novel-stats/novel_stats/novel_stats.py

#!/usr/bin/python


import argparse
import collections
import tempfile

CHAPTER_MARKER = '## '
STATUS_MARKER = '[status]: # '
ACT_MARKER = '[act]: # '
# Standard markdown comment marker, supported by Pandoc and Calibre's ebook-convert.
COMMENT_MARKER = '[//]: # '


def count_words(line):
    count = 0

    for word in line.strip().split(' '):
        if not word.strip() or word == '*' or word.startswith('#'):
            continue

        count += 1

    return count


def main():
    # Better argument parsing
    parser = argparse.ArgumentParser()
    parser.add_argument(
        '-c',
        '--chapter',
        action='store_true',
        help='output chapter-by-chapter breakdown of word counts, including how many words in each chapter are tagged with which status',
    )
    parser.add_argument(
        '-a',
        '--act',
        action='store_true',
        help='output act-by-act breakdown of word counts (total only)',
    )
    parser.add_argument(
        '-pp',
        action='store_true',
        help='run markdown pre-processor, this allows for a multi-file input (e.g. each chapter in its own file), but requires the MarkdownPP python library',
    )
    parser.add_argument(
        'markdown_file',
        type=argparse.FileType('r'),
        help='The markdown file for the novel, main file if a multi-file novel',
    )
    arguments = parser.parse_args()

    mdfile = None

    if arguments.pp:
        # -pp flag to allow Markdown Preprocessing primarily to allow multi-file novel formatting
        # this is implemented using a temporary file created using python's buit-in tempfile library
        import MarkdownPP

        mdfile = tempfile.TemporaryFile(mode='w+')
        MarkdownPP.MarkdownPP(
            input=arguments.markdown_file, output=mdfile, modules=list(MarkdownPP.modules)
        )
        mdfile.seek(0)
    else:
        mdfile = arguments.markdown_file

    chapter_heading = None
    act_heading = None
    total_word_count = 0
    word_count_by_chapter = collections.defaultdict(int)
    word_count_by_status = collections.defaultdict(int)
    word_count_by_act = collections.defaultdict(int)
    status_by_chapter = collections.defaultdict(lambda: collections.defaultdict(int))
    current_status = None

    for line in mdfile.readlines():
        if line.startswith(CHAPTER_MARKER):
            word_count_by_act[act_heading] += word_count_by_chapter[chapter_heading]
            total_word_count += word_count_by_chapter[chapter_heading]

            chapter_heading = line[len(CHAPTER_MARKER) :].strip('()\n')

            # Count the words in chapter heading, because the chapter number and title count as words.
            if chapter_heading:
                word_count_by_chapter[chapter_heading] = count_words(chapter_heading)
                current_status = None
        # Modified to allow multiple statuses in a single chapter, can swap back and forth.
        elif line.startswith(STATUS_MARKER):
            if current_status is None:
                current_status = line[len(STATUS_MARKER) :].strip('()\n')
                if chapter_heading:
                    status_by_chapter[chapter_heading][current_status] = count_words(
                        chapter_heading
                    )
            else:
                current_status = line[len(STATUS_MARKER) :].strip('()\n')
        elif line.startswith(ACT_MARKER):
            act_heading = line[len(ACT_MARKER) :].strip('()\n')
            word_count_by_act[act_heading] = count_words(act_heading)
        elif line.startswith(COMMENT_MARKER):  # Don't count the words in a comment.
            pass
        else:
            line_word_count = count_words(line)
            word_count_by_chapter[chapter_heading] += line_word_count

            if current_status:
                word_count_by_status[current_status] += line_word_count
                status_by_chapter[chapter_heading][current_status] += line_word_count

    mdfile.close()

    # Do some final accounting after the last chapter.
    word_count_by_act[act_heading] += word_count_by_chapter[chapter_heading]
    total_word_count += word_count_by_chapter[chapter_heading]

    # -c or --chapter to give a chapter-by-chapter word count summary.
    if arguments.chapter:
        for chapter_heading, chapter_word_count in word_count_by_chapter.items():
            if chapter_heading is None:
                continue

            if len(status_by_chapter[chapter_heading]) > 1:
                print(f'chapter {chapter_heading}:')

                for chapter_status, status_count in status_by_chapter[chapter_heading].items():
                    print(f'\t {status_count:,} ({chapter_status})')
                print(f'\t {chapter_word_count:,} words (total)')
            elif len(status_by_chapter[chapter_heading]) == 1:
                chapter_status = list(status_by_chapter[chapter_heading].keys())[0]
                print(f'chapter {chapter_heading}: {chapter_word_count:,} ({chapter_status})')
            else:
                print(f'chapter {chapter_heading}: {chapter_word_count:,}')

        print()

    # -a or --act to give an act-by-act word count summary.
    if arguments.act:
        for act_heading, act_word_count in word_count_by_act.items():
            if act_heading is None:
                continue

            print(
                f'act {act_heading}: {act_word_count:,} words (~{act_word_count * 100// total_word_count}%)'
            )

        print()

    for status, status_word_count in word_count_by_status.items():
        print(
            f'{status}: {status_word_count:,} words (~{status_word_count * 100 // total_word_count}%)'
        )

    print(f'total: {total_word_count:,} words')


if __name__ == '__main__':
    main()