Sourcery AI

Sourcery AI

Instant Python refactoring

5.0•2 reviews•

268 followers

Visit website
•
•
Sourcery is your AI code reviewer who lives in your GitHub repos - instantly reviewing every pull request.
This is the 3rd launch from Sourcery AI. View more

Sourcery 2.0

Continuous code review to make your code more maintainable
Sourcery makes sure all of your code is following best practices from standard Pythonic rules to issues that pop up during code reviews. Add to your IDE to get instant feedback, use the CLI to review existing code, or integrate it into your CI.
Sourcery 2.0 gallery image
Sourcery 2.0 gallery image
Sourcery 2.0 gallery image
Free Options
Launch tags:Productivity•Developer Tools•Tech
Launch Team
AppSignal
AppSignal
Built for dev teams, not Fortune 500s.
Promoted

What do you think? …

Tim Gilboy
Sourcery AI
Maker
Hey Product Hunt 👋 - @brendan_maginnis, @nick_thapen, and I are back and we’re very excited to be sharing the new version of Sourcery with you all. Sourcery sits in the background while you work analyzing your code, finding problems, and suggesting improvements to make sure it’s following best practices. You can think of it as the first layer of code review, catching mistakes before a human reviewer needs to take a look. As always - Sourcery runs fully locally, so your code is private. Here’s what’s new: 1. Extend Sourcery to handle your best practices. You shouldn’t have to make the same comment more than one time in a code review. When you see issues coming up frequently you can create a Sourcery rule, and these issues will get caught for your whole team before the code is merged. 2. Leverage Pre-Built Rulesets. We’ve taken the Google Python Style Guide and made it available as a (set of Sourcery rules). Pick and choose the rules that make sense for you (and customize them how you want) or add in the full ruleset. 3. CLI for bulk review. We’ve extended the Sourcery CLI to make it more powerful (and much quicker) for reviewing multiple files, or your entire repo. With a single command you can scan hundreds of files, identify common issues, and make bulk changes to improve your code. 4. Review every commit with Sourcery in CI. Let Sourcery handle the first layer of your code reviews by adding it to your CI. Sourcery is easily configurable to run only on changed code to help cut down on noise from legacy code. Anyone can start using Sourcery for free, and teams coming through ProductHunt with promo code PRODUCTHUNT2022 can use the full team version (includes full use in CI or as a pre—commit hook - see more details at https://sourcery.ai/team/) for free for 2 months.
Sean Kruzel
I use sorcery everyday for the last 2 years. I love combining it with GitHub Copilot to really speed up my development process! For example CoPilot generates code, then Sourcery improves it and fits it into our style guides. Finally there is nothing quite as satisfying as using the CLI to refactor a huge out-of-date repo. @brendan_maginnis @nick_thapen @tim_sourcery
Brendan Maginnis
Sourcery AI
Maker
Thanks @seankruzel! That's really cool how you use Sourcery and CoPilot together 🚀
Adil shaikh
Congrats on the launch
Tim Gilboy
Sourcery AI
Maker
@kaira_khan thank you!
Reka
Sourcery AI
Maker
@kaira_khan Thanks a lot.
Ali Sayed
Congrats on the launch
Tim Gilboy
Sourcery AI
Maker
@scorpi0n thank you! Let us know what you think about the new version :)
Lukas Kemkes
Love the progress, team Sourcery! Excited for our engineering team to test this out 🚀
Tim Gilboy
Sourcery AI
Maker
@lukas_kemkes thank you! Let us know what you think :)
Pranav Prakash
Farcaster Store
We have been using Sourcery for more than a year now. Our team is able to find some really good and time saving optimisations using this. I would recommend this.
Brendan Maginnis
Sourcery AI
Maker
@pranavprakash Thanks Pranav, really glad Sourcery is helpful for you
Dr. Christian Poensgen
Love it! Also, I'm a bit curious: Do you use Sourcery for Sourcery’s code?
Tim Gilboy
Sourcery AI
Maker
@dr_christian_poensgen we do! We make sure all our code is following our core rules (standard Python best practices) and then we've built out a bunch of additional rules to tackle things like making sure certain sections of our codebase are following the boundarires we want in our monorepo, making sure we're not using/extending deprecated classes in any way, enforcing that we use helper functions when we can, etc. Basically anytime something comes up more than once during a code review we try to turn it into a rule for ourselves in Sourcery Each of us has it in our IDE (including @brendan_maginnis in Vim) and then we also have it built into our IDE.
Dr. Christian Poensgen
... makes total sense, thanks for sharing these insights!
Jérôme Bau
Love the product and the team. Our team has been using it for quite some time. I'm especially excited about the best practices and rulesets.
Brendan Maginnis
Sourcery AI
Maker
Thanks @jerome_bau !
1234
Next
Last