Which Data Formats Can VisiData Open?

Updated

VisiData opens a wide range of tabular and structured data formats, including CSV, TSV, Excel (.xls/.xlsx), JSON/JSON-Lines, YAML, HTML, SQLite, HDF5, NumPy, Pandas DataFrames, SAS/SPSS/Stata files, Shapefiles, packet captures, zip archives, git repositories, fixed-width text, and structured log files. If your data is tabular or can be treated as rows and columns, VisiData is likely to load it directly from the command line.

Core tabular and text formats

These are the formats most users encounter day to day. VisiData treats them as first-class inputs and lets you explore, sort, filter, and compute without writing a script.

Format Typical file Notes
CSV .csv Comma-separated values; the most common starting point
TSV .tsv Tab-separated values
Fixed-width text .txt Columns defined by character positions
Structured log files varies Rows parsed from log lines
Markdown .md Tables in Markdown, including org-mode variants
Plain text .txt Loaded as a single column or split as needed

The site's own example uses a batch command on a CSV file:

vd -b input.csv

The -b flag runs VisiData in batch mode, which is useful for scripted pipelines rather than interactive exploration. For interactive work you would launch vd input.csv and navigate with keystrokes.

Spreadsheet and document formats

  • Excel — .xls and .xlsx files open directly, so you can inspect sheets without Excel.
  • JSON and JSON-Lines — both nested JSON and newline-delimited JSON are supported.
  • HTML — tables embedded in HTML pages can be loaded.
  • YAML — structured YAML documents are treated as tabular data.
  • GeoJSON — geographic feature collections load as rows.

This matters when your source is a web export or a config file rather than a clean spreadsheet. For example, if you need to compare rows in a downloaded HTML table against a CSV, both can be opened in the same tool.

Databases and scientific formats

VisiData reaches beyond flat files into data stores and analysis ecosystems:

  • SQLite — .db files open as tables you can browse.
  • HDF5 — hierarchical scientific data.
  • NumPy — .npy and .npz arrays.
  • Pandas DataFrame — usable within the Python-integrated workflow.
  • SAS/SPSS/Stata — .xpt, .sas7bdat, .sav, .dta files.
  • Shapefile — .shp geospatial data.
  • Packet capture — .pcap network captures.

The common thread is that each format maps to rows and columns. The site describes VisiData as combining "the clarity of a spreadsheet, the efficiency of the terminal, and the power of Python," and claims it can "handle millions of rows with ease" and "open 1m+ rows in the blink of an eye."

Archives and repositories

  • Zip archives — .zip files can be opened, letting you reach the files inside.
  • Git repositories — a repository can be loaded as a data source.

These are less about a single table and more about reaching data that is bundled or versioned. If your dataset ships as a zip of CSVs, you can open the archive rather than extracting first.

What to check before relying on a format

The format list is broad, but loading is only the first step. Before committing to VisiData for a given file, confirm:

  1. The file actually parses as tabular. Nested JSON or deeply hierarchical HDF5 may need flattening or a specific loader option.
  2. Your terminal supports the interface. The site notes the demo requires JavaScript and describes a "delightful terminal interface"; interactive use depends on a capable terminal.
  3. Large files behave as expected. The million-row claim is a good benchmark, but test with your own file size and column count.
  4. Batch vs. interactive mode fits your task. Use -b for pipelines, interactive mode for exploration.

Practical starting points

  • To explore interactively: vd yourfile.csv
  • To run a batch pipeline: vd -b input.csv
  • To check a format you are unsure about, open a small sample first and confirm the columns and row count look right.

The site points to an Install page, a Tutorial, a Quick Reference Guide, Documentation, and a "10 Ways to Use VisiData" article for deeper walkthroughs. If your data is tabular in any of the formats above, VisiData is designed to open it without a separate conversion step.