A path towards removing URL literals #437
Labels
No labels
Area/build-packaging
Area/cli
Area/evaluator
Area/fetching
Area/flakes
Area/language
Area/profiles
Area/protocol
Area/releng
Area/remote-builds
Area/repl
Area/store
bug
crash 💥
Cross Compilation
devx
docs
Downstream Dependents
E/easy
E/hard
E/help wanted
E/reproducible
E/requires rearchitecture
imported
Needs Langver
OS/Linux
OS/macOS
performance
regression
release-blocker
RFD
stability
Status
blocked
Status
invalid
Status
postponed
Status
wontfix
testing
testing/flakey
ux
No milestone
No project
No assignees
2 participants
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference: lix-project/lix#437
Loading…
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
URL literals are a notoriously bad feature and we all agree that one day they should be no more. Despite probably nobody using them, the mere existence of the feature in the parser causes significant performance overhead (TODO quote @pennae for some numbers).
Currently URL literals can be merely disabled with
--extra-experimental-features=no-url-literals
, however doing so currently gives no performance benefits.The cleanest solution would be to implement proper language versioning which allows us to do breaking changes (see also RFC 137). However, this will not be implemented in the near future, therefore a different graceful mechanism for its deprecation is needed.
To remove URL literals in the language, the following steps are proposed:
no-url-literals
by default (explicitly specifying it is a no-op or maybe prints a warning), and add an option to disable it via a new--extra-deprecated-features=url-literals
.--extra-deprecated-features=url-literals
will throw an error and--extra-experimental-features=no-url-literals
will remain a no-op for a while until eventually being removed.This issue was mentioned on Gerrit on the following CLs: