
import mwparserfromhell
from mwviews.api import PageviewsClient
import mwapi


def get_id_page(d):
	return list(d.keys())[0]

def most_popular_pages(p, wikilinks):
	dictionary = {}
	pages = p.article_views('en.wikipedia', wikilinks, start='20220101', end='20221231')
	for k, v in pages.items():
		for kk, vv in v.items():
			if vv != None and kk in dictionary:
				dictionary[kk] += vv
			if vv != None and kk not in dictionary:
				dictionary[kk] = vv
	return dictionary

def get_most_popular_pages_from_text(text):
	# 3. List all the Wikilinks on that paragraph using the mwparserfromhell.
	wikicode = mwparserfromhell.parse(text)
	wikilinks = wikicode.filter_wikilinks()
	wikilinks_ = [link.title.strip().lower() for link in wikilinks]
	#print('List of links:')
	#print(wikilinks_)
	# 4. Using the Pageview API find the most popular pages on the A’s neighbordhood, during 2022.
	p = PageviewsClient('all-access',parallelism=10)
	most_popular = most_popular_pages(p, wikilinks_)
	most_popular = dict(sorted(most_popular.items(), key=lambda item: item[1], reverse=True))
	print('Count the most Popular links:')
	print(most_popular)
	#return most_popular

def get_last_revisions_from_title(session, title):
	revisions = session.get(action='query', prop='revisions', rvlimit=20, titles=title)
	page_id = get_id_page(revisions['query']['pages'])
	return revisions['query']['pages'][page_id]['revisions']


# 1. Select one Wikipedia Article (A).
title = 'Satellite_navigation'
# 2. Considering the current version of A, get the Wikitext from the first paragraph.
# I copy and paste the first paragraph of the Satellite_navigation page
text = "A '''satellite navigation''' or '''satnav''' system is a system that uses [[satellite]]s to provide autonomous geo-spatial positioning. It allows small [[electronics|electronic]] receivers to determine their location ([[longitude]], [[latitude]], and [[altitude]]/[[elevation]]) to high precision (within a few centimetres to metres) using [[time signal]]s transmitted along a [[Line-of-sight propagation|line of sight]] by [[radio]] from satellites. The system can be used for providing position, navigation or for tracking the position of something fitted with a receiver (satellite tracking). The signals also allow the electronic receiver to calculate the current local time to high precision, which allows time synchronisation. These uses are collectively known as Positioning, Navigation and Timing (PNT). Satnav systems operate independently of any telephonic or internet reception, though these technologies can enhance the usefulness of the positioning information generated."


# 3. List all the Wikilinks on that paragraph using the mwparserfromhell.
# 4. Using the Pageview API find the most popular pages on the A’s neighbordhood, during 2022.
get_most_popular_pages_from_text(text)

Count the most Popular links:
{'latitude': 476, 'longitude': 97, 'satellite': 45, 'radio': 32, 'electronics': 32, 'elevation': 18, 'line-of-sight_propagation': 18, 'altitude': 6}


# 5. Using the MediaWiki API get the last 20 revisions of A and repeat the procedure above for all those revisions.
# get the last 20 revisions of Satellite_navigation
session = mwapi.Session('https://en.wikipedia.org')
revisions = get_last_revisions_from_title(session, title)

for index, r in enumerate(revisions):
	tmp_rev = session.get(action='query', prop='revisions', ids=r['revid'], rvprop='content', titles='Satellite_navigation')
	page_id = get_id_page(tmp_rev['query']['pages'])
	tmp_text = tmp_rev['query']['pages'][page_id]['revisions'][0]['*']
	print("Most popular pages of version number", index+1)
	get_most_popular_pages_from_text(tmp_text)
	print("\n\n")

Sending requests with default User-Agent.  Set 'user_agent' on mwapi.Session to quiet this message.
The following query raised warnings: {'action': 'query', 'prop': 'revisions', 'ids': 1087332950, 'rvprop': 'content', 'titles': 'Satellite_navigation', 'format': 'json'}
	- main -- {'*': 'Subscribe to the mediawiki-api-announce mailing list at <https://lists.wikimedia.org/postorius/lists/mediawiki-api-announce.lists.wikimedia.org/> for notice of API deprecations and breaking changes. Use [[Special:ApiFeatureUsage]] to see usage of deprecated features by your application.\nUnrecognized parameter: ids.'}
	- revisions -- {'*': 'Because "rvslots" was not specified, a legacy format has been used for the output. This format is deprecated, and in the future the new format will always be used.'}



