[clang-tidy] Add new check 'llvm-use-ranges' (#152047)

First iteration of the check, mostly reused logic from
#97764 without adding any
LLVM-specific iterator-methods.
Successfully run on `LLVM` codebase with ~100 findings and a couple of
odd FPs: when we have `std::sort(this->begin(), this->end())` or
`std::sort(begin(), end())`.
I didn't fix this cases since it will be a separate task for the core
`utils::UseRangesCheck`.

Fixes #38468.

---------

Co-authored-by: EugeneZelenko <eugene.zelenko@gmail.com>

Author: vbvictor

clang-tools-extra/clang-tidy/llvm/CMakeLists.txt

@@ -12,6 +12,7 @@ add_clang_library(clangTidyLLVMModule STATIC
   PreferStaticOverAnonymousNamespaceCheck.cpp
   TwineLocalCheck.cpp
   UseNewMLIROpBuilderCheck.cpp
+  UseRangesCheck.cpp
 
   LINK_LIBS
   clangTidy

clang-tools-extra/clang-tidy/llvm/LLVMTidyModule.cpp

@@ -19,6 +19,7 @@
 #include "PreferStaticOverAnonymousNamespaceCheck.h"
 #include "TwineLocalCheck.h"
 #include "UseNewMLIROpBuilderCheck.h"
+#include "UseRangesCheck.h"
 
 namespace clang::tidy {
 namespace llvm_check {
@@ -43,6 +44,7 @@ class LLVMModule : public ClangTidyModule {
     CheckFactories.registerCheck<TwineLocalCheck>("llvm-twine-local");
     CheckFactories.registerCheck<UseNewMlirOpBuilderCheck>(
         "llvm-use-new-mlir-op-builder");
+    CheckFactories.registerCheck<UseRangesCheck>("llvm-use-ranges");
   }
 
   ClangTidyOptions getModuleOptions() override {

clang-tools-extra/clang-tidy/llvm/UseRangesCheck.cpp

@@ -0,0 +1,97 @@
+//===--- UseRangesCheck.cpp - clang-tidy ----------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "UseRangesCheck.h"
+
+namespace clang::tidy::llvm_check {
+
+namespace {
+
+class StdToLLVMReplacer : public utils::UseRangesCheck::Replacer {
+public:
+  explicit StdToLLVMReplacer(
+      ArrayRef<utils::UseRangesCheck::Signature> Signatures)
+      : Signatures(Signatures) {}
+
+  ArrayRef<utils::UseRangesCheck::Signature>
+  getReplacementSignatures() const override {
+    return Signatures;
+  }
+
+  std::optional<std::string>
+  getReplaceName(const NamedDecl &OriginalName) const override {
+    return ("llvm::" + OriginalName.getName()).str();
+  }
+
+  std::optional<std::string>
+  getHeaderInclusion(const NamedDecl &) const override {
+    return "llvm/ADT/STLExtras.h";
+  }
+
+private:
+  ArrayRef<utils::UseRangesCheck::Signature> Signatures;
+};
+
+} // namespace
+
+utils::UseRangesCheck::ReplacerMap UseRangesCheck::getReplacerMap() const {
+  ReplacerMap Results;
+
+  static const Signature SingleSig = {{0}};
+  static const Signature TwoSig = {{0}, {2}};
+
+  const auto AddStdToLLVM =
+      [&Results](llvm::IntrusiveRefCntPtr<Replacer> Replacer,
+                 std::initializer_list<StringRef> Names) {
+        for (const auto &Name : Names) {
+          Results.try_emplace(("::std::" + Name).str(), Replacer);
+        }
+      };
+
+  // Single range algorithms
+  AddStdToLLVM(llvm::makeIntrusiveRefCnt<StdToLLVMReplacer>(SingleSig),
+               {"all_of",      "any_of",
+                "none_of",     "for_each",
+                "find",        "find_if",
+                "find_if_not", "fill",
+                "count",       "count_if",
+                "copy",        "copy_if",
+                "transform",   "replace",
+                "remove_if",   "stable_sort",
+                "partition",   "partition_point",
+                "is_sorted",   "min_element",
+                "max_element", "binary_search",
+                "lower_bound", "upper_bound",
+                "unique",      "uninitialized_copy"});
+
+  // Two range algorithms
+  AddStdToLLVM(llvm::makeIntrusiveRefCnt<StdToLLVMReplacer>(TwoSig),
+               {"equal", "mismatch", "includes"});
+
+  return Results;
+}
+
+UseRangesCheck::UseRangesCheck(StringRef Name, ClangTidyContext *Context)
+    : utils::UseRangesCheck(Name, Context) {}
+
+DiagnosticBuilder UseRangesCheck::createDiag(const CallExpr &Call) {
+  return diag(Call.getBeginLoc(), "use an LLVM range-based algorithm");
+}
+
+ArrayRef<std::pair<StringRef, StringRef>>
+UseRangesCheck::getFreeBeginEndMethods() const {
+  static constexpr std::pair<StringRef, StringRef> Refs[] = {
+      {"::std::begin", "::std::end"},
+      {"::std::cbegin", "::std::cend"},
+      {"::std::rbegin", "::std::rend"},
+      {"::std::crbegin", "::std::crend"},
+  };
+  return Refs;
+}
+
+} // namespace clang::tidy::llvm_check

