ECMA-262 Normative: formalize module keying host invariant predicates and variables#59
Open
guybedford wants to merge 2 commits intomainfrom
Open
ECMA-262 Normative: formalize module keying host invariant predicates and variables#59guybedford wants to merge 2 commits intomainfrom
guybedford wants to merge 2 commits intomainfrom
Conversation
guybedford
force-pushed
the
host-formalize
branch
from
0479ac1 to
620109d
Compare
guybedford
force-pushed
the
module-source
branch
from
ae5e047 to
6cacca6
Compare
guybedford
force-pushed
the
host-formalize
branch
from
620109d to
9f9c8c9
Compare
guybedford
changed the title
guybedford
changed the title
guybedford
force-pushed
the
host-formalize
branch
from
8de19d7 to
e454d9a
Compare
guybedford
force-pushed
the
host-formalize
branch
3 times, most recently
from
604bd28 to
62d30c3
Compare
guybedford
changed the title
nicolo-ribaudo
approved these changes
Mar 10, 2026
Member
nicolo-ribaudo
left a comment
nicolo-ribaudo
left a comment
There was a problem hiding this comment.
Minor comments, but it looks overall good.
| <li>the first _referrer_ is the same as the second _referrer_;</li> | ||
| <li>ModuleRequestsEqual(the first _moduleRequest_, the second _moduleRequest_) is *true*;</li> | ||
| <li><ins>_moduleRequestKey_.[[Phase]] is equal to ~source~ (the lowest possible phase);</ins></li> | ||
| <li><ins>FinishLoadingImportedModule will return the same NormalCompletion(_module_) _result_ for all (_otherReferrer_, _otherModuleRequest_) pairs whose Canonical ModuleRequest Record is also _moduleRequestKey_;</ins></li> |
Member
There was a problem hiding this comment.
Note that two normal completions containing the same module are not "the same", as spec record have identity.
Collaborator
Author
There was a problem hiding this comment.
Do you mean instead here spec records don't have identity?
In this case, how are we able to equate _result_ today in the host invariant at all? What is the suggestion here?
legendecas
reviewed
Mar 11, 2026
legendecas
reviewed
Mar 11, 2026
| <ul> | ||
| <li>the first _referrer_ is the same as the second _referrer_;</li> | ||
| <li>ModuleRequestsEqual(the first _moduleRequest_, the second _moduleRequest_) is *true*;</li> | ||
| <li><ins>_moduleRequestKey_.[[Phase]] is equal to ~source~ (the lowest possible phase);</ins></li> |
Member
There was a problem hiding this comment.
IIUC, _moduleRequestKey_.[[Phase]] could be ~evaluation~. What's the reason of this restriction?
Collaborator
Author
There was a problem hiding this comment.
The reason for this is that a canonical key must be independent of the phase - thus we pick a constant phase for canonicalization. And this aligns semantically with the concept that source phases can be used as keys in dynamic import.
Co-authored-by: Nicolò Ribaudo <hello@nicr.dev>
guybedford
force-pushed
the
host-formalize
branch
from
bb9d202 to
fe37af6
Compare
Collaborator
Author
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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.
The host hook invariants for module loading are a very important concept in the spec and this specification extends them with support for the source phase module invariants (
import(source)of a ModuleSource object).We require that
import(source)and import(url)` where URL and source point to the same module, to always return the same instance.In order to achieve this we need to define a concept of a module key.
Currently today, the host invariant of ES Modules is not fully defined - in the sense that we do not define that
import('./foo.js')andimport('https://site.com/foo.js')resolve to the same module, since we do not have any concept of a module key.This change now defines, the concept of a Canonical ModuleRequest Record, where all
(referrer, moduleRequest)pairs have a single canonical ModuleRequest Record they can be related to.By defining this as part of the host invariants, we are then able to talk in terms of module keys, laying the groundwork for source phase module keys.
In the process, we define all variables of the constraint explicitly, and then we move all predicates of the condition to the formal predicate list to ensure the explicit wording is as short as possible and this is as close as possible to a mathematical constraint as we can make it.
//cc @nicolo-ribaudo