import csv, datetime as dt

rows = csv.DictReader(open("trade_fairs_de_2027.csv", encoding="utf-8"))
out = ["BEGIN:VCALENDAR", "VERSION:2.0", "PRODID:-//fairs2027//EN"]
seen = set()
for r in rows:
    key = (r["name"], r["start_date"])
    if key in seen:          # skip duplicates
        continue
    seen.add(key)
    end = dt.date.fromisoformat(r["end_date"]) + dt.timedelta(days=1)  # DTEND is exclusive
    place = r["city"] or r["venue"]   # city is blank for many rows
    out += ["BEGIN:VEVENT",
            f"UID:{abs(hash(key))}-{r['start_date']}@fairs2027",
            f"DTSTART;VALUE=DATE:{r['start_date'].replace('-', '')}",
            f"DTEND;VALUE=DATE:{end.strftime('%Y%m%d')}",
            f"SUMMARY:{r['name']}",
            f"LOCATION:{place}",
            "END:VEVENT"]
out.append("END:VCALENDAR")
open("fairs2027.ics", "w", encoding="utf-8").write("\r\n".join(out) + "\r\n")

Gotchas worth knowing:

• For all-day events DTEND is exclusive, so add one day.
• Lines must end with CRLF.
• Every UID must be unique, which is why duplicates are skipped first.
• Commas and semicolons in text fields should be escaped with a backslash if your calendar app is strict.

The input is the open (CC BY 4.0) "Trade Fairs in Europe 2027" dataset: https://huggingface.co/datasets/retiodisplays/trade-fairs-europe-2027 (also on Kaggle). There is one file per country. The readable version with all fairs is here: https://retio.eu/trade-fairs-in-europe-2027

Dates come from the exhibition centres and can change, so check with the organiser before you book anything.

Przemek (RETIO)