#!/usr/bin/env python
# -*- coding: utf-8 -*-
import re
import pandoc


class TextFile:
    '''
    Basic text file class
    '''
    def __init__(self, filename, source):
        self.filename = filename
        self.source = source

    @property
    def lines(self):
        '''
        Returns list of lines from source.
        Leading and trailing blank lines are removed.
        '''
        return self.source.strip().split('\n')

    @property
    def firstline(self):
        return self.lines[0]


class TOCItem:
    '''
    Table of Contents Line Item
    '''
    def __init__(self, heading, level):
        self.heading = heading
        self.level = level

    @property
    def header_id(self):
        # - Remove all formatting, links, etc.
        # - Remove all punctuation, except underscores, hyphens, and periods.
        # - Replace all spaces and newlines with hyphens.
        # - Convert all alphabetic characters to lowercase.
        # - Remove everything up to the first letter (identifiers may not begin
        #   with a number or punctuation mark).

        remove = re.compile('^[^a-zA-Z]|[+~!@#$%^&*\(\){}\[\];:"\',<>?/\`]')
        header_id = self.heading.lower().replace('\n', ' ').replace(' ', '-')
        return remove.sub('', header_id)


class Page(TextFile):
    '''
    Represents single page source file.
    '''

    def __init__(self, filename, source, format='markdown'):
        TextFile.__init__(self, filename, source)
        self.doc = pandoc.Document()
        self.doc._input(self.source, format=format)

    @property
    def markdown(self):
        return self.doc.markdown

    @property
    def title(self):
        '''
        Returns a tuple containing a string representing the page title
        and an integer representing the level in the document outline
        hierarchy.

        The level is preserved so that page titles can be properly
        positioned (indented) in the table of contents.

        If the page has a pandoc title block, the title is retreived
        from there and returned with a level of 1

        If a title block is not found, the first heading is returned
        with the corrisponding heading level.

        Finally, if no title block or headings are found in the page,
        the filename is returned as the title with underscores changed
        to spaces.
        '''
        if self.lines[0].startswith('%'):
            title = (self.lines[0].split(' ', 1)[1].strip(), 1)
        elif self.toc:
            title = self.toc[0]
        else:
            title = (self.filename.replace('_', ' '), 1)

        return title

    @property
    def toc(self):
        '''
        Returns a list of TOCItems of the page.
        '''
        toc = []
        heading = ''
        for line in self.lines:
            if heading and re.match('[=]{2}', line):
                toc.append(TOCItem(heading, 1))
                continue
            if heading and re.match('[-]{2}', line):
                toc.append(TOCItem(heading, 2))
                continue
            if re.match('#+.+[A-z|0-9]', line):
                level = len(re.match('#+', line).group())
                toc.append(TOCItem(line.split(' ', 1)[1].strip(), level))
                continue
            heading = line.strip()

        return toc

    @property
    def htmlfile(self):
        return '.'.join(self.filename.split('.')[:-1]) + '.html'