diff --git a/sort.py b/sort.py index 431bd02b..dc13ce3e 100755 --- a/sort.py +++ b/sort.py @@ -2,82 +2,111 @@ # coding: utf-8 """ - The approach taken is explained below. I decided to do it simply. - Initially I was considering parsing the data into some sort of - structure and then generating an appropriate README. I am still - considering doing it - but for now this should work. The only issue - I see is that it only sorts the entries at the lowest level, and that - the order of the top-level contents do not match the order of the actual - entries. - - This could be extended by having nested blocks, sorting them recursively - and flattening the end structure into a list of lines. Revision 2 maybe ^.^. +This script sorts the entries in the README.md file. It clusters the lines +into blocks based on indentation and sorts each block individually. """ -def sort_blocks(): - # First, we load the current README into memory - with open('README.md', 'r') as read_me_file: - read_me = read_me_file.read() - - # Separating the 'table of contents' from the contents (blocks) - table_of_contents = ''.join(read_me.split('- - -')[0]) - blocks = ''.join(read_me.split('- - -')[1]).split('\n# ') - for i in range(len(blocks)): - if i == 0: - blocks[i] = blocks[i] + '\n' - else: - blocks[i] = '# ' + blocks[i] + '\n' - - # Sorting the libraries - inner_blocks = sorted(blocks[0].split('##')) - for i in range(1, len(inner_blocks)): - if inner_blocks[i][0] != '#': - inner_blocks[i] = '##' + inner_blocks[i] - inner_blocks = ''.join(inner_blocks) - - # Replacing the non-sorted libraries by the sorted ones and gathering all at the final_README file - blocks[0] = inner_blocks - final_README = table_of_contents + '- - -' + ''.join(blocks) - - with open('README.md', 'w+') as sorted_file: - sorted_file.write(final_README) - -def main(): - # First, we load the current README into memory as an array of lines +def sort_readme(): + # Load the current README into memory as an array of lines with open('README.md', 'r') as read_me_file: read_me = read_me_file.readlines() - # Then we cluster the lines together as blocks - # Each block represents a collection of lines that should be sorted - # This was done by assuming only links ([...](...)) are meant to be sorted - # Clustering is done by indentation + # Cluster lines into blocks based on indentation blocks = [] + current_block = [] last_indent = None + for line in read_me: - s_line = line.lstrip() - indent = len(line) - len(s_line) - - if any([s_line.startswith(s) for s in ['* [', '- [']]): + stripped_line = line.lstrip() + indent = len(line) - len(stripped_line) + + # Detecting list items by starting character + if stripped_line.startswith(('* [', '- [')): if indent == last_indent: - blocks[-1].append(line) + current_block.append(line) else: - blocks.append([line]) + if current_block: + blocks.append(current_block) + current_block = [line] last_indent = indent else: + if current_block: + blocks.append(current_block) + current_block = [] blocks.append([line]) last_indent = None - with open('README.md', 'w+') as sorted_file: - # Then all of the blocks are sorted individually - blocks = [ - ''.join(sorted(block, key=str.lower)) for block in blocks - ] - # And the result is written back to README.md - sorted_file.write(''.join(blocks)) - - # Then we call the sorting method - sort_blocks() - + if current_block: + blocks.append(current_block) + + # Sort each block individually + sorted_blocks = [] + for block in blocks: + if block[0].lstrip().startswith(('* [', '- [')): + sorted_blocks.append(sorted(block, key=lambda s: s.lower())) + else: + sorted_blocks.append(block) + + # Write the sorted blocks back to the README.md file + with open('README.md', 'w') as sorted_file: + for block in sorted_blocks: + sorted_file.write(''.join(block)) if __name__ == "__main__": - main() + sort_readme() +#!/usr/bin/env python +# coding: utf-8 + +""" +This script sorts the entries in the README.md file. It clusters the lines +into blocks based on indentation and sorts each block individually. +""" + +def sort_readme(): + # Load the current README into memory as an array of lines + with open('README.md', 'r') as read_me_file: + read_me = read_me_file.readlines() + + # Cluster lines into blocks based on indentation + blocks = [] + current_block = [] + last_indent = None + + for line in read_me: + stripped_line = line.lstrip() + indent = len(line) - len(stripped_line) + + # Detecting list items by starting character + if stripped_line.startswith(('* [', '- [')): + if indent == last_indent: + current_block.append(line) + else: + if current_block: + blocks.append(current_block) + current_block = [line] + last_indent = indent + else: + if current_block: + blocks.append(current_block) + current_block = [] + blocks.append([line]) + last_indent = None + + if current_block: + blocks.append(current_block) + + # Sort each block individually + sorted_blocks = [] + for block in blocks: + if block[0].lstrip().startswith(('* [', '- [')): + sorted_blocks.append(sorted(block, key=lambda s: s.lower())) + else: + sorted_blocks.append(block) + + # Write the sorted blocks back to the README.md file + with open('README.md', 'w') as sorted_file: + for block in sorted_blocks: + sorted_file.write(''.join(block)) + +if __name__ == "__main__": + sort_readme()