Auto merge of #46907 - varkor:contrib-8, r=nagisa

Allow non-alphabetic underscores in camel case

Certain identifiers, such as `X86_64`, cannot currently be unambiguously represented in camel case (`X8664`, `X86_64`, `X8_664`, etc. are all transformed to the same identifier). This change relaxes the rules so that underscores are permitted between two non-alphabetic characters under `#[forbid(non_camel_case_types)]`. Fixes #34633 and fixes #41621.

src/librustc_lint/bad_style.rs

@@ -53,6 +53,10 @@ pub struct NonCamelCaseTypes;
 
 impl NonCamelCaseTypes {
     fn check_case(&self, cx: &LateContext, sort: &str, name: ast::Name, span: Span) {
+        fn char_has_case(c: char) -> bool {
+            c.is_lowercase() || c.is_uppercase()
+        }
+
         fn is_camel_case(name: ast::Name) -> bool {
             let name = name.as_str();
             if name.is_empty() {
@@ -62,20 +66,38 @@ impl NonCamelCaseTypes {
 
             // start with a non-lowercase letter rather than non-uppercase
             // ones (some scripts don't have a concept of upper/lowercase)
-            !name.is_empty() && !name.chars().next().unwrap().is_lowercase() && !name.contains('_')
+            !name.is_empty() && !name.chars().next().unwrap().is_lowercase() &&
+                !name.contains("__") && !name.chars().collect::<Vec<_>>().windows(2).any(|pair| {
+                    // contains a capitalisable character followed by, or preceded by, an underscore
+                    char_has_case(pair[0]) && pair[1] == '_' ||
+                    char_has_case(pair[1]) && pair[0] == '_'
+                })
         }
 
         fn to_camel_case(s: &str) -> String {
-            s.split('_')
-                .flat_map(|word| {
+            s.trim_matches('_')
+                .split('_')
+                .map(|word| {
                     word.chars().enumerate().map(|(i, c)| if i == 0 {
                         c.to_uppercase().collect::<String>()
                     } else {
                         c.to_lowercase().collect()
                     })
+                    .collect::<Vec<_>>()
+                    .concat()
                 })
+                .filter(|x| !x.is_empty())
                 .collect::<Vec<_>>()
-                .concat()
+                .iter().fold((String::new(), None), |(acc, prev): (String, Option<&String>), next| {
+                    // separate two components with an underscore if their boundary cannot
+                    // be distinguished using a uppercase/lowercase case distinction
+                    let join = if let Some(prev) = prev {
+                                    let l = prev.chars().last().unwrap();
+                                    let f = next.chars().next().unwrap();
+                                    !char_has_case(l) && !char_has_case(f)
+                                } else { false };
+                    (acc + if join { "_" } else { "" } + next, Some(next))
+                }).0
         }
 
         if !is_camel_case(name) {

src/test/compile-fail/lint-non-camel-case-types.rs

@@ -45,4 +45,12 @@ struct foo7 {
 
 type __ = isize; //~ ERROR type `__` should have a camel case name such as `CamelCase`
 
+struct X86_64;
+
+struct X86__64; //~ ERROR type `X86__64` should have a camel case name such as `X86_64`
+
+struct Abc_123; //~ ERROR type `Abc_123` should have a camel case name such as `Abc123`
+
+struct A1_b2_c3; //~ ERROR type `A1_b2_c3` should have a camel case name such as `A1B2C3`
+
 fn main() { }