clang-tools-extra/clang-tidy/llvm/UseRangesCheck.h

@@ -0,0 +1,33 @@
+//===--- UseRangesCheck.h - clang-tidy --------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_LLVM_USERANGESCHECK_H
+#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_LLVM_USERANGESCHECK_H
+
+#include "../utils/UseRangesCheck.h"
+
+namespace clang::tidy::llvm_check {
+
+/// Finds calls to STL iterator algorithms that can be replaced with LLVM
+/// range-based algorithms from `llvm/ADT/STLExtras.h`.
+///
+/// For the user-facing documentation see:
+/// http://clang.llvm.org/extra/clang-tidy/checks/llvm/use-ranges.html
+class UseRangesCheck : public utils::UseRangesCheck {
+public:
+  UseRangesCheck(StringRef Name, ClangTidyContext *Context);
+
+  ReplacerMap getReplacerMap() const override;
+  DiagnosticBuilder createDiag(const CallExpr &Call) override;
+  ArrayRef<std::pair<StringRef, StringRef>>
+  getFreeBeginEndMethods() const override;
+};
+
+} // namespace clang::tidy::llvm_check
+
+#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_LLVM_USERANGESCHECK_H

clang-tools-extra/docs/ReleaseNotes.rst

@@ -157,6 +157,12 @@ New checks
   Checks for uses of MLIR's old/to be deprecated ``OpBuilder::create<T>`` form
   and suggests using ``T::create`` instead.
 
+- New :doc:`llvm-use-ranges
+  <clang-tidy/checks/llvm/use-ranges>` check.
+
+  Finds calls to STL library iterator algorithms that could be replaced with
+  LLVM range-based algorithms from ``llvm/ADT/STLExtras.h``.
+
 - New :doc:`misc-override-with-different-visibility
   <clang-tidy/checks/misc/override-with-different-visibility>` check.
 

clang-tools-extra/docs/clang-tidy/checks/list.rst

@@ -249,12 +249,13 @@ Clang-Tidy Checks
    :doc:`linuxkernel-must-check-errs <linuxkernel/must-check-errs>`,
    :doc:`llvm-header-guard <llvm/header-guard>`,
    :doc:`llvm-include-order <llvm/include-order>`, "Yes"
-   :doc:`llvm-use-new-mlir-op-builder <llvm/use-new-mlir-op-builder>`, "Yes"
    :doc:`llvm-namespace-comment <llvm/namespace-comment>`,
    :doc:`llvm-prefer-isa-or-dyn-cast-in-conditionals <llvm/prefer-isa-or-dyn-cast-in-conditionals>`, "Yes"
    :doc:`llvm-prefer-register-over-unsigned <llvm/prefer-register-over-unsigned>`, "Yes"
    :doc:`llvm-prefer-static-over-anonymous-namespace <llvm/prefer-static-over-anonymous-namespace>`,
    :doc:`llvm-twine-local <llvm/twine-local>`, "Yes"
+   :doc:`llvm-use-new-mlir-op-builder <llvm/use-new-mlir-op-builder>`, "Yes"
+   :doc:`llvm-use-ranges <llvm/use-ranges>`, "Yes"
    :doc:`llvmlibc-callee-namespace <llvmlibc/callee-namespace>`,
    :doc:`llvmlibc-implementation-in-namespace <llvmlibc/implementation-in-namespace>`,
    :doc:`llvmlibc-inline-function-decl <llvmlibc/inline-function-decl>`, "Yes"

