lambda x: "(%s,%s)" % (escape_char(x[0]), escape_char(x[1])), is_const=True):
145 |     pub_string = "const"
146 |     if not is_const:
147 |         pub_string = "let"
148 |     if is_pub:
149 |         pub_string = "static"
150 |     f.write("    %s %s: %s = &[\n" % (pub_string, name, t_type))
151 |     data = ""
152 |     first = True
153 |     for dat in t_data:
154 |         if not first:
155 |             data += ","
156 |         first = False
157 |         data += pfun(dat)
158 |     format_table_content(f, data, 8)
159 |     f.write("\n    ];\n\n")
160 | 
161 | def emit_property_module(f, mod, tbl, emit):
162 |     f.write("pub mod %s {\n" % mod)
163 |     for cat in sorted(emit):
164 |         emit_table(f, "%s_table" % cat, tbl[cat])
165 |         f.write("    pub fn %s(c: char) -> bool {\n" % cat)
166 |         f.write("        super::bsearch_range_table(c, %s_table)\n" % cat)
167 |         f.write("    }\n\n")
168 |     f.write("}\n\n")
169 | 
170 | if __name__ == "__main__":
171 |     r = "tables.rs"
172 |     if os.path.exists(r):
173 |         os.remove(r)
174 |     with open(r, "w") as rf:
175 |         # write the file's preamble
176 |         rf.write(preamble)
177 | 
178 |         # download and parse all the data
179 |         fetch("ReadMe.txt")
180 |         with open("ReadMe.txt") as readme:
181 |             pattern = "for Version (\d+)\.(\d+)\.(\d+) of the Unicode"
182 |             unicode_version = re.search(pattern, readme.read()).groups()
183 |         rf.write("""
184 | /// The version of [Unicode](http://www.unicode.org/)
185 | /// that this version of unicode-xid is based on.
186 | pub const UNICODE_VERSION: (u64, u64, u64) = (%s, %s, %s);
187 | """ % unicode_version)
188 |         emit_bsearch_range_table(rf)
189 | 
190 |         want_derived = ["XID_Start", "XID_Continue"]
191 |         derived = load_properties("DerivedCoreProperties.txt", want_derived)
192 |         emit_property_module(rf, "derived_property", derived, want_derived)
193 | 


--------------------------------------------------------------------------------
/src/lib.rs:
--------------------------------------------------------------------------------
 1 | // Copyright 2012-2025 The Rust Project Developers. See the COPYRIGHT
 2 | // file at the top-level directory of this distribution and at
 3 | // https://www.rust-lang.org/policies/licenses.
 4 | //
 5 | // Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or
 6 | // https://www.apache.org/licenses/LICENSE-2.0> or the MIT license
 7 | // <LICENSE-MIT or https://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 | //! Determine if a `char` is a valid identifier for a parser and/or lexer according to
12 | //! [Unicode Standard Annex #31](http://www.unicode.org/reports/tr31/) rules.
13 | //!
14 | //! ```rust
15 | //! use unicode_xid::UnicodeXID;
16 | //!
17 | //! fn main() {
18 | //!     assert_eq!(UnicodeXID::is_xid_start('a'), true); // 'a' is a valid start of an identifier
19 | //!     assert_eq!(UnicodeXID::is_xid_start('△'), false); // '△' is a NOT valid start of an identifier
20 | //! }
21 | //! ```
22 | //!
23 | //! # features
24 | //!
25 | //! unicode-xid supports a `no_std` feature. This eliminates dependence
26 | //! on std, and instead uses equivalent functions from core.
27 | //!
28 | 
29 | #![forbid(unsafe_code)]
30 | #![deny(missing_docs)]
31 | #![doc(
32 |     html_logo_url = "https://unicode-rs.github.io/unicode-rs_sm.png",
33 |     html_favicon_url = "https://unicode-rs.github.io/unicode-rs_sm.png"
34 | )]
35 | #![no_std]
36 | #![cfg_attr(feature = "bench", feature(test, unicode_internals))]
37 | 
38 | #[cfg(test)]
39 | #[macro_use]
40 | extern crate std;
41 | 
42 | #[cfg(feature = "bench")]
43 | extern crate test;
44 | 
45 | use tables::derived_property;
46 | pub use tables::UNICODE_VERSION;
47 | 
48 | mod tables;
49 | 
50 | #[cfg(test)]
51 | mod tests;
52 | 
53 | /// Methods for determining if a character is a valid identifier character.
54 | pub trait UnicodeXID {
55 |     /// Returns whether the specified character satisfies the 'XID_Start'
56 |     /// Unicode property.
57 |     ///
58 |     /// 'XID_Start' is a Unicode Derived Property specified in
59 |     /// [UAX #31](http://unicode.org/reports/tr31/#NFKC_Modifications),
60 |     /// mostly similar to ID_Start but modified for closure under NFKx.
61 |     fn is_xid_start(self) -> bool;
62 | 
63 |     /// Returns whether the specified `char` satisfies the 'XID_Continue'
64 |     /// Unicode property.
65 |     ///
66 |     /// 'XID_Continue' is a Unicode Derived Property specified in
67 |     /// [UAX #31](http://unicode.org/reports/tr31/#NFKC_Modifications),
68 |     /// mostly similar to 'ID_Continue' but modified for closure under NFKx.
69 |     fn is_xid_continue(self) -> bool;
70 | }
71 | 
72 | impl UnicodeXID for char {
73 |     #[inline]
74 |     fn is_xid_start(self) -> bool {
75 |         // Fast-path for ascii idents
76 |         ('a' <= self && self <= 'z')
77 |             || ('A' <= self && self <= 'Z')
78 |             || (self > '\x7f' && derived_property::XID_Start(self))
79 |     }
80 | 
81 |     #[inline]
82 |     fn is_xid_continue(self) -> bool {
83 |         // Fast-path for ascii idents
84 |         ('a' <= self && self <= 'z')
85 |             || ('A' <= self && self <= 'Z')
86 |             || ('0' <= self && self <= '9')
87 |             || self == '_'
88 |             || (self > '\x7f' && derived_property::XID_Continue(self))
89 |     }
90 | }
91 | 


--------------------------------------------------------------------------------
/src/tables.rs:
--------------------------------------------------------------------------------
   1 | // Copyright 2012-2025 The Rust Project Developers. See the COPYRIGHT
   2 | // file at the top-level directory of this distribution and at
   3 | // https://www.rust-lang.org/policies/licenses.
   4 | //
   5 | // Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or
   6 | // https://www.apache.org/licenses/LICENSE-2.0> or the MIT license
   7 | // <LICENSE-MIT or https://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 | // NOTE: The following code was generated by "scripts/unicode.py", do not edit directly
  12 | 
  13 | #![allow(missing_docs, non_upper_case_globals, non_snake_case)]
  14 | 
  15 | /// The version of [Unicode](http://www.unicode.org/)
  16 | /// that this version of unicode-xid is based on.
  17 | pub const UNICODE_VERSION: (u64, u64, u64) = (16, 0, 0);
  18 | 
  19 | fn bsearch_range_table(c: char, r: &[(char, char)]) -> bool {
  20 |     use core::cmp::Ordering::{Equal, Greater, Less};
  21 | 
  22 |     r.binary_search_by(|&(lo, hi)| {
  23 |         // Because ASCII ranges are at the start of the tables, a search for an
  24 |         // ASCII char will involve more `Greater` results (i.e. the `(lo,hi)`
  25 |         // table entry is greater than `c`) than `Less` results. And given that
  26 |         // ASCII chars are so common, it makes sense to favor them. Therefore,
  27 |         // the `Greater` case is tested for before the `Less` case.
  28 |         if lo > c {
  29 |             Greater
  30 |         } else if hi < c {
  31 |             Less
  32 |         } else {
  33 |             Equal
  34 |         }
  35 |     })
  36 |     .is_ok()
  37 | }
  38 | 
  39 | pub mod derived_property {
  40 |     static XID_Continue_table: &[(char, char)] = &[
  41 |         ('\u{30}', '\u{39}'),
  42 |         ('\u{41}', '\u{5a}'),
  43 |         ('\u{5f}', '\u{5f}'),
  44 |         ('\u{61}', '\u{7a}'),
  45 |         ('\u{aa}', '\u{aa}'),
  46 |         ('\u{b5}', '\u{b5}'),
  47 |         ('\u{b7}', '\u{b7}'),
  48 |         ('\u{ba}', '\u{ba}'),
  49 |         ('\u{c0}', '\u{d6}'),
  50 |         ('\u{d8}', '\u{f6}'),
  51 |         ('\u{f8}', '\u{2c1}'),
  52 |         ('\u{2c6}', '\u{2d1}'),
  53 |         ('\u{2e0}', '\u{2e4}'),
  54 |         ('\u{2ec}', '\u{2ec}'),
  55 |         ('\u{2ee}', '\u{2ee}'),
  56 |         ('\u{300}', '\u{374}'),
  57 |         ('\u{376}', '\u{377}'),
  58 |         ('\u{37b}', '\u{37d}'),
  59 |         ('\u{37f}', '\u{37f}'),
  60 |         ('\u{386}', '\u{38a}'),
  61 |         ('\u{38c}', '\u{38c}'),
  62 |         ('\u{38e}', '\u{3a1}'),
  63 |         ('\u{3a3}', '\u{3f5}'),
  64 |         ('\u{3f7}', '\u{481}'),
  65 |         ('\u{483}', '\u{487}'),
  66 |         ('\u{48a}', '\u{52f}'),
  67 |         ('\u{531}', '\u{556}'),
  68 |         ('\u{559}', '\u{559}'),
  69 |         ('\u{560}', '\u{588}'),
  70 |         ('\u{591}', '\u{5bd}'),
  71 |         ('\u{5bf}', '\u{5bf}'),
  72 |         ('\u{5c1}', '\u{5c2}'),
  73 |         ('\u{5c4}', '\u{5c5}'),
  74 |         ('\u{5c7}', '\u{5c7}'),
  75 |         ('\u{5d0}', '\u{5ea}'),
  76 |         ('\u{5ef}', '\u{5f2}'),
  77 |         ('\u{610}', '\u{61a}'),
  78 |         ('\u{620}', '\u{669}'),
  79 |         ('\u{66e}', '\u{6d3}'),
  80 |         ('\u{6d5}', '\u{6dc}'),
  81 |         ('\u{6df}', '\u{6e8}'),
  82 |         ('\u{6ea}', '\u{6fc}'),
  83 |         ('\u{6ff}', '\u{6ff}'),
  84 |         ('\u{710}', '\u{74a}'),
  85 |         ('\u{74d}', '\u{7b1}'),
  86 |         ('\u{7c0}', '\u{7f5}'),
