first unicode lint: zero_width_space

README.md

@@ -29,6 +29,7 @@ Lints included in this crate:
  - `redundant_closure` warns on creating a closure where none is needed, e.g. `|x| foo(x)`, where `foo` can be used directly
  - `inline_always`: Warns on `#[inline(always)]`, because in most cases it is a bad idea
  - `collapsible_if`: Warns on cases where two nested `if`-expressions can be collapsed into one, e.g. `if x { if y { foo() } }` can be written as `if x && y { foo() }`
+ - `zero_width_space`: Warns on encountering a unicode zero-width space
 
 To use, add the following lines to your Cargo.toml:
 

src/lib.rs

@@ -26,6 +26,7 @@ pub mod mut_mut;
 pub mod len_zero;
 pub mod attrs;
 pub mod collapsible_if;
+pub mod unicode;
 pub mod utils;
 
 #[plugin_registrar]
@@ -49,6 +50,7 @@ pub fn plugin_registrar(reg: &mut Registry) {
     reg.register_lint_pass(box misc::CmpOwned as LintPassObject);
     reg.register_lint_pass(box attrs::AttrPass as LintPassObject);
     reg.register_lint_pass(box collapsible_if::CollapsibleIf as LintPassObject);
+    reg.register_lint_pass(box unicode::Unicode as LintPassObject);
     
     reg.register_lint_group("clippy", vec![types::BOX_VEC, types::LINKEDLIST,
                                            misc::SINGLE_MATCH, misc::STR_TO_STRING,
@@ -67,5 +69,6 @@ pub fn plugin_registrar(reg: &mut Registry) {
                                            len_zero::LEN_WITHOUT_IS_EMPTY,
                                            attrs::INLINE_ALWAYS,
                                            collapsible_if::COLLAPSIBLE_IF,
+                                           unicode::ZERO_WIDTH_SPACE,
                                            ]);
 }

src/unicode.rs

@@ -0,0 +1,45 @@
+use rustc::lint::*;
+use syntax::ast::*;
+use syntax::codemap::{BytePos, Span};
+
+declare_lint!{ pub ZERO_WIDTH_SPACE, Deny, "Zero-width space is confusing" }
+
+#[derive(Copy, Clone)]
+pub struct Unicode;
+
+impl LintPass for Unicode {
+	fn get_lints(&self) -> LintArray {
+        lint_array!(ZERO_WIDTH_SPACE)
+    }
+    
+    fn check_expr(&mut self, cx: &Context, expr: &Expr) {
+		if let ExprLit(ref lit) = expr.node {
+			if let LitStr(ref string, _) = lit.node {
+				check_str(cx, string, lit.span)
+			}
+		}
+	}
+}
+
+fn check_str(cx: &Context, string: &str, span: Span) {
+	let mut start: Option<usize> = None;
+	for (i, c) in string.char_indices() {
+		if c == '\u{200B}' {
+			if start.is_none() { start = Some(i); }
+		} else {
+			lint_zero_width(cx, span, start);
+			start = None;
+		}
+	}
+	lint_zero_width(cx, span, start);
+}
+
+fn lint_zero_width(cx: &Context, span: Span, start: Option<usize>) {
+	start.map(|index| {
+		cx.span_lint(ZERO_WIDTH_SPACE, Span {
+			lo: span.lo + BytePos(index as u32),
+			hi: span.lo + BytePos(index as u32),
+			expn_id: span.expn_id,
+		}, "Zero-width space detected. Consider using \\u{200B}")
+	});
+}

tests/compile-fail/unicode.rs

@@ -0,0 +1,25 @@
+#![feature(plugin)]
+#![plugin(clippy)]
+
+#[deny(zero_width_space)]
+fn zero() {
+	print!("Here >​< is a ZWS, and ​another"); 
+				//~^ ERROR Zero-width space detected. Consider using \u{200B}
+								//~^^ ERROR Zero-width space detected. Consider using \u{200B}
+}
+
+//#[deny(unicode_canon)]
+fn canon() {
+	print!("̀ah?"); //not yet ~ERROR Non-canonical unicode sequence detected. Consider using à
+}
+
+//#[deny(ascii_only)]
+fn uni() {
+	println!("Üben!"); //not yet ~ERROR Unicode literal detected. Consider using \u{FC}
+}
+
+fn main() {
+	zero();
+	uni();
+	canon();
+}