[ruff] Auto-add r prefix when string has no backslashes for unraw-re-pattern (RUF039) (#14536)

This PR adds a sometimes-available, safe autofix for [unraw-re-pattern
(RUF039)](https://docs.astral.sh/ruff/rules/unraw-re-pattern/#unraw-re-pattern-ruf039),
which prepends an `r` prefix. It is used only when the string in
question has no backslahses (and also does not have a `u` prefix, since
that causes a syntax error.)

Closes #14527

Notes: 
- Test fixture unchanged, but snapshot changed to include fix messages.
- This fix is automatically only available in preview since the rule
itself is in preview

crates/ruff_linter/src/rules/ruff/rules/unraw_re_pattern.rs

@@ -1,13 +1,15 @@
 use std::fmt::{Display, Formatter};
 use std::str::FromStr;
 
-use ruff_diagnostics::{Diagnostic, Violation};
+use ruff_diagnostics::{Diagnostic, Edit, Fix, FixAvailability, Violation};
 use ruff_macros::{derive_message_formats, violation};
 use ruff_python_ast::{
     BytesLiteral, Expr, ExprBytesLiteral, ExprCall, ExprStringLiteral, StringLiteral,
 };
 use ruff_python_semantic::{Modules, SemanticModel};
 
+use ruff_text_size::Ranged;
+
 use crate::checkers::ast::Checker;
 
 /// ## What it does
@@ -41,6 +43,7 @@ pub struct UnrawRePattern {
 }
 
 impl Violation for UnrawRePattern {
+    const FIX_AVAILABILITY: FixAvailability = FixAvailability::Sometimes;
     #[derive_message_formats]
     fn message(&self) -> String {
         let Self { module, func, kind } = &self;
@@ -158,8 +161,22 @@ fn check_string(checker: &mut Checker, literal: &StringLiteral, module: RegexMod
     let kind = PatternKind::String;
     let func = func.to_string();
     let range = literal.range;
-    let diagnostic = Diagnostic::new(UnrawRePattern { module, func, kind }, range);
+    let mut diagnostic = Diagnostic::new(UnrawRePattern { module, func, kind }, range);
 
+    if
+    // The (no-op) `u` prefix is a syntax error when combined with `r`
+    !literal.flags.prefix().is_unicode()
+    // We are looking for backslash characters
+    // in the raw source code here, because `\n`
+    // gets converted to a single character already
+    // at the lexing stage.
+    &&!checker.locator().slice(literal.range()).contains('\\')
+    {
+        diagnostic.set_fix(Fix::safe_edit(Edit::insertion(
+            "r".to_string(),
+            literal.range().start(),
+        )));
+    }
     checker.diagnostics.push(diagnostic);
 }
 

crates/ruff_linter/src/rules/ruff/snapshots/ruff_linter__rules__ruff__tests__preview__RUF039_RUF039.py.snap

@@ -1,8 +1,7 @@
 ---
 source: crates/ruff_linter/src/rules/ruff/mod.rs
-snapshot_kind: text
 ---
-RUF039.py:5:12: RUF039 First argument to `re.compile()` is not raw string
+RUF039.py:5:12: RUF039 [*] First argument to `re.compile()` is not raw string
   |
 4 | # Errors
 5 | re.compile('single free-spacing', flags=re.X)
@@ -12,6 +11,16 @@ RUF039.py:5:12: RUF039 First argument to `re.compile()` is not raw string
   |
   = help: Replace with raw string
 
+ℹ Safe fix
+2 2 | import regex
+3 3 | 
+4 4 | # Errors
+5   |-re.compile('single free-spacing', flags=re.X)
+  5 |+re.compile(r'single free-spacing', flags=re.X)
+6 6 | re.findall('si\ngle')
+7 7 | re.finditer("dou\ble")
+8 8 | re.fullmatch('''t\riple single''')
+
 RUF039.py:6:12: RUF039 First argument to `re.findall()` is not raw string
   |
 4 | # Errors
@@ -56,7 +65,7 @@ RUF039.py:9:10: RUF039 First argument to `re.match()` is not raw string
    |
    = help: Replace with raw string
 
