feat: auto-fix naked identifiers when prefer_quoted_identifiers is enabled (RF06)

[Copilot]

Rule RF06 (references.quoting) could detect unquoted identifiers when prefer_quoted_identifiers = true but couldn't fix them — the violations were reported with an empty fix list.

Changes

• rf06.rs: Added LintFix::replace in the prefer_quoted_identifiers branch, wrapping NakedIdentifier segments in dialect-appropriate quotes via two new helpers:
  • identifier_quote_chars(dialect) — maps dialect → (open, close) quote pair
  • make_quoted_identifier(dialect, contents, id) — builds the QuotedIdentifier segment
• RF06.yml: Added fix_str to all 9 previously unfixable test_fail_*_prefer_quoted_* cases

Quote character mapping

Dialect	Before	After
ANSI, Postgres, Snowflake, …	name	"name"
T-SQL	name	[name]
BigQuery, MySQL, SparkSQL	name	`name`

Example

-- Input
SELECT name FROM users

-- sqruff fix (ANSI, prefer_quoted_identifiers = true)
SELECT "name" FROM "users"

[github-actions]

Benchmark for 9cee641

Click to view benchmark

Test	Base	PR	%
DepthMap::from_parent	59.1±1.12µs	58.5±0.85µs	-1.02%
fix_complex_query	12.3±0.09ms	12.4±0.06ms	+0.81%
fix_superlong	203.7±7.85ms	204.3±3.67ms	+0.29%
parse_complex_query	4.2±0.20µs	4.2±0.11µs	0.00%
parse_expression_recursion	7.2±0.17µs	7.3±0.13µs	+1.39%
parse_simple_query	1053.1±24.86ns	1062.1±10.99ns	+0.85%

[codspeed-hq]

Merging this PR will not alter performance

✅ 6 untouched benchmarks
⏩ 6 skipped benchmarks¹

---

_{Comparing copilot/add-automatic-quoted-identifier-formatting (8b80f41) with main (f20b401)}

Footnotes

1. 6 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

[Copilot]

Pull request overview

This PR enables auto-fixing for rule RF06 (references.quoting) when prefer_quoted_identifiers = true by generating replacement segments that wrap naked identifiers in dialect-appropriate quotes. It also updates the RF06 fixture cases to assert the new fixes and removes outdated documentation stating the mode isn’t fix-compatible.

Changes:

• Add a LintFix::replace for the prefer_quoted_identifiers branch in RF06 to quote NakedIdentifier segments.
• Introduce helper functions to choose quote characters by dialect and build a QuotedIdentifier segment.
• Update RF06 YAML fixtures with fix_str for previously unfixable cases and remove the “not fix compatible” note from the docs.

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 1 comment.

File	Description
docs/reference/rules.md	Removes outdated note claiming prefer_quoted_identifiers mode is not fix-compatible.
crates/lib/test/fixtures/rules/std_rule_cases/RF06.yml	Adds expected fix_str outputs for prefer_quoted_identifiers failures (ANSI/TSQL/BigQuery).
crates/lib/src/rules/references/rf06.rs	Implements fixes for naked identifiers under prefer_quoted_identifiers and adds dialect quote helpers.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[github-actions]

Benchmark for daeb214

Click to view benchmark

Test	Base	PR	%
DepthMap::from_parent	61.6±0.56µs	60.9±2.55µs	-1.14%
fix_complex_query	12.7±0.04ms	13.0±0.05ms	+2.36%
fix_superlong	173.7±3.29ms	179.9±8.16ms	+3.57%
parse_complex_query	4.5±0.04µs	4.4±0.08µs	-2.22%
parse_expression_recursion	7.6±0.11µs	7.5±0.07µs	-1.32%
parse_simple_query	1102.8±20.35ns	1065.9±13.06ns	-3.35%