sleepydatapeek

A quick way to peek at local datafiles.

Welcome to sleepydatapeek!

One often needs to spit out a configurable preview of a data file. It would also be nice if said tool could detect and read several formats automatically.
sleepydatapeek has entered the chat!

Quickly summarize data files of type:

• csv
• parquet
• json
• pkl
• xlsx

And glance metadata for files:

• pdf
• png
• jpg|jpeg

ℹ️ Note that this tool presumes format by file extension. If you leave out extensions, or give csv data a .json extension for funsies, then you're being silly.

ℹ️ Due to how metadata formats vary across file types, how metadata is presented varies.

ℹ️ For further configuration options, see the sleepyconfig section below.

Get Started 🚀

pip install sleepydatapeek
pip install --upgrade sleepydatapeek

python -m sleepydatapeek --help
python -m sleepydatapeek data.csv
python -m sleepydatapeek doc.pdf

Usage ⚙

Set a function in your shell environment to run a script like:

alias datapeek='python -m sleepydatapeek'

Presuming you've named said macro datapeek, print the help message:

$ datapeek data.xlsx

════════════════════ data.xlsx ════════════════════
      Unnamed: 0    CustomerID  ProductName      Quantity  OrderDate      Price
--  ------------  ------------  -------------  ----------  -----------  -------
 0             0           101  Laptop                  2  2023-10-26      1200
 1             1           102  Mouse                   1  2023-10-26        25
 2             2           103  Keyboard                1  2023-10-27        50
 3             3           104  Monitor                 1  2023-10-27       300
 4             4           105  Headphones              3  2023-10-28        80

═══Summary Stats
╭──────────────┬─────────────────╮
│ Index Column │ (no_name):int64 │
├──────────────┼─────────────────┤
│ Row Count    │ 30              │
├──────────────┼─────────────────┤
│ Column Count │ 6               │
├──────────────┼─────────────────┤
│ Memory Usage │ < 0.00 bytes    │
╰──────────────┴─────────────────╯

═══Schema
╭─────────────┬────────╮
│ Unnamed: 0  │ int64  │
├─────────────┼────────┤
│ CustomerID  │ int64  │
├─────────────┼────────┤
│ ProductName │ object │
├─────────────┼────────┤
│ Quantity    │ int64  │
├─────────────┼────────┤
│ OrderDate   │ object │
├─────────────┼────────┤
│ Price       │ int64  │
╰─────────────┴────────╯
═══════════════════════════════════════════════════

Optionally, you can also get group-by counts for distinct values of a given column:

$ datapeek test.xlsx --groupby-count-column=ProductName

# typical output (elided)

═══Groupby Counts
  (row counts for distinct values of ProductName)
╭──────────────┬───╮
│ Laptop       │ 3 │
├──────────────┼───┤
│ Mouse        │ 3 │
├──────────────┼───┤
│ Keyboard     │ 3 │
├──────────────┼───┤
│ Monitor      │ 3 │
├──────────────┼───┤
│ Headphones   │ 3 │
├──────────────┼───┤
│ USB Drive    │ 3 │
├──────────────┼───┤
│ Printer      │ 3 │
├──────────────┼───┤
│ Webcam       │ 3 │
├──────────────┼───┤
│ Speakers     │ 3 │
├──────────────┼───┤
│ External HDD │ 3 │
╰──────────────┴───╯
═══════════════════════════════════════════════════

You can check metadata for certain file types too:

$ datapeek resume.pdf

📄test.pdf
╭──────────────┬─────────────────────────────────╮
│ CreationDate │ D:20250306111007-06'00'         │
├──────────────┼─────────────────────────────────┤
│ Creator      │ Adobe InDesign 20.1 (Macintosh) │
├──────────────┼─────────────────────────────────┤
│ ModDate      │ D:20250306111048-06'00'         │
├──────────────┼─────────────────────────────────┤
│ Producer     │ Adobe PDF Library 17.0          │
├──────────────┼─────────────────────────────────┤
│ Trapped      │ /False                          │
├──────────────┼─────────────────────────────────┤
│ Length       │ 48 pages                        │
╰──────────────┴─────────────────────────────────╯

SleepyConfig

You can personalize a few aspects of datapeek's behavior via a file strictly named ~/.sleepyconfig/params.yml. Paste the following into said file, and tinker to your liking:

datapeek_sample_size: 5
datapeek_table_style: 'rounded_grid'
datapeek_max_terminal_width: 80

All other sleepytools use this file as well. Browse my PyPI if you're interested!

Technologies 🧰

• Pandas
• Tabulate
• Typer
• PyArrow
• openpyxl
• PyPDF2
• PIllow

Contribute 🤝

If you have thoughts on how to make the tool more pragmatic, submit a PR 😊.

To add support for more data/file types:

1. append extension name to supported_formats in sleepydatapeek_toolchain.params.py
2. add detection logic branch to the main function in sleepydatapeek_toolchain/command_logic.py
3. update this readme

License, Stats, Author 📜

See License for the full license text.

This package was authored by Isaac Yep. 👉 GitHub
👉 PyPI