ソート済みテキストを二分探索する look コマンドの pure Python 実装

looks.py

#!/usr/bin/env python3
# see https://chalow.net/2023-10-16-1.html

import sys

class Searcher:
        def __init__(self, filename, buffer_size=10000):
                self.f = open(filename, 'rb')
                self.f.seek(0,2)
                self.length = self.f.tell()
                self.buffer_size = buffer_size

        def go_to_line_top(self, p):
                self.f.seek(p)
                self.f.readline()
                line_last = self.f.tell()
                self.f.seek(max(0, p - self.buffer_size))
                while True:
                        line = self.f.readline()
                        if not line or line_last == self.f.tell():
                                break

                self.f.seek(line_last - len(line))
                return line

        def find(self, string):
                low = 0
                high = self.length
                while low < high:
                        mid = (low+high)//2
                        line = self.go_to_line_top(mid)
                        if line < string:
                                low = mid + 1
                        else:
                                high = mid

                self.go_to_line_top(low)
                result = []
                while True:
                        line = self.f.readline()
                        if not line or not line.startswith(string):
                                break
                        result.append(line.rstrip(b'\r\n'))

                return result


def main():
        sortedFile = sys.argv[1]
        lookup = Searcher(sortedFile)

        for searchString in sys.stdin:
                searchString = searchString.rstrip().encode('utf8')
                for line in lookup.find(searchString):
                        print(line.decode())

        return 0

if __name__ == '__main__':
        exit(main())