Add world age hint for UndefVarError #58572
Conversation
Similar to the existing world age hint for MethodError, this adds a helpful message when an UndefVarError occurs because a binding was defined in a newer world age than the code trying to access it. The implementation checks if a binding that was undefined at the error's world age is now defined in the current world, and displays: "The binding may be too new: running in world age X, while current world is Y." Additionally, for all binding kinds, the error hint now notes when the binding state has changed between the error's world and the current world. 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <[email protected]>
|
|
||
| # Get the current world's binding partition for comparison | ||
| curworld = tls_world_age() | ||
| cur_bpart = lookup_binding_partition(curworld, GlobalRef(scope, var)) |
There was a problem hiding this comment.
The printing world doesn't really have anything in common with the error world, especially in the REPL. Can Claud add the field to UndefVarError?
There was a problem hiding this comment.
UndefVarError already has the field.
There was a problem hiding this comment.
The current world is just for the extra warning if the current kind is different from the old kind.
There was a problem hiding this comment.
Ah, great. I didn't realize we'd already done that. I assumed that Claude was just hallucinating it based on MethodError. I think we want get_world_counter() here though, not tls_world_age().
Lines 347 to 349 in 44c3813
There was a problem hiding this comment.
Claude thought so too, but I told it not to. I want to change the MethodError also, because otherwise there can be a TOCTOU issue with the hasmethod and the printing using different worlds. I think that's just confusing.
There was a problem hiding this comment.
I thought hasmethod is a tls_world query these days to prevent the TOCTOU issue (looks like that is used incorrectly above). The methods listing query though is a latest world query, so it should try to be consistent with that.
There was a problem hiding this comment.
Will make a separate PR to discuss that, but in any case, I think this should be tls_world_age
This rolls up a couple of bug fixes with some tweaks to documentation and AGENTS.md that came in useful getting the AI to do #58572.
This rolls up a couple of bug fixes with some tweaks to documentation and AGENTS.md that came in useful getting the AI to do #58572.
vtjnash
added
error messages
This rolls up a couple of bug fixes with some tweaks to documentation and AGENTS.md that came in useful getting the AI to do #58572.
Similar to the existing world age hint for MethodError, this adds a helpful message when an UndefVarError occurs because a binding was defined in a newer world age than the code trying to access it. The implementation checks if a binding that was undefined at the error's world age is now defined in the current world, and displays: "The binding may be too new: running in world age X, while current world is Y." Additionally, for all binding kinds, the error hint now notes when the binding state has changed between the error's world and the current world. 🤖 Generated with [Claude Code](https://claude.ai/code) Co-authored-by: Claude <[email protected]> (cherry picked from commit 2e39f64)
3 participants
Similar to the existing world age hint for MethodError, this adds
a helpful message when an UndefVarError occurs because a binding
was defined in a newer world age than the code trying to access it.
The implementation checks if a binding that was undefined at the
error's world age is now defined in the current world, and displays:
"The binding may be too new: running in world age X, while current
world is Y."
Additionally, for all binding kinds, the error hint now notes when
the binding state has changed between the error's world and the
current world.
🤖 Generated with Claude Code