blob: a44adc938559f61b7816839bdb13bf09cdaf947d [file] [log] [blame]
use clippy_utils::diagnostics::{span_lint, span_lint_and_help, span_lint_and_sugg};
use clippy_utils::source::{snippet, snippet_with_applicability};
use clippy_utils::ty::is_type_lang_item;
use clippy_utils::{
get_expr_use_or_unification_node, get_parent_expr, is_lint_allowed, is_path_diagnostic_item, method_calls,
peel_blocks, SpanlessEq,
};
use if_chain::if_chain;
use rustc_errors::Applicability;
use rustc_hir::def_id::DefId;
use rustc_hir::{BinOpKind, BorrowKind, Expr, ExprKind, LangItem, Node, QPath};
use rustc_lint::{LateContext, LateLintPass, LintContext};
use rustc_middle::lint::in_external_macro;
use rustc_middle::ty;
use rustc_session::{declare_lint_pass, declare_tool_lint};
use rustc_span::source_map::Spanned;
use rustc_span::sym;
declare_clippy_lint! {
/// ### What it does
/// Checks for string appends of the form `x = x + y` (without
/// `let`!).
///
/// ### Why is this bad?
/// It's not really bad, but some people think that the
/// `.push_str(_)` method is more readable.
///
/// ### Example
/// ```no_run
/// let mut x = "Hello".to_owned();
/// x = x + ", World";
///
/// // More readable
/// x += ", World";
/// x.push_str(", World");
/// ```
#[clippy::version = "pre 1.29.0"]
pub STRING_ADD_ASSIGN,
pedantic,
"using `x = x + ..` where x is a `String` instead of `push_str()`"
}
declare_clippy_lint! {
/// ### What it does
/// Checks for all instances of `x + _` where `x` is of type
/// `String`, but only if [`string_add_assign`](#string_add_assign) does *not*
/// match.
///
/// ### Why is this bad?
/// It's not bad in and of itself. However, this particular
/// `Add` implementation is asymmetric (the other operand need not be `String`,
/// but `x` does), while addition as mathematically defined is symmetric, also
/// the `String::push_str(_)` function is a perfectly good replacement.
/// Therefore, some dislike it and wish not to have it in their code.
///
/// That said, other people think that string addition, having a long tradition
/// in other languages is actually fine, which is why we decided to make this
/// particular lint `allow` by default.
///
/// ### Example
/// ```no_run
/// let x = "Hello".to_owned();
/// x + ", World";
/// ```
///
/// Use instead:
/// ```no_run
/// let mut x = "Hello".to_owned();
/// x.push_str(", World");
/// ```
#[clippy::version = "pre 1.29.0"]
pub STRING_ADD,
restriction,
"using `x + ..` where x is a `String` instead of `push_str()`"
}
declare_clippy_lint! {
/// ### What it does
/// Checks for the `as_bytes` method called on string literals
/// that contain only ASCII characters.
///
/// ### Why is this bad?
/// Byte string literals (e.g., `b"foo"`) can be used
/// instead. They are shorter but less discoverable than `as_bytes()`.
///
/// ### Known problems
/// `"str".as_bytes()` and the suggested replacement of `b"str"` are not
/// equivalent because they have different types. The former is `&[u8]`
/// while the latter is `&[u8; 3]`. That means in general they will have a
/// different set of methods and different trait implementations.
///
/// ```compile_fail
/// fn f(v: Vec<u8>) {}
///
/// f("...".as_bytes().to_owned()); // works
/// f(b"...".to_owned()); // does not work, because arg is [u8; 3] not Vec<u8>
///
/// fn g(r: impl std::io::Read) {}
///
/// g("...".as_bytes()); // works
/// g(b"..."); // does not work
/// ```
///
/// The actual equivalent of `"str".as_bytes()` with the same type is not
/// `b"str"` but `&b"str"[..]`, which is a great deal of punctuation and not
/// more readable than a function call.
///
/// ### Example
/// ```no_run
/// let bstr = "a byte string".as_bytes();
/// ```
///
/// Use instead:
/// ```no_run
/// let bstr = b"a byte string";
/// ```
#[clippy::version = "pre 1.29.0"]
pub STRING_LIT_AS_BYTES,
nursery,
"calling `as_bytes` on a string literal instead of using a byte string literal"
}
declare_clippy_lint! {
/// ### What it does
/// Checks for slice operations on strings
///
/// ### Why is this bad?