Most popular pages of version number 1

The following query raised warnings: {'action': 'query', 'prop': 'revisions', 'ids': 1087230413, 'rvprop': 'content', 'titles': 'Satellite_navigation', 'format': 'json'}
	- main -- {'*': 'Subscribe to the mediawiki-api-announce mailing list at <https://lists.wikimedia.org/postorius/lists/mediawiki-api-announce.lists.wikimedia.org/> for notice of API deprecations and breaking changes. Use [[Special:ApiFeatureUsage]] to see usage of deprecated features by your application.\nUnrecognized parameter: ids.'}
	- revisions -- {'*': 'Because "rvslots" was not specified, a legacy format has been used for the output. This format is deprecated, and in the future the new format will always be used.'}

Count the most Popular links:
{'50th_parallel_north': 21990, '30th_parallel_south': 8551, '30th_meridian_east': 8210, 'india': 6721, 'european_union': 6215, 'list_of_emerging_technologies': 5050, '130th_meridian_east': 3351, 'eu': 2611, 'general_relativity': 2551, 'united_states': 2242, 'galileo_positioning_system': 1767, 'global_positioning_system': 1583, 'galileo_(satellite_navigation)': 1240, 'japan': 765, 'kalman_filter': 524, 'latitude': 476, 'java_applet': 313, 'russia': 284, 'geostationary_orbit': 281, 'satellite_phone': 267, 'china': 221, 'ephemeris': 126, 'longitude': 97, 'orbital_inclination': 80, 'graphical_user_interface': 76, 'soviet_union': 59, 'refraction': 56, 'satellite': 45, 'radio': 32, 'electronics': 32, 'longwave': 30, 'indian_space_research_organisation': 28, 'fog_of_war': 25, 'elevation': 18, 'line-of-sight_propagation': 18, 'doppler_effect': 11, 'navigation': 9, 'geoinformatics': 7, 'altitude': 6, 'einstein': 5, 'gps': 5, 'automatic_dependent_surveillance–broadcast': 5, 'encryption': 4, 'loran': 4, 'inertial_navigation_system': 4, 'category:avionics': 4, 'guided_bomb': 4, 'european_space_agency': 4, 'transit_(satellite)': 3, 'radio_frequency': 3, 'safety_service': 3, 'transmitter': 3, 'orbital_plane_(astronomy)': 2, 'navic': 1, 'acronyms_and_abbreviations_in_avionics': 1, 'almanac': 1, 'satellite_constellation': 1, 'atomic_clock': 1}


Most popular pages of version number 2

The following query raised warnings: {'action': 'query', 'prop': 'revisions', 'ids': 1087227747, 'rvprop': 'content', 'titles': 'Satellite_navigation', 'format': 'json'}
	- main -- {'*': 'Subscribe to the mediawiki-api-announce mailing list at <https://lists.wikimedia.org/postorius/lists/mediawiki-api-announce.lists.wikimedia.org/> for notice of API deprecations and breaking changes. Use [[Special:ApiFeatureUsage]] to see usage of deprecated features by your application.\nUnrecognized parameter: ids.'}
	- revisions -- {'*': 'Because "rvslots" was not specified, a legacy format has been used for the output. This format is deprecated, and in the future the new format will always be used.'}

Count the most Popular links:
{'50th_parallel_north': 21990, '30th_parallel_south': 8551, '30th_meridian_east': 8210, 'india': 6721, 'european_union': 6215, 'list_of_emerging_technologies': 5050, '130th_meridian_east': 3351, 'eu': 2611, 'general_relativity': 2551, 'united_states': 2242, 'galileo_positioning_system': 1767, 'global_positioning_system': 1583, 'galileo_(satellite_navigation)': 1240, 'japan': 765, 'kalman_filter': 524, 'latitude': 476, 'java_applet': 313, 'russia': 284, 'geostationary_orbit': 281, 'satellite_phone': 267, 'china': 221, 'ephemeris': 126, 'longitude': 97, 'orbital_inclination': 80, 'graphical_user_interface': 76, 'soviet_union': 59, 'refraction': 56, 'satellite': 45, 'radio': 32, 'electronics': 32, 'longwave': 30, 'indian_space_research_organisation': 28, 'fog_of_war': 25, 'elevation': 18, 'line-of-sight_propagation': 18, 'doppler_effect': 11, 'navigation': 9, 'geoinformatics': 7, 'altitude': 6, 'einstein': 5, 'gps': 5, 'automatic_dependent_surveillance–broadcast': 5, 'encryption': 4, 'loran': 4, 'inertial_navigation_system': 4, 'category:avionics': 4, 'guided_bomb': 4, 'european_space_agency': 4, 'transit_(satellite)': 3, 'radio_frequency': 3, 'safety_service': 3, 'transmitter': 3, 'orbital_plane_(astronomy)': 2, 'navic': 1, 'acronyms_and_abbreviations_in_avionics': 1, 'almanac': 1, 'satellite_constellation': 1, 'atomic_clock': 1}


