Dynastie/generators/index.py

import datetime
from xml.dom.minidom import parse
from dynastie.generators.generator import DynastieGenerator
from django.db import models

# TODO : links between indexes
# TODO : gzip

class Index(DynastieGenerator):

    cur_page = 0
    cur_article = 0
    articles_per_page = 0

    def createArticle(self, article, dom, article_elem, root):
        values = {}
        values['title'] = article.title
        values['author'] = article.author.first_name
        values['date'] = article.creation_date.strftime("%d/%m/%Y")
        values['content'] = ''

        self.simpleTransform(values, dom, article_elem, root)

    def parse(self, articles, dom, root):
        for node in root.childNodes:
            if node.prefix == 'dyn':
                if node.localName == 'articles':
                    articles_elem = dom.createElement("articles")
                    for i in range(0, self.articles_per_page):
                        article_elem = dom.createElementNS(self.URI, "article")
                        self.createArticle(articles[self.cur_article], dom, article_elem, node)
                        articles_elem.appendChild(article_elem)

                        self.cur_article = self.cur_article + 1
                        if self.cur_article == articles.count():
                            break
                    root.replaceChild(articles_elem, node)
                continue
            if node.hasChildNodes():
                self.parse(articles, dom, node)
        return

    def generate(self, blog, src, output):
        from dynastie.models import *

        self.report = ''

        if not os.path.exists(src + '/_index.html'):
            self.addError('No _index.html found, exiting')
            return self.report

        try:
            dom = parse(src + '/_index.html')
        except xml.dom.DOMException as e:
            self.addError('Error parsing _index.html : ' + e)
            return self.report
            
        article_nodes = dom.getElementsByTagNameNS(self.URI, "articles")

        if not article_nodes is None:
            self.articles_per_page = int(article_nodes[0].getAttribute("limit"))
        else:
            self.addError('No tag dyn:articles found')

        now = datetime.datetime.now()
        cur_year = now.year
        articles = Article.objects.filter(creation_date__year=cur_year)

        if articles.count() < self.articles_per_page:
            articles = Article.objects.all()[:self.articles_per_page]

        filename = 'index.html'
        while self.cur_article < articles.count():
            nodes = dom.getElementsByTagName("*")
            self.parse(articles, dom, nodes[0])
            self.writeIfNotTheSame(output + '/' + filename, nodes[0].toxml('utf8'))
            self.cur_page = self.cur_page + 1
            filename = 'index' + str(self.cur_page) + '.html'

        while os.path.exists(filename):
            self.addReport('Removing unused ' + filename)
            os.unlink(filename)
            self.cur_page = self.cur_page + 1
            filename = 'index' + str(self.cur_page) + '.html'

        return self.report