/// UTF-8 characters span multiple bytes, and it is easy to inadvertently confuse character
/// counts and string indices. This may lead to panics, and should warrant some test cases
/// containing wide UTF-8 characters. This lint is most useful in code that should avoid
/// panics at all costs.
///
/// ### Known problems
/// Probably lots of false positives. If an index comes from a known valid position (e.g.
/// obtained via `char_indices` over the same string), it is totally OK.
///
/// ### Example
/// ```rust,should_panic
/// &"Ölkanne"[1..];
/// ```
#[clippy::version = "1.58.0"]
pub STRING_SLICE,
restriction,
"slicing a string"
}
declare_lint_pass!(StringAdd => [STRING_ADD, STRING_ADD_ASSIGN, STRING_SLICE]);
impl<'tcx> LateLintPass<'tcx> for StringAdd {
fn check_expr(&mut self, cx: &LateContext<'tcx>, e: &'tcx Expr<'_>) {
if in_external_macro(cx.sess(), e.span) {
return;
}
match e.kind {
ExprKind::Binary(
Spanned {
node: BinOpKind::Add, ..
},
left,
_,
) => {
if is_string(cx, left) {
if !is_lint_allowed(cx, STRING_ADD_ASSIGN, e.hir_id) {
let parent = get_parent_expr(cx, e);
if let Some(p) = parent {
if let ExprKind::Assign(target, _, _) = p.kind {
// avoid duplicate matches
if SpanlessEq::new(cx).eq_expr(target, left) {
return;
}
}
}
}
span_lint(
cx,
STRING_ADD,
e.span,
"you added something to a string. Consider using `String::push_str()` instead",
);
}
},
ExprKind::Assign(target, src, _) => {
if is_string(cx, target) && is_add(cx, src, target) {
span_lint(
cx,
STRING_ADD_ASSIGN,
e.span,
"you assigned the result of adding something to this string. Consider using \
`String::push_str()` instead",
);
}
},
ExprKind::Index(target, _idx, _) => {
let e_ty = cx.typeck_results().expr_ty(target).peel_refs();
if e_ty.is_str() || is_type_lang_item(cx, e_ty, LangItem::String) {
span_lint(
cx,
STRING_SLICE,
e.span,
"indexing into a string may panic if the index is within a UTF-8 character",
);
}
},
_ => {},
}
}
}
fn is_string(cx: &LateContext<'_>, e: &Expr<'_>) -> bool {
is_type_lang_item(cx, cx.typeck_results().expr_ty(e).peel_refs(), LangItem::String)
}
fn is_add(cx: &LateContext<'_>, src: &Expr<'_>, target: &Expr<'_>) -> bool {
match peel_blocks(src).kind {
ExprKind::Binary(
Spanned {
node: BinOpKind::Add, ..
},
left,
_,
) => SpanlessEq::new(cx).eq_expr(target, left),
_ => false,
}
}
declare_clippy_lint! {
/// ### What it does
/// Check if the string is transformed to byte array and casted back to string.
///
/// ### Why is this bad?
/// It's unnecessary, the string can be used directly.
///
/// ### Example
/// ```no_run
/// std::str::from_utf8(&"Hello World!".as_bytes()[6..11]).unwrap();
/// ```
///
/// Use instead:
/// ```no_run
/// &"Hello World!"[6..11];
/// ```
#[clippy::version = "1.50.0"]
pub STRING_FROM_UTF8_AS_BYTES,
complexity,
"casting string slices to byte slices and back"
}
// Max length a b"foo" string can take
const MAX_LENGTH_BYTE_STRING_LIT: usize = 32;
declare_lint_pass!(StringLitAsBytes => [STRING_LIT_AS_BYTES, STRING_FROM_UTF8_AS_BYTES]);
impl<'tcx> LateLintPass<'tcx> for StringLitAsBytes {
#[expect(clippy::too_many_lines)]
fn check_expr(&mut self, cx: &LateContext<'tcx>, e: &'tcx Expr<'_>) {
use rustc_ast::LitKind;
if_chain! {
// Find std::str::converts::from_utf8
if let ExprKind::Call(fun, args) = e.kind;
if is_path_diagnostic_item(cx, fun, sym::str_from_utf8);
// Find string::as_bytes
if let ExprKind::AddrOf(BorrowKind::Ref, _, args) = args[0].kind;