  87 |         ('\u{7fa}', '\u{7fa}'),
  88 |         ('\u{7fd}', '\u{7fd}'),
  89 |         ('\u{800}', '\u{82d}'),
  90 |         ('\u{840}', '\u{85b}'),
  91 |         ('\u{860}', '\u{86a}'),
  92 |         ('\u{870}', '\u{887}'),
  93 |         ('\u{889}', '\u{88e}'),
  94 |         ('\u{897}', '\u{8e1}'),
  95 |         ('\u{8e3}', '\u{963}'),
  96 |         ('\u{966}', '\u{96f}'),
  97 |         ('\u{971}', '\u{983}'),
  98 |         ('\u{985}', '\u{98c}'),
  99 |         ('\u{98f}', '\u{990}'),
 100 |         ('\u{993}', '\u{9a8}'),
 101 |         ('\u{9aa}', '\u{9b0}'),
 102 |         ('\u{9b2}', '\u{9b2}'),
 103 |         ('\u{9b6}', '\u{9b9}'),
 104 |         ('\u{9bc}', '\u{9c4}'),
 105 |         ('\u{9c7}', '\u{9c8}'),
 106 |         ('\u{9cb}', '\u{9ce}'),
 107 |         ('\u{9d7}', '\u{9d7}'),
 108 |         ('\u{9dc}', '\u{9dd}'),
 109 |         ('\u{9df}', '\u{9e3}'),
 110 |         ('\u{9e6}', '\u{9f1}'),
 111 |         ('\u{9fc}', '\u{9fc}'),
 112 |         ('\u{9fe}', '\u{9fe}'),
 113 |         ('\u{a01}', '\u{a03}'),
 114 |         ('\u{a05}', '\u{a0a}'),
 115 |         ('\u{a0f}', '\u{a10}'),
 116 |         ('\u{a13}', '\u{a28}'),
 117 |         ('\u{a2a}', '\u{a30}'),
 118 |         ('\u{a32}', '\u{a33}'),
 119 |         ('\u{a35}', '\u{a36}'),
 120 |         ('\u{a38}', '\u{a39}'),
 121 |         ('\u{a3c}', '\u{a3c}'),
 122 |         ('\u{a3e}', '\u{a42}'),
 123 |         ('\u{a47}', '\u{a48}'),
 124 |         ('\u{a4b}', '\u{a4d}'),
 125 |         ('\u{a51}', '\u{a51}'),
 126 |         ('\u{a59}', '\u{a5c}'),
 127 |         ('\u{a5e}', '\u{a5e}'),
 128 |         ('\u{a66}', '\u{a75}'),
 129 |         ('\u{a81}', '\u{a83}'),
 130 |         ('\u{a85}', '\u{a8d}'),
 131 |         ('\u{a8f}', '\u{a91}'),
 132 |         ('\u{a93}', '\u{aa8}'),
 133 |         ('\u{aaa}', '\u{ab0}'),
 134 |         ('\u{ab2}', '\u{ab3}'),
 135 |         ('\u{ab5}', '\u{ab9}'),
 136 |         ('\u{abc}', '\u{ac5}'),
 137 |         ('\u{ac7}', '\u{ac9}'),
 138 |         ('\u{acb}', '\u{acd}'),
 139 |         ('\u{ad0}', '\u{ad0}'),
 140 |         ('\u{ae0}', '\u{ae3}'),
 141 |         ('\u{ae6}', '\u{aef}'),
 142 |         ('\u{af9}', '\u{aff}'),
 143 |         ('\u{b01}', '\u{b03}'),
 144 |         ('\u{b05}', '\u{b0c}'),
 145 |         ('\u{b0f}', '\u{b10}'),
 146 |         ('\u{b13}', '\u{b28}'),
 147 |         ('\u{b2a}', '\u{b30}'),
 148 |         ('\u{b32}', '\u{b33}'),
 149 |         ('\u{b35}', '\u{b39}'),
 150 |         ('\u{b3c}', '\u{b44}'),
 151 |         ('\u{b47}', '\u{b48}'),
 152 |         ('\u{b4b}', '\u{b4d}'),
 153 |         ('\u{b55}', '\u{b57}'),
 154 |         ('\u{b5c}', '\u{b5d}'),
 155 |         ('\u{b5f}', '\u{b63}'),
 156 |         ('\u{b66}', '\u{b6f}'),
 157 |         ('\u{b71}', '\u{b71}'),
 158 |         ('\u{b82}', '\u{b83}'),
 159 |         ('\u{b85}', '\u{b8a}'),
 160 |         ('\u{b8e}', '\u{b90}'),
 161 |         ('\u{b92}', '\u{b95}'),
 162 |         ('\u{b99}', '\u{b9a}'),
 163 |         ('\u{b9c}', '\u{b9c}'),
 164 |         ('\u{b9e}', '\u{b9f}'),
 165 |         ('\u{ba3}', '\u{ba4}'),
 166 |         ('\u{ba8}', '\u{baa}'),
 167 |         ('\u{bae}', '\u{bb9}'),
 168 |         ('\u{bbe}', '\u{bc2}'),
 169 |         ('\u{bc6}', '\u{bc8}'),
 170 |         ('\u{bca}', '\u{bcd}'),
 171 |         ('\u{bd0}', '\u{bd0}'),
 172 |         ('\u{bd7}', '\u{bd7}'),
 173 |         ('\u{be6}', '\u{bef}'),
 174 |         ('\u{c00}', '\u{c0c}'),
 175 |         ('\u{c0e}', '\u{c10}'),
 176 |         ('\u{c12}', '\u{c28}'),
 177 |         ('\u{c2a}', '\u{c39}'),
 178 |         ('\u{c3c}', '\u{c44}'),
 179 |         ('\u{c46}', '\u{c48}'),
 180 |         ('\u{c4a}', '\u{c4d}'),
 181 |         ('\u{c55}', '\u{c56}'),
 182 |         ('\u{c58}', '\u{c5a}'),
 183 |         ('\u{c5d}', '\u{c5d}'),
 184 |         ('\u{c60}', '\u{c63}'),
 185 |         ('\u{c66}', '\u{c6f}'),
 186 |         ('\u{c80}', '\u{c83}'),
 187 |         ('\u{c85}', '\u{c8c}'),
 188 |         ('\u{c8e}', '\u{c90}'),
 189 |         ('\u{c92}', '\u{ca8}'),
 190 |         ('\u{caa}', '\u{cb3}'),
 191 |         ('\u{cb5}', '\u{cb9}'),
 192 |         ('\u{cbc}', '\u{cc4}'),
 193 |         ('\u{cc6}', '\u{cc8}'),
 194 |         ('\u{cca}', '\u{ccd}'),
 195 |         ('\u{cd5}', '\u{cd6}'),
 196 |         ('\u{cdd}', '\u{cde}'),
 197 |         ('\u{ce0}', '\u{ce3}'),
 198 |         ('\u{ce6}', '\u{cef}'),
 199 |         ('\u{cf1}', '\u{cf3}'),
 200 |         ('\u{d00}', '\u{d0c}'),
 201 |         ('\u{d0e}', '\u{d10}'),
 202 |         ('\u{d12}', '\u{d44}'),
 203 |         ('\u{d46}', '\u{d48}'),
 204 |         ('\u{d4a}', '\u{d4e}'),
 205 |         ('\u{d54}', '\u{d57}'),
 206 |         ('\u{d5f}', '\u{d63}'),
 207 |         ('\u{d66}', '\u{d6f}'),
 208 |         ('\u{d7a}', '\u{d7f}'),
 209 |         ('\u{d81}', '\u{d83}'),
 210 |         ('\u{d85}', '\u{d96}'),
 211 |         ('\u{d9a}', '\u{db1}'),
 212 |         ('\u{db3}', '\u{dbb}'),
 213 |         ('\u{dbd}', '\u{dbd}'),
 214 |         ('\u{dc0}', '\u{dc6}'),
 215 |         ('\u{dca}', '\u{dca}'),
 216 |         ('\u{dcf}', '\u{dd4}'),
 217 |         ('\u{dd6}', '\u{dd6}'),
 218 |         ('\u{dd8}', '\u{ddf}'),
 219 |         ('\u{de6}', '\u{def}'),
 220 |         ('\u{df2}', '\u{df3}'),
 221 |         ('\u{e01}', '\u{e3a}'),
 222 |         ('\u{e40}', '\u{e4e}'),
 223 |         ('\u{e50}', '\u{e59}'),
 224 |         ('\u{e81}', '\u{e82}'),
 225 |         ('\u{e84}', '\u{e84}'),
 226 |         ('\u{e86}', '\u{e8a}'),
 227 |         ('\u{e8c}', '\u{ea3}'),
 228 |         ('\u{ea5}', '\u{ea5}'),
 229 |         ('\u{ea7}', '\u{ebd}'),
 230 |         ('\u{ec0}', '\u{ec4}'),
 231 |         ('\u{ec6}', '\u{ec6}'),
 232 |         ('\u{ec8}', '\u{ece}'),
 233 |         ('\u{ed0}', '\u{ed9}'),
 234 |         ('\u{edc}', '\u{edf}'),
 235 |         ('\u{f00}', '\u{f00}'),
 236 |         ('\u{f18}', '\u{f19}'),
 237 |         ('\u{f20}', '\u{f29}'),
 238 |         ('\u{f35}', '\u{f35}'),
 239 |         ('\u{f37}', '\u{f37}'),
 240 |         ('\u{f39}', '\u{f39}'),
 241 |         ('\u{f3e}', '\u{f47}'),
 242 |         ('\u{f49}', '\u{f6c}'),
 243 |         ('\u{f71}', '\u{f84}'),
 244 |         ('\u{f86}', '\u{f97}'),
 245 |         ('\u{f99}', '\u{fbc}'),
 246 |         ('\u{fc6}', '\u{fc6}'),
 247 |         ('\u{1000}', '\u{1049}'),
 248 |         ('\u{1050}', '\u{109d}'),
 249 |         ('\u{10a0}', '\u{10c5}'),
 250 |         ('\u{10c7}', '\u{10c7}'),
 251 |         ('\u{10cd}', '\u{10cd}'),
 252 |         ('\u{10d0}', '\u{10fa}'),
 253 |         ('\u{10fc}', '\u{1248}'),
 254 |         ('\u{124a}', '\u{124d}'),
 255 |         ('\u{1250}', '\u{1256}'),
 256 |         ('\u{1258}', '\u{1258}'),
 257 |         ('\u{125a}', '\u{125d}'),
 258 |         ('\u{1260}', '\u{1288}'),
 259 |         ('\u{128a}', '\u{128d}'),
 260 |         ('\u{1290}', '\u{12b0}'),
 261 |         ('\u{12b2}', '\u{12b5}'),
 262 |         ('\u{12b8}', '\u{12be}'),
 263 |         ('\u{12c0}', '\u{12c0}'),
 264 |         ('\u{12c2}', '\u{12c5}'),
 265 |         ('\u{12c8}', '\u{12d6}'),
 266 |         ('\u{12d8}', '\u{1310}'),
 267 |         ('\u{1312}', '\u{1315}'),
 268 |         ('\u{1318}', '\u{135a}'),
 269 |         ('\u{135d}', '\u{135f}'),
 270 |         ('\u{1369}', '\u{1371}'),
 271 |         ('\u{1380}', '\u{138f}'),
 272 |         ('\u{13a0}', '\u{13f5}'),
 273 |         ('\u{13f8}', '\u{13fd}'),
 274 |         ('\u{1401}', '\u{166c}'),
 275 |         ('\u{166f}', '\u{167f}'),
 276 |         ('\u{1681}', '\u{169a}'),
 277 |         ('\u{16a0}', '\u{16ea}'),
 278 |         ('\u{16ee}', '\u{16f8}'),
 279 |         ('\u{1700}', '\u{1715}'),
 280 |         ('\u{171f}', '\u{1734}'),
 281 |         ('\u{1740}', '\u{1753}'),
 282 |         ('\u{1760}', '\u{176c}'),
 283 |         ('\u{176e}', '\u{1770}'),
 284 |         ('\u{1772}', '\u{1773}'),
 285 |         ('\u{1780}', '\u{17d3}'),
 286 |         ('\u{17d7}', '\u{17d7}'),
 287 |         ('\u{17dc}', '\u{17dd}'),
 288 |         ('\u{17e0}', '\u{17e9}'),
 289 |         ('\u{180b}', '\u{180d}'),
 290 |         ('\u{180f}', '\u{1819}'),
 291 |         ('\u{1820}', '\u{1878}'),
 292 |         ('\u{1880}', '\u{18aa}'),
 293 |         ('\u{18b0}', '\u{18f5}'),
 294 |         ('\u{1900}', '\u{191e}'),
 295 |         ('\u{1920}', '\u{192b}'),
 296 |         ('\u{1930}', '\u{193b}'),
 297 |         ('\u{1946}', '\u{196d}'),
 298 |         ('\u{1970}', '\u{1974}'),
 299 |         ('\u{1980}', '\u{19ab}'),
 300 |         ('\u{19b0}', '\u{19c9}'),
 301 |         ('\u{19d0}', '\u{19da}'),
 302 |         ('\u{1a00}', '\u{1a1b}'),
 303 |         ('\u{1a20}', '\u{1a5e}'),
 304 |         ('\u{1a60}', '\u{1a7c}'),
 305 |         ('\u{1a7f}', '\u{1a89}'),
 306 |         ('\u{1a90}', '\u{1a99}'),
 307 |         ('\u{1aa7}', '\u{1aa7}'),
 308 |         ('\u{1ab0}', '\u{1abd}'),
 309 |         ('\u{1abf}', '\u{1ace}'),
 310 |         ('\u{1b00}', '\u{1b4c}'),
 311 |         ('\u{1b50}', '\u{1b59}'),
 312 |         ('\u{1b6b}', '\u{1b73}'),
 313 |         ('\u{1b80}', '\u{1bf3}'),
 314 |         ('\u{1c00}', '\u{1c37}'),
 315 |         ('\u{1c40}', '\u{1c49}'),
 316 |         ('\u{1c4d}', '\u{1c7d}'),
 317 |         ('\u{1c80}', '\u{1c8a}'),
