Generator for https://glyphwiki.org/wiki/Group:文字種別グリフ数

count_by_pattern.py

#!/usr/bin/env python3

# usage : ./count_by_pattern.py path/to/dump_newest_only.txt < template.txt

import argparse
import os.path
import re
import sys
import time
from typing import NamedTuple, Optional, Pattern, Sequence, TextIO, Union


class Stat:
    __slots__ = ("pattern", "count", "sample")

    def __init__(self, pattern: Pattern[str]):
        self.pattern = pattern
        self.count = 0
        self.sample = "notfound"


class StatLine(NamedTuple):
    desc_type: str
    desc_pattern: str
    stat: Stat


PageLine = Union[StatLine, str]


def parse_stat_line(line: str) -> Optional[StatLine]:
    if not line.startswith(","):
        return None
    columns = line.split(",")
    if len(columns) != 6:
        return None

    desc_type = ",".join(columns[:3])
    desc_pattern = columns[5]

    if not (desc_pattern.startswith(">") and desc_pattern.endswith("<")):
        return None

    pattern = re.compile(desc_pattern[1:-1].replace("，", ","))
    stat = Stat(pattern)

    return StatLine(
        desc_type=desc_type,
        desc_pattern=desc_pattern,
        stat=stat)


def print_output(lines: Sequence[PageLine], dumppath: str) -> None:
    for line in lines:
        if isinstance(line, StatLine):
            print(
                f"{line.desc_type},{line.stat.count},[[{line.stat.sample}]],"
                f"{line.desc_pattern}")
            continue

        if line.startswith("filename : "):
            filename = os.path.basename(dumppath)
            print(f"filename : {filename}")
        elif line.startswith("filesize : "):
            filesize = os.path.getsize(dumppath)
            print(f"filesize : {filesize}")
        elif line.startswith("timestamp : "):
            mtime = os.path.getmtime(dumppath)
            timestamp = time.ctime(mtime)
            print(f"timestamp : {timestamp}")
        else:
            print(line, end="")


def count_by_pattern(dumppath: str, template: TextIO) -> None:
    lines: list[PageLine] = [
        parse_stat_line(line.rstrip()) or line
        for line in template
    ]
    stats = [
        line.stat
        for line in lines
        if isinstance(line, StatLine)
    ]

    for line in open(dumppath):
        stripped_line = line.strip()
        columns = stripped_line.split("|")
        if len(columns) != 3:
            continue
        name = columns[0].strip()
        for stat in stats:
            if stat.pattern.match(stripped_line):
                stat.count += 1
                stat.sample = name

    print_output(lines, dumppath)


def main():
    parser = argparse.ArgumentParser()

    parser.add_argument("dump")
    parser.add_argument(
        "--template", nargs="?", type=argparse.FileType("r"),
        default=sys.stdin)

    args = parser.parse_args()
    count_by_pattern(args.dump, args.template)


if __name__ == "__main__":
    main()