q README

The project's README from github.com/harelba/q. As one markdown file: README.md.

Build and Package

q - Text as Data

q's purpose is to bring SQL expressive power to the Linux command line and to provide easy access to text as actual data.

q allows the following:

The following table shows the impact of using caching:

Rows Columns File Size Query time without caching Query time with caching Speed Improvement
5,000,000 100 4.8GB 4 minutes, 47 seconds 1.92 seconds x149
1,000,000 100 983MB 50.9 seconds 0.461 seconds x110
1,000,000 50 477MB 27.1 seconds 0.272 seconds x99
100,000 100 99MB 5.2 seconds 0.141 seconds x36
100,000 50 48MB 2.7 seconds 0.105 seconds x25

Notice that for the current version, caching is not enabled by default, since the caches take disk space. Use -C readwrite or -C read to enable it for a query, or add caching_mode to .qrc to set a new default.

q's web site is https://harelba.github.io/q/ or https://q.textasdata.wiki It contains everything you need to download and use q immediately.

Usage Examples

q treats ordinary files as database tables, and supports all SQL constructs, such as WHERE, GROUP BY, JOINs, etc. It supports automatic column name and type detection, and provides full support for multiple character encodings.

Here are some example commands to get the idea:

$ q "SELECT COUNT(*) FROM ./clicks_file.csv WHERE c3 > 32.3"

$ ps -ef | q -H "SELECT UID, COUNT(*) cnt FROM - GROUP BY UID ORDER BY cnt DESC LIMIT 3"

$ q "select count(*) from some_db.sqlite3:::albums a left join another_db.sqlite3:::tracks t on (a.album_id = t.album_id)"

Detailed examples are in here

Installation.

New Major Version 3.1.6 is out with a lot of significant additions.

Instructions for all OSs are here.

The previous version 2.0.19 Can still be downloaded from here

Contact

Any feedback/suggestions/complaints regarding this tool would be much appreciated. Contributions are most welcome as well, of course.

Linkedin: Harel Ben Attia

Twitter @harelba

Email [email protected]

q on twitter: #qtextasdata

Patreon: harelba - All the money received is donated to the Center for the Prevention and Treatment of Domestic Violence in my hometown - Ramla, Israel.

q (harelba/q)

sqlite3
Searching Database…

10.4K ·GitHub
Q
What is this?q is a command-line tool that runs SQL directly on delimited text files (CSV, TSV) and on multi-file SQLite databases, treating files as tables you can query.
Is it a library I import into code, or something I install and run?It's a standalone command-line tool written in Python — you install it once and then run queries straight from your shell, no code required.
How do I get it onto my machine?Download it from q's website, which contains everything you need to use it immediately — there are installation instructions for all OSs, and the current major version is 3.1.6.
Will it run on my Mac or Windows box, or is it Linux-only?It's cross-platform: the README points to installation instructions for all operating systems, and it's packaged for Windows, macOS and Linux.
Is q free to use?Yes — it's open source under the GPL-3.0 license with no paid tiers. The author accepts Patreon donations, and all of that money is donated to a domestic violence prevention center in Ramla, Israel.
Does my data leave my machine when I query files?No — q is a local command-line tool that reads your files directly; nothing in the project involves uploading data to a server.
What are similar projects?visidata, miller, usql, datasette, xsv, pandas, prediction-market-analysis, pgcli
Which topics is it in?sqlite3, text-processing, csv, data-analysis, sql, data-science, sqlite, command-line, utility, database, gplv3, command-line-tool
Creator
harelba
Size
3.3 MB
Stars
10.4K
Updated
CommitDec 21Releasev3.1.6Nov 26
Age
14y
Commits · 90 Days