Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 94.26%. Comparing base (4876807) to head (6d4abbe).
Report is 1 commits behind head on master.

@@           Coverage Diff           @@
##           master     #713   +/-   ##
=======================================
  Coverage   94.26%   94.26%           
=======================================
  Files           8        8           
  Lines        2094     2094           
  Branches      537      537           
=======================================
  Hits         1974     1974           
  Misses        120      120           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

I will file a release within the next few days.

@krystofwoldrich FYI: it will still take some time, since I want to avoid a ton of releases with just type fixes, so I'm working on #717 right now, which will fix all currently known type issues.

As part of that I also widened the accepted type to a literal type | string, just in case somebody only has that as a type for the mimeType to pass.

Would be really nice to get a new release out soon, as the latest version throws an error when you don't supply the mimeType arg, which makes it incompatible with things like saml2-js? That seems more important to fix than typings?

@papandreou this is intentional breaking change in version 0.9.0 which aligns the behavior of xmldom with the specs. This is not going to be "fixed", since it works as expected.
If the previous behavior is required you can not upgrade and have to stick to 0.8.*!
I do not recommend sticking to this behavior, since it means the content that you are parsing is used to determine the mime type. Which is only secure if you know what kind of content you expect, in which case you can also put the mime type as an argument.

The issue described here, is that this change is not reflected in the types of 0.9.0, which will be fixed along with all other type issues as part of the PR linked in my previous comment.

for further discussion on this topic I would prefer to have it in the release discussion #435

Ah, apologies, I thought it was the other way around 🙈

hmmm... side note - should such changes be in major releases?
Today catch production issues because of this change after "silent" upgrade of this package.

In 0.* (aka unstable) versions, breaking changes are happening in minor versions.
There are tools like Renovate, that are aware is aware of this.
Sorry that you were not aware of it, and were affected by this.