fix(sqlite): correct StmtLocation/StmtLen for non-ASCII characters in comments#4314
Open
mem wants to merge 1 commit intosqlc-dev:mainfrom
Open
fix(sqlite): correct StmtLocation/StmtLen for non-ASCII characters in comments#4314mem wants to merge 1 commit intosqlc-dev:mainfrom
mem wants to merge 1 commit intosqlc-dev:mainfrom
Conversation
… comments ANTLR4-go stores its input stream as []rune, so all token positions returned by GetStart().GetStart() and GetStop().GetStop() are rune indices, not byte offsets. The SQLite parser was storing these values directly as StmtLocation and StmtLen, which are later consumed by source.Pluck() using byte-based Go string slicing (source[head:tail]). For source files that contain multi-byte UTF-8 characters (non-ASCII) in comments, the rune index diverges from the byte offset, causing the plucked query text to be truncated. Each 2-byte character (e.g. Ü, é) caused one byte to be dropped from the end of the query; each 3-byte character (e.g. ♥) caused two bytes to be dropped; and so on. Fix this by building a rune-index to byte-offset map from the source string before processing the ANTLR parse tree, then converting the ANTLR rune positions to byte offsets before storing them in the AST. The internal loc tracking variable continues to use rune indices (for consistency with the ANTLR token positions), while only the values written into StmtLocation and StmtLen are converted to byte offsets. Add TestParseNonASCIIComment covering 2-, 3-, and 4-byte characters in dash comments, multiple non-ASCII characters, and the multi-statement case where an incorrect loc for one statement would propagate and corrupt the StmtLocation of the following statement.
dosubot
bot
added
size:L
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.
ANTLR4-go stores its input stream as []rune, so all token positions returned by GetStart().GetStart() and GetStop().GetStop() are rune indices, not byte offsets. The SQLite parser was storing these values directly as StmtLocation and StmtLen, which are later consumed by source.Pluck() using byte-based Go string slicing (source[head:tail]).
For source files that contain multi-byte UTF-8 characters (non-ASCII) in comments, the rune index diverges from the byte offset, causing the plucked query text to be truncated. Each 2-byte character (e.g. Ü, é) caused one byte to be dropped from the end of the query; each 3-byte character (e.g. ♥) caused two bytes to be dropped; and so on.
Fix this by building a rune-index to byte-offset map from the source string before processing the ANTLR parse tree, then converting the ANTLR rune positions to byte offsets before storing them in the AST. The internal loc tracking variable continues to use rune indices (for consistency with the ANTLR token positions), while only the values written into StmtLocation and StmtLen are converted to byte offsets.
Add TestParseNonASCIIComment covering 2-, 3-, and 4-byte characters in dash comments, multiple non-ASCII characters, and the multi-statement case where an incorrect loc for one statement would propagate and corrupt the StmtLocation of the following statement.