# -*- coding: utf-8 -*-

'''

Created on Jun 7, 2011

@author: ymh

'''

from django.conf import settings

from django.core.management.base import NoArgsCommand

from django.core.management.color import no_style

from hdabo.models import Tag

from optparse import make_option

from wikitools import wiki

import math

import sys

class Command(NoArgsCommand):

    '''

    query and update wikipedia for tag title.

    '''

    options = ''

    help = """query and update wikipedia for tag title."""

    option_list = NoArgsCommand.option_list + (

        make_option('--force',

            action='store_true',

            dest='force',

            default=False,

            help='force all tags to be updated, not only those not yet processed'),

        make_option('--random',

            action='store_true',

            dest='random',

            default=False,

            help='randomize query on tags'),

        make_option('--site',

            action='store',

            type='string',

            dest='site_url',

            default="http://fr.wikipedia.org/w/api.php",

            help='the url for the wikipedia site'),

        make_option('--limit',

            action='store',

            type='int',

            dest='limit',

            default= -1,

            help='number of tag to process'),

        make_option('--start',

            action='store',

            type='int',

            dest='start',

            default=0,

            help='number of tag to ignore'),

        )

    def __is_homonymie(self, page_dict):

        for cat in page_dict.get(u"categories", []):

            if u'Catégorie:Homonymie' in cat.get(u"title", u"") or u'Category:Disambiguation pages' in cat.get(u"title", u""):

                return True

        return False

    def process_wp_response(self, label, response):

        query_dict = response['query']

        # get page if multiple pages or none -> return Tag.null_result

        pages = query_dict.get("pages", {})

        if len(pages) > 1 or len(pages) == 0:

            return None, Tag.TAG_URL_STATUS_DICT["null_result"], None, None

        page = pages.values()[0]

        if u"invalid" in page or u"missing" in page:

            return None, Tag.TAG_URL_STATUS_DICT["null_result"], None, None

        url = page.get(u'fullurl', None)

        pageid = page.get(u'pageid', None)

        new_label = page[u'title']

        if self.__is_homonymie(page):

            status = Tag.TAG_URL_STATUS_DICT["homonyme"]

        elif u"redirect" in page:

            status = Tag.TAG_URL_STATUS_DICT["redirection"]

        else:

            status = Tag.TAG_URL_STATUS_DICT["match"]

        return new_label, status, url, pageid 

    def show_progress(self, current_line, total_line, label, width):

        percent = (float(current_line) / float(total_line)) * 100.0

        marks = math.floor(width * (percent / 100.0))

        spaces = math.floor(width - marks)

        loader = u'[' + (u'=' * int(marks)) + (u' ' * int(spaces)) + u']'

        sys.stdout.write(u"%s %d%% %d/%d - %s\r" % (loader, percent, current_line - 1, total_line - 1, repr(label))) #takes the header into account

        if percent >= 100:

            sys.stdout.write("\n")

        sys.stdout.flush()

    def handle_noargs(self, **options):

        self.style = no_style()

        interactive = options.get('interactive', True)

        verbosity = int(options.get('verbosity', '1'))

        force = options.get('force', False)

        limit = options.get("limit", -1)

        start = options.get("start", 0)

        site_url = options.get('site_url', settings.WIKIPEDIA_API_URL)

        random = options.get('random', False)

        if verbosity > 2:

            print "option passed : " + repr(options)

        if force and interactive:

            confirm = raw_input("""You have requested to query and replace the wikipedia information for all datasheets.

Are you sure you want to do this ?

    Type 'yes' to continue, or 'no' to cancel: """)

        else:

            confirm = 'yes'

        if confirm != "yes":

            print "wikipedia query cancelled"

            return

        if force:

            queryset = Tag.objects.all()

        else:

            queryset = Tag.objects.filter(url_status=None)                    

        if random:

            queryset = queryset.order_by("?")

        else:

            queryset = queryset.order_by("label")

        if limit >= 0:

            queryset = queryset[start:limit]

        else:

            queryset = queryset[start:]

        if verbosity > 2 :

            print "Tag Query is %s" % (queryset.query)

        site = wiki.Wiki(site_url) #@UndefinedVariable

        count = queryset.count()

        if verbosity > 1:

            print "Processing %d tags" % (count)

        for i, tag in enumerate(queryset):

            if verbosity > 1:

                print "processing tag %s (%d/%d)" % (tag.label, i + 1, count)

            else:

                self.show_progress(i + 1, count, tag.label, 60)                            

            process_tag(site, tag, verbosity)