diff --git a/tools/maimai_convert.py b/tools/maimai_convert.py
new file mode 100644
index 00000000..080eb189
--- /dev/null
+++ b/tools/maimai_convert.py
@@ -0,0 +1,75 @@
+import argparse
+from pathlib import Path
+from typing import NamedTuple
+
+import pandas as pd
+import xmltodict
+from hypy_utils import write
+
+
+def read_list(search_path: str, pattern: str, parse_fn) -> list:
+    search_path = path / search_path
+    data = [parse_fn(xmltodict.parse(t.read_text('utf-8')))
+            for t in list(search_path.glob(pattern))]
+    Path('csv').mkdir(exist_ok=True, parents=True)
+    pd.DataFrame(data).to_csv(f'csv/{search_path.name}.csv', index=False)
+    return data
+
+
+class Ticket(NamedTuple):
+    id: int  # TicketData.name.id
+    name: str  # TicketData.name.str
+    credits: int  # TicketData.creditNum
+    kind: str  # TicketData.ticketKind.str
+    max: int  # TicketData.maxCount
+    detail: str  # TicketData.detail
+    eventId: int  # TicketData.ticketEvent.id
+    eventName: str  # TicketData.ticketEvent.str
+
+
+def parse_ticket(d: dict) -> Ticket:
+    return Ticket(
+        id=int(d['TicketData']['name']['id']),
+        name=d['TicketData']['name']['str'],
+        credits=int(d['TicketData']['creditNum']),
+        kind=d['TicketData']['ticketKind']['str'],
+        max=int(d['TicketData']['maxCount']),
+        detail=d['TicketData']['detail'],
+        eventId=int(d['TicketData']['ticketEvent']['id']),
+        eventName=d['TicketData']['ticketEvent']['str']
+    )
+
+
+class Event(NamedTuple):
+    id: int  # EventData.name.id
+    type: int  # EventData.infoType
+    detail: str  # EventData.name.str
+    alwaysOpen: bool  # EventData.alwaysOpen
+
+
+def parse_event(d: dict) -> Event:
+    return Event(
+        id=int(d['EventData']['name']['id']),
+        type=int(d['EventData']['infoType']),
+        detail=d['EventData']['name']['str'],
+        alwaysOpen=bool(d['EventData']['alwaysOpen'])
+    )
+
+
+if __name__ == '__main__':
+    agupa = argparse.ArgumentParser(description='Convert maimai data to csv')
+    agupa.add_argument('path', type=Path, help='Path to A000 data folder')
+    args = agupa.parse_args()
+    path = Path(args.path)
+
+    tickets = read_list('ticket', '*/Ticket.xml', parse_ticket)
+
+    events = read_list('event', '*/Event.xml', parse_event)
+
+    # Write incremental sql
+    ids = [int(v.split(",")[0]) for v in (Path(__file__).parent / 'maimai2_game_event.csv').read_text().splitlines()]
+    new_events = [e for e in events if e.id not in ids]
+    sql = "INSERT INTO `maimai2_game_event` (`id`, `end_date`, `start_date`, `type`, `enable`) VALUES \n" + \
+        ",\n".join([f"({e.id}, '2029-01-01 00:00:00.000000', '2019-01-01 00:00:00.000000', {e.type}, '1')" for e in new_events])
+    sql += ";\n"
+    write('sql/maimai2_game_event.sql', sql)