Profiling polymer polyfills

[Don Olmstead]

I'm exporting my Polymer elements to a platform that doesn't have native implementations of the Web Components spec. Its a WebKit port roughly equivalent to Safari 7.

I wanted to see how much time the polyfills, and the Polymer sugaring layer, were taking on the platform so I was wondering if there are any particular areas that I should take a look at and place some markers around. My main concerns are the startup time and the custom element creation but I'd also like to get some idea on the cost of other things such as the Shadow DOM polyfills.

If anyone has run Polymer through a structural profiler I'd be interested in what they found.

Thanks!

[Matthew McNulty]

First off, you're on the right track that startup time in the polyfilled environment is a hotspot. But this is an area of extremely active development - especially the shadow DOM polyfill, as we work on trying to minimize this hotspot in the short-term. It's more of a concern on mobile, which may or may not be relevant to your environment. We should have more news to share on this shortly.

One note of caution - generally when profiling, it's pretty easy to fall into the trap of micro benchmarks. Certain DOM operations can be significantly slower in the polyfills when tested in isolation, but much less so in real world scenarios. 

We'd be happy to take a look at whatever you find, or scenarios you've found to be particularly problematic.

Follow Polymer on Google+: plus.google.com/107187849809354688692
---
You received this message because you are subscribed to the Google Groups "Polymer" group.
To unsubscribe from this group and stop receiving emails from it, send an email to polymer-dev...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/polymer-dev/b6961d2e-0964-40d2-ab32-c4e018cc26ef%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.