 318 |         ('\u{1c90}', '\u{1cba}'),
 319 |         ('\u{1cbd}', '\u{1cbf}'),
 320 |         ('\u{1cd0}', '\u{1cd2}'),
 321 |         ('\u{1cd4}', '\u{1cfa}'),
 322 |         ('\u{1d00}', '\u{1f15}'),
 323 |         ('\u{1f18}', '\u{1f1d}'),
 324 |         ('\u{1f20}', '\u{1f45}'),
 325 |         ('\u{1f48}', '\u{1f4d}'),
 326 |         ('\u{1f50}', '\u{1f57}'),
 327 |         ('\u{1f59}', '\u{1f59}'),
 328 |         ('\u{1f5b}', '\u{1f5b}'),
 329 |         ('\u{1f5d}', '\u{1f5d}'),
 330 |         ('\u{1f5f}', '\u{1f7d}'),
 331 |         ('\u{1f80}', '\u{1fb4}'),
 332 |         ('\u{1fb6}', '\u{1fbc}'),
 333 |         ('\u{1fbe}', '\u{1fbe}'),
 334 |         ('\u{1fc2}', '\u{1fc4}'),
 335 |         ('\u{1fc6}', '\u{1fcc}'),
 336 |         ('\u{1fd0}', '\u{1fd3}'),
 337 |         ('\u{1fd6}', '\u{1fdb}'),
 338 |         ('\u{1fe0}', '\u{1fec}'),
 339 |         ('\u{1ff2}', '\u{1ff4}'),
 340 |         ('\u{1ff6}', '\u{1ffc}'),
 341 |         ('\u{200c}', '\u{200d}'),
 342 |         ('\u{203f}', '\u{2040}'),
 343 |         ('\u{2054}', '\u{2054}'),
 344 |         ('\u{2071}', '\u{2071}'),
 345 |         ('\u{207f}', '\u{207f}'),
 346 |         ('\u{2090}', '\u{209c}'),
 347 |         ('\u{20d0}', '\u{20dc}'),
 348 |         ('\u{20e1}', '\u{20e1}'),
 349 |         ('\u{20e5}', '\u{20f0}'),
 350 |         ('\u{2102}', '\u{2102}'),
 351 |         ('\u{2107}', '\u{2107}'),
 352 |         ('\u{210a}', '\u{2113}'),
 353 |         ('\u{2115}', '\u{2115}'),
 354 |         ('\u{2118}', '\u{211d}'),
 355 |         ('\u{2124}', '\u{2124}'),
 356 |         ('\u{2126}', '\u{2126}'),
 357 |         ('\u{2128}', '\u{2128}'),
 358 |         ('\u{212a}', '\u{2139}'),
 359 |         ('\u{213c}', '\u{213f}'),
 360 |         ('\u{2145}', '\u{2149}'),
 361 |         ('\u{214e}', '\u{214e}'),
 362 |         ('\u{2160}', '\u{2188}'),
 363 |         ('\u{2c00}', '\u{2ce4}'),
 364 |         ('\u{2ceb}', '\u{2cf3}'),
 365 |         ('\u{2d00}', '\u{2d25}'),
 366 |         ('\u{2d27}', '\u{2d27}'),
 367 |         ('\u{2d2d}', '\u{2d2d}'),
 368 |         ('\u{2d30}', '\u{2d67}'),
 369 |         ('\u{2d6f}', '\u{2d6f}'),
 370 |         ('\u{2d7f}', '\u{2d96}'),
 371 |         ('\u{2da0}', '\u{2da6}'),
 372 |         ('\u{2da8}', '\u{2dae}'),
 373 |         ('\u{2db0}', '\u{2db6}'),
 374 |         ('\u{2db8}', '\u{2dbe}'),
 375 |         ('\u{2dc0}', '\u{2dc6}'),
 376 |         ('\u{2dc8}', '\u{2dce}'),
 377 |         ('\u{2dd0}', '\u{2dd6}'),
 378 |         ('\u{2dd8}', '\u{2dde}'),
 379 |         ('\u{2de0}', '\u{2dff}'),
 380 |         ('\u{3005}', '\u{3007}'),
 381 |         ('\u{3021}', '\u{302f}'),
 382 |         ('\u{3031}', '\u{3035}'),
 383 |         ('\u{3038}', '\u{303c}'),
 384 |         ('\u{3041}', '\u{3096}'),
 385 |         ('\u{3099}', '\u{309a}'),
 386 |         ('\u{309d}', '\u{309f}'),
 387 |         ('\u{30a1}', '\u{30ff}'),
 388 |         ('\u{3105}', '\u{312f}'),
 389 |         ('\u{3131}', '\u{318e}'),
 390 |         ('\u{31a0}', '\u{31bf}'),
 391 |         ('\u{31f0}', '\u{31ff}'),
 392 |         ('\u{3400}', '\u{4dbf}'),
 393 |         ('\u{4e00}', '\u{a48c}'),
 394 |         ('\u{a4d0}', '\u{a4fd}'),
 395 |         ('\u{a500}', '\u{a60c}'),
 396 |         ('\u{a610}', '\u{a62b}'),
 397 |         ('\u{a640}', '\u{a66f}'),
 398 |         ('\u{a674}', '\u{a67d}'),
 399 |         ('\u{a67f}', '\u{a6f1}'),
 400 |         ('\u{a717}', '\u{a71f}'),
 401 |         ('\u{a722}', '\u{a788}'),
 402 |         ('\u{a78b}', '\u{a7cd}'),
 403 |         ('\u{a7d0}', '\u{a7d1}'),
 404 |         ('\u{a7d3}', '\u{a7d3}'),
 405 |         ('\u{a7d5}', '\u{a7dc}'),
 406 |         ('\u{a7f2}', '\u{a827}'),
 407 |         ('\u{a82c}', '\u{a82c}'),
 408 |         ('\u{a840}', '\u{a873}'),
 409 |         ('\u{a880}', '\u{a8c5}'),
 410 |         ('\u{a8d0}', '\u{a8d9}'),
 411 |         ('\u{a8e0}', '\u{a8f7}'),
 412 |         ('\u{a8fb}', '\u{a8fb}'),
 413 |         ('\u{a8fd}', '\u{a92d}'),
 414 |         ('\u{a930}', '\u{a953}'),
 415 |         ('\u{a960}', '\u{a97c}'),
 416 |         ('\u{a980}', '\u{a9c0}'),
 417 |         ('\u{a9cf}', '\u{a9d9}'),
 418 |         ('\u{a9e0}', '\u{a9fe}'),
 419 |         ('\u{aa00}', '\u{aa36}'),
 420 |         ('\u{aa40}', '\u{aa4d}'),
 421 |         ('\u{aa50}', '\u{aa59}'),
 422 |         ('\u{aa60}', '\u{aa76}'),
 423 |         ('\u{aa7a}', '\u{aac2}'),
 424 |         ('\u{aadb}', '\u{aadd}'),
 425 |         ('\u{aae0}', '\u{aaef}'),
 426 |         ('\u{aaf2}', '\u{aaf6}'),
 427 |         ('\u{ab01}', '\u{ab06}'),
 428 |         ('\u{ab09}', '\u{ab0e}'),
 429 |         ('\u{ab11}', '\u{ab16}'),
 430 |         ('\u{ab20}', '\u{ab26}'),
 431 |         ('\u{ab28}', '\u{ab2e}'),
 432 |         ('\u{ab30}', '\u{ab5a}'),
 433 |         ('\u{ab5c}', '\u{ab69}'),
 434 |         ('\u{ab70}', '\u{abea}'),
 435 |         ('\u{abec}', '\u{abed}'),
 436 |         ('\u{abf0}', '\u{abf9}'),
 437 |         ('\u{ac00}', '\u{d7a3}'),
 438 |         ('\u{d7b0}', '\u{d7c6}'),
 439 |         ('\u{d7cb}', '\u{d7fb}'),
 440 |         ('\u{f900}', '\u{fa6d}'),
 441 |         ('\u{fa70}', '\u{fad9}'),
 442 |         ('\u{fb00}', '\u{fb06}'),
 443 |         ('\u{fb13}', '\u{fb17}'),
 444 |         ('\u{fb1d}', '\u{fb28}'),
 445 |         ('\u{fb2a}', '\u{fb36}'),
 446 |         ('\u{fb38}', '\u{fb3c}'),
 447 |         ('\u{fb3e}', '\u{fb3e}'),
 448 |         ('\u{fb40}', '\u{fb41}'),
 449 |         ('\u{fb43}', '\u{fb44}'),
 450 |         ('\u{fb46}', '\u{fbb1}'),
 451 |         ('\u{fbd3}', '\u{fc5d}'),
 452 |         ('\u{fc64}', '\u{fd3d}'),
 453 |         ('\u{fd50}', '\u{fd8f}'),
 454 |         ('\u{fd92}', '\u{fdc7}'),
 455 |         ('\u{fdf0}', '\u{fdf9}'),
 456 |         ('\u{fe00}', '\u{fe0f}'),
 457 |         ('\u{fe20}', '\u{fe2f}'),
 458 |         ('\u{fe33}', '\u{fe34}'),
 459 |         ('\u{fe4d}', '\u{fe4f}'),
 460 |         ('\u{fe71}', '\u{fe71}'),
 461 |         ('\u{fe73}', '\u{fe73}'),
 462 |         ('\u{fe77}', '\u{fe77}'),
 463 |         ('\u{fe79}', '\u{fe79}'),
 464 |         ('\u{fe7b}', '\u{fe7b}'),
 465 |         ('\u{fe7d}', '\u{fe7d}'),
 466 |         ('\u{fe7f}', '\u{fefc}'),
 467 |         ('\u{ff10}', '\u{ff19}'),
 468 |         ('\u{ff21}', '\u{ff3a}'),
 469 |         ('\u{ff3f}', '\u{ff3f}'),
 470 |         ('\u{ff41}', '\u{ff5a}'),
 471 |         ('\u{ff65}', '\u{ffbe}'),
 472 |         ('\u{ffc2}', '\u{ffc7}'),
 473 |         ('\u{ffca}', '\u{ffcf}'),
 474 |         ('\u{ffd2}', '\u{ffd7}'),
 475 |         ('\u{ffda}', '\u{ffdc}'),
 476 |         ('\u{10000}', '\u{1000b}'),
 477 |         ('\u{1000d}', '\u{10026}'),
 478 |         ('\u{10028}', '\u{1003a}'),
 479 |         ('\u{1003c}', '\u{1003d}'),
 480 |         ('\u{1003f}', '\u{1004d}'),
 481 |         ('\u{10050}', '\u{1005d}'),
 482 |         ('\u{10080}', '\u{100fa}'),
 483 |         ('\u{10140}', '\u{10174}'),
 484 |         ('\u{101fd}', '\u{101fd}'),
 485 |         ('\u{10280}', '\u{1029c}'),
 486 |         ('\u{102a0}', '\u{102d0}'),
 487 |         ('\u{102e0}', '\u{102e0}'),
 488 |         ('\u{10300}', '\u{1031f}'),
 489 |         ('\u{1032d}', '\u{1034a}'),
 490 |         ('\u{10350}', '\u{1037a}'),
 491 |         ('\u{10380}', '\u{1039d}'),
 492 |         ('\u{103a0}', '\u{103c3}'),
 493 |         ('\u{103c8}', '\u{103cf}'),
 494 |         ('\u{103d1}', '\u{103d5}'),
 495 |         ('\u{10400}', '\u{1049d}'),
 496 |         ('\u{104a0}', '\u{104a9}'),
 497 |         ('\u{104b0}', '\u{104d3}'),
 498 |         ('\u{104d8}', '\u{104fb}'),
 499 |         ('\u{10500}', '\u{10527}'),
 500 |         ('\u{10530}', '\u{10563}'),
 501 |         ('\u{10570}', '\u{1057a}'),
 502 |         ('\u{1057c}', '\u{1058a}'),
 503 |         ('\u{1058c}', '\u{10592}'),
 504 |         ('\u{10594}', '\u{10595}'),
 505 |         ('\u{10597}', '\u{105a1}'),
 506 |         ('\u{105a3}', '\u{105b1}'),
 507 |         ('\u{105b3}', '\u{105b9}'),
 508 |         ('\u{105bb}', '\u{105bc}'),
 509 |         ('\u{105c0}', '\u{105f3}'),
 510 |         ('\u{10600}', '\u{10736}'),
 511 |         ('\u{10740}', '\u{10755}'),
 512 |         ('\u{10760}', '\u{10767}'),
 513 |         ('\u{10780}', '\u{10785}'),
 514 |         ('\u{10787}', '\u{107b0}'),
 515 |         ('\u{107b2}', '\u{107ba}'),
 516 |         ('\u{10800}', '\u{10805}'),
 517 |         ('\u{10808}', '\u{10808}'),
 518 |         ('\u{1080a}', '\u{10835}'),
 519 |         ('\u{10837}', '\u{10838}'),
 520 |         ('\u{1083c}', '\u{1083c}'),
 521 |         ('\u{1083f}', '\u{10855}'),
 522 |         ('\u{10860}', '\u{10876}'),
 523 |         ('\u{10880}', '\u{1089e}'),
 524 |         ('\u{108e0}', '\u{108f2}'),
 525 |         ('\u{108f4}', '\u{108f5}'),
 526 |         ('\u{10900}', '\u{10915}'),
 527 |         ('\u{10920}', '\u{10939}'),
 528 |         ('\u{10980}', '\u{109b7}'),
 529 |         ('\u{109be}', '\u{109bf}'),
 530 |         ('\u{10a00}', '\u{10a03}'),
 531 |         ('\u{10a05}', '\u{10a06}'),
 532 |         ('\u{10a0c}', '\u{10a13}'),
 533 |         ('\u{10a15}', '\u{10a17}'),
 534 |         ('\u{10a19}', '\u{10a35}'),
 535 |         ('\u{10a38}', '\u{10a3a}'),
 536 |         ('\u{10a3f}', '\u{10a3f}'),
 537 |         ('\u{10a60}', '\u{10a7c}'),
 538 |         ('\u{10a80}', '\u{10a9c}'),
