# Extracts titles from a front page of the Sydney Morning Herald.
#
# Written by Eric Martin for COMP9021

# Note that some nonASCII characters do not display properly
# when the program is from from Idle, but do display properly
# when the program is run from the command line.


import re


def extract_text(line, pattern):
    extracted_text = pattern.search(line)
    if extracted_text:
        title = extracted_text.groups()[0]
        print(title.replace(' ', ''))

# We look for text of the form title=....>TITLE</a></h3>
full_title_pattern = re.compile('title=[^>]*>([^<]*)</a></h3>')
# In some cases, </a></3> is at the beginning of the next line
title_at_end_of_line_pattern = re.compile('[^>]*>([^<]*)\n$')
end_tags_at_start_of_next_line_pattern = re.compile('^</a></h3>')

with open('SMH.txt', 'r') as file:
    line = file.readline()
    for next_line in file:
        if end_tags_at_start_of_next_line_pattern.search(next_line):
            extract_text(line, title_at_end_of_line_pattern)
        else:
            extract_text(line, full_title_pattern)
        line = next_line
    # Process last line in the unique possible way
    extract_text(line, full_title_pattern)