'''

Created on 27 juin 2012

@author: gerard

'''

import MySQLdb

import logging

import warnings

logger = logging.getLogger('document')

class MediapartReader() :

    def __init__(self):

        self.conn = MySQLdb.connect (host = 'localhost',

            user = 'root',

            passwd = '',

            db = 'www_mediapart_fr')

        warnings.filterwarnings('ignore', category=MySQLdb.Warning)

    def __del__(self):

        self.conn.close ()

    def get_date(self, document_id):     

        self.conn.query('SELECT changed FROM node where nid=\''+document_id+'\'')

        r = self.conn.use_result()

        row = r.fetch_row()

        if row:

            date = (row)[0][0]

        else:

            date = '0'

        '''logger.info('get_date')

        logger.info('document_id = '+document_id)

        logger.info('date = '+str(date))'''

        #changed_date = datetime.datetime.fromtimestamp(int(change_date_unix[0])).strftime('%Y-%m-%d %H:%M:%S')

        return date

    def get_title(self, document_id): 

        self.conn.query('SELECT title FROM node where nid=\''+document_id+'\'')

        '''logger.info('get_title')

        logger.info('document_id = '+document_id)'''

        r = self.conn.use_result()

        row = r.fetch_row()

        if row:

            title = (row)[0][0].decode("windows-1252").encode("utf8")

        else:

            title = ''

        return title

    '''def get_category(self, document_id):

        self.conn.query('SELECT dst FROM url_alias where src="node/'+document_id+'"')

        r = self.conn.use_result()

        row = r.fetch_row()

        if row:

            dst = (row)[0][0]

            fields = dst.split('/')

            category = fields[0]

        else:

            category = ''

        return category'''

    def get_category(self, document_id):

        cursor =self.conn.cursor()

        cursor.execute('SELECT tid FROM term_node where nid=\''+document_id+'\'')

        tids = cursor.fetchall()

        category = ''

        for tid in tids:

            if str(tid[0]) == '27':

                category = 'France'

                return category

            elif str(tid[0]) == '28':

                category = 'International'

                return category

            elif str(tid[0]) == '29':

                category = 'Economie'

                return category

            elif str(tid[0]) == '30':

                category = 'Culture'

                return category 

            else:

                category = '' 

        return category

    def get_tags(self, document_id):

        tags=[]

        cursor = self.conn.cursor()

        cursor.execute('SELECT tid FROM term_node where nid=\''+document_id+'\'')

        tids = cursor.fetchall()

        for tid in tids:

            if str(tid[0]) != '27' and str(tid[0]) != '28' and str(tid[0]) != '29' and str(tid[0]) != '30':

                cursor.execute('SELECT name FROM term_data where tid=' + str(tid[0]))

                names = cursor.fetchall()

                for name in names:

                    tags.append(str(name).decode("windows-1252").encode("utf8"))

        return tags

    def get_url(self, document_id):

        logger.info('get_url')

        logger.info('document_id = '+document_id)

        self.conn.query('SELECT dst FROM url_alias where src="node/'+document_id+'"')

        '''logger.info('get_url')

        logger.info('document_id = '+document_id)'''

        r = self.conn.use_result()

        row = r.fetch_row()

        if row:

            url = 'http://www.mediapart.fr/'+(row)[0][0]

        else:

            url = ''

        logger.info('url = '+url)

        return url

    def get_author(self, document_id):

        self.conn.query('SELECT u.uid, u.name FROM users u, node n where u.uid=n.uid AND n.nid=\''+document_id+'\'')

        '''logger.info('get_author')

        logger.info('document_id = '+document_id)'''

        r = self.conn.use_result()

        row = r.fetch_row()

        if row:

            name = (row)[0][1].decode("windows-1252").encode("utf8")

            uid = (row)[0][0]

        else:

            name = ''

            uid=''

        return {'id':uid, 'name':name}