 539 |         ('\u{10ac0}', '\u{10ac7}'),
 540 |         ('\u{10ac9}', '\u{10ae6}'),
 541 |         ('\u{10b00}', '\u{10b35}'),
 542 |         ('\u{10b40}', '\u{10b55}'),
 543 |         ('\u{10b60}', '\u{10b72}'),
 544 |         ('\u{10b80}', '\u{10b91}'),
 545 |         ('\u{10c00}', '\u{10c48}'),
 546 |         ('\u{10c80}', '\u{10cb2}'),
 547 |         ('\u{10cc0}', '\u{10cf2}'),
 548 |         ('\u{10d00}', '\u{10d27}'),
 549 |         ('\u{10d30}', '\u{10d39}'),
 550 |         ('\u{10d40}', '\u{10d65}'),
 551 |         ('\u{10d69}', '\u{10d6d}'),
 552 |         ('\u{10d6f}', '\u{10d85}'),
 553 |         ('\u{10e80}', '\u{10ea9}'),
 554 |         ('\u{10eab}', '\u{10eac}'),
 555 |         ('\u{10eb0}', '\u{10eb1}'),
 556 |         ('\u{10ec2}', '\u{10ec4}'),
 557 |         ('\u{10efc}', '\u{10f1c}'),
 558 |         ('\u{10f27}', '\u{10f27}'),
 559 |         ('\u{10f30}', '\u{10f50}'),
 560 |         ('\u{10f70}', '\u{10f85}'),
 561 |         ('\u{10fb0}', '\u{10fc4}'),
 562 |         ('\u{10fe0}', '\u{10ff6}'),
 563 |         ('\u{11000}', '\u{11046}'),
 564 |         ('\u{11066}', '\u{11075}'),
 565 |         ('\u{1107f}', '\u{110ba}'),
 566 |         ('\u{110c2}', '\u{110c2}'),
 567 |         ('\u{110d0}', '\u{110e8}'),
 568 |         ('\u{110f0}', '\u{110f9}'),
 569 |         ('\u{11100}', '\u{11134}'),
 570 |         ('\u{11136}', '\u{1113f}'),
 571 |         ('\u{11144}', '\u{11147}'),
 572 |         ('\u{11150}', '\u{11173}'),
 573 |         ('\u{11176}', '\u{11176}'),
 574 |         ('\u{11180}', '\u{111c4}'),
 575 |         ('\u{111c9}', '\u{111cc}'),
 576 |         ('\u{111ce}', '\u{111da}'),
 577 |         ('\u{111dc}', '\u{111dc}'),
 578 |         ('\u{11200}', '\u{11211}'),
 579 |         ('\u{11213}', '\u{11237}'),
 580 |         ('\u{1123e}', '\u{11241}'),
 581 |         ('\u{11280}', '\u{11286}'),
 582 |         ('\u{11288}', '\u{11288}'),
 583 |         ('\u{1128a}', '\u{1128d}'),
 584 |         ('\u{1128f}', '\u{1129d}'),
 585 |         ('\u{1129f}', '\u{112a8}'),
 586 |         ('\u{112b0}', '\u{112ea}'),
 587 |         ('\u{112f0}', '\u{112f9}'),
 588 |         ('\u{11300}', '\u{11303}'),
 589 |         ('\u{11305}', '\u{1130c}'),
 590 |         ('\u{1130f}', '\u{11310}'),
 591 |         ('\u{11313}', '\u{11328}'),
 592 |         ('\u{1132a}', '\u{11330}'),
 593 |         ('\u{11332}', '\u{11333}'),
 594 |         ('\u{11335}', '\u{11339}'),
 595 |         ('\u{1133b}', '\u{11344}'),
 596 |         ('\u{11347}', '\u{11348}'),
 597 |         ('\u{1134b}', '\u{1134d}'),
 598 |         ('\u{11350}', '\u{11350}'),
 599 |         ('\u{11357}', '\u{11357}'),
 600 |         ('\u{1135d}', '\u{11363}'),
 601 |         ('\u{11366}', '\u{1136c}'),
 602 |         ('\u{11370}', '\u{11374}'),
 603 |         ('\u{11380}', '\u{11389}'),
 604 |         ('\u{1138b}', '\u{1138b}'),
 605 |         ('\u{1138e}', '\u{1138e}'),
 606 |         ('\u{11390}', '\u{113b5}'),
 607 |         ('\u{113b7}', '\u{113c0}'),
 608 |         ('\u{113c2}', '\u{113c2}'),
 609 |         ('\u{113c5}', '\u{113c5}'),
 610 |         ('\u{113c7}', '\u{113ca}'),
 611 |         ('\u{113cc}', '\u{113d3}'),
 612 |         ('\u{113e1}', '\u{113e2}'),
 613 |         ('\u{11400}', '\u{1144a}'),
 614 |         ('\u{11450}', '\u{11459}'),
 615 |         ('\u{1145e}', '\u{11461}'),
 616 |         ('\u{11480}', '\u{114c5}'),
 617 |         ('\u{114c7}', '\u{114c7}'),
 618 |         ('\u{114d0}', '\u{114d9}'),
 619 |         ('\u{11580}', '\u{115b5}'),
 620 |         ('\u{115b8}', '\u{115c0}'),
 621 |         ('\u{115d8}', '\u{115dd}'),
 622 |         ('\u{11600}', '\u{11640}'),
 623 |         ('\u{11644}', '\u{11644}'),
 624 |         ('\u{11650}', '\u{11659}'),
 625 |         ('\u{11680}', '\u{116b8}'),
 626 |         ('\u{116c0}', '\u{116c9}'),
 627 |         ('\u{116d0}', '\u{116e3}'),
 628 |         ('\u{11700}', '\u{1171a}'),
 629 |         ('\u{1171d}', '\u{1172b}'),
 630 |         ('\u{11730}', '\u{11739}'),
 631 |         ('\u{11740}', '\u{11746}'),
 632 |         ('\u{11800}', '\u{1183a}'),
 633 |         ('\u{118a0}', '\u{118e9}'),
 634 |         ('\u{118ff}', '\u{11906}'),
 635 |         ('\u{11909}', '\u{11909}'),
 636 |         ('\u{1190c}', '\u{11913}'),
 637 |         ('\u{11915}', '\u{11916}'),
 638 |         ('\u{11918}', '\u{11935}'),
 639 |         ('\u{11937}', '\u{11938}'),
 640 |         ('\u{1193b}', '\u{11943}'),
 641 |         ('\u{11950}', '\u{11959}'),
 642 |         ('\u{119a0}', '\u{119a7}'),
 643 |         ('\u{119aa}', '\u{119d7}'),
 644 |         ('\u{119da}', '\u{119e1}'),
 645 |         ('\u{119e3}', '\u{119e4}'),
 646 |         ('\u{11a00}', '\u{11a3e}'),
 647 |         ('\u{11a47}', '\u{11a47}'),
 648 |         ('\u{11a50}', '\u{11a99}'),
 649 |         ('\u{11a9d}', '\u{11a9d}'),
 650 |         ('\u{11ab0}', '\u{11af8}'),
 651 |         ('\u{11bc0}', '\u{11be0}'),
 652 |         ('\u{11bf0}', '\u{11bf9}'),
 653 |         ('\u{11c00}', '\u{11c08}'),
 654 |         ('\u{11c0a}', '\u{11c36}'),
 655 |         ('\u{11c38}', '\u{11c40}'),
 656 |         ('\u{11c50}', '\u{11c59}'),
 657 |         ('\u{11c72}', '\u{11c8f}'),
 658 |         ('\u{11c92}', '\u{11ca7}'),
 659 |         ('\u{11ca9}', '\u{11cb6}'),
 660 |         ('\u{11d00}', '\u{11d06}'),
 661 |         ('\u{11d08}', '\u{11d09}'),
 662 |         ('\u{11d0b}', '\u{11d36}'),
 663 |         ('\u{11d3a}', '\u{11d3a}'),
 664 |         ('\u{11d3c}', '\u{11d3d}'),
 665 |         ('\u{11d3f}', '\u{11d47}'),
 666 |         ('\u{11d50}', '\u{11d59}'),
 667 |         ('\u{11d60}', '\u{11d65}'),
 668 |         ('\u{11d67}', '\u{11d68}'),
 669 |         ('\u{11d6a}', '\u{11d8e}'),
 670 |         ('\u{11d90}', '\u{11d91}'),
 671 |         ('\u{11d93}', '\u{11d98}'),
 672 |         ('\u{11da0}', '\u{11da9}'),
 673 |         ('\u{11ee0}', '\u{11ef6}'),
 674 |         ('\u{11f00}', '\u{11f10}'),
 675 |         ('\u{11f12}', '\u{11f3a}'),
 676 |         ('\u{11f3e}', '\u{11f42}'),
 677 |         ('\u{11f50}', '\u{11f5a}'),
 678 |         ('\u{11fb0}', '\u{11fb0}'),
 679 |         ('\u{12000}', '\u{12399}'),
 680 |         ('\u{12400}', '\u{1246e}'),
 681 |         ('\u{12480}', '\u{12543}'),
 682 |         ('\u{12f90}', '\u{12ff0}'),
 683 |         ('\u{13000}', '\u{1342f}'),
 684 |         ('\u{13440}', '\u{13455}'),
 685 |         ('\u{13460}', '\u{143fa}'),
 686 |         ('\u{14400}', '\u{14646}'),
 687 |         ('\u{16100}', '\u{16139}'),
 688 |         ('\u{16800}', '\u{16a38}'),
 689 |         ('\u{16a40}', '\u{16a5e}'),
 690 |         ('\u{16a60}', '\u{16a69}'),
 691 |         ('\u{16a70}', '\u{16abe}'),
 692 |         ('\u{16ac0}', '\u{16ac9}'),
 693 |         ('\u{16ad0}', '\u{16aed}'),
 694 |         ('\u{16af0}', '\u{16af4}'),
 695 |         ('\u{16b00}', '\u{16b36}'),
 696 |         ('\u{16b40}', '\u{16b43}'),
 697 |         ('\u{16b50}', '\u{16b59}'),
 698 |         ('\u{16b63}', '\u{16b77}'),
 699 |         ('\u{16b7d}', '\u{16b8f}'),
 700 |         ('\u{16d40}', '\u{16d6c}'),
 701 |         ('\u{16d70}', '\u{16d79}'),
 702 |         ('\u{16e40}', '\u{16e7f}'),
 703 |         ('\u{16f00}', '\u{16f4a}'),
 704 |         ('\u{16f4f}', '\u{16f87}'),
 705 |         ('\u{16f8f}', '\u{16f9f}'),
 706 |         ('\u{16fe0}', '\u{16fe1}'),
 707 |         ('\u{16fe3}', '\u{16fe4}'),
 708 |         ('\u{16ff0}', '\u{16ff1}'),
 709 |         ('\u{17000}', '\u{187f7}'),
 710 |         ('\u{18800}', '\u{18cd5}'),
 711 |         ('\u{18cff}', '\u{18d08}'),
 712 |         ('\u{1aff0}', '\u{1aff3}'),
 713 |         ('\u{1aff5}', '\u{1affb}'),
 714 |         ('\u{1affd}', '\u{1affe}'),
 715 |         ('\u{1b000}', '\u{1b122}'),
 716 |         ('\u{1b132}', '\u{1b132}'),
 717 |         ('\u{1b150}', '\u{1b152}'),
 718 |         ('\u{1b155}', '\u{1b155}'),
 719 |         ('\u{1b164}', '\u{1b167}'),
 720 |         ('\u{1b170}', '\u{1b2fb}'),
 721 |         ('\u{1bc00}', '\u{1bc6a}'),
 722 |         ('\u{1bc70}', '\u{1bc7c}'),
 723 |         ('\u{1bc80}', '\u{1bc88}'),
 724 |         ('\u{1bc90}', '\u{1bc99}'),
 725 |         ('\u{1bc9d}', '\u{1bc9e}'),
 726 |         ('\u{1ccf0}', '\u{1ccf9}'),
 727 |         ('\u{1cf00}', '\u{1cf2d}'),
 728 |         ('\u{1cf30}', '\u{1cf46}'),
 729 |         ('\u{1d165}', '\u{1d169}'),
 730 |         ('\u{1d16d}', '\u{1d172}'),
 731 |         ('\u{1d17b}', '\u{1d182}'),
 732 |         ('\u{1d185}', '\u{1d18b}'),
 733 |         ('\u{1d1aa}', '\u{1d1ad}'),
 734 |         ('\u{1d242}', '\u{1d244}'),
 735 |         ('\u{1d400}', '\u{1d454}'),
 736 |         ('\u{1d456}', '\u{1d49c}'),
 737 |         ('\u{1d49e}', '\u{1d49f}'),
 738 |         ('\u{1d4a2}', '\u{1d4a2}'),
 739 |         ('\u{1d4a5}', '\u{1d4a6}'),
 740 |         ('\u{1d4a9}', '\u{1d4ac}'),
 741 |         ('\u{1d4ae}', '\u{1d4b9}'),
 742 |         ('\u{1d4bb}', '\u{1d4bb}'),
 743 |         ('\u{1d4bd}', '\u{1d4c3}'),
 744 |         ('\u{1d4c5}', '\u{1d505}'),
 745 |         ('\u{1d507}', '\u{1d50a}'),
 746 |         ('\u{1d50d}', '\u{1d514}'),
 747 |         ('\u{1d516}', '\u{1d51c}'),
 748 |         ('\u{1d51e}', '\u{1d539}'),
 749 |         ('\u{1d53b}', '\u{1d53e}'),
 750 |         ('\u{1d540}', '\u{1d544}'),
 751 |         ('\u{1d546}', '\u{1d546}'),