-RUF039.py:10:11: RUF039 First argument to `re.search()` is not raw string
+RUF039.py:10:11: RUF039 [*] First argument to `re.search()` is not raw string
    |
  8 | re.fullmatch('''t\riple single''')
  9 | re.match("""\triple double""")
@@ -67,7 +76,17 @@ RUF039.py:10:11: RUF039 First argument to `re.search()` is not raw string
    |
    = help: Replace with raw string
 
-RUF039.py:11:10: RUF039 First argument to `re.split()` is not raw string
+ℹ Safe fix
+7  7  | re.finditer("dou\ble")
+8  8  | re.fullmatch('''t\riple single''')
+9  9  | re.match("""\triple double""")
+10    |-re.search('two', 'args')
+   10 |+re.search(r'two', 'args')
+11 11 | re.split("raw", r'second')
+12 12 | re.sub(u'''nicode''', u"f(?i)rst")
+13 13 | re.subn(b"""ytes are""", f"\u006e")
+
+RUF039.py:11:10: RUF039 [*] First argument to `re.split()` is not raw string
    |
  9 | re.match("""\triple double""")
 10 | re.search('two', 'args')
@@ -78,6 +97,16 @@ RUF039.py:11:10: RUF039 First argument to `re.split()` is not raw string
    |
    = help: Replace with raw string
 
+ℹ Safe fix
+8  8  | re.fullmatch('''t\riple single''')
+9  9  | re.match("""\triple double""")
+10 10 | re.search('two', 'args')
+11    |-re.split("raw", r'second')
+   11 |+re.split(r"raw", r'second')
+12 12 | re.sub(u'''nicode''', u"f(?i)rst")
+13 13 | re.subn(b"""ytes are""", f"\u006e")
+14 14 | 
+
 RUF039.py:12:8: RUF039 First argument to `re.sub()` is not raw string
    |
 10 | re.search('two', 'args')
@@ -99,7 +128,7 @@ RUF039.py:13:9: RUF039 First argument to `re.subn()` is not raw bytes literal
    |
    = help: Replace with raw bytes literal
 
-RUF039.py:15:15: RUF039 First argument to `regex.compile()` is not raw string
+RUF039.py:15:15: RUF039 [*] First argument to `regex.compile()` is not raw string
    |
 13 | re.subn(b"""ytes are""", f"\u006e")
 14 | 
@@ -110,6 +139,16 @@ RUF039.py:15:15: RUF039 First argument to `regex.compile()` is not raw string
    |
    = help: Replace with raw string
 
+ℹ Safe fix
+12 12 | re.sub(u'''nicode''', u"f(?i)rst")
+13 13 | re.subn(b"""ytes are""", f"\u006e")
+14 14 | 
+15    |-regex.compile('single free-spacing', flags=regex.X)
+   15 |+regex.compile(r'single free-spacing', flags=regex.X)
+16 16 | regex.findall('si\ngle')
+17 17 | regex.finditer("dou\ble")
+18 18 | regex.fullmatch('''t\riple single''')
+
 RUF039.py:16:15: RUF039 First argument to `regex.findall()` is not raw string
    |
 15 | regex.compile('single free-spacing', flags=regex.X)
@@ -153,7 +192,7 @@ RUF039.py:19:13: RUF039 First argument to `regex.match()` is not raw string
    |
    = help: Replace with raw string
 
-RUF039.py:20:14: RUF039 First argument to `regex.search()` is not raw string
+RUF039.py:20:14: RUF039 [*] First argument to `regex.search()` is not raw string
    |
 18 | regex.fullmatch('''t\riple single''')
 19 | regex.match("""\triple double""")
@@ -164,7 +203,17 @@ RUF039.py:20:14: RUF039 First argument to `regex.search()` is not raw string
    |
    = help: Replace with raw string
 
