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)
print 'memberpages', memberpages

#memberpages = api_pagesincategories('Graduation work', '2015') #list, containing dictionary of all pages ids. Example: [{u'ns': 0, u'pageid': 15974, u'title': u'Ahhhh'}, {u'ns': 0, u'pageid': 16005, u'title': u'Artyom-graduation-work'}]
#memberpages = [{u'ns': 0, u'pageid': 15982, u'title': u'The Aesthetics of Ethics'}]
#memberpages = [{u'ns': 0, u'pageid': 16005, u'title': u'Artyom-graduation-work'}]
#memberpages =   [{u'ns': 0, u'pageid': 16007, u'title': u'U ntitled'}]
#memberpages = [{u'ns': 0, u'pageid': 15965, u'title': u'Qq'}]
## output: memberpages [{u'ns': 0, u'pageid': 15982, u'title': u'The Aesthetics of Ethics'}]


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

index_file = 'web/index-template.html'
index_file = open(index_file, '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')

    # parse workpage_mw
    workpage_mw = replace_gallery(workpage_mw)
    workpage_mw = replace_video(workpage_mw)
    workdict = parse_work(member, workpage_mw) # create dictionary workpage_mw template
    
    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')) )



    print 'EXTRA', type (workdict['Extra']), workdict['Extra']
    # fill template with dictionary/mw_page values
    pprint.pprint(workdict)

    if len(workdict['Creator'])>1 and len(workdict['Creator'])>1:

        workpage_html =  page_template.format(title=('Title'),#workdict['Title']),
                                              creator=(workdict['Creator']),
                                              date=workdict['Date'],
                                              website=workdict['Website'],
                                              thumbnail=workdict['Thumbnail'],
                                              bio=(workdict['Bio']),
                                              description=('desc'),#workdict['Description']),
                                              extra=( '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:
            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)                    

        website = tree.find('.//p[@class="hightlightSidebar"]/a')
        if not website.get('href'): # if no website link. 
            #remove empty <a> .//p[@class="hightlightSidebar"]/a
            website_parent = tree.find('.//p[@class="hightlightSidebar"]')
            website_parent.remove(website)

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


        print 'DICT', type(workdict['Title'])
        # 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'],
        )

write_html_file(index_tree, 'web/index.html')
print
print