fix: refuse a json5 comment that is not valid utf-8 - #546
Merged
josegonzalez merged 1 commit intoSep 16, 2026
Merged
josegonzalez merged 1 commit into
josegonzalez merged 1 commit into
Conversation
The lexer keeps a comment's bytes verbatim, which is what lets `fmt` carry one across a conversion, and nothing between there and yaml.v3's emitter checked that they were text: a comment is written out raw and the writer panics outright on a byte that starts no rune, so `docket fmt --format yaml` died rather than reporting a problem with the file. A comment is now screened as it is lexed, and the file refused with the offending byte and its offset. A string value needs no such screen and does not get one, since an invalid byte there already reads as U+FFFD, which is what the loader reads it as.
josegonzalez
deleted the
544-fmt-panics-converting-a-json5-comment-that-is-not-valid-utf-8
branch
September 16, 2026 16:39
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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 lexer keeps a comment's bytes verbatim, which is what lets
fmtcarry one across a conversion, and nothing between there and yaml.v3's emitter checked that they were text: a comment is written out raw and the writer panics outright on a byte that starts no rune, sodocket fmt --format yamldied rather than reporting a problem with the file. A comment is now screened as it is lexed, and the file refused with the offending byte and its offset. A string value needs no such screen and does not get one, since an invalid byte there already reads as U+FFFD, which is what the loader reads it as.Closes #544