-RUF039.py:21:13: RUF039 First argument to `regex.split()` is not raw string
+ℹ Safe fix
+17 17 | regex.finditer("dou\ble")
+18 18 | regex.fullmatch('''t\riple single''')
+19 19 | regex.match("""\triple double""")
+20    |-regex.search('two', 'args')
+   20 |+regex.search(r'two', 'args')
+21 21 | regex.split("raw", r'second')
+22 22 | regex.sub(u'''nicode''', u"f(?i)rst")
+23 23 | regex.subn(b"""ytes are""", f"\u006e")
+
+RUF039.py:21:13: RUF039 [*] First argument to `regex.split()` is not raw string
    |
 19 | regex.match("""\triple double""")
 20 | regex.search('two', 'args')
@@ -175,6 +224,16 @@ RUF039.py:21:13: RUF039 First argument to `regex.split()` is not raw string
    |
    = help: Replace with raw string
 
+ℹ Safe fix
+18 18 | regex.fullmatch('''t\riple single''')
+19 19 | regex.match("""\triple double""")
+20 20 | regex.search('two', 'args')
+21    |-regex.split("raw", r'second')
+   21 |+regex.split(r"raw", r'second')
+22 22 | regex.sub(u'''nicode''', u"f(?i)rst")
+23 23 | regex.subn(b"""ytes are""", f"\u006e")
+24 24 | 
+
 RUF039.py:22:11: RUF039 First argument to `regex.sub()` is not raw string
    |
 20 | regex.search('two', 'args')
@@ -196,7 +255,7 @@ RUF039.py:23:12: RUF039 First argument to `regex.subn()` is not raw bytes litera
    |
    = help: Replace with raw bytes literal
 
-RUF039.py:25:16: RUF039 First argument to `regex.template()` is not raw string
+RUF039.py:25:16: RUF039 [*] First argument to `regex.template()` is not raw string
    |
 23 |   regex.subn(b"""ytes are""", f"\u006e")
 24 |   
@@ -209,3 +268,13 @@ RUF039.py:25:16: RUF039 First argument to `regex.template()` is not raw string
    | |___^ RUF039
    |
    = help: Replace with raw string
+
+ℹ Safe fix
+22 22 | regex.sub(u'''nicode''', u"f(?i)rst")
+23 23 | regex.subn(b"""ytes are""", f"\u006e")
+24 24 | 
+25    |-regex.template("""(?m)
+   25 |+regex.template(r"""(?m)
+26 26 |     (?:ulti)?
+27 27 |     (?=(?<!(?<=(?!l)))
+28 28 |     l(?i:ne)

crates/ruff_linter/src/rules/ruff/snapshots/ruff_linter__rules__ruff__tests__preview__RUF039_RUF039_concat.py.snap

@@ -1,8 +1,7 @@
 ---
 source: crates/ruff_linter/src/rules/ruff/mod.rs
-snapshot_kind: text
 ---
-RUF039_concat.py:5:5: RUF039 First argument to `re.compile()` is not raw string
+RUF039_concat.py:5:5: RUF039 [*] First argument to `re.compile()` is not raw string
   |
 4 | re.compile(
 5 |     'implicit'
@@ -12,7 +11,17 @@ RUF039_concat.py:5:5: RUF039 First argument to `re.compile()` is not raw string
   |
   = help: Replace with raw string
 
-RUF039_concat.py:6:5: RUF039 First argument to `re.compile()` is not raw string
+ℹ Safe fix
+2 2 | 
+3 3 | 
+4 4 | re.compile(
+5   |-    'implicit'
+  5 |+    r'implicit'
+6 6 |     'concatenation'
+7 7 | )
+8 8 | re.findall(
+
+RUF039_concat.py:6:5: RUF039 [*] First argument to `re.compile()` is not raw string
   |
 4 | re.compile(
 5 |     'implicit'
@@ -23,7 +32,17 @@ RUF039_concat.py:6:5: RUF039 First argument to `re.compile()` is not raw string
   |
   = help: Replace with raw string
 
-RUF039_concat.py:12:5: RUF039 First argument to `re.findall()` is not raw string
+ℹ Safe fix
+3 3 | 
+4 4 | re.compile(
+5 5 |     'implicit'
+6   |-    'concatenation'
+  6 |+    r'concatenation'
+7 7 | )
+8 8 | re.findall(
+9 9 |     r'''
+
+RUF039_concat.py:12:5: RUF039 [*] First argument to `re.findall()` is not raw string
    |
 10 |       multiline
 11 |       '''
@@ -37,6 +56,16 @@ RUF039_concat.py:12:5: RUF039 First argument to `re.findall()` is not raw string
    |
    = help: Replace with raw string
 
