Support optional semicolon between statements #1937
Conversation
| @@ -2480,3 +2492,15 @@ fn parse_mssql_grant() { | |||
| fn parse_mssql_deny() { | |||
| ms().verified_stmt("DENY SELECT ON my_table TO public, db_admin"); | |||
| } | |||
|
|
|||
| #[test] | |||
There was a problem hiding this comment.
Could we have a similar test or move this over to common.rs? thinking since this flag is being added to the parser itself, it feels a bit like it could be lost if only tested here within mssql
src/parser/mod.rs
Outdated
| @@ -222,13 +222,17 @@ pub struct ParserOptions { | |||
| /// Controls how literal values are unescaped. See | |||
| /// [`Tokenizer::with_unescape`] for more details. | |||
| pub unescape: bool, | |||
| /// Controls if the parser expects a semi-colon token | |||
| /// between statements. | |||
There was a problem hiding this comment.
Thinking it might be helpful if we include in the doc that the default is true?
iffyio
changed the title
|
I went to rebase #1843 and noticed this PR had been merged. Great! I see we had similar thoughts on how to approach it. I closed my PR. However, my PR also introduced a lot more testing for the feature missing from this PR. I'll open a separate PR for more test coverage if it turns out to be needed. |
Havign some more test coverage would be great -- if you have the time to make one please ping me and I will be happy to review it |
T-SQL scripts do not require a semicolon as a statement delimiter. Added a new parser option to not require semicolons, as introducing this at the dialect level seems to be very intrusive. Also added a
tsql()function to create an MS SQL dialect for testing with the new option.