Most popular pages of version number 3

The following query raised warnings: {'action': 'query', 'prop': 'revisions', 'ids': 1087227583, 'rvprop': 'content', 'titles': 'Satellite_navigation', 'format': 'json'}
	- main -- {'*': 'Subscribe to the mediawiki-api-announce mailing list at <https://lists.wikimedia.org/postorius/lists/mediawiki-api-announce.lists.wikimedia.org/> for notice of API deprecations and breaking changes. Use [[Special:ApiFeatureUsage]] to see usage of deprecated features by your application.\nUnrecognized parameter: ids.'}
	- revisions -- {'*': 'Because "rvslots" was not specified, a legacy format has been used for the output. This format is deprecated, and in the future the new format will always be used.'}

Count the most Popular links:
{'50th_parallel_north': 21990, '30th_parallel_south': 8551, '30th_meridian_east': 8210, 'india': 6721, 'european_union': 6215, 'list_of_emerging_technologies': 5050, '130th_meridian_east': 3351, 'eu': 2611, 'general_relativity': 2551, 'united_states': 2242, 'galileo_positioning_system': 1767, 'global_positioning_system': 1583, 'galileo_(satellite_navigation)': 1240, 'japan': 765, 'kalman_filter': 524, 'latitude': 476, 'java_applet': 313, 'russia': 284, 'geostationary_orbit': 281, 'satellite_phone': 267, 'china': 221, 'ephemeris': 126, 'longitude': 97, 'orbital_inclination': 80, 'graphical_user_interface': 76, 'soviet_union': 59, 'refraction': 56, 'satellite': 45, 'radio': 32, 'electronics': 32, 'longwave': 30, 'indian_space_research_organisation': 28, 'fog_of_war': 25, 'elevation': 18, 'line-of-sight_propagation': 18, 'doppler_effect': 11, 'navigation': 9, 'geoinformatics': 7, 'altitude': 6, 'einstein': 5, 'gps': 5, 'automatic_dependent_surveillance–broadcast': 5, 'encryption': 4, 'loran': 4, 'inertial_navigation_system': 4, 'category:avionics': 4, 'guided_bomb': 4, 'european_space_agency': 4, 'transit_(satellite)': 3, 'radio_frequency': 3, 'safety_service': 3, 'transmitter': 3, 'orbital_plane_(astronomy)': 2, 'navic': 1, 'acronyms_and_abbreviations_in_avionics': 1, 'almanac': 1, 'satellite_constellation': 1, 'atomic_clock': 1}


Most popular pages of version number 4

The following query raised warnings: {'action': 'query', 'prop': 'revisions', 'ids': 1087227529, 'rvprop': 'content', 'titles': 'Satellite_navigation', 'format': 'json'}
	- main -- {'*': 'Subscribe to the mediawiki-api-announce mailing list at <https://lists.wikimedia.org/postorius/lists/mediawiki-api-announce.lists.wikimedia.org/> for notice of API deprecations and breaking changes. Use [[Special:ApiFeatureUsage]] to see usage of deprecated features by your application.\nUnrecognized parameter: ids.'}
	- revisions -- {'*': 'Because "rvslots" was not specified, a legacy format has been used for the output. This format is deprecated, and in the future the new format will always be used.'}

Count the most Popular links:
{'50th_parallel_north': 21990, '30th_parallel_south': 8551, '30th_meridian_east': 8210, 'india': 6721, 'european_union': 6215, 'list_of_emerging_technologies': 5050, '130th_meridian_east': 3351, 'eu': 2611, 'general_relativity': 2551, 'united_states': 2242, 'galileo_positioning_system': 1767, 'global_positioning_system': 1583, 'galileo_(satellite_navigation)': 1240, 'japan': 765, 'kalman_filter': 524, 'latitude': 476, 'java_applet': 313, 'russia': 284, 'geostationary_orbit': 281, 'satellite_phone': 267, 'china': 221, 'ephemeris': 126, 'longitude': 97, 'orbital_inclination': 80, 'graphical_user_interface': 76, 'soviet_union': 59, 'refraction': 56, 'satellite': 45, 'radio': 32, 'electronics': 32, 'longwave': 30, 'indian_space_research_organisation': 28, 'fog_of_war': 25, 'elevation': 18, 'line-of-sight_propagation': 18, 'doppler_effect': 11, 'navigation': 9, 'geoinformatics': 7, 'altitude': 6, 'einstein': 5, 'gps': 5, 'automatic_dependent_surveillance–broadcast': 5, 'encryption': 4, 'loran': 4, 'inertial_navigation_system': 4, 'category:avionics': 4, 'guided_bomb': 4, 'european_space_agency': 4, 'transit_(satellite)': 3, 'radio_frequency': 3, 'safety_service': 3, 'transmitter': 3, 'orbital_plane_(astronomy)': 2, 'navic': 1, 'acronyms_and_abbreviations_in_avionics': 1, 'almanac': 1, 'satellite_constellation': 1, 'atomic_clock': 1}


Most popular pages of version number 5