 752 |         ('\u{1d54a}', '\u{1d550}'),
 753 |         ('\u{1d552}', '\u{1d6a5}'),
 754 |         ('\u{1d6a8}', '\u{1d6c0}'),
 755 |         ('\u{1d6c2}', '\u{1d6da}'),
 756 |         ('\u{1d6dc}', '\u{1d6fa}'),
 757 |         ('\u{1d6fc}', '\u{1d714}'),
 758 |         ('\u{1d716}', '\u{1d734}'),
 759 |         ('\u{1d736}', '\u{1d74e}'),
 760 |         ('\u{1d750}', '\u{1d76e}'),
 761 |         ('\u{1d770}', '\u{1d788}'),
 762 |         ('\u{1d78a}', '\u{1d7a8}'),
 763 |         ('\u{1d7aa}', '\u{1d7c2}'),
 764 |         ('\u{1d7c4}', '\u{1d7cb}'),
 765 |         ('\u{1d7ce}', '\u{1d7ff}'),
 766 |         ('\u{1da00}', '\u{1da36}'),
 767 |         ('\u{1da3b}', '\u{1da6c}'),
 768 |         ('\u{1da75}', '\u{1da75}'),
 769 |         ('\u{1da84}', '\u{1da84}'),
 770 |         ('\u{1da9b}', '\u{1da9f}'),
 771 |         ('\u{1daa1}', '\u{1daaf}'),
 772 |         ('\u{1df00}', '\u{1df1e}'),
 773 |         ('\u{1df25}', '\u{1df2a}'),
 774 |         ('\u{1e000}', '\u{1e006}'),
 775 |         ('\u{1e008}', '\u{1e018}'),
 776 |         ('\u{1e01b}', '\u{1e021}'),
 777 |         ('\u{1e023}', '\u{1e024}'),
 778 |         ('\u{1e026}', '\u{1e02a}'),
 779 |         ('\u{1e030}', '\u{1e06d}'),
 780 |         ('\u{1e08f}', '\u{1e08f}'),
 781 |         ('\u{1e100}', '\u{1e12c}'),
 782 |         ('\u{1e130}', '\u{1e13d}'),
 783 |         ('\u{1e140}', '\u{1e149}'),
 784 |         ('\u{1e14e}', '\u{1e14e}'),
 785 |         ('\u{1e290}', '\u{1e2ae}'),
 786 |         ('\u{1e2c0}', '\u{1e2f9}'),
 787 |         ('\u{1e4d0}', '\u{1e4f9}'),
 788 |         ('\u{1e5d0}', '\u{1e5fa}'),
 789 |         ('\u{1e7e0}', '\u{1e7e6}'),
 790 |         ('\u{1e7e8}', '\u{1e7eb}'),
 791 |         ('\u{1e7ed}', '\u{1e7ee}'),
 792 |         ('\u{1e7f0}', '\u{1e7fe}'),
 793 |         ('\u{1e800}', '\u{1e8c4}'),
 794 |         ('\u{1e8d0}', '\u{1e8d6}'),
 795 |         ('\u{1e900}', '\u{1e94b}'),
 796 |         ('\u{1e950}', '\u{1e959}'),
 797 |         ('\u{1ee00}', '\u{1ee03}'),
 798 |         ('\u{1ee05}', '\u{1ee1f}'),
 799 |         ('\u{1ee21}', '\u{1ee22}'),
 800 |         ('\u{1ee24}', '\u{1ee24}'),
 801 |         ('\u{1ee27}', '\u{1ee27}'),
 802 |         ('\u{1ee29}', '\u{1ee32}'),
 803 |         ('\u{1ee34}', '\u{1ee37}'),
 804 |         ('\u{1ee39}', '\u{1ee39}'),
 805 |         ('\u{1ee3b}', '\u{1ee3b}'),
 806 |         ('\u{1ee42}', '\u{1ee42}'),
 807 |         ('\u{1ee47}', '\u{1ee47}'),
 808 |         ('\u{1ee49}', '\u{1ee49}'),
 809 |         ('\u{1ee4b}', '\u{1ee4b}'),
 810 |         ('\u{1ee4d}', '\u{1ee4f}'),
 811 |         ('\u{1ee51}', '\u{1ee52}'),
 812 |         ('\u{1ee54}', '\u{1ee54}'),
 813 |         ('\u{1ee57}', '\u{1ee57}'),
 814 |         ('\u{1ee59}', '\u{1ee59}'),
 815 |         ('\u{1ee5b}', '\u{1ee5b}'),
 816 |         ('\u{1ee5d}', '\u{1ee5d}'),
 817 |         ('\u{1ee5f}', '\u{1ee5f}'),
 818 |         ('\u{1ee61}', '\u{1ee62}'),
 819 |         ('\u{1ee64}', '\u{1ee64}'),
 820 |         ('\u{1ee67}', '\u{1ee6a}'),
 821 |         ('\u{1ee6c}', '\u{1ee72}'),
 822 |         ('\u{1ee74}', '\u{1ee77}'),
 823 |         ('\u{1ee79}', '\u{1ee7c}'),
 824 |         ('\u{1ee7e}', '\u{1ee7e}'),
 825 |         ('\u{1ee80}', '\u{1ee89}'),
 826 |         ('\u{1ee8b}', '\u{1ee9b}'),
 827 |         ('\u{1eea1}', '\u{1eea3}'),
 828 |         ('\u{1eea5}', '\u{1eea9}'),
 829 |         ('\u{1eeab}', '\u{1eebb}'),
 830 |         ('\u{1fbf0}', '\u{1fbf9}'),
 831 |         ('\u{20000}', '\u{2a6df}'),
 832 |         ('\u{2a700}', '\u{2b739}'),
 833 |         ('\u{2b740}', '\u{2b81d}'),
 834 |         ('\u{2b820}', '\u{2cea1}'),
 835 |         ('\u{2ceb0}', '\u{2ebe0}'),
 836 |         ('\u{2ebf0}', '\u{2ee5d}'),
 837 |         ('\u{2f800}', '\u{2fa1d}'),
 838 |         ('\u{30000}', '\u{3134a}'),
 839 |         ('\u{31350}', '\u{323af}'),
 840 |         ('\u{e0100}', '\u{e01ef}'),
 841 |     ];
 842 | 
 843 |     pub fn XID_Continue(c: char) -> bool {
 844 |         super::bsearch_range_table(c, XID_Continue_table)
 845 |     }
 846 | 
 847 |     static XID_Start_table: &[(char, char)] = &[
 848 |         ('\u{41}', '\u{5a}'),
 849 |         ('\u{61}', '\u{7a}'),
 850 |         ('\u{aa}', '\u{aa}'),
 851 |         ('\u{b5}', '\u{b5}'),
 852 |         ('\u{ba}', '\u{ba}'),
 853 |         ('\u{c0}', '\u{d6}'),
 854 |         ('\u{d8}', '\u{f6}'),
 855 |         ('\u{f8}', '\u{2c1}'),
 856 |         ('\u{2c6}', '\u{2d1}'),
 857 |         ('\u{2e0}', '\u{2e4}'),
 858 |         ('\u{2ec}', '\u{2ec}'),
 859 |         ('\u{2ee}', '\u{2ee}'),
 860 |         ('\u{370}', '\u{374}'),
 861 |         ('\u{376}', '\u{377}'),
 862 |         ('\u{37b}', '\u{37d}'),
 863 |         ('\u{37f}', '\u{37f}'),
 864 |         ('\u{386}', '\u{386}'),
 865 |         ('\u{388}', '\u{38a}'),
 866 |         ('\u{38c}', '\u{38c}'),
 867 |         ('\u{38e}', '\u{3a1}'),
 868 |         ('\u{3a3}', '\u{3f5}'),
 869 |         ('\u{3f7}', '\u{481}'),
 870 |         ('\u{48a}', '\u{52f}'),
 871 |         ('\u{531}', '\u{556}'),
 872 |         ('\u{559}', '\u{559}'),
 873 |         ('\u{560}', '\u{588}'),
 874 |         ('\u{5d0}', '\u{5ea}'),
 875 |         ('\u{5ef}', '\u{5f2}'),
 876 |         ('\u{620}', '\u{64a}'),
 877 |         ('\u{66e}', '\u{66f}'),
 878 |         ('\u{671}', '\u{6d3}'),
 879 |         ('\u{6d5}', '\u{6d5}'),
 880 |         ('\u{6e5}', '\u{6e6}'),
 881 |         ('\u{6ee}', '\u{6ef}'),
 882 |         ('\u{6fa}', '\u{6fc}'),
 883 |         ('\u{6ff}', '\u{6ff}'),
 884 |         ('\u{710}', '\u{710}'),
 885 |         ('\u{712}', '\u{72f}'),
 886 |         ('\u{74d}', '\u{7a5}'),
 887 |         ('\u{7b1}', '\u{7b1}'),
 888 |         ('\u{7ca}', '\u{7ea}'),
 889 |         ('\u{7f4}', '\u{7f5}'),
 890 |         ('\u{7fa}', '\u{7fa}'),
 891 |         ('\u{800}', '\u{815}'),
 892 |         ('\u{81a}', '\u{81a}'),
 893 |         ('\u{824}', '\u{824}'),
 894 |         ('\u{828}', '\u{828}'),
 895 |         ('\u{840}', '\u{858}'),
 896 |         ('\u{860}', '\u{86a}'),
 897 |         ('\u{870}', '\u{887}'),
 898 |         ('\u{889}', '\u{88e}'),
 899 |         ('\u{8a0}', '\u{8c9}'),
 900 |         ('\u{904}', '\u{939}'),
 901 |         ('\u{93d}', '\u{93d}'),
 902 |         ('\u{950}', '\u{950}'),
 903 |         ('\u{958}', '\u{961}'),
 904 |         ('\u{971}', '\u{980}'),
 905 |         ('\u{985}', '\u{98c}'),
 906 |         ('\u{98f}', '\u{990}'),
 907 |         ('\u{993}', '\u{9a8}'),
 908 |         ('\u{9aa}', '\u{9b0}'),
 909 |         ('\u{9b2}', '\u{9b2}'),
 910 |         ('\u{9b6}', '\u{9b9}'),
 911 |         ('\u{9bd}', '\u{9bd}'),
 912 |         ('\u{9ce}', '\u{9ce}'),
 913 |         ('\u{9dc}', '\u{9dd}'),
 914 |         ('\u{9df}', '\u{9e1}'),
 915 |         ('\u{9f0}', '\u{9f1}'),
 916 |         ('\u{9fc}', '\u{9fc}'),
 917 |         ('\u{a05}', '\u{a0a}'),
 918 |         ('\u{a0f}', '\u{a10}'),
 919 |         ('\u{a13}', '\u{a28}'),
 920 |         ('\u{a2a}', '\u{a30}'),
 921 |         ('\u{a32}', '\u{a33}'),
 922 |         ('\u{a35}', '\u{a36}'),
 923 |         ('\u{a38}', '\u{a39}'),
 924 |         ('\u{a59}', '\u{a5c}'),
 925 |         ('\u{a5e}', '\u{a5e}'),
 926 |         ('\u{a72}', '\u{a74}'),
 927 |         ('\u{a85}', '\u{a8d}'),
 928 |         ('\u{a8f}', '\u{a91}'),
 929 |         ('\u{a93}', '\u{aa8}'),
 930 |         ('\u{aaa}', '\u{ab0}'),
 931 |         ('\u{ab2}', '\u{ab3}'),
 932 |         ('\u{ab5}', '\u{ab9}'),
 933 |         ('\u{abd}', '\u{abd}'),
 934 |         ('\u{ad0}', '\u{ad0}'),
 935 |         ('\u{ae0}', '\u{ae1}'),
 936 |         ('\u{af9}', '\u{af9}'),
 937 |         ('\u{b05}', '\u{b0c}'),
 938 |         ('\u{b0f}', '\u{b10}'),
 939 |         ('\u{b13}', '\u{b28}'),
 940 |         ('\u{b2a}', '\u{b30}'),
 941 |         ('\u{b32}', '\u{b33}'),
 942 |         ('\u{b35}', '\u{b39}'),
 943 |         ('\u{b3d}', '\u{b3d}'),
 944 |         ('\u{b5c}', '\u{b5d}'),
 945 |         ('\u{b5f}', '\u{b61}'),
 946 |         ('\u{b71}', '\u{b71}'),
 947 |         ('\u{b83}', '\u{b83}'),
 948 |         ('\u{b85}', '\u{b8a}'),
 949 |         ('\u{b8e}', '\u{b90}'),
 950 |         ('\u{b92}', '\u{b95}'),
 951 |         ('\u{b99}', '\u{b9a}'),
 952 |         ('\u{b9c}', '\u{b9c}'),
 953 |         ('\u{b9e}', '\u{b9f}'),
 954 |         ('\u{ba3}', '\u{ba4}'),
 955 |         ('\u{ba8}', '\u{baa}'),
 956 |         ('\u{bae}', '\u{bb9}'),
 957 |         ('\u{bd0}', '\u{bd0}'),
 958 |         ('\u{c05}', '\u{c0c}'),
 959 |         ('\u{c0e}', '\u{c10}'),
 960 |         ('\u{c12}', '\u{c28}'),
 961 |         ('\u{c2a}', '\u{c39}'),
 962 |         ('\u{c3d}', '\u{c3d}'),
 963 |         ('\u{c58}', '\u{c5a}'),
 964 |         ('\u{c5d}', '\u{c5d}'),
 965 |         ('\u{c60}', '\u{c61}'),
 966 |         ('\u{c80}', '\u{c80}'),
 967 |         ('\u{c85}', '\u{c8c}'),
 968 |         ('\u{c8e}', '\u{c90}'),
 969 |         ('\u{c92}', '\u{ca8}'),
 970 |         ('\u{caa}', '\u{cb3}'),
 971 |         ('\u{cb5}', '\u{cb9}'),
 972 |         ('\u{cbd}', '\u{cbd}'),
 973 |         ('\u{cdd}', '\u{cde}'),
 974 |         ('\u{ce0}', '\u{ce1}'),
 975 |         ('\u{cf1}', '\u{cf2}'),
 976 |         ('\u{d04}', '\u{d0c}'),
 977 |         ('\u{d0e}', '\u{d10}'),
 978 |         ('\u{d12}', '\u{d3a}'),
 979 |         ('\u{d3d}', '\u{d3d}'),
 980 |         ('\u{d4e}', '\u{d4e}'),
