2015-06-11 04:35:00 -05:00
|
|
|
use rustc::lint::*;
|
2015-09-03 09:42:17 -05:00
|
|
|
use rustc_front::hir::*;
|
2015-09-04 02:08:07 -05:00
|
|
|
use syntax::codemap::Span;
|
2015-08-16 01:54:43 -05:00
|
|
|
|
2015-09-16 19:01:41 -05:00
|
|
|
use syntax::ast::Lit_::*;
|
|
|
|
|
2015-09-04 02:08:07 -05:00
|
|
|
use unicode_normalization::UnicodeNormalization;
|
|
|
|
|
2015-09-04 07:24:49 -05:00
|
|
|
use utils::{snippet, span_help_and_lint};
|
2015-06-11 04:35:00 -05:00
|
|
|
|
2015-08-13 03:32:35 -05:00
|
|
|
declare_lint!{ pub ZERO_WIDTH_SPACE, Deny,
|
|
|
|
"using a zero-width space in a string literal, which is confusing" }
|
|
|
|
declare_lint!{ pub NON_ASCII_LITERAL, Allow,
|
|
|
|
"using any literal non-ASCII chars in a string literal; suggests \
|
|
|
|
using the \\u escape instead" }
|
2015-09-04 02:08:07 -05:00
|
|
|
declare_lint!{ pub UNICODE_NOT_NFC, Allow,
|
|
|
|
"using a unicode literal not in NFC normal form (see \
|
|
|
|
http://www.unicode.org/reports/tr15/ for further information)" }
|
|
|
|
|
2015-06-11 04:35:00 -05:00
|
|
|
|
|
|
|
#[derive(Copy, Clone)]
|
|
|
|
pub struct Unicode;
|
|
|
|
|
|
|
|
impl LintPass for Unicode {
|
2015-08-11 13:22:20 -05:00
|
|
|
fn get_lints(&self) -> LintArray {
|
2015-09-04 02:08:07 -05:00
|
|
|
lint_array!(ZERO_WIDTH_SPACE, NON_ASCII_LITERAL, UNICODE_NOT_NFC)
|
2015-06-11 04:35:00 -05:00
|
|
|
}
|
2015-08-11 13:22:20 -05:00
|
|
|
|
2015-06-11 04:35:00 -05:00
|
|
|
fn check_expr(&mut self, cx: &Context, expr: &Expr) {
|
2015-08-11 13:22:20 -05:00
|
|
|
if let ExprLit(ref lit) = expr.node {
|
2015-09-04 07:24:49 -05:00
|
|
|
if let LitStr(_, _) = lit.node {
|
|
|
|
check_str(cx, lit.span)
|
2015-08-11 13:22:20 -05:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
2015-06-11 04:35:00 -05:00
|
|
|
}
|
|
|
|
|
2015-09-04 02:08:07 -05:00
|
|
|
fn escape<T: Iterator<Item=char>>(s: T) -> String {
|
|
|
|
let mut result = String::new();
|
|
|
|
for c in s {
|
2015-08-12 13:36:10 -05:00
|
|
|
if c as u32 > 0x7F {
|
2015-09-04 02:08:07 -05:00
|
|
|
for d in c.escape_unicode() { result.push(d) };
|
|
|
|
} else {
|
|
|
|
result.push(c);
|
2015-08-11 13:22:20 -05:00
|
|
|
}
|
|
|
|
}
|
2015-09-04 02:08:07 -05:00
|
|
|
result
|
2015-06-11 04:35:00 -05:00
|
|
|
}
|
|
|
|
|
2015-09-04 07:24:49 -05:00
|
|
|
fn check_str(cx: &Context, span: Span) {
|
|
|
|
let string = snippet(cx, span, "");
|
2015-09-04 02:08:07 -05:00
|
|
|
if string.contains('\u{200B}') {
|
|
|
|
span_help_and_lint(cx, ZERO_WIDTH_SPACE, span,
|
|
|
|
"zero-width space detected",
|
|
|
|
&format!("Consider replacing the string with:\n\"{}\"",
|
|
|
|
string.replace("\u{200B}", "\\u{200B}")));
|
|
|
|
}
|
|
|
|
if string.chars().any(|c| c as u32 > 0x7F) {
|
|
|
|
span_help_and_lint(cx, NON_ASCII_LITERAL, span,
|
|
|
|
"literal non-ASCII character detected",
|
|
|
|
&format!("Consider replacing the string with:\n\"{}\"",
|
|
|
|
if cx.current_level(UNICODE_NOT_NFC) == Level::Allow {
|
|
|
|
escape(string.chars())
|
|
|
|
} else {
|
|
|
|
escape(string.nfc())
|
|
|
|
}));
|
|
|
|
}
|
2015-09-04 07:24:49 -05:00
|
|
|
if cx.current_level(NON_ASCII_LITERAL) == Level::Allow &&
|
|
|
|
string.chars().zip(string.nfc()).any(|(a, b)| a != b) {
|
|
|
|
span_help_and_lint(cx, UNICODE_NOT_NFC, span,
|
|
|
|
"non-nfc unicode sequence detected",
|
|
|
|
&format!("Consider replacing the string with:\n\"{}\"",
|
|
|
|
string.nfc().collect::<String>()));
|
2015-09-04 02:08:07 -05:00
|
|
|
}
|
2015-06-11 04:35:00 -05:00
|
|
|
}
|