if let ExprKind::Index(left, right, _) = args.kind;
let (method_names, expressions, _) = method_calls(left, 1);
if method_names.len() == 1;
if expressions.len() == 1;
if expressions[0].1.is_empty();
if method_names[0] == sym!(as_bytes);
// Check for slicer
if let ExprKind::Struct(QPath::LangItem(LangItem::Range, ..), _, _) = right.kind;
then {
let mut applicability = Applicability::MachineApplicable;
let string_expression = &expressions[0].0;
let snippet_app = snippet_with_applicability(
cx,
string_expression.span, "..",
&mut applicability,
);
span_lint_and_sugg(
cx,
STRING_FROM_UTF8_AS_BYTES,
e.span,
"calling a slice of `as_bytes()` with `from_utf8` should be not necessary",
"try",
format!("Some(&{snippet_app}[{}])", snippet(cx, right.span, "..")),
applicability
)
}
}
if_chain! {
if !in_external_macro(cx.sess(), e.span);
if let ExprKind::MethodCall(path, receiver, ..) = &e.kind;
if path.ident.name == sym!(as_bytes);
if let ExprKind::Lit(lit) = &receiver.kind;
if let LitKind::Str(lit_content, _) = &lit.node;
then {
let callsite = snippet(cx, receiver.span.source_callsite(), r#""foo""#);
let mut applicability = Applicability::MachineApplicable;
if callsite.starts_with("include_str!") {
span_lint_and_sugg(
cx,
STRING_LIT_AS_BYTES,
e.span,
"calling `as_bytes()` on `include_str!(..)`",
"consider using `include_bytes!(..)` instead",
snippet_with_applicability(cx, receiver.span, r#""foo""#, &mut applicability).replacen(
"include_str",
"include_bytes",
1,
),
applicability,
);
} else if lit_content.as_str().is_ascii()
&& lit_content.as_str().len() <= MAX_LENGTH_BYTE_STRING_LIT
&& !receiver.span.from_expansion()
{
if let Some((parent, id)) = get_expr_use_or_unification_node(cx.tcx, e)
&& let Node::Expr(parent) = parent
&& let ExprKind::Match(scrutinee, ..) = parent.kind
&& scrutinee.hir_id == id
{
// Don't lint. Byte strings produce `&[u8; N]` whereas `as_bytes()` produces
// `&[u8]`. This change would prevent matching with different sized slices.
} else if !callsite.starts_with("env!") {
span_lint_and_sugg(
cx,
STRING_LIT_AS_BYTES,
e.span,
"calling `as_bytes()` on a string literal",
"consider using a byte string literal instead",
format!(
"b{}",
snippet_with_applicability(cx, receiver.span, r#""foo""#, &mut applicability)
),
applicability,
);
}
}
}
}
if_chain! {
if let ExprKind::MethodCall(path, recv, [], _) = &e.kind;
if path.ident.name == sym!(into_bytes);
if let ExprKind::MethodCall(path, recv, [], _) = &recv.kind;
if matches!(path.ident.name.as_str(), "to_owned" | "to_string");
if let ExprKind::Lit(lit) = &recv.kind;
if let LitKind::Str(lit_content, _) = &lit.node;
if lit_content.as_str().is_ascii();
if lit_content.as_str().len() <= MAX_LENGTH_BYTE_STRING_LIT;
if !recv.span.from_expansion();
then {
let mut applicability = Applicability::MachineApplicable;
span_lint_and_sugg(
cx,
STRING_LIT_AS_BYTES,
e.span,
"calling `into_bytes()` on a string literal",
"consider using a byte string literal instead",
format!(
"b{}.to_vec()",
snippet_with_applicability(cx, recv.span, r#""..""#, &mut applicability)
),
applicability,
);
}
}
}
}
declare_clippy_lint! {
/// ### What it does
/// This lint checks for `.to_string()` method calls on values of type `&str`.
///
/// ### Why is this bad?
/// The `to_string` method is also used on other types to convert them to a string.
/// When called on a `&str` it turns the `&str` into the owned variant `String`, which can be better
/// expressed with `.to_owned()`.
