Closed davido closed 5 years ago
It's possible that CC optimizing goog.object.equals reordered the two loops since less work is done in the second loop, they both return the same value, and don't side-effect without proxy traps or setters.
That's pretty hand wavey though, so I'll dig in when I get a chance.
@mikesamuel Thanks for looking into it.
I belive that polymer-resin is broken with polymer 1.x now :( (Or rather closure compiler broke it).
I am seeing alot of this
Failed to sanitize attribute of <li>: <li aria-label="
"2222"
">
Though i cannot reproduce this on https://gerrit-review.googlesource.com/c/plugins/owners/+/223319 (which makes me think they are using a different version of resin?)
You can reproduce this at https://gerrit.git.wmflabs.org/r/c/All-Projects/+/2946
resin 2.0 fixes it. Im going to upgrade gerrit to resin 2.0.
After the upgrade of
closure-compiler
from v20180805 to the version v20190325 inrules_closure
in https://github.com/bazelbuild/rules_closure/pull/347, Gerrit Code Review cannot built any more: [1].The workaround is to add new suppression to the
JSC_REFERENCE_BEFORE_DECLARE
to theclosure_js_library
function invocation:The affected part of the code seems to be:
for(b in a)d[b]=a[b]
in https://github.com/Polymer/polymer-resin/blob/master/standalone/polymer-resin.js. FWIW, I don't see the corresponding part in https://github.com/Polymer/polymer-resin/blob/master/standalone/polymer-resin-debug.js.[1] http://paste.openstack.org/show/748913.