 981 |         ('\u{d54}', '\u{d56}'),
 982 |         ('\u{d5f}', '\u{d61}'),
 983 |         ('\u{d7a}', '\u{d7f}'),
 984 |         ('\u{d85}', '\u{d96}'),
 985 |         ('\u{d9a}', '\u{db1}'),
 986 |         ('\u{db3}', '\u{dbb}'),
 987 |         ('\u{dbd}', '\u{dbd}'),
 988 |         ('\u{dc0}', '\u{dc6}'),
 989 |         ('\u{e01}', '\u{e30}'),
 990 |         ('\u{e32}', '\u{e32}'),
 991 |         ('\u{e40}', '\u{e46}'),
 992 |         ('\u{e81}', '\u{e82}'),
 993 |         ('\u{e84}', '\u{e84}'),
 994 |         ('\u{e86}', '\u{e8a}'),
 995 |         ('\u{e8c}', '\u{ea3}'),
 996 |         ('\u{ea5}', '\u{ea5}'),
 997 |         ('\u{ea7}', '\u{eb0}'),
 998 |         ('\u{eb2}', '\u{eb2}'),
 999 |         ('\u{ebd}', '\u{ebd}'),
1000 |         ('\u{ec0}', '\u{ec4}'),
1001 |         ('\u{ec6}', '\u{ec6}'),
1002 |         ('\u{edc}', '\u{edf}'),
1003 |         ('\u{f00}', '\u{f00}'),
1004 |         ('\u{f40}', '\u{f47}'),
1005 |         ('\u{f49}', '\u{f6c}'),
1006 |         ('\u{f88}', '\u{f8c}'),
1007 |         ('\u{1000}', '\u{102a}'),
1008 |         ('\u{103f}', '\u{103f}'),
1009 |         ('\u{1050}', '\u{1055}'),
1010 |         ('\u{105a}', '\u{105d}'),
1011 |         ('\u{1061}', '\u{1061}'),
1012 |         ('\u{1065}', '\u{1066}'),
1013 |         ('\u{106e}', '\u{1070}'),
1014 |         ('\u{1075}', '\u{1081}'),
1015 |         ('\u{108e}', '\u{108e}'),
1016 |         ('\u{10a0}', '\u{10c5}'),
1017 |         ('\u{10c7}', '\u{10c7}'),
1018 |         ('\u{10cd}', '\u{10cd}'),
1019 |         ('\u{10d0}', '\u{10fa}'),
1020 |         ('\u{10fc}', '\u{1248}'),
1021 |         ('\u{124a}', '\u{124d}'),
1022 |         ('\u{1250}', '\u{1256}'),
1023 |         ('\u{1258}', '\u{1258}'),
1024 |         ('\u{125a}', '\u{125d}'),
1025 |         ('\u{1260}', '\u{1288}'),
1026 |         ('\u{128a}', '\u{128d}'),
1027 |         ('\u{1290}', '\u{12b0}'),
1028 |         ('\u{12b2}', '\u{12b5}'),
1029 |         ('\u{12b8}', '\u{12be}'),
1030 |         ('\u{12c0}', '\u{12c0}'),
1031 |         ('\u{12c2}', '\u{12c5}'),
1032 |         ('\u{12c8}', '\u{12d6}'),
1033 |         ('\u{12d8}', '\u{1310}'),
1034 |         ('\u{1312}', '\u{1315}'),
1035 |         ('\u{1318}', '\u{135a}'),
1036 |         ('\u{1380}', '\u{138f}'),
1037 |         ('\u{13a0}', '\u{13f5}'),
1038 |         ('\u{13f8}', '\u{13fd}'),
1039 |         ('\u{1401}', '\u{166c}'),
1040 |         ('\u{166f}', '\u{167f}'),
1041 |         ('\u{1681}', '\u{169a}'),
1042 |         ('\u{16a0}', '\u{16ea}'),
1043 |         ('\u{16ee}', '\u{16f8}'),
1044 |         ('\u{1700}', '\u{1711}'),
1045 |         ('\u{171f}', '\u{1731}'),
1046 |         ('\u{1740}', '\u{1751}'),
1047 |         ('\u{1760}', '\u{176c}'),
1048 |         ('\u{176e}', '\u{1770}'),
1049 |         ('\u{1780}', '\u{17b3}'),
1050 |         ('\u{17d7}', '\u{17d7}'),
1051 |         ('\u{17dc}', '\u{17dc}'),
1052 |         ('\u{1820}', '\u{1878}'),
1053 |         ('\u{1880}', '\u{18a8}'),
1054 |         ('\u{18aa}', '\u{18aa}'),
1055 |         ('\u{18b0}', '\u{18f5}'),
1056 |         ('\u{1900}', '\u{191e}'),
1057 |         ('\u{1950}', '\u{196d}'),
1058 |         ('\u{1970}', '\u{1974}'),
1059 |         ('\u{1980}', '\u{19ab}'),
1060 |         ('\u{19b0}', '\u{19c9}'),
1061 |         ('\u{1a00}', '\u{1a16}'),
1062 |         ('\u{1a20}', '\u{1a54}'),
1063 |         ('\u{1aa7}', '\u{1aa7}'),
1064 |         ('\u{1b05}', '\u{1b33}'),
1065 |         ('\u{1b45}', '\u{1b4c}'),
1066 |         ('\u{1b83}', '\u{1ba0}'),
1067 |         ('\u{1bae}', '\u{1baf}'),
1068 |         ('\u{1bba}', '\u{1be5}'),
1069 |         ('\u{1c00}', '\u{1c23}'),
1070 |         ('\u{1c4d}', '\u{1c4f}'),
1071 |         ('\u{1c5a}', '\u{1c7d}'),
1072 |         ('\u{1c80}', '\u{1c8a}'),
1073 |         ('\u{1c90}', '\u{1cba}'),
1074 |         ('\u{1cbd}', '\u{1cbf}'),
1075 |         ('\u{1ce9}', '\u{1cec}'),
1076 |         ('\u{1cee}', '\u{1cf3}'),
1077 |         ('\u{1cf5}', '\u{1cf6}'),
1078 |         ('\u{1cfa}', '\u{1cfa}'),
1079 |         ('\u{1d00}', '\u{1dbf}'),
1080 |         ('\u{1e00}', '\u{1f15}'),
1081 |         ('\u{1f18}', '\u{1f1d}'),
1082 |         ('\u{1f20}', '\u{1f45}'),
1083 |         ('\u{1f48}', '\u{1f4d}'),
1084 |         ('\u{1f50}', '\u{1f57}'),
1085 |         ('\u{1f59}', '\u{1f59}'),
1086 |         ('\u{1f5b}', '\u{1f5b}'),
1087 |         ('\u{1f5d}', '\u{1f5d}'),
1088 |         ('\u{1f5f}', '\u{1f7d}'),
1089 |         ('\u{1f80}', '\u{1fb4}'),
1090 |         ('\u{1fb6}', '\u{1fbc}'),
1091 |         ('\u{1fbe}', '\u{1fbe}'),
1092 |         ('\u{1fc2}', '\u{1fc4}'),
1093 |         ('\u{1fc6}', '\u{1fcc}'),
1094 |         ('\u{1fd0}', '\u{1fd3}'),
1095 |         ('\u{1fd6}', '\u{1fdb}'),
1096 |         ('\u{1fe0}', '\u{1fec}'),
1097 |         ('\u{1ff2}', '\u{1ff4}'),
1098 |         ('\u{1ff6}', '\u{1ffc}'),
1099 |         ('\u{2071}', '\u{2071}'),
1100 |         ('\u{207f}', '\u{207f}'),
1101 |         ('\u{2090}', '\u{209c}'),
1102 |         ('\u{2102}', '\u{2102}'),
1103 |         ('\u{2107}', '\u{2107}'),
1104 |         ('\u{210a}', '\u{2113}'),
1105 |         ('\u{2115}', '\u{2115}'),
1106 |         ('\u{2118}', '\u{211d}'),
1107 |         ('\u{2124}', '\u{2124}'),
1108 |         ('\u{2126}', '\u{2126}'),
1109 |         ('\u{2128}', '\u{2128}'),
1110 |         ('\u{212a}', '\u{2139}'),
1111 |         ('\u{213c}', '\u{213f}'),
1112 |         ('\u{2145}', '\u{2149}'),
1113 |         ('\u{214e}', '\u{214e}'),
1114 |         ('\u{2160}', '\u{2188}'),
1115 |         ('\u{2c00}', '\u{2ce4}'),
1116 |         ('\u{2ceb}', '\u{2cee}'),
1117 |         ('\u{2cf2}', '\u{2cf3}'),
1118 |         ('\u{2d00}', '\u{2d25}'),
1119 |         ('\u{2d27}', '\u{2d27}'),
1120 |         ('\u{2d2d}', '\u{2d2d}'),
1121 |         ('\u{2d30}', '\u{2d67}'),
1122 |         ('\u{2d6f}', '\u{2d6f}'),
1123 |         ('\u{2d80}', '\u{2d96}'),
1124 |         ('\u{2da0}', '\u{2da6}'),
1125 |         ('\u{2da8}', '\u{2dae}'),
1126 |         ('\u{2db0}', '\u{2db6}'),
1127 |         ('\u{2db8}', '\u{2dbe}'),
1128 |         ('\u{2dc0}', '\u{2dc6}'),
1129 |         ('\u{2dc8}', '\u{2dce}'),
1130 |         ('\u{2dd0}', '\u{2dd6}'),
1131 |         ('\u{2dd8}', '\u{2dde}'),
1132 |         ('\u{3005}', '\u{3007}'),
1133 |         ('\u{3021}', '\u{3029}'),
1134 |         ('\u{3031}', '\u{3035}'),
1135 |         ('\u{3038}', '\u{303c}'),
1136 |         ('\u{3041}', '\u{3096}'),
1137 |         ('\u{309d}', '\u{309f}'),
1138 |         ('\u{30a1}', '\u{30fa}'),
1139 |         ('\u{30fc}', '\u{30ff}'),
1140 |         ('\u{3105}', '\u{312f}'),
1141 |         ('\u{3131}', '\u{318e}'),
1142 |         ('\u{31a0}', '\u{31bf}'),
1143 |         ('\u{31f0}', '\u{31ff}'),
1144 |         ('\u{3400}', '\u{4dbf}'),
1145 |         ('\u{4e00}', '\u{a48c}'),
1146 |         ('\u{a4d0}', '\u{a4fd}'),
1147 |         ('\u{a500}', '\u{a60c}'),
1148 |         ('\u{a610}', '\u{a61f}'),
1149 |         ('\u{a62a}', '\u{a62b}'),
1150 |         ('\u{a640}', '\u{a66e}'),
1151 |         ('\u{a67f}', '\u{a69d}'),
1152 |         ('\u{a6a0}', '\u{a6ef}'),
1153 |         ('\u{a717}', '\u{a71f}'),
1154 |         ('\u{a722}', '\u{a788}'),
1155 |         ('\u{a78b}', '\u{a7cd}'),
1156 |         ('\u{a7d0}', '\u{a7d1}'),
1157 |         ('\u{a7d3}', '\u{a7d3}'),
1158 |         ('\u{a7d5}', '\u{a7dc}'),
1159 |         ('\u{a7f2}', '\u{a801}'),
1160 |         ('\u{a803}', '\u{a805}'),
1161 |         ('\u{a807}', '\u{a80a}'),
1162 |         ('\u{a80c}', '\u{a822}'),
1163 |         ('\u{a840}', '\u{a873}'),
1164 |         ('\u{a882}', '\u{a8b3}'),
1165 |         ('\u{a8f2}', '\u{a8f7}'),
1166 |         ('\u{a8fb}', '\u{a8fb}'),
1167 |         ('\u{a8fd}', '\u{a8fe}'),
1168 |         ('\u{a90a}', '\u{a925}'),
1169 |         ('\u{a930}', '\u{a946}'),
1170 |         ('\u{a960}', '\u{a97c}'),
1171 |         ('\u{a984}', '\u{a9b2}'),
1172 |         ('\u{a9cf}', '\u{a9cf}'),
1173 |         ('\u{a9e0}', '\u{a9e4}'),
1174 |         ('\u{a9e6}', '\u{a9ef}'),
1175 |         ('\u{a9fa}', '\u{a9fe}'),
1176 |         ('\u{aa00}', '\u{aa28}'),
1177 |         ('\u{aa40}', '\u{aa42}'),
1178 |         ('\u{aa44}', '\u{aa4b}'),
1179 |         ('\u{aa60}', '\u{aa76}'),
1180 |         ('\u{aa7a}', '\u{aa7a}'),
1181 |         ('\u{aa7e}', '\u{aaaf}'),
1182 |         ('\u{aab1}', '\u{aab1}'),
1183 |         ('\u{aab5}', '\u{aab6}'),
1184 |         ('\u{aab9}', '\u{aabd}'),
1185 |         ('\u{aac0}', '\u{aac0}'),
1186 |         ('\u{aac2}', '\u{aac2}'),
1187 |         ('\u{aadb}', '\u{aadd}'),
1188 |         ('\u{aae0}', '\u{aaea}'),
1189 |         ('\u{aaf2}', '\u{aaf4}'),
1190 |         ('\u{ab01}', '\u{ab06}'),
1191 |         ('\u{ab09}', '\u{ab0e}'),
1192 |         ('\u{ab11}', '\u{ab16}'),
1193 |         ('\u{ab20}', '\u{ab26}'),
1194 |         ('\u{ab28}', '\u{ab2e}'),
1195 |         ('\u{ab30}', '\u{ab5a}'),
1196 |         ('\u{ab5c}', '\u{ab69}'),
1197 |         ('\u{ab70}', '\u{abe2}'),
1198 |         ('\u{ac00}', '\u{d7a3}'),
1199 |         ('\u{d7b0}', '\u{d7c6}'),
1200 |         ('\u{d7cb}', '\u{d7fb}'),
1201 |         ('\u{f900}', '\u{fa6d}'),
1202 |         ('\u{fa70}', '\u{fad9}'),
1203 |         ('\u{fb00}', '\u{fb06}'),
1204 |         ('\u{fb13}', '\u{fb17}'),
1205 |         ('\u{fb1d}', '\u{fb1d}'),
