Performance is definitely the bottleneck to applying this kind of polyfill. I don't have anything concrete as a benchmark yet but in earlier versions I was manually testing with Bootstrap added to the page (100kb of CSS rules) and there was a small delay. I will be figuring out performance specifics a lot more before I try to convince anybody to actually use this approach. I was looking through the examples on the polymer homepage and thought that since they did work already in IE9/10 (surprised that the calculator worked in IE8!) that they would not be good examples to showcase this script so I'll probably come up with an elaborate page that ensures tons of rules will need to be negated.
My inspiration for development was a project that overlays a dialog (of test results) on top of unknown websites. Others on the team balked at my idea of wrapping the entire thing in an iframe (still what I consider the most prudent approach) and suggested I look into WebComponents. Until I can prove that this approach is usable, it's only wishful thinking. My Windows VM for IE testing will show any bottleneck though as I'm running on an I3 1.4ghz dual core laptop. If performance is usable in this environment, it is probably decent elsewhere. You have caught me on multiple nested shadow root style applications. At this point, one instance of '[shadow]' in a rule will select for any depth of shadow nesting, kind of like a /deep/ selector. The operation is currently very simple: https://github.com/numtel/shadowstyles/blob/master/src/shadowStyles.js#L108 In the near future, I would like to determine a way to improve this. My intuition tells me to split the selector on '[shadow]' then check to see if each ancestral shadowed element matches the leading part of the selector before/after checking if the child element matches the full selector. I will continue to post updates as I progress. Thanks for your interest! On Wednesday, November 12, 2014 10:07:49 AM UTC-8, Steve Orvell wrote: > > This is an interesting idea. We didn't consider this type of approach for > Polymer because we thought the performance not be good enough. Do you have > a sense of how your approach impacts performance? > > Also, does this script isolate shadowRoots from styles applying to > shadowRoots of ancestor elements? (e.g. a rule like x-foo[shadow] p { > color: red; } blocked from applying to x-foo -> SR -> x-bar -> SR -> p) > > On Tue, Nov 11, 2014 at 4:20 PM, 'Scott Miles' via Polymer < > [email protected] <javascript:>> wrote: > >> Sorry you haven't had any reply. Your project is very interesting. The >> only reason you aren't getting feedback right away is because at the moment >> we are focused on non-CSS problems, and we are all at 100% cpu utilization. >> =P >> >> If we were in the middle of working on CSS systems, you probably would >> have received a reply sooner. For now, please hang in there, we will take a >> look at your code as soon as we are able. >> >> Scott >> >> >> >> On Mon, Nov 10, 2014 at 8:36 PM, numtel <[email protected] >> <javascript:>> wrote: >> >>> Those two hacks are now gone... The source and the UX are much better >>> now. >>> >>> Is this a good place to post about this script? I'm surprised to not >>> have any reply. >>> >>> 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 [email protected] <javascript:>. >>> To view this discussion on the web visit >>> https://groups.google.com/d/msgid/polymer-dev/3b9b280e-2e7e-4681-ac86-8571d9687825%40googlegroups.com >>> >>> <https://groups.google.com/d/msgid/polymer-dev/3b9b280e-2e7e-4681-ac86-8571d9687825%40googlegroups.com?utm_medium=email&utm_source=footer> >>> . >>> >>> For more options, visit https://groups.google.com/d/optout. >>> >> >> 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 [email protected] <javascript:>. >> To view this discussion on the web visit >> https://groups.google.com/d/msgid/polymer-dev/CAHbmOLYFxR27Jxj%3D5Qxfy9bnMfHygzo2Ttb4%2BrnVG1WGQVRyrQ%40mail.gmail.com >> >> <https://groups.google.com/d/msgid/polymer-dev/CAHbmOLYFxR27Jxj%3D5Qxfy9bnMfHygzo2Ttb4%2BrnVG1WGQVRyrQ%40mail.gmail.com?utm_medium=email&utm_source=footer> >> . >> >> For more options, visit https://groups.google.com/d/optout. >> > > 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 [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/polymer-dev/d6b121bf-46d4-42a8-a9cf-e23c34cd26e6%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
