1 use crate::utils::{is_allowed, snippet, span_lint_and_sugg};
2 use rustc::declare_lint_pass;
4 use rustc::lint::{LateContext, LateLintPass, LintArray, LintPass};
5 use rustc_errors::Applicability;
6 use rustc_session::declare_tool_lint;
7 use syntax::ast::LitKind;
8 use syntax::source_map::Span;
9 use unicode_normalization::UnicodeNormalization;
11 declare_clippy_lint! {
12 /// **What it does:** Checks for the Unicode zero-width space in the code.
14 /// **Why is this bad?** Having an invisible character in the code makes for all
15 /// sorts of April fools, but otherwise is very much frowned upon.
17 /// **Known problems:** None.
19 /// **Example:** You don't see it, but there may be a zero-width space
20 /// somewhere in this text.
23 "using a zero-width space in a string literal, which is confusing"
26 declare_clippy_lint! {
27 /// **What it does:** Checks for non-ASCII characters in string literals.
29 /// **Why is this bad?** Yeah, we know, the 90's called and wanted their charset
30 /// back. Even so, there still are editors and other programs out there that
31 /// don't work well with Unicode. So if the code is meant to be used
32 /// internationally, on multiple operating systems, or has other portability
33 /// requirements, activating this lint could be useful.
35 /// **Known problems:** None.
39 /// let x = String::from("€");
41 /// Could be written as:
43 /// let x = String::from("\u{20ac}");
45 pub NON_ASCII_LITERAL,
47 "using any literal non-ASCII chars in a string literal instead of using the `\\u` escape"
50 declare_clippy_lint! {
51 /// **What it does:** Checks for string literals that contain Unicode in a form
52 /// that is not equal to its
53 /// [NFC-recomposition](http://www.unicode.org/reports/tr15/#Norm_Forms).
55 /// **Why is this bad?** If such a string is compared to another, the results
56 /// may be surprising.
58 /// **Known problems** None.
60 /// **Example:** You may not see it, but "à"" and "à"" aren't the same string. The
61 /// former when escaped is actually `"a\u{300}"` while the latter is `"\u{e0}"`.
64 "using a Unicode literal not in NFC normal form (see [Unicode tr15](http://www.unicode.org/reports/tr15/) for further information)"
67 declare_lint_pass!(Unicode => [ZERO_WIDTH_SPACE, NON_ASCII_LITERAL, UNICODE_NOT_NFC]);
69 impl LateLintPass<'_, '_> for Unicode {
70 fn check_expr(&mut self, cx: &LateContext<'_, '_>, expr: &'_ Expr) {
71 if let ExprKind::Lit(ref lit) = expr.kind {
72 if let LitKind::Str(_, _) = lit.node {
73 check_str(cx, lit.span, expr.hir_id)
79 fn escape<T: Iterator<Item = char>>(s: T) -> String {
80 let mut result = String::new();
83 for d in c.escape_unicode() {
93 fn check_str(cx: &LateContext<'_, '_>, span: Span, id: HirId) {
94 let string = snippet(cx, span, "");
95 if string.contains('\u{200B}') {
100 "zero-width space detected",
101 "consider replacing the string with",
102 string.replace("\u{200B}", "\\u{200B}"),
103 Applicability::MachineApplicable,
106 if string.chars().any(|c| c as u32 > 0x7F) {
111 "literal non-ASCII character detected",
112 "consider replacing the string with",
113 if is_allowed(cx, UNICODE_NOT_NFC, id) {
114 escape(string.chars())
118 Applicability::MachineApplicable,
121 if is_allowed(cx, NON_ASCII_LITERAL, id) && string.chars().zip(string.nfc()).any(|(a, b)| a != b) {
126 "non-NFC Unicode sequence detected",
127 "consider replacing the string with",
128 string.nfc().collect::<String>(),
129 Applicability::MachineApplicable,