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

###########
# prototyping downloading and converting mw page content to html
###########

# OVERVIEW:
# * creating one single html page
#     * replace {{youtube/vimeo}} with video tags
#     * replace galleries with rows of images
# request  all the pages
# build index
# build all pages

import xml.etree.ElementTree as ET
import html5lib, re, pprint
from mmdc_modules import api_request, api_page, api_thumb_url, pandoc2html, parse_work, api_file_url, replace_gallery, replace_video, gallery_exp, video_exp

template = open("web/page-template.html", "r")
template =  template.read()

# download mw work page
pageid='15965'#Qq  #'16025'  #'15986'Jozeph #'16025'Mina
work = 'Q' #'Mina'#'User:Joak/graduation/catalog1'
workpage_mw = api_page(pageid, 'content')

# parse workpage_mw
workpage_mw = replace_gallery(workpage_mw)
workpage_mw = replace_video(workpage_mw)
workdict = parse_work(work, workpage_mw) # create dictionary workpage_mw template 
for key in workdict.keys():
    if key in ['Extra', 'Description', 'Bio']:        
        workdict[key] =  pandoc2html(workdict[key].encode('utf-8'))

# fill template with dictionary/mw_page values
workpage_html =  template.format(title=workdict['Title'], creator=workdict['Creator'], date=workdict['Date'], website=workdict['Website'], thumbnail=workdict['Thumbnail'], bio=workdict['Bio'],description=workdict['Description'], extra=workdict['Extra']  )

# parse workpage_html # process html: img full url
tree = html5lib.parse(workpage_html, namespaceHTMLElements=False)
imgs = tree.findall('.//img')
for img in  imgs:
    src = img.get('src')
    newsrc = api_file_url(src)
    if newsrc:
        img.set('src', newsrc)                    

# save workpage_html
workpage_html = ET.tostring(tree)      
work_filename = 'web/{}-{}-{}.html'.format(workdict['Date'], (workdict['Creator'].encode('ascii', 'ignore')).replace(' ','_'), pageid)
work_file = open(work_filename, "w")
work_file.write(workpage_html)
work_file.close()