scopus/ak_wordpress.py

""" Create posts via wordpress API
    
*A. Kopmann 6.2.2017 (ak)*
"""


from datetime import datetime
import json
from pprint import pprint

from wordpress_xmlrpc import Client
from wordpress_xmlrpc import WordPressPost, WordPressComment
from wordpress_xmlrpc.methods.posts import GetPost, NewPost, EditPost
from wordpress_xmlrpc.methods.comments import NewComment, EditComment
from wordpress_xmlrpc.methods.taxonomies import GetTerms

from config import *

# Use Wordpress account - not the mysql credentials
wp = Client(wp_api_url, wp_user, wp_password) 


#
# Get category from slug name used in the configuration file
#

def wordpress_get_category(slug):
     """ Load taxonomy and search for the slug """
     catlist = wp.call(GetTerms('category'))

     for cat in catlist:
         if cat.slug == slug:
             return cat


#
# query post
#

def wordpress_get_post(wpid):
    """ Query post """
 
    try:
        post = wp.call(GetPost(wpid))
        #print post.title
    
        ret = 1
    except:
        #print "Post seems to be not available"
        ret = 0

    return ret


#
# create a post from a scopus query
#
def wordpress_post_by_scopus(data, category = []):
    """ Create a new post based on the Scopus information """

    #print data['abstracts-retrieval-response']
 
    try:
        coredata = data['abstracts-retrieval-response']['coredata']
        authors = data['abstracts-retrieval-response']['authors']['author']

    except KeyError:
        pprint(data)
        print ""
        print "Have not found authors in dataset"
        print " -> Is the connection to scopus broken???"
        return(0)

    # decode date
    tsstring = coredata['prism:coverDate'].encode('utf-8')
    ts = datetime.strptime(tsstring, "%Y-%m-%d").timetuple()
    year = ts.tm_year

    # Display cover date and title
    print("%s  -- %s" % (tsstring, coredata['dc:title']))


    # define post structure
    post = WordPressPost()
    post.title = coredata['dc:title']
    post.date = ts

    # set the name of the post different to the title
    post.slug = coredata['dc:identifier']
    
    post.excerpt = authors[0]['ce:indexed-name']
    if len(authors) > 2:
        post.excerpt += u' et al.'
    elif len(authors) == 2:
        post.excerpt += u', ' + authors[1]['ce:indexed-name']
    post.excerpt += u', in <em>' + coredata['prism:publicationName'] + u'</em>'
    if 'prism:volume' in coredata:
        post.excerpt += u', ' + coredata['prism:volume']
    post.excerpt += u' (' + str(year).encode('utf-8') + u')'
    if 'prism:pageRange' in coredata:
        post.excerpt += u' ' + coredata['prism:pageRange']
    if 'article-number' in coredata:
        post.excerpt += u', ' + coredata['article-number']
    post.excerpt += u'.'

    post.content = u'<p>' + authors[0]['ce:indexed-name']
    authors.pop(0)
    if len(authors) > 20:
        post.content += u' et al.'
    else:
        for author in authors:
            post.content += u', ' + author['ce:indexed-name']
    post.content += u'</p>'
    post.content += u'<p>in <em>' + coredata['prism:publicationName'] + u'</em>'
    if 'prism:volume' in coredata:
        post.content += u', ' + coredata['prism:volume']
    post.content += u' (' + str(year).encode('utf-8') + u')'
    if 'prism:pageRange' in coredata:
        post.content += u' ' + coredata['prism:pageRange']
    if 'article-number' in coredata:
        post.content += u', ' + coredata['article-number']
    post.content += u'.'
    if 'prism:doi' in coredata:
        post.content += u' DOI:' + coredata['prism:doi']
    post.content += u'</p>\n\n'
    if 'dc:description' in coredata:
        post.content += u'<div class="accordion-inner"><h4>Abstract</h4>' + coredata['dc:description']
        if 'authkeywords' in coredata:
            post.content += u'\n<b>Keywords:</b> ' + coredata['authkeywords']
        post.content += u'</div>'
    if 'prism:doi' in coredata:
        link = u'http://dx.doi.org/' + coredata['prism:doi']
        post.content += u'\n\n<div class="accordion-inner"><a class="btn btn-primary" href="' + link + u'"><i class="icon-download icon-white"></i> Get it</a></div>'