1206 |         ('\u{fb1f}', '\u{fb28}'),
1207 |         ('\u{fb2a}', '\u{fb36}'),
1208 |         ('\u{fb38}', '\u{fb3c}'),
1209 |         ('\u{fb3e}', '\u{fb3e}'),
1210 |         ('\u{fb40}', '\u{fb41}'),
1211 |         ('\u{fb43}', '\u{fb44}'),
1212 |         ('\u{fb46}', '\u{fbb1}'),
1213 |         ('\u{fbd3}', '\u{fc5d}'),
1214 |         ('\u{fc64}', '\u{fd3d}'),
1215 |         ('\u{fd50}', '\u{fd8f}'),
1216 |         ('\u{fd92}', '\u{fdc7}'),
1217 |         ('\u{fdf0}', '\u{fdf9}'),
1218 |         ('\u{fe71}', '\u{fe71}'),
1219 |         ('\u{fe73}', '\u{fe73}'),
1220 |         ('\u{fe77}', '\u{fe77}'),
1221 |         ('\u{fe79}', '\u{fe79}'),
1222 |         ('\u{fe7b}', '\u{fe7b}'),
1223 |         ('\u{fe7d}', '\u{fe7d}'),
1224 |         ('\u{fe7f}', '\u{fefc}'),
1225 |         ('\u{ff21}', '\u{ff3a}'),
1226 |         ('\u{ff41}', '\u{ff5a}'),
1227 |         ('\u{ff66}', '\u{ff9d}'),
1228 |         ('\u{ffa0}', '\u{ffbe}'),
1229 |         ('\u{ffc2}', '\u{ffc7}'),
1230 |         ('\u{ffca}', '\u{ffcf}'),
1231 |         ('\u{ffd2}', '\u{ffd7}'),
1232 |         ('\u{ffda}', '\u{ffdc}'),
1233 |         ('\u{10000}', '\u{1000b}'),
1234 |         ('\u{1000d}', '\u{10026}'),
1235 |         ('\u{10028}', '\u{1003a}'),
1236 |         ('\u{1003c}', '\u{1003d}'),
1237 |         ('\u{1003f}', '\u{1004d}'),
1238 |         ('\u{10050}', '\u{1005d}'),
1239 |         ('\u{10080}', '\u{100fa}'),
1240 |         ('\u{10140}', '\u{10174}'),
1241 |         ('\u{10280}', '\u{1029c}'),
1242 |         ('\u{102a0}', '\u{102d0}'),
1243 |         ('\u{10300}', '\u{1031f}'),
1244 |         ('\u{1032d}', '\u{1034a}'),
1245 |         ('\u{10350}', '\u{10375}'),
1246 |         ('\u{10380}', '\u{1039d}'),
1247 |         ('\u{103a0}', '\u{103c3}'),
1248 |         ('\u{103c8}', '\u{103cf}'),
1249 |         ('\u{103d1}', '\u{103d5}'),
1250 |         ('\u{10400}', '\u{1049d}'),
1251 |         ('\u{104b0}', '\u{104d3}'),
1252 |         ('\u{104d8}', '\u{104fb}'),
1253 |         ('\u{10500}', '\u{10527}'),
1254 |         ('\u{10530}', '\u{10563}'),
1255 |         ('\u{10570}', '\u{1057a}'),
1256 |         ('\u{1057c}', '\u{1058a}'),
1257 |         ('\u{1058c}', '\u{10592}'),
1258 |         ('\u{10594}', '\u{10595}'),
1259 |         ('\u{10597}', '\u{105a1}'),
1260 |         ('\u{105a3}', '\u{105b1}'),
1261 |         ('\u{105b3}', '\u{105b9}'),
1262 |         ('\u{105bb}', '\u{105bc}'),
1263 |         ('\u{105c0}', '\u{105f3}'),
1264 |         ('\u{10600}', '\u{10736}'),
1265 |         ('\u{10740}', '\u{10755}'),
1266 |         ('\u{10760}', '\u{10767}'),
1267 |         ('\u{10780}', '\u{10785}'),
1268 |         ('\u{10787}', '\u{107b0}'),
1269 |         ('\u{107b2}', '\u{107ba}'),
1270 |         ('\u{10800}', '\u{10805}'),
1271 |         ('\u{10808}', '\u{10808}'),
1272 |         ('\u{1080a}', '\u{10835}'),
1273 |         ('\u{10837}', '\u{10838}'),
1274 |         ('\u{1083c}', '\u{1083c}'),
1275 |         ('\u{1083f}', '\u{10855}'),
1276 |         ('\u{10860}', '\u{10876}'),
1277 |         ('\u{10880}', '\u{1089e}'),
1278 |         ('\u{108e0}', '\u{108f2}'),
1279 |         ('\u{108f4}', '\u{108f5}'),
1280 |         ('\u{10900}', '\u{10915}'),
1281 |         ('\u{10920}', '\u{10939}'),
1282 |         ('\u{10980}', '\u{109b7}'),
1283 |         ('\u{109be}', '\u{109bf}'),
1284 |         ('\u{10a00}', '\u{10a00}'),
1285 |         ('\u{10a10}', '\u{10a13}'),
1286 |         ('\u{10a15}', '\u{10a17}'),
1287 |         ('\u{10a19}', '\u{10a35}'),
1288 |         ('\u{10a60}', '\u{10a7c}'),
1289 |         ('\u{10a80}', '\u{10a9c}'),
1290 |         ('\u{10ac0}', '\u{10ac7}'),
1291 |         ('\u{10ac9}', '\u{10ae4}'),
1292 |         ('\u{10b00}', '\u{10b35}'),
1293 |         ('\u{10b40}', '\u{10b55}'),
1294 |         ('\u{10b60}', '\u{10b72}'),
1295 |         ('\u{10b80}', '\u{10b91}'),
1296 |         ('\u{10c00}', '\u{10c48}'),
1297 |         ('\u{10c80}', '\u{10cb2}'),
1298 |         ('\u{10cc0}', '\u{10cf2}'),
1299 |         ('\u{10d00}', '\u{10d23}'),
1300 |         ('\u{10d4a}', '\u{10d65}'),
1301 |         ('\u{10d6f}', '\u{10d85}'),
1302 |         ('\u{10e80}', '\u{10ea9}'),
1303 |         ('\u{10eb0}', '\u{10eb1}'),
1304 |         ('\u{10ec2}', '\u{10ec4}'),
1305 |         ('\u{10f00}', '\u{10f1c}'),
1306 |         ('\u{10f27}', '\u{10f27}'),
1307 |         ('\u{10f30}', '\u{10f45}'),
1308 |         ('\u{10f70}', '\u{10f81}'),
1309 |         ('\u{10fb0}', '\u{10fc4}'),
1310 |         ('\u{10fe0}', '\u{10ff6}'),
1311 |         ('\u{11003}', '\u{11037}'),
1312 |         ('\u{11071}', '\u{11072}'),
1313 |         ('\u{11075}', '\u{11075}'),
1314 |         ('\u{11083}', '\u{110af}'),
1315 |         ('\u{110d0}', '\u{110e8}'),
1316 |         ('\u{11103}', '\u{11126}'),
1317 |         ('\u{11144}', '\u{11144}'),
1318 |         ('\u{11147}', '\u{11147}'),
1319 |         ('\u{11150}', '\u{11172}'),
1320 |         ('\u{11176}', '\u{11176}'),
1321 |         ('\u{11183}', '\u{111b2}'),
1322 |         ('\u{111c1}', '\u{111c4}'),
1323 |         ('\u{111da}', '\u{111da}'),
1324 |         ('\u{111dc}', '\u{111dc}'),
1325 |         ('\u{11200}', '\u{11211}'),
1326 |         ('\u{11213}', '\u{1122b}'),
1327 |         ('\u{1123f}', '\u{11240}'),
1328 |         ('\u{11280}', '\u{11286}'),
1329 |         ('\u{11288}', '\u{11288}'),
1330 |         ('\u{1128a}', '\u{1128d}'),
1331 |         ('\u{1128f}', '\u{1129d}'),
1332 |         ('\u{1129f}', '\u{112a8}'),
1333 |         ('\u{112b0}', '\u{112de}'),
1334 |         ('\u{11305}', '\u{1130c}'),
1335 |         ('\u{1130f}', '\u{11310}'),
1336 |         ('\u{11313}', '\u{11328}'),
1337 |         ('\u{1132a}', '\u{11330}'),
1338 |         ('\u{11332}', '\u{11333}'),
1339 |         ('\u{11335}', '\u{11339}'),
1340 |         ('\u{1133d}', '\u{1133d}'),
1341 |         ('\u{11350}', '\u{11350}'),
1342 |         ('\u{1135d}', '\u{11361}'),
1343 |         ('\u{11380}', '\u{11389}'),
1344 |         ('\u{1138b}', '\u{1138b}'),
1345 |         ('\u{1138e}', '\u{1138e}'),
1346 |         ('\u{11390}', '\u{113b5}'),
1347 |         ('\u{113b7}', '\u{113b7}'),
1348 |         ('\u{113d1}', '\u{113d1}'),
1349 |         ('\u{113d3}', '\u{113d3}'),
1350 |         ('\u{11400}', '\u{11434}'),
1351 |         ('\u{11447}', '\u{1144a}'),
1352 |         ('\u{1145f}', '\u{11461}'),
1353 |         ('\u{11480}', '\u{114af}'),
1354 |         ('\u{114c4}', '\u{114c5}'),
1355 |         ('\u{114c7}', '\u{114c7}'),
1356 |         ('\u{11580}', '\u{115ae}'),
1357 |         ('\u{115d8}', '\u{115db}'),
1358 |         ('\u{11600}', '\u{1162f}'),
1359 |         ('\u{11644}', '\u{11644}'),
1360 |         ('\u{11680}', '\u{116aa}'),
1361 |         ('\u{116b8}', '\u{116b8}'),
1362 |         ('\u{11700}', '\u{1171a}'),
1363 |         ('\u{11740}', '\u{11746}'),
1364 |         ('\u{11800}', '\u{1182b}'),
1365 |         ('\u{118a0}', '\u{118df}'),
1366 |         ('\u{118ff}', '\u{11906}'),
1367 |         ('\u{11909}', '\u{11909}'),
1368 |         ('\u{1190c}', '\u{11913}'),
1369 |         ('\u{11915}', '\u{11916}'),
1370 |         ('\u{11918}', '\u{1192f}'),
1371 |         ('\u{1193f}', '\u{1193f}'),
1372 |         ('\u{11941}', '\u{11941}'),
1373 |         ('\u{119a0}', '\u{119a7}'),
1374 |         ('\u{119aa}', '\u{119d0}'),
1375 |         ('\u{119e1}', '\u{119e1}'),
1376 |         ('\u{119e3}', '\u{119e3}'),
1377 |         ('\u{11a00}', '\u{11a00}'),
1378 |         ('\u{11a0b}', '\u{11a32}'),
1379 |         ('\u{11a3a}', '\u{11a3a}'),
1380 |         ('\u{11a50}', '\u{11a50}'),
1381 |         ('\u{11a5c}', '\u{11a89}'),
1382 |         ('\u{11a9d}', '\u{11a9d}'),
1383 |         ('\u{11ab0}', '\u{11af8}'),
1384 |         ('\u{11bc0}', '\u{11be0}'),
1385 |         ('\u{11c00}', '\u{11c08}'),
1386 |         ('\u{11c0a}', '\u{11c2e}'),
1387 |         ('\u{11c40}', '\u{11c40}'),
1388 |         ('\u{11c72}', '\u{11c8f}'),
1389 |         ('\u{11d00}', '\u{11d06}'),
1390 |         ('\u{11d08}', '\u{11d09}'),
1391 |         ('\u{11d0b}', '\u{11d30}'),
1392 |         ('\u{11d46}', '\u{11d46}'),
1393 |         ('\u{11d60}', '\u{11d65}'),
1394 |         ('\u{11d67}', '\u{11d68}'),
1395 |         ('\u{11d6a}', '\u{11d89}'),
1396 |         ('\u{11d98}', '\u{11d98}'),
1397 |         ('\u{11ee0}', '\u{11ef2}'),
1398 |         ('\u{11f02}', '\u{11f02}'),
1399 |         ('\u{11f04}', '\u{11f10}'),
1400 |         ('\u{11f12}', '\u{11f33}'),
1401 |         ('\u{11fb0}', '\u{11fb0}'),
1402 |         ('\u{12000}', '\u{12399}'),
1403 |         ('\u{12400}', '\u{1246e}'),
1404 |         ('\u{12480}', '\u{12543}'),
1405 |         ('\u{12f90}', '\u{12ff0}'),
1406 |         ('\u{13000}', '\u{1342f}'),
1407 |         ('\u{13441}', '\u{13446}'),
1408 |         ('\u{13460}', '\u{143fa}'),
1409 |         ('\u{14400}', '\u{14646}'),
1410 |         ('\u{16100}', '\u{1611d}'),
1411 |         ('\u{16800}', '\u{16a38}'),
1412 |         ('\u{16a40}', '\u{16a5e}'),
1413 |         ('\u{16a70}', '\u{16abe}'),
1414 |         ('\u{16ad0}', '\u{16aed}'),
1415 |         ('\u{16b00}', '\u{16b2f}'),
1416 |         ('\u{16b40}', '\u{16b43}'),
1417 |         ('\u{16b63}', '\u{16b77}'),
1418 |         ('\u{16b7d}', '\u{16b8f}'),
1419 |         ('\u{16d40}', '\u{16d6c}'),
