pzimediadesign/mmdc_wiki2web.py

#! /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, pprint
from mmdc_modules import api_page, pandoc2html, parse_work, api_file_url, replace_gallery, replace_video, index_addwork, write_html_file, mw_cats
from argparse import ArgumentParser

p = ArgumentParser()
p.add_argument("--host", default="pzwiki.wdka.nl")
p.add_argument("--path", default="/mw-mediadesign/", help="nb: should end with /")
p.add_argument("--category", "-c", nargs="*", default=[], action="append", help="category to query, use -c foo -c bar to intersect multiple categories")
args = p.parse_args()
print args
########
# QUERY API
########
sid = '1234'
useragent = "Mozilla/5.001 (windows; U; NT4.0; en-US; rv:1.0) Gecko/25250101"
endpoint = "http://pzwiki.wdka.nl/mw-mediadesign/api.php?format=json&"

########
# CREATE INDEX
########
memberpages=mw_cats(args)
#memberpages['Ctrl-F Reader','As We Speak']
print 'memberpages', memberpages


########
# Templates
########
page_template = open("web/page-template.html", "r")
index_file = open('web/index-template.html', 'r') 
index_tree = html5lib.parse(index_file, namespaceHTMLElements=False)
index_container = index_tree.find(".//div[@class='isotope']") #maybe id is important, to destinguish it

########
# CREATE PAGE 
########
for member in memberpages:
    print ' member', member
    # download mw work page
#    pageid=member['pageid']
#    pagetitle=(member['title'].encode('utf-8'))
    workpage_mw = api_page(member, 'content')
    workpage_mw = replace_gallery(workpage_mw)
    workpage_mw = replace_video(workpage_mw)
    workdict = parse_work(member, workpage_mw) # create dictionary workpage_mw template

    # Only parse pages with Creator, Title, Thumbnail
    
    if len(workdict['Creator'])>1 and len(workdict['Title'])>1 and len(workdict['Description'])>1 and len(workdict['Thumbnail'])>1:    
        for key in workdict.keys(): # convert Extra, Description, Bio to HTML
            if key in ['Extra', 'Description', 'Bio'] and workdict[key]:
                workdict[key] =  pandoc2html( (workdict[key].decode('utf-8')))
            elif key in ['Creator']:
                workdict[key] =  workdict[key].replace(',','' ) #remove comma        

        #replace empty dict values with ' ' # to avoid empty tags
        for key in workdict.keys():
            if workdict[key] is '' and key is not 'Thumbnail':
                workdict[key] = ' '
            elif key is 'Thumbnail' and workdict[key]:
                img = '<img id="thumnail" src="{}" />'.format(workdict[key])
                # append img to text
                workdict[key] =  workdict[key] # + img
                #print 'THUMB', workdict[key]
    
            if type(workdict[key]) is unicode:
                workdict[key]=workdict[key].encode('utf-8')

        # create work page
        page_tree = html5lib.parse(page_template, namespaceHTMLElements=False)
        page_title = page_tree.find('.//title')
        page_title.text=workdict['Title']
        page_creator = page_tree.find('.//h2[@id="creator"]')
        page_creator.text=workdict['Creator']
        page_title_date = page_tree.find('.//p[@id="title"]')
        page_title_date.text="{} {}".format(workdict['Title'], workdict['Date'])
        page_description = page_tree.find('.//div[@id="description"]')
        page_description_el = ET.fromstring('<div>'+workdict['Description']+'</div>')
        page_description.extend(page_description_el)
        page_bio = page_tree.find('.//div[@id="bio"]')
        page_bio_el = ET.fromstring('<div>'+workdict['Bio']+'</div>')
        page_bio.extend(page_bio_el)
        page_sortArea_title = page_tree.find('.//div[@id="sortArea"]/p')
        page_sortArea_title.text =workdict['Title'] 
        page_extra = page_tree.find('.//div[@id="extra"]')
        page_extra_el = ET.fromstring('<div>'+workdict['Extra']+'</div>')
        page_extra.extend(page_extra_el)
        page_website = page_tree.find('.//p[@class="hightlightSidebar"]/a')
        page_website.set('href', workdict['Website'])
        page_website.text=workdict['Website']
        page_thumb = page_tree.find('.//img[@id="thumbnail"]')
        page_thumb.set('src', workdict['Thumbnail'])
        # give work page's imgs full url
        imgs = page_tree.findall('.//img')
        for img in  imgs:
            if img.get('id') is not 'logo':
                src = img.get('src')
                newsrc = api_file_url(src) ## MOVE FULL URl OPERATION TO MW CONTENT 
                if newsrc:
                    img.set('src', newsrc)                    
        # save work page
        creator = workdict['Creator']#.decode('ascii', 'ignore')
        creator = creator.replace(' ','_')
        work_filename = 'web/{}-{}.html'.format(workdict['Date'], creator)
        write_html_file(page_tree, work_filename)

        #######
        # INDEX
        #######
        # insert work to index        
        index_addwork( parent=index_container,
                       workid=key,
                       href=work_filename.replace('web/',''),
                       title=workdict['Title'],#.decode('utf-8'),
                       creator=workdict['Creator'],#.decode('utf-8'),
                       date=workdict['Date'],                  
                       thumbnail=workdict['Thumbnail']
        )
     #   print '----', workdict['Title']
     #   print ET.tostring(tree)
        
print index_tree, type(index_tree)        
write_html_file(index_tree, 'web/index.html')
print
print