Skip to content
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

src: fix vm enumerator callbacks #22836

Closed
wants to merge 2 commits into from
Closed

Conversation

addaleax
Copy link
Member

Some of the choices here are odd, including that symbols are missing.
However, that matches previous behaviour.

What had to be changed was that inherited properties are no longer
included; the alternative would be to also refactor the descriptor
callbacks to provide data for inherited properties.

Fixes: #22723
Refs: #22390

Checklist
  • make -j4 test (UNIX), or vcbuild test (Windows) passes
  • tests and/or benchmarks are included
  • commit message follows commit guidelines

@nodejs/vm

@nodejs-github-bot nodejs-github-bot added c++ Issues and PRs that require attention from people who are familiar with C++. vm Issues and PRs related to the vm subsystem. labels Sep 13, 2018
@addaleax
Copy link
Member Author

Looks like this also fixes a known-issues test? :)

CI: https://ci.nodejs.org/job/node-test-pull-request/17166/


assert.deepStrictEqual(
Object.keys(proxied).sort(),
['0', '1', 'not', '0.5', '-1'].sort());
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is the sorting necessary? Object.keys() and Object.getOwnPropertyNames() should always return the properties in the order you already expect them to be. Everything else is likely a bug.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It fails if I don’t do this, even though this is the order in which they were declared. I’m not sure on the “why” of that, though. (i.e. this is also a question for @nodejs/vm)

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

By spec the return value is:

  1. all indices (integers >= 0) in ascending order
  2. all other non-symbol properties in their insertion order
  3. all symbol properties in their insertion order

So if something else is returned, we should probably look into that.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Would you be so kind and add a comment that this is against the spec and should be fixed?

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@BridgeAR does b9ec6b825844 sound okay?

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yes :)

@jdalton
Copy link
Member

jdalton commented Sep 13, 2018

@addaleax

What had to be changed was that inherited properties are no longer
included;

No longer included in/from what?
Is this related to #22390 and the V8 change (@camillobruni) or the backport (@TimothyGu)?

Copy link
Member

@jdalton jdalton left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The property order issue raised is an indicator that there's a bug hiding somewhere.

@addaleax
Copy link
Member Author

@jdalton Just to be clear, the current behavior is to hard-crash.


assert.deepStrictEqual(
Object.keys(proxied).sort(),
['0', '1', 'not', '0.5', '-1'].sort());
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Would you be so kind and add a comment that this is against the spec and should be fixed?

@addaleax
Copy link
Member Author

Ping @jdalton – is your objection standing, even though I’d say this is strictly an improvement over crashing the process?

@jdalton
Copy link
Member

jdalton commented Sep 17, 2018

I believe my objection stands. If I understand correctly the VM enumerator callback patch was reverted because of issues this PR is attempting to fix. So there is no crash in production or stable bits? Now you're presenting a fix but one that is buggy and asking if I think it's better than the crash scenario that isn't in production or a pending release?

@addaleax
Copy link
Member Author

@jdalton The VM enumerator callback patch was not reverted at this point, no – but we could still do that. Would you prefer to do that, then for me to rebase this PR on top of it and wait until we’ve ironed everything out? That would be okay with me, just indicate your preference.

@jdalton
Copy link
Member

jdalton commented Sep 17, 2018

@addaleax Ah ok! Thank you for clarifying things for me. I think that it would be best to revert the patch, and then iron out things to a greater extent, before pushing it out in a release.

addaleax added a commit to addaleax/node that referenced this pull request Sep 17, 2018
This reverts commit 85c356c
from PR nodejs#22390.

See the discussion in the (proposed) fix at
nodejs#22836.

Refs: nodejs#22836
Refs: nodejs#22390
Fixes: nodejs#22723
addaleax added a commit that referenced this pull request Sep 18, 2018
This reverts commit 85c356c
from PR #22390.

See the discussion in the (proposed) fix at
#22836.

Refs: #22836
Refs: #22390
Fixes: #22723

PR-URL: #22911
Reviewed-By: Colin Ihrig <[email protected]>
Reviewed-By: Gus Caplan <[email protected]>
Reviewed-By: Tiancheng "Timothy" Gu <[email protected]>
Reviewed-By: John-David Dalton <[email protected]>
Reviewed-By: Ruben Bridgewater <[email protected]>
TimothyGu and others added 2 commits September 18, 2018 15:50
This is a re-land of a commit landed as part of
nodejs#22390.

---

This allows using a Proxy object as the sandbox for a VM context.

Refs: nodejs#22390
Fixes: nodejs#17480
Fixes: nodejs#17481
Reviewed-By: Anna Henningsen <[email protected]>
Reviewed-By: Gus Caplan <[email protected]>
Reviewed-By: James M Snell <[email protected]>
Some of the choices here are odd, including that symbols are missing.
However, that matches previous behaviour.

What had to be changed was that inherited properties are no longer
included; the alternative would be to also refactor the descriptor
callbacks to provide data for inherited properties.

Fixes: nodejs#22723
Refs: nodejs#22390
@addaleax addaleax added the wip Issues and PRs that are still a work in progress. label Sep 24, 2018
@ryzokuken
Copy link
Contributor

@addaleax is this abandoned? Do you want someone else to pick this up?

@addaleax addaleax added the help wanted Issues that need assistance from volunteers or PRs that need help to proceed. label Nov 15, 2018
@fhinkel
Copy link
Member

fhinkel commented Oct 28, 2019

I'm cleaning out a few old PRs 🧹. I'm closing this due to inactivity. Please re-open if needed!

@fhinkel fhinkel closed this Oct 28, 2019
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
c++ Issues and PRs that require attention from people who are familiar with C++. help wanted Issues that need assistance from volunteers or PRs that need help to proceed. vm Issues and PRs related to the vm subsystem. wip Issues and PRs that are still a work in progress.
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Abort during Object.keys after vm.runInContext
9 participants