+ℹ Safe fix
+9  9  |     r'''
+10 10 |     multiline
+11 11 |     '''
+12    |-    """
+   12 |+    r"""
+13 13 |     concatenation
+14 14 |     """
+15 15 | )
+
 RUF039_concat.py:26:5: RUF039 First argument to `re.match()` is not raw bytes literal
    |
 24 | )
@@ -59,7 +88,7 @@ RUF039_concat.py:30:8: RUF039 First argument to `re.search()` is not raw string
    |
    = help: Replace with raw string
 
-RUF039_concat.py:31:5: RUF039 First argument to `re.search()` is not raw string
+RUF039_concat.py:31:5: RUF039 [*] First argument to `re.search()` is not raw string
    |
 29 | re.search(
 30 |     r''u''
@@ -70,7 +99,17 @@ RUF039_concat.py:31:5: RUF039 First argument to `re.search()` is not raw string
    |
    = help: Replace with raw string
 
-RUF039_concat.py:33:10: RUF039 First argument to `re.split()` is not raw string
+ℹ Safe fix
+28 28 | )
+29 29 | re.search(
+30 30 |     r''u''
+31    |-    '''okay?'''
+   31 |+    r'''okay?'''
+32 32 | )
+33 33 | re.split(''U"""w"""U'')
+34 34 | re.sub(
+
+RUF039_concat.py:33:10: RUF039 [*] First argument to `re.split()` is not raw string
    |
 31 |     '''okay?'''
 32 | )
@@ -81,6 +120,16 @@ RUF039_concat.py:33:10: RUF039 First argument to `re.split()` is not raw string
    |
    = help: Replace with raw string
 
+ℹ Safe fix
+30 30 |     r''u''
+31 31 |     '''okay?'''
+32 32 | )
+33    |-re.split(''U"""w"""U'')
+   33 |+re.split(r''U"""w"""U'')
+34 34 | re.sub(
+35 35 |     "I''m o"
+36 36 |     'utta ideas'
+
 RUF039_concat.py:33:12: RUF039 First argument to `re.split()` is not raw string
    |
 31 |     '''okay?'''
@@ -103,7 +152,7 @@ RUF039_concat.py:33:20: RUF039 First argument to `re.split()` is not raw string
    |
    = help: Replace with raw string
 
-RUF039_concat.py:35:5: RUF039 First argument to `re.sub()` is not raw string
+RUF039_concat.py:35:5: RUF039 [*] First argument to `re.sub()` is not raw string
    |
 33 | re.split(''U"""w"""U'')
 34 | re.sub(
@@ -114,7 +163,17 @@ RUF039_concat.py:35:5: RUF039 First argument to `re.sub()` is not raw string
    |
    = help: Replace with raw string
 
-RUF039_concat.py:36:5: RUF039 First argument to `re.sub()` is not raw string
+ℹ Safe fix
+32 32 | )
+33 33 | re.split(''U"""w"""U'')
+34 34 | re.sub(
+35    |-    "I''m o"
+   35 |+    r"I''m o"
+36 36 |     'utta ideas'
+37 37 | )
+38 38 | re.subn("()"r' am I'"??")
+
+RUF039_concat.py:36:5: RUF039 [*] First argument to `re.sub()` is not raw string
    |
 34 | re.sub(
 35 |     "I''m o"
@@ -125,7 +184,17 @@ RUF039_concat.py:36:5: RUF039 First argument to `re.sub()` is not raw string
    |
    = help: Replace with raw string
 
-RUF039_concat.py:38:9: RUF039 First argument to `re.subn()` is not raw string
+ℹ Safe fix
+33 33 | re.split(''U"""w"""U'')
+34 34 | re.sub(
+35 35 |     "I''m o"
+36    |-    'utta ideas'
+   36 |+    r'utta ideas'
+37 37 | )
+38 38 | re.subn("()"r' am I'"??")
+39 39 | 
+
+RUF039_concat.py:38:9: RUF039 [*] First argument to `re.subn()` is not raw string
    |
 36 |     'utta ideas'
 37 | )
@@ -134,7 +203,17 @@ RUF039_concat.py:38:9: RUF039 First argument to `re.subn()` is not raw string
    |
    = help: Replace with raw string
 
-RUF039_concat.py:38:21: RUF039 First argument to `re.subn()` is not raw string
+ℹ Safe fix
+35 35 |     "I''m o"
+36 36 |     'utta ideas'
+37 37 | )
+38    |-re.subn("()"r' am I'"??")
+   38 |+re.subn(r"()"r' am I'"??")
+39 39 | 
+40 40 | 
+41 41 | import regex
+
+RUF039_concat.py:38:21: RUF039 [*] First argument to `re.subn()` is not raw string
    |
 36 |     'utta ideas'
 37 | )
@@ -143,7 +222,17 @@ RUF039_concat.py:38:21: RUF039 First argument to `re.subn()` is not raw string
    |
    = help: Replace with raw string
 
-RUF039_concat.py:45:5: RUF039 First argument to `regex.compile()` is not raw string
+ℹ Safe fix
+35 35 |     "I''m o"
+36 36 |     'utta ideas'
+37 37 | )
+38    |-re.subn("()"r' am I'"??")
+   38 |+re.subn("()"r' am I'r"??")
+39 39 | 
+40 40 | 
+41 41 | import regex
+
+RUF039_concat.py:45:5: RUF039 [*] First argument to `regex.compile()` is not raw string
    |
 44 | regex.compile(
 45 |     'implicit'
@@ -153,7 +242,17 @@ RUF039_concat.py:45:5: RUF039 First argument to `regex.compile()` is not raw str
    |
    = help: Replace with raw string
 
-RUF039_concat.py:46:5: RUF039 First argument to `regex.compile()` is not raw string
+ℹ Safe fix
+42 42 | 
+43 43 | 
+44 44 | regex.compile(
+45    |-    'implicit'
+   45 |+    r'implicit'
+46 46 |     'concatenation'
+47 47 | )
+48 48 | regex.findall(
+
+RUF039_concat.py:46:5: RUF039 [*] First argument to `regex.compile()` is not raw string
    |
 44 | regex.compile(
 45 |     'implicit'
@@ -164,7 +263,17 @@ RUF039_concat.py:46:5: RUF039 First argument to `regex.compile()` is not raw str
    |
    = help: Replace with raw string
 
-RUF039_concat.py:52:5: RUF039 First argument to `regex.findall()` is not raw string
+ℹ Safe fix
+43 43 | 
+44 44 | regex.compile(
+45 45 |     'implicit'
+46    |-    'concatenation'
+   46 |+    r'concatenation'
+47 47 | )
+48 48 | regex.findall(
+49 49 |     r'''
+
+RUF039_concat.py:52:5: RUF039 [*] First argument to `regex.findall()` is not raw string
    |
 50 |       multiline
 51 |       '''
@@ -178,6 +287,16 @@ RUF039_concat.py:52:5: RUF039 First argument to `regex.findall()` is not raw str
    |
    = help: Replace with raw string
 
+ℹ Safe fix
+49 49 |     r'''
+50 50 |     multiline
+51 51 |     '''
+52    |-    """
+   52 |+    r"""
+53 53 |     concatenation
+54 54 |     """
+55 55 | )
+
 RUF039_concat.py:66:5: RUF039 First argument to `regex.match()` is not raw bytes literal
    |
 64 | )
@@ -200,7 +319,7 @@ RUF039_concat.py:70:8: RUF039 First argument to `regex.search()` is not raw stri
    |
    = help: Replace with raw string
 
-RUF039_concat.py:71:5: RUF039 First argument to `regex.search()` is not raw string
+RUF039_concat.py:71:5: RUF039 [*] First argument to `regex.search()` is not raw string
    |
 69 | regex.search(
 70 |     r''u''
@@ -211,7 +330,17 @@ RUF039_concat.py:71:5: RUF039 First argument to `regex.search()` is not raw stri
    |
    = help: Replace with raw string
 