///
/// ### Example
/// ```no_run
/// // example code where clippy issues a warning
/// let _ = "str".to_string();
/// ```
/// Use instead:
/// ```no_run
/// // example code which does not raise clippy warning
/// let _ = "str".to_owned();
/// ```
#[clippy::version = "pre 1.29.0"]
pub STR_TO_STRING,
restriction,
"using `to_string()` on a `&str`, which should be `to_owned()`"
}
declare_lint_pass!(StrToString => [STR_TO_STRING]);
impl<'tcx> LateLintPass<'tcx> for StrToString {
fn check_expr(&mut self, cx: &LateContext<'tcx>, expr: &Expr<'_>) {
if_chain! {
if let ExprKind::MethodCall(path, self_arg, ..) = &expr.kind;
if path.ident.name == sym::to_string;
let ty = cx.typeck_results().expr_ty(self_arg);
if let ty::Ref(_, ty, ..) = ty.kind();
if ty.is_str();
then {
span_lint_and_help(
cx,
STR_TO_STRING,
expr.span,
"`to_string()` called on a `&str`",
None,
"consider using `.to_owned()`",
);
}
}
}
}
declare_clippy_lint! {
/// ### What it does
/// This lint checks for `.to_string()` method calls on values of type `String`.
///
/// ### Why is this bad?
/// The `to_string` method is also used on other types to convert them to a string.
/// When called on a `String` it only clones the `String`, which can be better expressed with `.clone()`.
///
/// ### Example
/// ```no_run
/// // example code where clippy issues a warning
/// let msg = String::from("Hello World");
/// let _ = msg.to_string();
/// ```
/// Use instead:
/// ```no_run
/// // example code which does not raise clippy warning
/// let msg = String::from("Hello World");
/// let _ = msg.clone();
/// ```
#[clippy::version = "pre 1.29.0"]
pub STRING_TO_STRING,
restriction,
"using `to_string()` on a `String`, which should be `clone()`"
}
declare_lint_pass!(StringToString => [STRING_TO_STRING]);
impl<'tcx> LateLintPass<'tcx> for StringToString {
fn check_expr(&mut self, cx: &LateContext<'tcx>, expr: &Expr<'_>) {
if_chain! {
if let ExprKind::MethodCall(path, self_arg, ..) = &expr.kind;
if path.ident.name == sym::to_string;
let ty = cx.typeck_results().expr_ty(self_arg);
if is_type_lang_item(cx, ty, LangItem::String);
then {
span_lint_and_help(
cx,
STRING_TO_STRING,
expr.span,
"`to_string()` called on a `String`",
None,
"consider using `.clone()`",
);
}
}
}
}
declare_clippy_lint! {
/// ### What it does
/// Warns about calling `str::trim` (or variants) before `str::split_whitespace`.
///
/// ### Why is this bad?
/// `split_whitespace` already ignores leading and trailing whitespace.
///
/// ### Example
/// ```no_run
/// " A B C ".trim().split_whitespace();
/// ```
/// Use instead:
/// ```no_run
/// " A B C ".split_whitespace();
/// ```
#[clippy::version = "1.62.0"]
pub TRIM_SPLIT_WHITESPACE,
style,
"using `str::trim()` or alike before `str::split_whitespace`"
}
declare_lint_pass!(TrimSplitWhitespace => [TRIM_SPLIT_WHITESPACE]);
impl<'tcx> LateLintPass<'tcx> for TrimSplitWhitespace {
fn check_expr(&mut self, cx: &LateContext<'tcx>, expr: &Expr<'_>) {
let tyckres = cx.typeck_results();
if_chain! {
if let ExprKind::MethodCall(path, split_recv, [], split_ws_span) = expr.kind;
if path.ident.name == sym!(split_whitespace);
if let Some(split_ws_def_id) = tyckres.type_dependent_def_id(expr.hir_id);
if cx.tcx.is_diagnostic_item(sym::str_split_whitespace, split_ws_def_id);
if let ExprKind::MethodCall(path, _trim_recv, [], trim_span) = split_recv.kind;
if let trim_fn_name @ ("trim" | "trim_start" | "trim_end") = path.ident.name.as_str();
if let Some(trim_def_id) = tyckres.type_dependent_def_id(split_recv.hir_id);
if is_one_of_trim_diagnostic_items(cx, trim_def_id);
then {
span_lint_and_sugg(
cx,
TRIM_SPLIT_WHITESPACE,
trim_span.with_hi(split_ws_span.lo()),
&format!("found call to `str::{trim_fn_name}` before `str::split_whitespace`"),
&format!("remove `{trim_fn_name}()`"),
String::new(),
Applicability::MachineApplicable,
);
}
}
}
}
fn is_one_of_trim_diagnostic_items(cx: &LateContext<'_>, trim_def_id: DefId) -> bool {
cx.tcx.is_diagnostic_item(sym::str_trim, trim_def_id)
|| cx.tcx.is_diagnostic_item(sym::str_trim_start, trim_def_id)
|| cx.tcx.is_diagnostic_item(sym::str_trim_end, trim_def_id)
}