Datasets:
File size: 4,253 Bytes
5d3b219 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 | """Import Rick and Morty episode transcripts from the Fandom wiki API.
The wiki's HTML pages are behind a 403, but ``api.php`` answers, and every
episode has a ``<Episode>/Transcript`` page. Content is CC BY-SA, so it is a
cleaner source than a scraped dataset.
Usage::
python scripts/import_rm_fandom.py --max-episodes 120
"""
from __future__ import annotations
import argparse
import json
import re
import sys
import time
import urllib.parse
import urllib.request
from pathlib import Path
sys.path.insert(0, str(Path(__file__).resolve().parents[1]))
from kronllm.bilingual import append_jsonl, normalize, record_shard
API = "https://rickandmorty.fandom.com/api.php"
UA = {"User-Agent": "KronLLM-research/0.1 (corpus; github.com/Icold21/kron-llm)"}
ROOT = Path("data/bilingual")
SHARDS = ROOT / "shards"
MANIFEST = ROOT / "manifest.jsonl"
_RE_TEMPLATE = re.compile(r"\{\{[^{}]*\}\}")
_RE_FILE = re.compile(r"\[\[(?:File|Image):[^\]]*\]\]", re.I)
_RE_LINK = re.compile(r"\[\[([^\]|]+)(?:\|([^\]]+))?\]\]")
_RE_TAG = re.compile(r"<[^>]+>")
_RE_HEAD = re.compile(r"^=+.*?=+\s*$", re.M)
def api(params: dict) -> dict:
url = API + "?" + urllib.parse.urlencode(params)
req = urllib.request.Request(url, headers=UA)
with urllib.request.urlopen(req, timeout=30) as r:
return json.loads(r.read())
def clean_wikitext(text: str) -> str:
prev = None
while prev != text: # nested templates
prev, text = text, _RE_TEMPLATE.sub("", text)
text = _RE_FILE.sub("", text)
text = _RE_LINK.sub(lambda m: m.group(2) or m.group(1), text)
text = _RE_TAG.sub("", text)
text = _RE_HEAD.sub("", text)
text = text.replace("'''", "").replace("''", "")
text = re.sub(r"\[https?://\S+\s*([^\]]*)\]", r"\1", text)
text = re.sub(r"\n{3,}", "\n\n", text)
return normalize(text)
def main() -> None:
ap = argparse.ArgumentParser()
ap.add_argument("--domain", default="humour")
ap.add_argument("--max-episodes", type=int, default=120)
ap.add_argument("--min-chars", type=int, default=1500)
args = ap.parse_args()
titles: list[str] = []
cont: dict = {}
while True:
params = {"action": "query", "list": "categorymembers",
"cmtitle": "Category:Episodes", "cmlimit": 500,
"cmtype": "page", "format": "json", **cont}
data = api(params)
titles.extend(m["title"] for m in data["query"]["categorymembers"])
cont = data.get("continue") or {}
if not cont:
break
titles = titles[: args.max_episodes]
print(f"[fandom] {len(titles)} episode pages")
docs: list[dict] = []
for i, title in enumerate(titles, 1):
page = f"{title}/Transcript"
try:
data = api({"action": "parse", "page": page, "prop": "wikitext",
"format": "json"})
raw = data["parse"]["wikitext"]["*"]
except Exception: # missing page / parse error
continue
text = clean_wikitext(raw)
if len(text) < args.min_chars:
continue
docs.append({"title": page, "text": text, "language": "en",
"domain": args.domain, "source": "Fandom wiki (api.php)",
"licence": "CC-BY-SA-3.0",
"url": "https://rickandmorty.fandom.com/wiki/" + page,
"note": "easter egg: series dialogue"})
if i % 10 == 0:
print(f" {i}/{len(titles)} pages, {len(docs)} kept", flush=True)
time.sleep(0.15)
out = SHARDS / f"fandom_rm_{args.domain}.jsonl"
if out.exists():
out.unlink()
n = append_jsonl(out, docs)
shard = record_shard(out, MANIFEST, source="Fandom wiki (api.php)",
licence="CC-BY-SA-3.0",
url="https://rickandmorty.fandom.com/",
language="en", domain=args.domain, docs=n,
note="episode transcripts, easter egg")
print(f"[done] {n} transcripts, {shard.bytes/1e6:.2f} MB -> {out.name}")
if __name__ == "__main__":
main()
|