rust-clippy/clippy_lints/src/unicode.rs

132 lines
4.5 KiB
Rust
Raw Normal View History

use crate::utils::{is_allowed, snippet, span_lint_and_sugg};
2020-03-01 03:23:33 +00:00
use rustc_ast::ast::LitKind;
2019-05-20 13:23:38 +00:00
use rustc_errors::Applicability;
2020-02-21 08:39:38 +00:00
use rustc_hir::{Expr, ExprKind, HirId};
2020-01-12 06:08:41 +00:00
use rustc_lint::{LateContext, LateLintPass};
2020-01-11 11:37:08 +00:00
use rustc_session::{declare_lint_pass, declare_tool_lint};
use rustc_span::source_map::Span;
2018-11-27 20:14:15 +00:00
use unicode_normalization::UnicodeNormalization;
2015-06-11 09:35:00 +00:00
2018-03-28 13:24:26 +00:00
declare_clippy_lint! {
2020-10-02 09:54:31 +00:00
/// **What it does:** Checks for invisible Unicode characters in the code.
///
/// **Why is this bad?** Having an invisible character in the code makes for all
/// sorts of April fools, but otherwise is very much frowned upon.
///
/// **Known problems:** None.
///
2020-10-02 09:54:31 +00:00
/// **Example:** You don't see it, but there may be a zero-width space or soft hyphen
/// some­where in this text.
2020-10-02 22:03:33 +00:00
pub INVISIBLE_CHARACTERS,
2018-03-28 13:24:26 +00:00
correctness,
2020-10-02 09:54:31 +00:00
"using an invisible character in a string literal, which is confusing"
}
2018-03-28 13:24:26 +00:00
declare_clippy_lint! {
/// **What it does:** Checks for non-ASCII characters in string literals.
///
/// **Why is this bad?** Yeah, we know, the 90's called and wanted their charset
/// back. Even so, there still are editors and other programs out there that
/// don't work well with Unicode. So if the code is meant to be used
/// internationally, on multiple operating systems, or has other portability
/// requirements, activating this lint could be useful.
///
/// **Known problems:** None.
///
/// **Example:**
/// ```rust
/// let x = String::from("€");
/// ```
/// Could be written as:
/// ```rust
/// let x = String::from("\u{20ac}");
/// ```
2018-11-27 20:49:09 +00:00
pub NON_ASCII_LITERAL,
pedantic,
"using any literal non-ASCII chars in a string literal instead of using the `\\u` escape"
}
2018-03-28 13:24:26 +00:00
declare_clippy_lint! {
/// **What it does:** Checks for string literals that contain Unicode in a form
/// that is not equal to its
/// [NFC-recomposition](http://www.unicode.org/reports/tr15/#Norm_Forms).
///
/// **Why is this bad?** If such a string is compared to another, the results
/// may be surprising.
///
/// **Known problems** None.
///
2019-08-23 05:42:45 +00:00
/// **Example:** You may not see it, but "à"" and "à"" aren't the same string. The
/// former when escaped is actually `"a\u{300}"` while the latter is `"\u{e0}"`.
2018-11-27 20:49:09 +00:00
pub UNICODE_NOT_NFC,
pedantic,
2019-08-23 05:42:45 +00:00
"using a Unicode literal not in NFC normal form (see [Unicode tr15](http://www.unicode.org/reports/tr15/) for further information)"
}
2020-10-02 22:03:33 +00:00
declare_lint_pass!(Unicode => [INVISIBLE_CHARACTERS, NON_ASCII_LITERAL, UNICODE_NOT_NFC]);
impl LateLintPass<'_> for Unicode {
fn check_expr(&mut self, cx: &LateContext<'_>, expr: &'_ Expr<'_>) {
2019-09-27 15:16:06 +00:00
if let ExprKind::Lit(ref lit) = expr.kind {
2016-02-12 17:35:44 +00:00
if let LitKind::Str(_, _) = lit.node {
2019-02-24 18:43:15 +00:00
check_str(cx, lit.span, expr.hir_id)
}
}
}
2015-06-11 09:35:00 +00:00
}
2016-01-04 04:26:12 +00:00
fn escape<T: Iterator<Item = char>>(s: T) -> String {
let mut result = String::new();
for c in s {
if c as u32 > 0x7F {
2016-01-04 04:26:12 +00:00
for d in c.escape_unicode() {
result.push(d)
}
} else {
result.push(c);
}
}
result
2015-06-11 09:35:00 +00:00
}
fn check_str(cx: &LateContext<'_>, span: Span, id: HirId) {
let string = snippet(cx, span, "");
2020-10-02 22:03:33 +00:00
if string.chars().any(|c| ['\u{200B}', '\u{ad}'].contains(&c)) {
span_lint_and_sugg(
2017-08-09 07:30:56 +00:00
cx,
2020-10-02 22:03:33 +00:00
INVISIBLE_CHARACTERS,
2017-08-09 07:30:56 +00:00
span,
2020-10-02 22:03:33 +00:00
"invisible character detected",
"consider replacing the string with",
2020-10-02 09:54:31 +00:00
string.replace("\u{200B}", "\\u{200B}").replace("\u{ad}", "\\u{AD}"),
Applicability::MachineApplicable,
2017-08-09 07:30:56 +00:00
);
}
if string.chars().any(|c| c as u32 > 0x7F) {
2019-05-20 13:23:38 +00:00
span_lint_and_sugg(
2017-08-09 07:30:56 +00:00
cx,
NON_ASCII_LITERAL,
span,
"literal non-ASCII character detected",
"consider replacing the string with",
if is_allowed(cx, UNICODE_NOT_NFC, id) {
escape(string.chars())
} else {
escape(string.nfc())
},
2019-05-20 13:23:38 +00:00
Applicability::MachineApplicable,
2017-08-09 07:30:56 +00:00
);
}
2017-08-11 12:11:46 +00:00
if is_allowed(cx, NON_ASCII_LITERAL, id) && string.chars().zip(string.nfc()).any(|(a, b)| a != b) {
span_lint_and_sugg(
2017-08-09 07:30:56 +00:00
cx,
UNICODE_NOT_NFC,
span,
2019-08-23 05:42:45 +00:00
"non-NFC Unicode sequence detected",
"consider replacing the string with",
string.nfc().collect::<String>(),
Applicability::MachineApplicable,
2017-08-09 07:30:56 +00:00
);
}
2015-06-11 09:35:00 +00:00
}