Merge "Fix frameworks/base/tools/localedata/extract_icu_data.py" into main

import os.path

import sys

import xml.etree.ElementTree as ElementTree

def get_locale_parts(locale):

    """Split a locale into three parts, for langauge, script, and region."""

def read_likely_subtags(input_file_name):

    """Read and parse ICU's likelySubtags.txt."""

    with open(input_file_name) as input_file:

    likely_script_dict = {

        # Android's additions for pseudo-locales. These internal codes make

        # sure that the pseudo-locales would not match other English or

        # while. Fortunately, MX < US, so if both exist, MX

        # would be chosen.)

    }

        for line in input_file:

            line = line.strip(u' \n\uFEFF')

            if line.startswith('//'):

                continue

            if '{' in line and '}' in line:

                from_locale = line[:line.index('{')]

                to_locale = line[line.index('"')+1:line.rindex('"')]

    xml_tree = ElementTree.parse(input_file_name)

    likely_subtags = xml_tree.find('likelySubtags')

    for child in likely_subtags:

        from_locale = child.get('from')

        to_locale = child.get('to')

        # print(f'from: {from_locale} to: {to_locale}')

        from_lang, from_scr, from_region = get_locale_parts(from_locale)

        _, to_scr, to_region = get_locale_parts(to_locale)

        if to_locale == "FAIL":

            continue # "FAIL" cases are not useful here.

        if from_lang == 'und':

            continue  # not very useful for our purposes

        if from_region is None and to_region not in ['001', 'ZZ']:

            representative_locales.add(to_locale)

        if from_scr is None:

            likely_script_dict[from_locale] = to_scr

    return likely_script_dict, frozenset(representative_locales)

    elif len(inp) == 2:

        return ord(inp[0]), ord(inp[1])

    else:

        assert len(inp) == 3

        assert len(inp) == 3, f'Expects a 3-character string, but "{inp}" '

        base = ord(base)

        first = ord(inp[0]) - base

        second = ord(inp[1]) - base

    print('});')

def read_and_dump_likely_data(icu_data_dir):

def read_and_dump_likely_data(cldr_source_dir):

    """Read and dump the likely-script data."""

    likely_subtags_txt = os.path.join(icu_data_dir, 'misc', 'likelySubtags.txt')

    likely_subtags_txt = os.path.join(cldr_source_dir,

                                      'common', 'supplemental', 'likelySubtags.xml')

    likely_script_dict, representative_locales = read_likely_subtags(

        likely_subtags_txt)

    icu_data_dir = os.path.join(

        source_root,

        'external', 'icu', 'icu4c', 'source', 'data')

    cldr_source_dir = os.path.join(source_root, 'external', 'cldr')

    print('// Auto-generated by %s' % sys.argv[0])

    print()

    likely_script_dict = read_and_dump_likely_data(icu_data_dir)

    likely_script_dict = read_and_dump_likely_data(cldr_source_dir)

    read_and_dump_parent_data(icu_data_dir, likely_script_dict)