clang-tools-extra/docs/clang-tidy/checks/llvm/use-ranges.rst

@@ -0,0 +1,61 @@
+.. title:: clang-tidy - llvm-use-ranges
+
+llvm-use-ranges
+===============
+
+Finds calls to STL library iterator algorithms that could be replaced with
+LLVM range-based algorithms from ``llvm/ADT/STLExtras.h``.
+
+Example
+-------
+
+.. code-block:: c++
+
+  auto it = std::find(vec.begin(), vec.end(), value);
+  bool all = std::all_of(vec.begin(), vec.end(), 
+                         [](int x) { return x > 0; });
+
+Transforms to:
+
+.. code-block:: c++
+
+  auto it = llvm::find(vec, value);
+  bool all = llvm::all_of(vec, [](int x) { return x > 0; });
+
+Supported algorithms
+--------------------
+
+Calls to the following STL algorithms are checked:
+
+``std::all_of``,
+``std::any_of``,
+``std::binary_search``,
+``std::copy``,
+``std::copy_if``,
+``std::count``,
+``std::count_if``,
+``std::equal``,
+``std::fill``,
+``std::find``,
+``std::find_if``,
+``std::find_if_not``,
+``std::for_each``,
+``std::includes``,
+``std::is_sorted``,
+``std::lower_bound``,
+``std::max_element``,
+``std::min_element``,
+``std::mismatch``,
+``std::none_of``,
+``std::partition``,
+``std::partition_point``,
+``std::remove_if``,
+``std::replace``,
+``std::stable_sort``,
+``std::transform``,
+``std::uninitialized_copy``,
+``std::unique``,
+``std::upper_bound``.
+
+The check will add the necessary ``#include "llvm/ADT/STLExtras.h"`` directive
+when applying fixes.