1420 |         ('\u{16e40}', '\u{16e7f}'),
1421 |         ('\u{16f00}', '\u{16f4a}'),
1422 |         ('\u{16f50}', '\u{16f50}'),
1423 |         ('\u{16f93}', '\u{16f9f}'),
1424 |         ('\u{16fe0}', '\u{16fe1}'),
1425 |         ('\u{16fe3}', '\u{16fe3}'),
1426 |         ('\u{17000}', '\u{187f7}'),
1427 |         ('\u{18800}', '\u{18cd5}'),
1428 |         ('\u{18cff}', '\u{18d08}'),
1429 |         ('\u{1aff0}', '\u{1aff3}'),
1430 |         ('\u{1aff5}', '\u{1affb}'),
1431 |         ('\u{1affd}', '\u{1affe}'),
1432 |         ('\u{1b000}', '\u{1b122}'),
1433 |         ('\u{1b132}', '\u{1b132}'),
1434 |         ('\u{1b150}', '\u{1b152}'),
1435 |         ('\u{1b155}', '\u{1b155}'),
1436 |         ('\u{1b164}', '\u{1b167}'),
1437 |         ('\u{1b170}', '\u{1b2fb}'),
1438 |         ('\u{1bc00}', '\u{1bc6a}'),
1439 |         ('\u{1bc70}', '\u{1bc7c}'),
1440 |         ('\u{1bc80}', '\u{1bc88}'),
1441 |         ('\u{1bc90}', '\u{1bc99}'),
1442 |         ('\u{1d400}', '\u{1d454}'),
1443 |         ('\u{1d456}', '\u{1d49c}'),
1444 |         ('\u{1d49e}', '\u{1d49f}'),
1445 |         ('\u{1d4a2}', '\u{1d4a2}'),
1446 |         ('\u{1d4a5}', '\u{1d4a6}'),
1447 |         ('\u{1d4a9}', '\u{1d4ac}'),
1448 |         ('\u{1d4ae}', '\u{1d4b9}'),
1449 |         ('\u{1d4bb}', '\u{1d4bb}'),
1450 |         ('\u{1d4bd}', '\u{1d4c3}'),
1451 |         ('\u{1d4c5}', '\u{1d505}'),
1452 |         ('\u{1d507}', '\u{1d50a}'),
1453 |         ('\u{1d50d}', '\u{1d514}'),
1454 |         ('\u{1d516}', '\u{1d51c}'),
1455 |         ('\u{1d51e}', '\u{1d539}'),
1456 |         ('\u{1d53b}', '\u{1d53e}'),
1457 |         ('\u{1d540}', '\u{1d544}'),
1458 |         ('\u{1d546}', '\u{1d546}'),
1459 |         ('\u{1d54a}', '\u{1d550}'),
1460 |         ('\u{1d552}', '\u{1d6a5}'),
1461 |         ('\u{1d6a8}', '\u{1d6c0}'),
1462 |         ('\u{1d6c2}', '\u{1d6da}'),
1463 |         ('\u{1d6dc}', '\u{1d6fa}'),
1464 |         ('\u{1d6fc}', '\u{1d714}'),
1465 |         ('\u{1d716}', '\u{1d734}'),
1466 |         ('\u{1d736}', '\u{1d74e}'),
1467 |         ('\u{1d750}', '\u{1d76e}'),
1468 |         ('\u{1d770}', '\u{1d788}'),
1469 |         ('\u{1d78a}', '\u{1d7a8}'),
1470 |         ('\u{1d7aa}', '\u{1d7c2}'),
1471 |         ('\u{1d7c4}', '\u{1d7cb}'),
1472 |         ('\u{1df00}', '\u{1df1e}'),
1473 |         ('\u{1df25}', '\u{1df2a}'),
1474 |         ('\u{1e030}', '\u{1e06d}'),
1475 |         ('\u{1e100}', '\u{1e12c}'),
1476 |         ('\u{1e137}', '\u{1e13d}'),
1477 |         ('\u{1e14e}', '\u{1e14e}'),
1478 |         ('\u{1e290}', '\u{1e2ad}'),
1479 |         ('\u{1e2c0}', '\u{1e2eb}'),
1480 |         ('\u{1e4d0}', '\u{1e4eb}'),
1481 |         ('\u{1e5d0}', '\u{1e5ed}'),
1482 |         ('\u{1e5f0}', '\u{1e5f0}'),
1483 |         ('\u{1e7e0}', '\u{1e7e6}'),
1484 |         ('\u{1e7e8}', '\u{1e7eb}'),
1485 |         ('\u{1e7ed}', '\u{1e7ee}'),
1486 |         ('\u{1e7f0}', '\u{1e7fe}'),
1487 |         ('\u{1e800}', '\u{1e8c4}'),
1488 |         ('\u{1e900}', '\u{1e943}'),
1489 |         ('\u{1e94b}', '\u{1e94b}'),
1490 |         ('\u{1ee00}', '\u{1ee03}'),
1491 |         ('\u{1ee05}', '\u{1ee1f}'),
1492 |         ('\u{1ee21}', '\u{1ee22}'),
1493 |         ('\u{1ee24}', '\u{1ee24}'),
1494 |         ('\u{1ee27}', '\u{1ee27}'),
1495 |         ('\u{1ee29}', '\u{1ee32}'),
1496 |         ('\u{1ee34}', '\u{1ee37}'),
1497 |         ('\u{1ee39}', '\u{1ee39}'),
1498 |         ('\u{1ee3b}', '\u{1ee3b}'),
1499 |         ('\u{1ee42}', '\u{1ee42}'),
1500 |         ('\u{1ee47}', '\u{1ee47}'),
1501 |         ('\u{1ee49}', '\u{1ee49}'),
1502 |         ('\u{1ee4b}', '\u{1ee4b}'),
1503 |         ('\u{1ee4d}', '\u{1ee4f}'),
1504 |         ('\u{1ee51}', '\u{1ee52}'),
1505 |         ('\u{1ee54}', '\u{1ee54}'),
1506 |         ('\u{1ee57}', '\u{1ee57}'),
1507 |         ('\u{1ee59}', '\u{1ee59}'),
1508 |         ('\u{1ee5b}', '\u{1ee5b}'),
1509 |         ('\u{1ee5d}', '\u{1ee5d}'),
1510 |         ('\u{1ee5f}', '\u{1ee5f}'),
1511 |         ('\u{1ee61}', '\u{1ee62}'),
1512 |         ('\u{1ee64}', '\u{1ee64}'),
1513 |         ('\u{1ee67}', '\u{1ee6a}'),
1514 |         ('\u{1ee6c}', '\u{1ee72}'),
1515 |         ('\u{1ee74}', '\u{1ee77}'),
1516 |         ('\u{1ee79}', '\u{1ee7c}'),
1517 |         ('\u{1ee7e}', '\u{1ee7e}'),
1518 |         ('\u{1ee80}', '\u{1ee89}'),
1519 |         ('\u{1ee8b}', '\u{1ee9b}'),
1520 |         ('\u{1eea1}', '\u{1eea3}'),
1521 |         ('\u{1eea5}', '\u{1eea9}'),
1522 |         ('\u{1eeab}', '\u{1eebb}'),
1523 |         ('\u{20000}', '\u{2a6df}'),
1524 |         ('\u{2a700}', '\u{2b739}'),
1525 |         ('\u{2b740}', '\u{2b81d}'),
1526 |         ('\u{2b820}', '\u{2cea1}'),
1527 |         ('\u{2ceb0}', '\u{2ebe0}'),
1528 |         ('\u{2ebf0}', '\u{2ee5d}'),
1529 |         ('\u{2f800}', '\u{2fa1d}'),
1530 |         ('\u{30000}', '\u{3134a}'),
1531 |         ('\u{31350}', '\u{323af}'),
1532 |     ];
1533 | 
1534 |     pub fn XID_Start(c: char) -> bool {
1535 |         super::bsearch_range_table(c, XID_Start_table)
1536 |     }
1537 | }
1538 | 


--------------------------------------------------------------------------------
/src/tests.rs:
--------------------------------------------------------------------------------
 1 | // Copyright 2012-2025 The Rust Project Developers. See the COPYRIGHT
 2 | // file at the top-level directory of this distribution and at
 3 | // https://www.rust-lang.org/policies/licenses.
 4 | //
 5 | // Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or
 6 | // https://www.apache.org/licenses/LICENSE-2.0> or the MIT license
 7 | // <LICENSE-MIT or https://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 | #[cfg(feature = "bench")]
12 | use std::iter;
13 | #[cfg(feature = "bench")]
14 | use std::prelude::v1::*;
15 | #[cfg(feature = "bench")]
16 | use test::Bencher;
17 | 
18 | #[cfg(feature = "bench")]
19 | use UnicodeXID;
20 | 
21 | #[cfg(feature = "bench")]
22 | #[bench]
23 | fn cargo_is_xid_start(b: &mut Bencher) {
24 |     let string = iter::repeat('a').take(4096).collect::<String>();
25 | 
26 |     b.bytes = string.len() as u64;
27 |     b.iter(|| string.chars().all(super::UnicodeXID::is_xid_start));
28 | }
29 | 
30 | #[cfg(feature = "bench")]
31 | #[bench]
32 | fn stdlib_is_xid_start(b: &mut Bencher) {
33 |     let string = iter::repeat('a').take(4096).collect::<String>();
34 | 
35 |     b.bytes = string.len() as u64;
36 |     b.iter(|| string.chars().all(char::is_xid_start));
37 | }
38 | 
39 | #[cfg(feature = "bench")]
40 | #[bench]
41 | fn cargo_xid_continue(b: &mut Bencher) {
42 |     let string = iter::repeat('a').take(4096).collect::<String>();
43 | 
44 |     b.bytes = string.len() as u64;
45 |     b.iter(|| string.chars().all(super::UnicodeXID::is_xid_continue));
46 | }
47 | 
48 | #[cfg(feature = "bench")]
49 | #[bench]
50 | fn stdlib_xid_continue(b: &mut Bencher) {
51 |     let string = iter::repeat('a').take(4096).collect::<String>();
52 | 
53 |     b.bytes = string.len() as u64;
54 |     b.iter(|| string.chars().all(char::is_xid_continue));
55 | }
56 | 
57 | #[test]
58 | fn test_is_xid_start() {
59 |     let chars = ['A', 'Z', 'a', 'z', '\u{1000d}', '\u{10026}'];
60 | 
61 |     for ch in &chars {
62 |         assert!(super::UnicodeXID::is_xid_start(*ch), "{}", ch);
63 |     }
64 | }
65 | 
66 | #[test]
67 | fn test_is_not_xid_start() {
68 |     let chars = [
69 |         '\x00', '\x01', '0', '9', ' ', '[', '<', '{', '(', '\u{02c2}', '\u{ffff}',
70 |     ];
71 | 
72 |     for ch in &chars {
73 |         assert!(!super::UnicodeXID::is_xid_start(*ch), "{}", ch);
74 |     }
75 | }
76 | 
77 | #[test]
78 | fn test_is_xid_continue() {
79 |     let chars = ['0', '9', 'A', 'Z', 'a', 'z', '_', '\u{1000d}', '\u{10026}'];
80 | 
81 |     for ch in &chars {
82 |         assert!(super::UnicodeXID::is_xid_continue(*ch), "{}", ch);
83 |     }
84 | }
85 | 
86 | #[test]
87 | fn test_is_not_xid_continue() {
88 |     let chars = [
89 |         '\x00', '\x01', ' ', '[', '<', '{', '(', '\u{02c2}', '\u{ffff}',
90 |     ];
91 | 
92 |     for &ch in &chars {
93 |         assert!(!super::UnicodeXID::is_xid_continue(ch), "{}", ch);
94 |     }
95 | }
96 | 


--------------------------------------------------------------------------------
/tests/exhaustive_tests.rs:
--------------------------------------------------------------------------------
 1 | extern crate unicode_xid;
 2 | use unicode_xid::UnicodeXID;
 3 | /// A `char` in Rust is a Unicode Scalar Value
 4 | ///
 5 | /// See: http://www.unicode.org/glossary/#unicode_scalar_value
 6 | fn all_valid_chars() -> impl Iterator<Item = char> {
 7 |     (0u32..=0xD7FF).chain(0xE000u32..=0x10FFFF).map(|u| {
 8 |         core::convert::TryFrom::try_from(u)
 9 |             .expect("The selected range should be infallible if the docs match impl")
10 |     })
11 | }
12 | 
13 | #[test]
14 | fn all_valid_chars_do_not_panic_for_is_xid_start() {
15 |     for c in all_valid_chars() {
16 |         let _ = UnicodeXID::is_xid_start(c);
17 |     }
18 | }
19 | 
20 | #[test]
21 | fn all_valid_chars_do_not_panic_for_is_xid_continue() {
22 |     for c in all_valid_chars() {
23 |         let _ = UnicodeXID::is_xid_continue(c);
24 |     }
25 | }
26 | 


--------------------------------------------------------------------------------