4 ms·
I was just going to comment that it would be interesting to see the size of a run-through with browserify+uglifyify+uglify but looks like there is already a PR
by kasbah 11y ago
I was just going to comment that it would be interesting to see the size of a run-through with browserify+uglifyify+uglify but looks like there is already a PR for it.
https://github.com/samccone/The-cost-of-transpiling-es2015-in-2016/pull/12 https://github.com/samccone/The-cost-of-transpiling-es2015-i...
- wesleytodd 11y agoI just ran that PR and this is what it got: ----------------------------------------- stat -f%z src/dist/bundle.js 15757 gzip -c src/dist/bundle.js | wc -c 4270 ----------------------------------------- So still not quite as small as the rollup, closure, or webpack
- joshschreuder 11y agoI'm new to the whole webpack / browserify / etc. stuff, but this seems like a supremely weird build pipeline to me. > When using uglifyify you should generally also use Uglify, to achieve the smallest output. Uglifyify provides an additional optimization when used with Uglify, but does not provide all of the optimization that using Uglify on its own does, so it's not a replacement. Why doesn't Uglify just do all the work and not require a separate component to do extra optimisation? Or is the extra optimisation specifically tailored to browserify?
- dplgk 11y agoThe current state of javascript is to employ 40 tools for build process and 2000 modules to build your product.