Free online tool
In developmentConvert Parquet to TSV
TSV (tab-separated values) is CSV's saner cousin: tabs are rare in real data, so quoting is rarely needed and the output is easier to grep, cut, and awk. The dedicated TSV converter is in development — meanwhile use the Parquet → CSV converter (most consumers accept either).
Why TSV over CSV?
Both formats serve the same purpose, but TSV has a few practical wins:
- Rare separator collisions. Free-text fields almost never contain tabs. Commas are everywhere — addresses, names, descriptions.
- Simpler parsing. Without quoting drama, you can split by
\tand get correct fields. - Unix-friendly.
cut -f,awk -F'\t', andpasteall default to tabs. - Excel still opens it. Excel auto-detects tab-delimited files (called "Text" in the import dialog).
The downside: copy-pasting from a webpage or chat tool can convert tabs into spaces, breaking the file. CSV survives copy-paste better.
Convert Parquet to TSV manually
DuckDB
duckdb -c "COPY (SELECT * FROM 'data.parquet')
TO 'data.tsv'
(DELIMITER E'\t', HEADER)"Python (pandas)
import pandas as pd
df = pd.read_parquet("data.parquet")
df.to_csv("data.tsv", sep="\t", index=False)In the meantime
The Parquet → CSV converter is live. Most TSV consumers also accept CSV; if you absolutely need a tab separator, run the CSV file through tr ',' '\t' locally (with the caveat that this won't handle quoted commas).