pzimediadesign/md_index.py

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

##############
# Edit the index sorting articles according to topic, section and issue
#####

#
import xml.etree.ElementTree as ET
import html5lib, urllib2, json, pprint, subprocess
from urllib import quote as quote
import re

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&"
allworks = {}
mainkeys = ['Thumbnail','Date','Creator']

def api_request(action, pagename): #get page: content, metadata, images, imageifnp 
    url = endpoint + action.format(pagename)
    request = urllib2.urlopen(url)
    jsonp = json.loads(request.read() )
    json_dic= (jsonp.get('query').get('pages'))
    page_id =  json_dic.keys()[0]
    page_content = json_dic.get(page_id)
    print 'API URL:', url
    return page_content


def api_page(pageid, info):
    if info == 'content':
        api_response = api_request('action=query&pageids={}&prop=revisions&rvprop=content', pageid)
        page_content = ((api_response.get('revisions'))[0])['*']
    elif info == 'metadata':
        page_content = api_request('action=query&pageids={}&prop=info', pageid)
    elif info == 'articleimgs':
        page_content = api_request('action=query&pageids={}&prop=images', pageid)
    elif info == 'imageinfo':
        page_content = api_request('action=query&pageids={}&prop=imageinfo&iiprop=url&iiurlwidth=300', pageid)  # iiurlwidht dermines with of thumbnail         
    return page_content

def api_page_content(pagename):
    content = ((page.get('revisions'))[0])['*']
    return content
#    print json.dumps( revisions, sort_keys=True, indent=4) ## see response



def api_thumb_url(filename):
   '''get thumbnail url of image'''
   page_content_dict = api_page(filename, 'imageinfo')   
   if 'imageinfo' in page_content_dict.keys():
       thumburl = ((page_content_dict.get('imageinfo'))[0].get('thumburl'))
       return page_content_dict


def find_imgs(article):
    '''get all internal images from published articles:
    * query all images in  a article
    * for each image request a thumbnail
    * get the thumburl    
    '''
    page_content_dict = api_page(article, 'articleimgs')
    if 'images' in page_content_dict.keys():
        images_list = page_content_dict.get('images')
#        print 'images_list', images_list
        thumbs_list = []
        for img in images_list: # all images in article
            title = img.get('title')
            thumburl_json = api_thumb_url(title)
            if thumburl_json != None:
                thumburl = (thumburl_json.get('imageinfo')[0]).get('thumburl')
                thumbs_list.append(thumburl)
            else:
                thumburl = None
                
        return thumbs_list 

    
def parse_work_page(title, content):
    content = content.encode('utf-8')
    if re.match('\{\{\Graduation work', content):
        work_dict = {}
        work_dict['Title']=title
        template, extra = (re.findall('\{\{Graduation work\n(.*?)\}\}(.*)', content, re.DOTALL))[0]
        # template's key/value pair
        keyval = re.findall('\|(.*?)\=(.*?\n)', template, re.DOTALL)
        extra = ('Extra', extra)
        keyval.append(extra)
        keys = [keyval[i][0] for i in range(len(keyval))]
        #checkkeys: list of mainkeys present, w/ values, in tuples [(key, val),(key, val)...]
        checkkeys = [keyval[i] for i in range(len(keyval)) if keyval[i][0] in mainkeys and len(keyval[i][1])>3] 
        if len(checkkeys) == 3 : # checkkeys contains all mainkeys and values 
            for pair in keyval:
                key = pair[0]
                val = pair[1]
                val = val.replace('\n','')            
                if 'Creator' in key:
                    val = val.replace(', ', '')
                work_dict[key]=val            
            return work_dict

def api_category(category, year):
    '''Finds all pages within  category and eact to allworks dictionary'''
    category =  category.replace(' ', '_')
    if year:
        api_url = endpoint + 'action=query&list=categorymembers&cmlimit=500&cmtitle=Category:{}&cmtitle=Category:{}'.format(category, year) #BUG: API only queries last cmtitle: YEAR
    else:
        api_url = endpoint + 'action=query&list=categorymembers&cmlimit=500&cmtitle=Category:{}'.format(category)
    
    request = urllib2.urlopen(api_url)
    jsonp = json.loads(request.read())    

    for page in  jsonp['query']['categorymembers']:
        print 'Page:', page
        title = ((page['title']).encode('utf-8') ).replace(" ", "_") #snakecase for page titles
        pageid = page['pageid']
        ## NOTE: instead of using page name to query page, use PAGE ID 
        article = api_page(pageid, 'content')
        print title
        #        pprint.pprint(article)
        work = parse_work_page(title, article)
        if work:
            allworks[pageid] = work #dictionary(allworks) entry
            print work
        else:
            print 'WORK DOES NOT CONTAIN REQUIRED CONTENT'
        print '-------------'
        print 

api_category('Graduation work', '2013')
#pprint.pprint(allworks)

# save json 
json_allworks = open('md_allworks.json', 'w')
json.dump(allworks, json_allworks )
#print "wrote json dictionary to:", 'md_allworks.json'

## TO DO
# How do handle work['Extra'] value? some tiles  work['Extra'] contains: <gallery>, [[Pages]], text, etc
# Do template values need to be converted to html?

# Thumbnails need a full url