site-aurore/gitlabCI.py

from markdown2 import markdown_path
from typing import List, Tuple
from datetime import date
from os import walk, system


markdown_dir = 'markdown'
html_dir = 'templates'
img_dir = 'images'
tab = ''


#################################
#          Build Stage          #
#################################

def build() -> None:
    """
    Build all files needed for the web site.

    Build HTML files from Markdown files.
    Copy images from Markdown directory to HTML directory.
    Build events template.
    """
    md_files = get_sources()
    create_html_files(md_files)
    copy_files(md_files)
    create_templates(md_files)

    exit(0)


def change_img_url(html: str, html_path: str) -> str:
    """
    Change the URL from each image tag inside the HTML code. The change is done only if it is a local address.

    :param html: The original HTML code
    :type html: str
    :param html_path: Path of the HTML code
    :type html_path: str
    :return: The modify HTML code
    :rtype: str

    :Example:

    >>> change_img_url('<img src="images/img.png" />', '13-04-2019-Example')
    '<img src="templates/13-04-2019-Example/images/img.png" />'
    >>> change_img_url('<img src="http://example.com" />', '13-04-2019-Example')
    '<img src="http://example.com" />'
    """
    img_pos = html.find("<img ")
    if img_pos != -1:
        new_html = html[:img_pos]
    else:
        new_html = html

    while img_pos != -1:
        src_pos = html.find("src=", img_pos+5)
        src_end = html.find(html[src_pos + 4], src_pos + 5)
        src = html[src_pos+5:src_end]
        if src[:4] != 'http':
            src = html_path[:html_path.rfind('/')] + '/' + src
        new_html += html[img_pos:src_pos+5] + src

        img_pos = html.find("<img ", src_end)
        if img_pos != -1:
            new_html += html[src_end:img_pos]
        else:
            new_html += html[src_end:]

    return new_html


def copy_files(md_files: List[str]) -> None:
    """
    Copy files and directories from Markdown directory to HTML directory.

    :param md_files: List of all Markdown files path in the Markdown directory
    :type md_files: List[str]
    """
    for md_file in md_files:
        directory = md_file[:md_file.rfind('/')]
        system('cp -r ' + directory.replace(' ', '\\ ') + '/* ' + directory.replace(' ', '_').replace(markdown_dir, html_dir))
        rm_file = md_file.replace(markdown_dir, html_dir).replace(' ', '\\ ')
        rm_file = rm_file[:rm_file.rfind('/')+1].replace('\\ ', '_') + rm_file[rm_file.rfind('/')+1:]
        system('rm ' + rm_file)


def create_html_files(md_files: List[str]) -> None:
    """
    Build HTML files from Markdown files.

    :param md_files: List of all Markdown files path in the Markdown directory
    :type md_files: List[str]
    """
    for md_file in md_files:
        original_html = markdown_path(md_file)  # Conversion from markdown to HTML
        html_directory = html_dir + '/' + md_file.split('/')[1].replace(' ', '_')
        html_filename = md_file[:-3].replace(markdown_dir, html_dir).replace(' ', '_') + '.html'
        system('mkdir ' + html_directory)  # Creation of the HTML directory
        final_html = change_img_url(original_html, html_filename)

        title = md_file[md_file.rfind('/') + 1:].split('-')[3][:-3]
        event_date = md_file[md_file.rfind('/')+1:][:10].replace('-', '/')
        final_html = '<h2 class="major"><div class="title">' + title + '</div><div class="date">' + event_date + \
                     '</div></h2>\n' + final_html

        with open(html_filename, 'w', encoding='utf-8') as html_file:
            html_file.write(final_html)  # Write HTML inside the file


def create_templates(md_files: List[str]) -> None:
    """
    Build events and index template.

