test: cover try/catch coercion difference in new do elaborator#13849
Merged
Conversation
This PR adds a test pinning down the breaking change in the new `do` elaborator: it no longer accepts a `try`/`catch` body whose result type only matches the surrounding expected type up to coercion, bringing it in line with the rest of `do` and the non-`do` term form, where the same expression has never been accepted.
github-actions
Bot
added
the
toolchain-available
|
Mathlib CI status (docs):
|
Collaborator
|
Reference manual CI status:
|
pull Bot
pushed a commit
to justinlietz93/lean4
that referenced
this pull request
Jun 1, 2026
This PR makes the new `do` elaborator (leanprover#12459) the default by flipping `backward.do.legacy` to `false`. Legacy behavior remains available via `set_option backward.do.legacy true`. New surface syntax that is now accepted (and may shadow user-defined macros of the same shape): * `while let pat := e do …`, `while let pat ← e do …`, and `while h : cond do …` via a unified condition syntax (leanprover#13534). * `let ⟨a, b⟩ : T ← act` propagates the type annotation to the monadic action (leanprover#12866). * `letConfig` options (`(eq := h)`, `+nondep`, `+usedOnly`, `+zeta`) on do `let`/`have` (leanprover#13250, leanprover#13255). They are rejected on `let mut`, and `+postponeValue`/`+generalize` are rejected outright. * Anonymous dependent `if _ : p then … else …` now elaborates inside `do` blocks (leanprover#13192). Breaking changes: * `do` notation now requires a `Pure` instance, not just `Bind`. * The arms of `do match` are non-dependent by default; write `do match (dependent := true)` to recover the legacy term-match expansion. The patterns are still dependently refined (leanprover#12673), which subsumes many former dependent uses. * `try`/`catch` no longer accepts a body whose result type matches the surrounding expected type only via coercion (leanprover#13849). The same expression has never been accepted outside `do`. * Unreachable code now triggers a warning instead of an error. Code written after a `break`-less `repeat` is now reported as unreachable (leanprover#13506). * The syntax for `let pat := rhs | otherwise` and similar now scope over the `doSeq` that follows. Furthermore, `otherwise` and the sequence that follows are now `doSeqIndented` in order not to steal syntax from record syntax.
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
2 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.
This PR adds a test pinning down a breaking change in the new
doelaborator: it no longer accepts atry/catchbody whose result type only matches the surrounding expected type up to coercion, bringing it in line with the rest ofdoand the non-doterm form, where the same expression has never been accepted.