Polymarket trades and volume

# replace each {placeholder} with your own value
curl -H "X-API-KEY: $VIKE_API_KEY" \
  "https://data.vike.io/v1/polymarket/markets/{condition_id}/trades?limit=50"

import os, requests

r = requests.get(
    "https://data.vike.io/v1/polymarket/markets/{condition_id}/trades?limit=50",
    headers={"X-API-KEY": os.environ["VIKE_API_KEY"]},
    timeout=30,
)
r.raise_for_status()
body = r.json()

import os, requests
import polars as pl

body = requests.get(
    "https://data.vike.io/v1/polymarket/markets/{condition_id}/trades?limit=50",
    headers={"X-API-KEY": os.environ["VIKE_API_KEY"]},
    timeout=30,
).json()
rows = body if isinstance(body, list) else next(v for v in body.values() if isinstance(v, list))
df = pl.DataFrame(rows)
print(df.head())

# replace each {placeholder} with your own value
curl -H "X-API-KEY: $VIKE_API_KEY" \
  "https://data.vike.io/v1/polymarket/events?token_id={token_id}&from_ms={from_ms}&to_ms={to_ms}&event_type=trade&limit=50"

# Full history is the Parquet archive. Pick a date from manifest.json (no key needed),
# download the whole file, then read it locally:
DAY=YYYY-MM-DD
curl -H "X-API-KEY: $VIKE_API_KEY" -O \
  "https://data.vike.io/archive/venue=polymarket/date=$DAY/trades.parquet"

# Or the free sample — one complete market, no key:
curl -O https://data.vike.io/archive/samples/trades.parquet

import os, requests
import pyarrow.parquet as pq

H = {"X-API-KEY": os.environ["VIKE_API_KEY"]}
day = "YYYY-MM-DD"   # any date listed in https://data.vike.io/archive/manifest.json
url = f"https://data.vike.io/archive/venue=polymarket/date={day}/trades.parquet"

# Download the whole file once, then read it locally — a remote Parquet
# reader re-fetches the footer and column chunks on every query.
with requests.get(url, headers=H, stream=True, timeout=600) as r:
    r.raise_for_status()
    with open("trades.parquet", "wb") as f:
        for chunk in r.iter_content(1 << 20):
            f.write(chunk)

table = pq.read_table("trades.parquet")
print(table.num_rows, table.schema.names)

import polars as pl

df = pl.read_parquet("trades.parquet")   # downloaded first — see the cURL tab
df = df.with_columns(pl.from_epoch("ts", time_unit="ms").alias("time"))
vwap = (
    df.group_by("token_id")
      .agg(((pl.col("price") * pl.col("size")).sum() / pl.col("size").sum()).alias("vwap"),
           pl.len().alias("prints"))
      .sort("prints", descending=True)
)
print(vwap.head())

-- trades.parquet downloaded first (see the cURL tab), then read locally
SELECT token_id,
       count(*)                      AS prints,
       sum(size)                     AS shares,
       sum(price * size) / sum(size) AS vwap,
       epoch_ms(min(ts))             AS first_print
FROM   'trades.parquet'
GROUP  BY token_id
ORDER  BY prints DESC
LIMIT  10;