clang-tools-extra/test/clang-tidy/checkers/llvm/use-ranges.cpp

@@ -0,0 +1,141 @@
+// RUN: %check_clang_tidy %s llvm-use-ranges %t
+
+// Test that the header is included
+// CHECK-FIXES: #include "llvm/ADT/STLExtras.h"
+
+namespace std {
+
+template <typename T> class vector {
+public:
+  using iterator = T *;
+  using const_iterator = const T *;
+
+  iterator begin();
+  iterator end();
+  const_iterator begin() const;
+  const_iterator end() const;
+  const_iterator cbegin() const;
+  const_iterator cend() const;
+};
+
+template <typename T> T* begin(T (&arr)[5]);
+template <typename T> T* end(T (&arr)[5]);
+
+template <class InputIt, class T>
+InputIt find(InputIt first, InputIt last, const T &value);
+
+template <class RandomIt>
+void sort(RandomIt first, RandomIt last);
+
+template <class RandomIt>
+void stable_sort(RandomIt first, RandomIt last);
+
+template <class InputIt, class UnaryPredicate>
+bool all_of(InputIt first, InputIt last, UnaryPredicate p);
+
+template <class InputIt, class UnaryFunction>
+UnaryFunction for_each(InputIt first, InputIt last, UnaryFunction f);
+
+template <class ForwardIt, class T>
+ForwardIt remove(ForwardIt first, ForwardIt last, const T& value);
+
+template <class ForwardIt>
+ForwardIt min_element(ForwardIt first, ForwardIt last);
+
+template <class InputIt1, class InputIt2>
+bool equal(InputIt1 first1, InputIt1 last1, InputIt2 first2, InputIt2 last2);
+
+template <class InputIt1, class InputIt2>
+bool equal(InputIt1 first1, InputIt1 last1, InputIt2 first2);
+
+template <class InputIt, class OutputIt>
+OutputIt copy(InputIt first, InputIt last, OutputIt d_first);
+
+template <class ForwardIt, class T>
+void fill(ForwardIt first, ForwardIt last, const T& value);
+
+template <class BidirIt>
+void reverse(BidirIt first, BidirIt last);
+
+template <class ForwardIt>
+ForwardIt unique(ForwardIt first, ForwardIt last);
+
+template <class ForwardIt>
+bool is_sorted(ForwardIt first, ForwardIt last);
+
+template <class InputIt1, class InputIt2>
+bool includes(InputIt1 first1, InputIt1 last1, InputIt2 first2, InputIt2 last2);
+
+} // namespace std
+
+bool is_even(int x);
+void double_ref(int& x);
+
+void test_positive() {
+  std::vector<int> vec;
+  int arr[5] = {1, 2, 3, 4, 5};
+  
+  auto it1 = std::find(vec.begin(), vec.end(), 3);
+  // CHECK-MESSAGES: :[[@LINE-1]]:14: warning: use an LLVM range-based algorithm
+  // CHECK-FIXES: auto it1 = llvm::find(vec, 3);
+
+  auto it2 = std::find(std::begin(arr), std::end(arr), 3);
+  // CHECK-MESSAGES: :[[@LINE-1]]:14: warning: use an LLVM range-based algorithm
+  // CHECK-FIXES: auto it2 = llvm::find(arr, 3);
+
+  std::stable_sort(vec.begin(), vec.end());
+  // CHECK-MESSAGES: :[[@LINE-1]]:3: warning: use an LLVM range-based algorithm
+  // CHECK-FIXES: llvm::stable_sort(vec);
+
+  bool all = std::all_of(vec.begin(), vec.end(), is_even);
+  // CHECK-MESSAGES: :[[@LINE-1]]:14: warning: use an LLVM range-based algorithm
+  // CHECK-FIXES: bool all = llvm::all_of(vec, is_even);
+
+  std::for_each(vec.begin(), vec.end(), double_ref);
+  // CHECK-MESSAGES: :[[@LINE-1]]:3: warning: use an LLVM range-based algorithm
+  // CHECK-FIXES: llvm::for_each(vec, double_ref);
+
+  auto min_it = std::min_element(vec.begin(), vec.end());
+  // CHECK-MESSAGES: :[[@LINE-1]]:17: warning: use an LLVM range-based algorithm
+  // CHECK-FIXES: auto min_it = llvm::min_element(vec);
+
+  std::vector<int> vec2;
+  bool eq = std::equal(vec.begin(), vec.end(), vec2.begin(), vec2.end());
+  // CHECK-MESSAGES: :[[@LINE-1]]:13: warning: use an LLVM range-based algorithm
+  // CHECK-FIXES: bool eq = llvm::equal(vec, vec2);
+
+  std::copy(vec.begin(), vec.end(), vec2.begin());
+  // CHECK-MESSAGES: :[[@LINE-1]]:3: warning: use an LLVM range-based algorithm
+  // CHECK-FIXES: llvm::copy(vec, vec2.begin());
+
+  std::fill(vec.begin(), vec.end(), 0);
+  // CHECK-MESSAGES: :[[@LINE-1]]:3: warning: use an LLVM range-based algorithm
+  // CHECK-FIXES: llvm::fill(vec, 0);
+  
+  auto last = std::unique(vec.begin(), vec.end());
+  // CHECK-MESSAGES: :[[@LINE-1]]:15: warning: use an LLVM range-based algorithm
+  // CHECK-FIXES: auto last = llvm::unique(vec);
+
+  bool sorted = std::is_sorted(vec.begin(), vec.end());
+  // CHECK-MESSAGES: :[[@LINE-1]]:17: warning: use an LLVM range-based algorithm
+  // CHECK-FIXES: bool sorted = llvm::is_sorted(vec);
+
+  std::includes(vec.begin(), vec.end(), std::begin(arr), std::end(arr));
+  // CHECK-MESSAGES: :[[@LINE-1]]:3: warning: use an LLVM range-based algorithm
+  // CHECK-FIXES: llvm::includes(vec, arr);
+}
+
+void test_negative() {
+  std::vector<int> v;
+  
+  // can not use `llvm::sort` because of potential different ordering from `std::sort`.
+  std::sort(v.begin(), v.end());
+
+  //non-begin/end iterators
+  auto it1 = std::find(v.begin() + 1, v.end(), 2);
+  auto it2 = std::find(v.begin(), v.end() - 1, 2);
+  
+  // Using different containers (3-arg equal)
+  std::vector<int> v2;
+  bool eq = std::equal(v.begin(), v.end(), v2.begin());
+}