    #print post.content
    #post.id = wp.call(NewPost(post)) # Creates a new post and returns the id!

    catlist = []
    for slug in category:
        try:
           cat = wordpress_get_category(slug)
           catlist.append(cat)
        except: 
           print "Slug %s not found in Wordpress" % slug
           exit
    post.terms = catlist
    print post.terms

    taglist = []
    try:
        for tag in data['abstracts-retrieval-response']['authkeywords']['author-keyword']:
            try:
                print "Keyword: ", tag
                taglist.append(tag['$'].decode('utf-8','ignore'))
            except:
                print "Keyword contains special characters - droped!"
                pass
    except:
        # No keywords given
	pass

    post.terms_names = {
        'category': ['Publications'],
        'post_tag': taglist
    }
    print post.terms_names

    # whoops, I forgot to publish it!
    if len(authors) > sc_max_authors:
        post.post_status = 'draft' # check how to handle publication in wordpress
        print "Too many authors %d - set to draft" % (len(authors))
    else:
        post.post_status = 'publish' # handled as a standard publication
    post.comment_status = 'closed' # allow comments - may be only for scopus

    # Todo: this can fail! Add proper error handling 
    post.id = wp.call(NewPost(post)) # Creates a new post and returns the id!
    #wp.call(EditPost(post.id, post))# Update the before created post
    print post.id
    
    # need to update the database !!!
    return post.id


#
# Create comments for all citations
#

def wordpress_comment_by_scopus(wpid, data):
    """ Create a new comment based on Scopus data """

    #print "Create Wordpress comment for post %d" % wpid

    #print json.dumps(data,sort_keys=True,indent=4, separators=(',', ': '))

    # decode date
    tsstring = data['prism:coverDate'].encode('utf-8')
    ts = datetime.strptime(tsstring, "%Y-%m-%d").timetuple()
    year = ts.tm_year

    # Display cover date and title
    print("%s  -- %s" % (tsstring, data['dc:title']))


    # Create WP comment
    # define post structure
    comment = WordPressComment()
    comment.id = 0
    comment.content = ""
    
    if 'dc:creator' in data and data['dc:creator'] is not None:
        comment.content = data['dc:creator'] + 'et al.: '
    
    if 'prism:doi' in data and data['prism:doi'] is not None:
        comment.content +='<a href="http://dx.doi.org/' + data['prism:doi'] + '">' + data['dc:title'] + '</a>'
    else:
        comment.content +='<b>' + data['dc:title'] + '</b>'

    comment.content += ' in ' + data['prism:publicationName']
    if 'prism:volume' in data and data['prism:volume'] is not None:
        comment.content += ', ' + data['prism:volume']
    comment.content += ' (' + str(year).encode('utf-8') + ')'
    if 'prism:pageRange' in data and data['prism:pageRange'] is not None:
        comment.content += ' ' + data['prism:pageRange']
    if 'article-number' in data and data['article-number'] is not None:
        comment.content += ' ' + data['article-number']
    comment.content += '.'

    # Enable comments
    post = WordPressPost()
    postorig = wp.call(GetPost(wpid))

    post.id = wpid
    post.date = postorig.date
    post.title = postorig.title
    post.content = postorig.content

    post.comment_status = 'open' # allow comments - may be only for scopus
    wp.call(EditPost(wpid, post))

    comment.id = wp.call(NewComment(wpid, comment))

    # Warning: Date can only be specified in edit command
    comment.date_created = ts
    wp.call(EditComment(comment.id, comment))# Update the before created post

    # Close comments for scopus posts
    post.comment_status = 'closed' # allow comments - may be only for scopus
    wp.call(EditPost(wpid, post))# Update the before created post

    return comment.id