Hermes Agent
90 lines
3.6 KiB
Python
90 lines
3.6 KiB
Python
#!/usr/bin/env python3
|
|
"""Parse kinderspiele_deals HTML and build Joomla article."""
|
|
import re, json, sys
|
|
|
|
with open("/home/hermes/workspace/kinderspiele_deals_2026-06-21.html") as f:
|
|
html = f.read()
|
|
|
|
# Extract deal blocks - each is a <div>...</div> with the deal info
|
|
# Pattern: <div style="display:flex;..."> <img src="URL"> ... <strong>TITLE</strong> ... price ... UVP ... -PCT% ... dp/ASIN?tag
|
|
blocks = re.split(r'<div style="display:flex;gap:1rem;padding:1rem 0;border-bottom:1px solid #eee;">', html)[1:]
|
|
|
|
deals = []
|
|
for block in blocks:
|
|
# Extract fields
|
|
img_match = re.search(r'<img src="([^"]+)"', block)
|
|
title_match = re.search(r'<strong>([^<]+)</strong>', block)
|
|
price_match = re.search(r'font-weight:bold;">([\d.]+)\s*€<', block)
|
|
uvp_match = re.search(r'UVP\s+([\d.]+)\s*€<', block)
|
|
pct_match = re.search(r'-(\d+)%<', block)
|
|
saved_match = re.search(r'>\(([\d.]+)\s*€\s*gespart\)<', block)
|
|
asin_match = re.search(r'/dp/([A-Z0-9]+)\?tag', block)
|
|
|
|
if all([img_match, title_match, price_match, uvp_match, pct_match, saved_match, asin_match]):
|
|
deals.append({
|
|
"img": img_match.group(1),
|
|
"title": title_match.group(1).strip(),
|
|
"price": float(price_match.group(1)),
|
|
"uvp": float(uvp_match.group(1)),
|
|
"pct": int(pct_match.group(1)),
|
|
"saved": float(saved_match.group(1)),
|
|
"asin": asin_match.group(1),
|
|
})
|
|
|
|
print(f"Parsed {len(deals)} deals", file=sys.stderr)
|
|
|
|
def categorize(title):
|
|
tl = title.lower()
|
|
if 'ab 2 jahr' in tl or 'meine ersten spiele' in tl or '2 jahre' in tl and 'ab 2' in tl:
|
|
return "2+"
|
|
if 'ab 3 jahr' in tl or '3 jahre' in tl and 'ab 3' in tl:
|
|
return "3+"
|
|
if 'ab 4 jahr' in tl or '4 jahre' in tl and 'ab 4' in tl or 'vorschul' in tl:
|
|
return "4+"
|
|
if 'ab 5 jahr' in tl or '5 jahre' in tl and 'ab 5' in tl:
|
|
return "5+"
|
|
if 'ab 6 jahr' in tl or '6-9 jahre' in tl or 'ab 6' in tl:
|
|
return "6+"
|
|
if 'ab 7 jahr' in tl:
|
|
return "7+"
|
|
known = {
|
|
'monsterjäger': '5+', 'paletti spaghetti': '5+', 'topp die torte': '5+',
|
|
'pupsparade': '4+', 'kinderspiele klassiker': '4+', 'sagaland': '6+',
|
|
'kristallica': '4+', 'lotti karotti': '4+', 'milly muffin': '4+',
|
|
'kroko doc': '4+', 'kakerlakak': '5+', 'sos affenalarm': '5+',
|
|
'scrabble junior': '6+', 'wer war': '6+', 'magischen schlüssel': '6+',
|
|
'magic 8 ball': '7+', 'wort für wort': '6+', 'erster obstgarten': '2+',
|
|
'hanni honigbiene': '2+', 'sachen suchen': '2+',
|
|
}
|
|
for key, age in known.items():
|
|
if key in tl:
|
|
return age
|
|
return "Allgemein"
|
|
|
|
age_groups = {}
|
|
for d in deals:
|
|
age = categorize(d['title'])
|
|
if age not in age_groups:
|
|
age_groups[age] = []
|
|
age_groups[age].append(d)
|
|
|
|
# Output structured data as JSON for next step
|
|
output = {
|
|
"deals": deals,
|
|
"age_groups": {k: [{"title": d['title'], "price": d['price'], "uvp": d['uvp'],
|
|
"pct": d['pct'], "saved": d['saved'], "asin": d['asin'],
|
|
"img": d['img']} for d in v] for k, v in age_groups.items()}
|
|
}
|
|
|
|
# Print summary
|
|
age_order = ["2+", "3+", "4+", "5+", "6+", "7+", "Allgemein"]
|
|
for age in age_order:
|
|
if age in age_groups:
|
|
games = age_groups[age]
|
|
print(f"\n{age}: {len(games)} Spiele")
|
|
for g in games:
|
|
print(f" -{g['pct']}% | {g['price']}€ (UVP {g['uvp']}€) | {g['asin']} | {g['title'][:80]}")
|
|
|
|
json.dump(output, open("/tmp/kinderspiele_data.json", "w"), ensure_ascii=False, indent=2)
|
|
print(f"\nTotal: {len(deals)} deals, {len(age_groups)} groups")
|