Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
node: warn for Object.prototype.__* accessors common in security warnings #39824
base: main
Are you sure you want to change the base?
node: warn for Object.prototype.__* accessors common in security warnings #39824
Changes from 4 commits
813dd33
c49435f
fd4e664
bc2ef67
05ef3f2
9f266d6
301d5ea
9198ad2
04fbd18
083c842
308eb0e
c55d04a
741f229
4b73dde
bd37bbc
07faa4b
File filter
Filter by extension
Conversations
Jump to
There are no files selected for viewing
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I get that
__proto__
is annoying, but it’s also widely used (which is why I had excluded it from #39576 as well). Runtime-deprecating that is a big breaking change.There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
__proto__
is the main problem for CVEs regarding prototype pollution, it is the most important to figure out how to address. I am not arguing popularity or utility, just that the API is problematic in practice.There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Right – that’s what
--disable-proto
is for, no? Anyway, if we do this, it should be communicated very clearly to users.(I also don’t think putting this behind
--pending-deprecation
is particularly useful, given that there already is a flag to opt-out of this behavior. If we do make the decision to runtime-deprecate fully eventually, then I guess that decision can also be made now.)There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
--disable-proto
is a bit different, it doesn't let you see that something is using__proto__
so you can fix it. It just removes it or makes it throw; also it is opt-in so the ecosystem noise is just permanent since it doesn't actually cause any sort of signal that security warning isn't a false positive.There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Well, I would say that throwing exceptions definitely lets you do that :)
In any case, to be clear I’m not -1 on this per se, I just think that this is a big change and we should call it out very explicitly.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Throwing alters behavior, so maybe --disable-proto could get a warning mode that lets programs run and you fix it when you see it rather than taking down a process potentially?
Seems fine to have whole blog posts before this and waiting for a major to me on this PR. Since this affects legacy codebases as well it will likely also take some effort to PR things.
We could also add a flag to re-add the accessors if we ever do remove them for people needing to run legacy code.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I mean – yes, throwing alters behavior, but practically speaking, people will notice whether they are using
__proto__
with either method, which is the point here anyw2ay.Yeah, I think in the long run that might be a good idea – just remove the accessors, but add a flag to add them for those who really need them.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Too bad that
--disable-proto=log
wasn't an option ?There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
On the other hand,
--disable-proto=throw
during development should spot all issues (if using a package lock).There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Perhaps we should make these proper runtime deprecations?
/cc @addaleax
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'm not opposed, is it just adding to the docs a new number to get the DEP###?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
added to DEP doc, idk if there is a process to do instead
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Look for other DEP codes in the code and you'll see how those are emitted :-)
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Example:
node/lib/sys.js
Lines 28 to 29 in e46c680
As soon as this is changed to emit a deprecation warning, I'll launch a CITGM run with
--throw-deprecation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
this seems difficult to wrangle without giving a unique DEP id to each accessor given the other feedback of giving more actionable feedback in #39824 (comment)
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
There is a
common.expectWarning()
utility.There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
done, but it seems
deprecate(...)
in lib seems to only fire once per dep code so having specialized messages per accessor seems to require unique DEP codes.