-RUF039_concat.py:73:13: RUF039 First argument to `regex.split()` is not raw string
+ℹ Safe fix
+68 68 | )
+69 69 | regex.search(
+70 70 |     r''u''
+71    |-    '''okay?'''
+   71 |+    r'''okay?'''
+72 72 | )
+73 73 | regex.split(''U"""w"""U'')
+74 74 | regex.sub(
+
+RUF039_concat.py:73:13: RUF039 [*] First argument to `regex.split()` is not raw string
    |
 71 |     '''okay?'''
 72 | )
@@ -222,6 +351,16 @@ RUF039_concat.py:73:13: RUF039 First argument to `regex.split()` is not raw stri
    |
    = help: Replace with raw string
 
+ℹ Safe fix
+70 70 |     r''u''
+71 71 |     '''okay?'''
+72 72 | )
+73    |-regex.split(''U"""w"""U'')
+   73 |+regex.split(r''U"""w"""U'')
+74 74 | regex.sub(
+75 75 |     "I''m o"
+76 76 |     'utta ideas'
+
 RUF039_concat.py:73:15: RUF039 First argument to `regex.split()` is not raw string
    |
 71 |     '''okay?'''
@@ -244,7 +383,7 @@ RUF039_concat.py:73:23: RUF039 First argument to `regex.split()` is not raw stri
    |
    = help: Replace with raw string
 
-RUF039_concat.py:75:5: RUF039 First argument to `regex.sub()` is not raw string
+RUF039_concat.py:75:5: RUF039 [*] First argument to `regex.sub()` is not raw string
    |
 73 | regex.split(''U"""w"""U'')
 74 | regex.sub(
@@ -255,7 +394,17 @@ RUF039_concat.py:75:5: RUF039 First argument to `regex.sub()` is not raw string
    |
    = help: Replace with raw string
 
-RUF039_concat.py:76:5: RUF039 First argument to `regex.sub()` is not raw string
+ℹ Safe fix
+72 72 | )
+73 73 | regex.split(''U"""w"""U'')
+74 74 | regex.sub(
+75    |-    "I''m o"
+   75 |+    r"I''m o"
+76 76 |     'utta ideas'
+77 77 | )
+78 78 | regex.subn("()"r' am I'"??")
+
+RUF039_concat.py:76:5: RUF039 [*] First argument to `regex.sub()` is not raw string
    |
 74 | regex.sub(
 75 |     "I''m o"
@@ -266,7 +415,17 @@ RUF039_concat.py:76:5: RUF039 First argument to `regex.sub()` is not raw string
    |
    = help: Replace with raw string
 
-RUF039_concat.py:78:12: RUF039 First argument to `regex.subn()` is not raw string
+ℹ Safe fix
+73 73 | regex.split(''U"""w"""U'')
+74 74 | regex.sub(
+75 75 |     "I''m o"
+76    |-    'utta ideas'
+   76 |+    r'utta ideas'
+77 77 | )
+78 78 | regex.subn("()"r' am I'"??")
+79 79 | 
+
+RUF039_concat.py:78:12: RUF039 [*] First argument to `regex.subn()` is not raw string
    |
 76 |     'utta ideas'
 77 | )
@@ -275,7 +434,17 @@ RUF039_concat.py:78:12: RUF039 First argument to `regex.subn()` is not raw strin
    |
    = help: Replace with raw string
 
-RUF039_concat.py:78:24: RUF039 First argument to `regex.subn()` is not raw string
+ℹ Safe fix
+75 75 |     "I''m o"
+76 76 |     'utta ideas'
+77 77 | )
+78    |-regex.subn("()"r' am I'"??")
+   78 |+regex.subn(r"()"r' am I'"??")
+79 79 | 
+80 80 | 
+81 81 | regex.template(
+
+RUF039_concat.py:78:24: RUF039 [*] First argument to `regex.subn()` is not raw string
    |
 76 |     'utta ideas'
 77 | )
@@ -283,3 +452,13 @@ RUF039_concat.py:78:24: RUF039 First argument to `regex.subn()` is not raw strin
    |                        ^^^^ RUF039
    |
    = help: Replace with raw string
+
+ℹ Safe fix
+75 75 |     "I''m o"
+76 76 |     'utta ideas'
+77 77 | )
+78    |-regex.subn("()"r' am I'"??")
+   78 |+regex.subn("()"r' am I'r"??")
+79 79 | 
+80 80 | 
+81 81 | regex.template(