Open
Conversation
Implements error recovery inspired by Chumsky's recovery system in Rust, addressing GitHub issue alecthomas#342. New features: - RecoveryStrategy interface with multiple implementations: - SkipUntil/SkipPast: classic panic-mode recovery - SkipThenRetryUntil: skip and retry parsing - NestedDelimiters: balanced delimiter recovery - TokenSyncStrategy: sync on token types - CompositeStrategy: try multiple strategies in order - RecoveryError type for multi-error reporting - ParseOptions: Recover(strategies...) and MaxRecoveryErrors(n) Integration points: - parseContext with tryRecover(), recoveryEnabled(), addRecoveryError() - Recovery hooks in strct.Parse and group.Parse (nodes.go) - Error aggregation in parser.go The implementation allows parsers to continue after errors, collecting multiple errors and producing partial ASTs - useful for IDE integration, linters, and comprehensive error messages. Includes comprehensive test coverage (100% on new code) and example at _examples/recovery/. Amp-Thread-ID: https://ampcode.com/threads/T-b95d5ac9-6023-4947-a477-53364956f203 Co-authored-by: Amp <amp@ampcode.com>
Extends the error recovery system with Chumsky-inspired features: Per-Node Recovery (via struct tags): - Add 'recover' struct tag for per-field recovery strategies - Syntax: \`parser:"@Ident" recover:"skip_until(;)"\` - Supports: skip_until(), skip_past(), nested(), retry_until() - Multiple strategies via pipe: \`recover:"nested((,))|skip_until(;)"\` - Labels for error messages: \`recover:"label:expression|skip_until(;)"\` Recovery Metadata Fields: - Recovered bool - set to true when struct was recovered - RecoveredSpan lexer.Position - position where recovery started - Auto-injected like Pos/EndPos fields Implementation: - recoveryNode wrapper applies recovery to any node - Handles both errors AND 'no match' cases (nil, nil) - Creates meaningful error messages for recovery reporting - Updated visit.go, ebnf.go for recoveryNode support This brings participle closer to Chumsky's recovery capabilities, enabling fine-grained recovery control per grammar rule. Amp-Thread-ID: https://ampcode.com/threads/T-b95d5ac9-6023-4947-a477-53364956f203 Co-authored-by: Amp <amp@ampcode.com>
alecthomas
reviewed
Nov 28, 2025
Owner
alecthomas
left a comment
alecthomas
left a comment
There was a problem hiding this comment.
This is awesome! I'll try to take a comprehensive look on the weekend.
alecthomas
reviewed
Dec 11, 2025
Owner
alecthomas
left a comment
alecthomas
left a comment
There was a problem hiding this comment.
Apologies for the delay, this is a great idea!
I've left a few comments, but I'll do a full pass in a bit.
Comment on lines
+421
to
+429
| if strings.HasPrefix(tag, "label:") { | ||
| parts := strings.SplitN(tag[6:], "|", 2) | ||
| config.label = strings.TrimSpace(parts[0]) | ||
| if len(parts) > 1 { | ||
| tag = parts[1] | ||
| } else { | ||
| return config, nil | ||
| } | ||
| } |
Owner
There was a problem hiding this comment.
This seems redundant given that the strategy loop also checks for the label?
|
|
||
| // recoveryNode wraps another node with recovery configuration. | ||
| // This allows any node to have per-node recovery without modifying all node types. | ||
| type recoveryNode struct { |
Owner
There was a problem hiding this comment.
There are a few locations where the code type switches on the node type that will need to be updated. You're probably going to have to make this public, and add a method that "unwraps" the underlying node for introspection.
Comment on lines
+600
to
+602
| // ============================================================================= | ||
| // Field Recovery Tag Extraction | ||
| // ============================================================================= |
Owner
There was a problem hiding this comment.
Can you get rid of these section markers please?
Owner
|
Apologies for the delay BTW, I've been absolutely swamped at work. |
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.
Implements error recovery inspired by Chumsky's recovery system in Rust, addressing GitHub issue #342.
Features:
The implementation allows parsers to continue after errors, collecting
multiple errors and producing partial ASTs - useful for IDE integration,
linters, and comprehensive error messages.
Includes comprehensive test coverage (100% on new code) and example at
_examples/recovery/.