    :param md_files: List of all Markdown files path in the Markdown directory
    :type md_files: List[str]
    """
    events_html = '<h2 class="major">Évènements et foyers</h2>\n'
    index_html = ''
    if len(md_files) != 0:
        for md_file in md_files:
            event_date, title, event = parse_md_filename(md_file.replace(' ', '_'))
            if (event_date, date, title) == (date.min, '', ''):
                continue

            events_html += '<a class="event" href="#event-' + event_date.strftime('%d-%m-%Y') + '-' + title + '">\n \
                        <div class="date">' + event_date.strftime('%a %d %b %Y') + '</div>\n \
                        <div class="title">' + title.replace('_', ' ') + '</div>\n \
                        <div class="description">\n' + event + '</div>\n \
                     </a>\n'

            index_html += '<article id="event-' + event_date.strftime(
                '%d-%m-%Y') + '-' + title + '"></article>\n\t\t\t\t'
    else:
        events_html += "<p>Pas d'évènements prévus à ce jour :(</p>\n"

    with open('templates/events.html', 'w', encoding='utf-8') as events_template:
        events_template.write(events_html)

    initial_index_html = ''
    with open("index.html", "r", encoding='utf-8') as index_template:
        for line in index_template:
            initial_index_html += line

    with open("index.html", "w", encoding='utf-8') as index_template:
        position = initial_index_html.find('<!-- Events articles -->')
        index_template.write(initial_index_html[:position] + index_html + initial_index_html[position:])


def del_links(html: str) -> str:
    """
    Delete the link tags (<a>) inside of the HTML code.

    :param html: HTML code with link tags
    :type html: str
    :return: HTML code without link tags
    :rtype: str

    :Example:

    >>> del_links('<p>An <a href="http://example.com">example</a></p>')
    '<p>An example</p>'
    """
    new_html = ""
    i = 0
    opened = False
    while i < len(html) - 4:  # Not going out of range
        if html[i:i + 3] == "<a ":  # Beginning of an opening link tag
            for j in range(3, len(html) - i - 3):
                if html[i + j] == ">":  # Ending of the opening link tag
                    i += j + 1
                    break
            opened = True
        elif html[i:i + 4] == "</a>" and opened:  # Closing link tag
            i += 4
            opened = False
        else:
            new_html += html[i]
            i += 1

    return new_html


def get_sources() -> List[str]:
    """
    Search and return markdown files path used as sources.

    :return: The markdown files
    :rtype: List[str]
    """
    md_files = []
    for (root, dirs, files) in walk(markdown_dir):
        for file in files:
            if file[-3:] == '.md':  # It is a markdown file
                md_files.append(root + '/' + file)

    return md_files


def parse_md_filename(filename: str) -> Tuple[date, str, str]:
    """
    Parse the Markdown file to have the event date, the title and the description.

    :param filename: Filename of the Markdown file.
                     Format must be: day-month-year-title.md
    :type filename: str
    :return: If the filename format is correct: (Event Date, Title, Description)
             If it is not correct: (date.min, '', '')
    :rtype: Tuple[date, str, str]

    :Example:

    >>> parse_md_filename('13-06-2019-Example.md')
    (date(2019, 06, 13), Example, *Content of the Markdown file without link tags*)
    >>> parse_md_filename('13-06-2019-BadExample')
    (date.min, '', '')
    >>> parse_md_filename('Bad-Example')
    (date.min, '', '')
    """
    split = filename[filename.rfind('/') + 1:].split('-')
    if len(split) == 4:  # Format is good : day_month_year_title.md
        day, month, year, title = split
        title = title[:-3]  # Remove the extension
        event_date = date(int(year), int(month), int(day))
    else:
        return date.min, '', ''

    event = ''
    nb_line = 0
    with open(filename.replace(markdown_dir, html_dir).replace('.md', '.html'), 'r', encoding='utf-8') as html_file:
        for line in html_file:
            if nb_line > 0:
                event += line
            nb_line += 1
    event = del_links(event)

    return event_date, title, event