Datasets:

MahimaSinghal
/

npov-onset

#!/usr/bin/env python3
"""
Minimal NPOV-Onset loader: read the three tables, reconstruct the 688 matched pairs,
and show how to pull the article text for a snapshot from the MediaWiki API.
Works directly on the CSVs in this folder (no `datasets` install needed). The Hugging Face
equivalent is one `load_dataset(...)` call per config — see the dataset card.
"""
import csv, os

HERE = os.path.dirname(os.path.abspath(__file__))


def load(name):
    with open(os.path.join(HERE, name), newline="", encoding="utf-8") as fh:
        return list(csv.DictReader(fh))


def main():
    events = load("full_npov_events_en.csv")
    controls = load("full_npov_controls_en.csv")

    # one control per event, joined on the event title
    ctrl_by_event = {c["matched_to"]: c for c in controls}
    pairs = [(e, ctrl_by_event[e["title"]]) for e in events if e["title"] in ctrl_by_event]
    print(f"{len(events)} events, {len(controls)} controls, {len(pairs)} matched pairs")

    # measurement-bias flags you will want to filter on
    bot = sum(1 for e in events if e["anchor_is_bot_dating"] == "True")
    cens = sum(1 for e in events if e["first_tag_censored"] == "True")
    print(f"bot-dated anchors: {bot} ({bot/len(events):.0%}); "
          f"left-censored: {cens} ({cens/len(events):.0%})")

    e, c = pairs[0]
    onset_revid = int(float(e["onset_revid"]))  # stored as float in CSV; revids are ints
    print(f"\nexample pair:\n  event   {e['title']!r} tagged {e['template']} "
          f"at rev {onset_revid} ({e['onset_ts']})\n"
          f"  control {c['title']!r} (shares {c['match_category']})")

    # To fetch the prose for any snapshot, resolve the revid against the API, e.g.:
    #   https://en.wikipedia.org/w/api.php?action=parse&oldid=<onset_revid>
    #       &prop=wikitext&format=json
    # For lead-time work, pick the revision closest to N days before onset_ts.
    print(f"\ntext for the onset snapshot: "
          f"https://en.wikipedia.org/w/api.php?action=parse&oldid={onset_revid}"
          f"&prop=wikitext&format=json")


if __name__ == "__main__":
    main()

	#!/usr/bin/env python3
	"""
	Minimal NPOV-Onset loader: read the three tables, reconstruct the 688 matched pairs,
	and show how to pull the article text for a snapshot from the MediaWiki API.

	Works directly on the CSVs in this folder (no `datasets` install needed). The Hugging Face
	equivalent is one `load_dataset(...)` call per config — see the dataset card.
	"""
	import csv, os

	HERE = os.path.dirname(os.path.abspath(__file__))


	def load(name):
	with open(os.path.join(HERE, name), newline="", encoding="utf-8") as fh:
	return list(csv.DictReader(fh))


	def main():
	events = load("full_npov_events_en.csv")
	controls = load("full_npov_controls_en.csv")

	# one control per event, joined on the event title
	ctrl_by_event = {c["matched_to"]: c for c in controls}
	pairs = [(e, ctrl_by_event[e["title"]]) for e in events if e["title"] in ctrl_by_event]
	print(f"{len(events)} events, {len(controls)} controls, {len(pairs)} matched pairs")

	# measurement-bias flags you will want to filter on
	bot = sum(1 for e in events if e["anchor_is_bot_dating"] == "True")
	cens = sum(1 for e in events if e["first_tag_censored"] == "True")
	print(f"bot-dated anchors: {bot} ({bot/len(events):.0%}); "
	f"left-censored: {cens} ({cens/len(events):.0%})")

	e, c = pairs[0]
	onset_revid = int(float(e["onset_revid"])) # stored as float in CSV; revids are ints
	print(f"\nexample pair:\n event {e['title']!r} tagged {e['template']} "
	f"at rev {onset_revid} ({e['onset_ts']})\n"
	f" control {c['title']!r} (shares {c['match_category']})")

	# To fetch the prose for any snapshot, resolve the revid against the API, e.g.:
	# https://en.wikipedia.org/w/api.php?action=parse&oldid=<onset_revid>
	# &prop=wikitext&format=json
	# For lead-time work, pick the revision closest to N days before onset_ts.
	print(f"\ntext for the onset snapshot: "
	f"https://en.wikipedia.org/w/api.php?action=parse&oldid={onset_revid}"
	f"&prop=wikitext&format=json")


	if __name__ == "__main__":
	main()