制作 rime 五笔86 单字词库

wubi86_single.py

# forked from https://github.com/yekingyan/rime-wubi-86-single/blob/master/wubi86_single.py


import re


INPUT_FILE_PATH = "wubi86.dict.yaml"
OUTPUT_FILE_PATH = "wubi86.dict.yaml.single"


def line_key(line):
    fields = line.split('\t')
    code = fields[1]
    weight = fields[2] or 0
    return (code, -int(weight, 10))


def get_trim_line():
    stay_line = []

    with open(INPUT_FILE_PATH, encoding="utf8") as f:
        trim_working = False
        for line in f:
            if line == '...\n':
                stay_line.append(line)
                trim_working = True
                continue
            if not trim_working or len(line.strip()) == 0:
                stay_line.append(line)
            else:
                line_list = line.split('\t')
                word = line_list[0]
                code = line_list[1]
                if not word.startswith("#") and (len(word) == 1 or code.startswith("z")):
                    stay_line.append(line)

    splitter_index = stay_line.index('...\n')
    config_lines = stay_line[0 : splitter_index + 2]
    word_lines = stay_line[splitter_index + 2 :]

    word_lines.sort(key=lambda line: float((line + '\t0').split('\t')[2]), reverse=True);
    word_lines.sort(key=lambda line: line.split('\t')[1]);
    word_lines.sort(key=lambda line: len(line.split('\t')[1]));

    return config_lines + word_lines


def main():
    stay_line = get_trim_line()
    with open(OUTPUT_FILE_PATH, "w+", encoding="utf8") as f:
        for line in stay_line:
            f.writelines(line)


if __name__ == '__main__':
    main()