Merging upstream version 25.16.1.

Signed-off-by: Daniel Baumann <daniel@debian.org>

sqlglotrs/Cargo.lock

@@ -188,7 +188,7 @@ checksum = "4dccd0940a2dcdf68d092b8cbab7dc0ad8fa938bf95787e1b916b0e3d0e8e970"
 
 [[package]]
 name = "sqlglotrs"
-version = "0.2.8"
+version = "0.2.9"
 dependencies = [
  "pyo3",
 ]

sqlglotrs/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "sqlglotrs"
-version = "0.2.8"
+version = "0.2.9"
 edition = "2021"
 
 [lib]

sqlglotrs/src/settings.rs

@@ -77,6 +77,7 @@ pub struct TokenizerSettings {
     pub command_prefix_tokens: HashSet<TokenType>,
     pub heredoc_tag_is_identifier: bool,
     pub string_escapes_allowed_in_raw_strings: bool,
+    pub nested_comments: bool,
 }
 
 #[pymethods]
@@ -100,6 +101,7 @@ impl TokenizerSettings {
         command_prefix_tokens: HashSet<TokenType>,
         heredoc_tag_is_identifier: bool,
         string_escapes_allowed_in_raw_strings: bool,
+        nested_comments: bool,
     ) -> Self {
         let to_char = |v: &String| {
             if v.len() == 1 {
@@ -150,6 +152,7 @@ impl TokenizerSettings {
             command_prefix_tokens,
             heredoc_tag_is_identifier,
             string_escapes_allowed_in_raw_strings,
+            nested_comments,
         }
     }
 }

sqlglotrs/src/tokenizer.rs

@@ -375,7 +375,7 @@ impl<'a> TokenizerState<'a> {
                 self.advance(1)?;
 
                 // Nested comments are allowed by some dialects, e.g. databricks, duckdb, postgres
-                if !self.is_end && self.chars(comment_start_size) == *comment_start {
+                if self.settings.nested_comments && !self.is_end && self.chars(comment_start_size) == *comment_start {
                     self.advance(comment_start_size as isize)?;
                     comment_count += 1
                 }