dsq VS jq

You might want to look at tsv-utils, or a similar project: https://github.com/eBay/tsv-utils

For the SQL part, but maybe a lot heavier, you can use one of the projects listed on this page: https://github.com/multiprocessio/dsq (No longer maintained, but has links to lots of other projects)

Welcome to the gang! :)

https://github.com/multiprocessio/dsq#comparisons

SPyQL is really cool and its design is very smart, with it being able to leverage normal Python functions!

As far as similar tools go, I recommend taking a look at DataFusion[0], dsq[1], and OctoSQL[2].

DataFusion is a very (very very) fast command-line SQL engine but with limited support for data formats.

dsq is based on SQLite which means it has to load data into SQLite first, but then gives you the whole breath of SQLite, it also supports many data formats, but is slower at the same time.

OctoSQL is faster, extensible through plugins, and supports incremental query execution, so you can i.e. calculate a running group by + count while tailing a log file. It also supports normal databases, not just file formats, so you can i.e. join with a Postgres table.

[0]: https://github.com/apache/arrow-datafusion

[1]: https://github.com/multiprocessio/dsq

[2]: https://github.com/cube2222/octosql

Disclaimer: Author of OctoSQL

> dsq registers go-sqlite3-stdlib so you get access to numerous statistics, url, math, string, and regexp functions that aren't part of the SQLite base. (https://github.com/multiprocessio/dsq#standard-library)

Ah, I wondered if they rolled their own SQL parser, but no, I now see the sqlite.go in the repo and all is made clear

I am currently evaluating dsq and its partner desktop app DataStation. AIUI, the developer of DataStation realised that it would be useful to extract the underlying pieces into a standalone CLI, so they both support the same range of sources.

dsq CLI - https://github.com/multiprocessio/dsq

OctoSQL is an awesome project and Kuba has a lot of great experience to share from building this project I'm excited to learn from.

And while building a custom database engine does allow you to do pretty quick queries, there are a few issues.

First, the SQL implemented is nonstandard. As I was looking for documentation and it pointed me to `SELECT * FROM docs.functions fs`. I tried to count the number of functions but octosql crashed (a Go panic) when I ran `SELECT count(1) FROM docs.functions fs` and `SELECT count() FROM docs.functions fs` which is what I lazily do in standard SQL databases. (`SELECT count(fs.name) FROM docs.function fs` worked.)

This kind of thing will keep happening because this project just doesn't have as much resources today as SQLite, Postgres, DuckDB, etc. It will support a limited subset of SQL.

Second, the standard library seems pretty small. When I counted the builtin functions there were only 29. Now this is an easy thing to rectify over time but just noting about the state today.

And third this project only has builtin support for querying CSV and JSON files. Again this could be easy to rectify over time but just mentioning the state today.

octosql is a great project but there are also different ways to do the same thing.

I build dsq [0] which runs all queries through SQLite so it avoids point 1. It has access to SQLite's standard builtin functions plus* a battery of extra statistic aggregation, string manipulation, url manipulation, date manipulation, hashing, and math functions custom built to help this kind of interactive querying developers commonly do [1].

And dsq supports not just CSV and JSON but parquet, excel, ODS, ORC, YAML, TSV, and Apache and nginx logs.

A downside to dsq is that it is slower for large files (say over 10GB) when you only want a few columns whereas octosql does better in some of those cases. I'm hoping to improve this over time by adding a SQL filtering frontend to dsq but in all cases dsq will ultimately use SQLite as the query engine.

You can find more info about similar projects in octosql's Benchmark section but I also have a comparison section in dsq [2] and an extension of the octosql benchmark with different set of tools [3] including duckdb.

Everyone should check out duckdb. :)

[0] https://github.com/multiprocessio/dsq

[1] https://github.com/multiprocessio/go-sqlite3-stdlib

[2] https://github.com/multiprocessio/dsq#comparisons

[3] https://github.com/multiprocessio/dsq#benchmark

What's annoying about this is that the PR doesn't even say it's trying to run tests. It says everything is passing and just doesn't list the actions.

For a second I thought someone must have deleted the actions yaml files.

This is a dangerous failure mode.

https://github.com/multiprocessio/dsq/pull/82

This is a cool project! But if you query Excel and ODS files with dsq you get the same thing plus a growing standard library of functions that don't come built into SQLite such as best-effort date parsing, URL parsing/extraction, statistical aggregation functions, math functions, string and regex helpers, hashing functions and so on [1].

[0] https://github.com/multiprocessio/dsq

[1] https://github.com/multiprocessio/go-sqlite3-stdlib

ob-dsq gives org source block support for the command line tool dsq. This lets you query an org-mode table with SQL. So that's good if you know SQL. And is more powerful than just simple column filtering.

I think like you. But also, one does not necessarily know beforehand that they will want to make money.

Like a project could be born out of pure generosity, but after the happy initial phase the project might get too heavy on the maintenance requirements, causing the author to approach burnout, and possibly deciding that they want to make money to continue pulling the cart forward.

However, here's something I do think: if you create something as Open Source, it should be out of a mentality of goodwill and for the greater good, regardless of how it ends up being used. OSS licenses do mean this with their terms. If you later get tired or burned out, you should just retire and allow the community to keep taking care of it. Just like it happened with the Jq tool [1].

[1]: https://github.com/jqlang/jq/releases/tag/jq-1.7

Official Documentation: jqlang.github.io/jq

View on GitHub

To handle JSON files and JSON outputs in a script or format and highlight it, jq can be very handy. Many command line tools provide a json output, so you don't have to write a custom parser for a table a list in a terminal. Instead of that, you can use jq to get a specific value from the output or even modify the output. For more information, you can visit https://jqlang.github.io/jq/

In some projects I've wanted to use HTTPie to test APIs and jq to work with some JSON data. Nix has been really helpful in managing those dependencies that I can't easily get from npm.

> I'd love to see programs communicate through a typed JSON/proto format that shed enough details to make this more independent, and get useful shell command structuring/completion or full blown GUIs from simply introspecting the expected input and output types.

You should try PowerShell. It's basically Microsoft's .NET ecosystem molded into an interactive command line. I'm not entirely sure if PoweShell can make full use of the static types that build up its core, but its ability to exchange objects in the command line is almost unmatched.

On Linux you can use `jc` (https://github.com/kellyjonbrazil/jc) combined with `jq` (https://jqlang.github.io/jq/) to glue together command lines.

Yeah, but muscle memory bites me all the time and I put the backslash on the closing paren, too, because I'm so used to the regex usage of that syntax which needs them to match

I also want to draw the reader's attention to the magic of |@uri <https://github.com/jqlang/jq/blob/jq-1.7/docs/content/manual...> for a bunch of cases, but doubly so in TFA's case where they're plugging strings into a URI context. Simple string concat often works great for "hello world", but the world is not always just hello, so one quick use of the filter and jq's got your back

  echo "the world's scary" | jq -Rr '"\(.)"'

I think the original devs just got burnt out for a while https://github.com/jqlang/jq/issues/2305#issuecomment-157263...

jq 1.7 do preserve large integers but will truncate if any operation is done on them. Unfortunetly it currently truncates to a decimal64 which is a bit confusing, this will be fixed in next release where it follow the suggestion from the JSON spec and truncates to binary64 (double) https://github.com/jqlang/jq/pull/2949

The fact that jq takes almost a second to run on a Pi is crazy[0]. And the tool is written in C.

[0] https://github.com/jqlang/jq/issues/1411