wangdx25/polymarket_orderbook_preview

Polymarket Microstructure — Free Preview (26–29 Jun + L2 depth sample) (TsFile)

Apache TsFile version of astrnvk/polymarket-orderbook-data-preview.

Overview

A free sample of the Polymarket CLOB dataset, so you can inspect the exact schema, file layout and manifest format before buying. Same Parquet + sha256-manifest packaging as the full dataset.

• Tables: 3 TsFile table(s), 7,793,489 rows in total.
  • polymarket_orderbook_preview_g0: 6,155,534 rows, 2026-06-26 09:19:08 鈫?2026-06-29 23:59:59.015000 (UTC)
  • polymarket_orderbook_preview_g1: 1,233,060 rows, 2026-06-26 00:03:52 鈫?2026-06-29 23:59:59.175000 (UTC)
  • polymarket_orderbook_preview_g2: 404,895 rows, 2026-06-26 09:18:38 鈫?2026-06-29 23:59:59.001000 (UTC)

Schema (TsFile structure)

polymarket_orderbook_preview_g0.tsfile

• Time (INT64, milliseconds) 鈥?the source ts column.
• TAG: market_id 鈥?device dimension; query one device with WHERE market_id=....
• FIELD (10): ts_ms (double), token (string), side (string), level (int32), price (double), size (double), slug (string), interval_sec (int32), spread (double), n_obs (int64)

polymarket_orderbook_preview_g1.tsfile

• Time (INT64, milliseconds) 鈥?the source ts column.
• TAG: none 鈥?the table is stored as a single device.
• FIELD (7): market_id (string), token (string), price (double), size (double), taker_side (string), slug (string), interval_sec (int32)

polymarket_orderbook_preview_g2.tsfile

• Time (INT64, milliseconds) 鈥?the source ts column.
• TAG: none 鈥?the table is stored as a single device.
• FIELD (7): asset (string), open (double), high (double), low (double), close (double), volume (double), interval_sec (int32)

Conversion notes

• One TsFile per source table (3 tables). Tables whose source carries several entities per file are stored as one device per entity, using market_id as the TAG.
• Time = the table's own clock column as INT64 milliseconds. The clock columns used here are ts.
• No column is dropped. Rows that share a timestamp inside one device are ordered by a 1 ms step in stable source order, so calendar columns keep their exact day.

Usage

Read this dataset with the Java SDK. It is written with the zstd compressor. The Apache TsFile Java SDK reads every compressor, but the Python SDK (pip install tsfile) implements only UNCOMPRESSED, SNAPPY, GZIP and LZ4; opening this file with it yields no rows.

Read a converted file with the Apache TsFile Java SDK (org.apache.tsfile:tsfile):

import org.apache.tsfile.file.metadata.IDeviceID;
import org.apache.tsfile.read.TsFileReader;
import org.apache.tsfile.read.TsFileSequenceReader;
import org.apache.tsfile.read.common.Path;
import org.apache.tsfile.read.expression.QueryExpression;
import org.apache.tsfile.read.query.dataset.QueryDataSet;
import org.apache.tsfile.write.schema.MeasurementSchema;

import java.util.ArrayList;
import java.util.List;

String path = "polymarket_orderbook_preview_g0_1.tsfile";
try (TsFileSequenceReader sequenceReader = new TsFileSequenceReader(path);
     TsFileReader reader = new TsFileReader(sequenceReader)) {
    // Devices are enumerated as IDeviceID objects. Tables with TAG columns
    // have multi-segment device ids that cannot be rebuilt from their string
    // form, so always use getAllDevices() rather than getAllDeviceIds().
    for (IDeviceID device : sequenceReader.getAllDevices()) {
        List<Path> paths = new ArrayList<>();
        for (MeasurementSchema schema : reader.getMeasurement(device)) {
            String name = schema.getMeasurementName();
            if (name != null && !name.isEmpty() && !name.equalsIgnoreCase("Time")) {
                paths.add(new Path(device, name, true));
            }
        }
        QueryDataSet dataSet = reader.query(QueryExpression.create(paths, null));
        while (dataSet.hasNext()) {
            System.out.println(dataSet.next());
        }
    }
}

Source & license

• Original dataset: astrnvk/polymarket-orderbook-data-preview
• Source card title: Polymarket Microstructure — Free Preview (26–29 Jun + L2 depth sample)
• License: cc-by-nc-4.0 (as declared by the original dataset).