tail-from-p3.py

#!/usr/bin/env python3
# -*- mode: python; coding: utf-8 -*-

assert str is not bytes

import sys

def main():
    if len(sys.argv) not in (3, 4):
        print('usage: {} <keyword> <path> [final_keyword]'.format(sys.argv[0]), file=sys.stderr)
        exit(2)
    
    keyword = sys.argv[1]
    path = sys.argv[2]
    if len(sys.argv) >= 4:
        final_keyword = sys.argv[3]
    else:
        final_keyword = None
    
    f_out = sys.stdout
    
    buf = []
    key_line_i = None
    
    with open(path, 'r', encoding='utf-8', errors='replace') as f_in:
        for line_i, line in enumerate(f_in):
            if keyword.lower() in line.lower():
                key_line_i = line_i
            
            buf.append(line)
    
    if key_line_i is None:
        print('error: keyword not found in file', file=sys.stderr)
        exit(1)
    
    out_buf = buf[key_line_i:]
    
    for line in out_buf:
        f_out.write(line + '\n')
        
        if final_keyword is not None and final_keyword.lower() in line.lower():
            break

if __name__ == '__main__':
    main()

tail-from.py

#!/usr/bin/env python2
# -*- mode: python; coding: utf-8 -*-

#   этот файл занимал бы в 2 раза меньше строчек,
#       если он бы был написан НЕ на древней версии Python.
#
#   так как можно было НЕ указывать операции связанные с перекодировками
#       (а их тут не менее половины от всего кода)

from __future__ import absolute_import

assert str is bytes
assert unicode is not str

import sys, locale

def main():
    if len(sys.argv) not in (3, 4):
        print >> sys.stderr, 'usage: %s <keyword> <path> [final_keyword]' % sys.argv[0]
        exit(2)
    
    keyword_bytes = sys.argv[1]
    path_bytes = sys.argv[2]
    if len(sys.argv) >= 4:
        final_keyword_bytes = sys.argv[3]
    else:
        final_keyword_bytes = None
    
    assert isinstance(keyword_bytes, bytes)
    assert isinstance(path_bytes, bytes)
    assert final_keyword_bytes is None or isinstance(final_keyword_bytes, bytes)
    
    locale_encoding = locale.getpreferredencoding() or 'utf-8'
    fs_encoding = sys.getfilesystemencoding() or 'utf-8'
    
    keyword = unicode(keyword_bytes, locale_encoding, 'replace')
    path = unicode(path_bytes, locale_encoding, 'replace')
    if final_keyword_bytes is not None:
        final_keyword = unicode(final_keyword_bytes, locale_encoding, 'replace')
    else:
        final_keyword = None
    
    f_out = sys.stdout
    f_out_encoding = getattr(f_out, 'encoding', None) or 'utf-8'
    
    buf = []
    key_line_i = None
    
    with open(path, 'rb') as f_in:
        for line_i, raw_line in enumerate(f_in):
            assert isinstance(raw_line, bytes)
            
            line = unicode(raw_line, 'utf-8', 'replace')
            
            if keyword.lower() in line.lower():
                key_line_i = line_i
            
            buf.append(line)
    
    if key_line_i is None:
        print >> sys.stderr, 'error: keyword not found in file'
        exit(1)
    
    out_buf = buf[key_line_i:]
    
    for line in out_buf:
        assert isinstance(line, unicode)
        
        line_bytes = line.encode(f_out_encoding, 'replace')
        f_out.write(line_bytes + '\n')
        
        if final_keyword is not None and final_keyword.lower() in line.lower():
            break

if __name__ == '__main__':
    main()