Fast DataFrames for Ruby(github.com) |
Fast DataFrames for Ruby(github.com) |
Modern Polars: A comparison of the Polars and Pandas dataframe libraries - https://news.ycombinator.com/item?id=34275818 - Jan 2023 (62 comments)
Also:
Polars: Fast DataFrame library for Rust and Python - https://news.ycombinator.com/item?id=29584698 - Dec 2021 (124 comments)
Polars: Rust DataFrames Based on Apache Arrow - https://news.ycombinator.com/item?id=23768227 - July 2020 (1 comment)
If we'd have just given the team a locked down Data Studio/PowerBI report none of that would have happened. Encouraging people to peek under the hood can be a huge benefit.
Most people don't have large datasets (even many people who think they have large datasets). Some do, or require more complex supporting libraries, and I get that Ruby then often isn't practical for them, and that's fine.
But it's nice to know I now have one more option reducing my need to consider another language.
I'd like to see spark bindings first though before I would seriously consider it.
https://github.com/ondra-m/ruby-spark
Hasn't been updated in a long time - no idea if that's because it's complete enough or if it's been abandoned, as I don't use Spark so I haven't tested this gem myself.
The use of the words "I'm trying to imagine" is really very provocative, because it suggests that the commenter has thought hard about it comprehensively eliminated any possible reason one would use Ruby. Which in turn implies there is literally nothing good about Ruby at all.
So someone has poured heart and soul into building a free library in their own time, giving it away to everyone and the response is to casually dismiss it with a remark suggesting there is literally no reason for it to exist.
Is it a big deal? no. But it would be good if people tried to be kind when commenting.
But there are often real world constraints that influence the choice of the tool. For example if you have a Rails app already, going with more Ruby code might fit better than branching out to Python or Julia.
> and some languages are better tools for certain tasks.
Yes, although in the case of Python, it's not the language that is better but the eco system support.
We process billions of events/records weekly with just Ruby and Sidekiq/Redis at my current job. Its way easier to extend what we have with Ruby than switching to Spark/Python/Scala/Kafka or whatever etc is complete overkill.