Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Roadmap to a more useable product #1

Closed
6 tasks done
pastelsky opened this issue Apr 2, 2017 · 11 comments
Closed
6 tasks done

Roadmap to a more useable product #1

pastelsky opened this issue Apr 2, 2017 · 11 comments

Comments

@pastelsky
Copy link
Owner

pastelsky commented Apr 2, 2017

Roadmap to 1.0 iff this idea catches on -

  • Evaluate using webpack instead of rollup (webpack seems to be more tolerant towards incorrect module exports. For eg. rollup will fail to bundle fail to bundle some popular packages like request) due to duplicate export names.
  • Show better looking / more actionable error messages and avoid alert()s
  • Implement a better queuing system for bundling is a CPU intensive process and Heroku times out after 30s.
  • Evaluate using babili vs. the current UglifyJS harmony minifier. Bundling some packages fail due to usage of ES6/7 syntax unsupported by Uglify JS.
  • Make the source code less ugly.
  • Better design, maybe?
@pastelsky pastelsky changed the title Roadmap to a useable product Roadmap to a more useable product Apr 3, 2017
@rickmed
Copy link

rickmed commented May 16, 2017

5 stars for this project.

Can I suggest:

  • Optionally having a list of modules for a grand total.
  • Adding an estimate cost for unpacking/parsing js (specially useful on mobile devices).

@pastelsky
Copy link
Owner Author

Hey, @rickmed. Could you explain what you mean by your first point? I did initially think of allowing users to add multiple packages in a single query, but bundling is a CPU intensive process, and Heroku would most probably time out.

Also I did think of adding parsing cost, but calculating so would require a dedicated machine so that times of different packages are comparable and independent of hosting environment and server load.

@rickmed
Copy link

rickmed commented May 16, 2017

  1. I have no idea how this works, is it possible to make it as a static page?
  2. I'm confused. How are you planning to calculate the parsing cost?

@pastelsky
Copy link
Owner Author

pastelsky commented May 17, 2017

  1. This cannot be a static page, as the calculations, bundling, minifying of modules happen on the server.
  2. A simple way to calculate the parse cost is to measure time taken to do a eval on the minified package code. But it would have to be done on an actual mobile device for good results.

@rickmed
Copy link

rickmed commented May 17, 2017

  1. I'm pretty sure it can be done but most likely would change the whole app -just throwing the idea out there...
  2. That sounds good. You could throttle CPU in devTools to get an estimate as well.

@jpdevries
Copy link

👍 for a feature request to somehow support multiple packages. Would be great to be able to do something like ?p=react@15.5.4,react-dom@15.5.4,rest of my packages... in your projects README.

Although there is the webpack visualizer for something similiar
https://chrisbateman.github.io/webpack-visualizer/

@zincli
Copy link

zincli commented Aug 18, 2017

How about supporting package sets and let users make comparison for their different strategy

@evan-scott-zocdoc
Copy link

Evaluate using babili vs. the current UglifyJS harmony minifier. Bundling some packages fail due to usage of ES6/7 syntax unsupported by Uglify JS.

I'd recommend using uglify-es instead, as it handles the newer syntaxes.

@pastelsky
Copy link
Owner Author

A new version is out now - https://github.com/pastelsky/bundlephobia/releases/tag/v1.0.1

Solves some of the concerns noted above. Also, support for batch requests is possible to implement with the current architecture.

@joshgillies
Copy link
Contributor

It would be cool if this project offered a way to compare two (or more?) packages.

@pastelsky
Copy link
Owner Author

Discounting a few quirks, Bundlephobia is pretty usable now. Closing this. Feature requests to go into a different issue.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

7 participants
@joshgillies @jpdevries @zincli @rickmed @pastelsky @evan-scott-zocdoc and others