1141401 - Robustness improvements for JSDOMParser

Status: RESOLVED WORKSFORME

(Toolkit :: Reader Mode, defect)

Description

[[:fabrice] Fabrice Desré]

I'd like it to be able to parse pages like http://www.nytimes.com/2014/12/09/science/a-one-way-trip-to-mars-many-would-sign-up.html

Comment 1

[:Gijs (he/him)]

What's currently not working about JSDOMParser? The page in question renders in reader mode fine - it misses paragraphs, but that's likely readability's fault, not jsdomparser. As it is, this report is incomplete because it lacks the information required to take action on it.

For some background: JSDOMParser requires an XML-compatible DOM string. There is no real reason to relax that requirement (we make sure the thing it gets from firefox is xml-compatible even if the original document wasn't by using an xml-based serializer). For a more robust parser, maybe see jsdom ( https://github.com/tmpvar/jsdom/ ).

Comment 2

[[:fabrice] Fabrice Desré]

When I filed this bug I was feeding it the raw http response, because I didn't want to go through a parse-then-reserialize-then-reparse chain. I'm not working on that anymore, so feel free to ignore.

Comment 3

[:Gijs (he/him)]

Thanks, going to resolve WFM then.