rust-url/make_idna_table.py

# Copyright 2013-2014 Valentin Gosu.
#
# Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or
# http://www.apache.org/licenses/LICENSE-2.0> or the MIT license
# <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your
# option. This file may not be copied, modified, or distributed
# except according to those terms.


# Run as: python make_idna_table.py idna_table.txt > src/idna_table.rs
# You can get the latest idna table from
# http://www.unicode.org/Public/idna/latest/IdnaMappingTable.txt

print('''\
// Copyright 2013-2014 Valentin Gosu.
//
// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or
// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license
// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your
// option. This file may not be copied, modified, or distributed
// except according to those terms.

// Generated by make_idna_table.py

pub enum Mapping {
    Valid,
    Ignored,
    Mapped(&'static str),
    Deviation(&'static str),
    Disallowed,
    DisallowedStd3Valid,
    DisallowedStd3Mapped(&'static str),
}

pub struct Range {
    pub from: char,
    pub to: char,
    pub mapping: Mapping,
}

use self::Mapping::*;

pub static TABLE: &'static [Range] = &[
''')

txt = open("IdnaMappingTable.txt")

def char(s):
    return (unichr(int(s, 16))
        .encode('utf8')
        .replace('\\', '\\\\')
        .replace('"', '\\"')
        .replace('\0', '\\0'))

for line in txt:
    # remove comments
    line, _, _ = line.partition('#')
    # skip empty lines
    if len(line.strip()) == 0:
        continue
    fields = line.split(';')
    if fields[0].strip() == 'D800..DFFF':
        continue  # Surrogates don't occur in Rust strings.
    first, _, last = fields[0].strip().partition('..')
    if not last:
        last = first
    mapping = fields[1].strip().replace('_', ' ').title().replace(' ', '')
    if len(fields) > 2:
        if fields[2].strip():
            mapping += '("%s")' % ''.join(char(c) for c in fields[2].strip().split(' '))
        elif mapping == "Deviation":
            mapping += '("")'
    print("    Range { from: '%s', to: '%s', mapping: %s }," % (char(first), char(last), mapping))

print("];")