[llvm] [clang-tools-extra] [clang] [clang-tidy] modernize-use-digit-separator issue #35414 (PR #76153)

[Piotr Zegar via cfe-commits]

================
@@ -0,0 +1,235 @@
+//===--- UseDigitSeparatorCheck.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 <algorithm>
+#include <iomanip>
+#include <numeric>
+#include <regex>
+#include <sstream>
+
+#include "UseDigitSeparatorCheck.h"
+#include "clang/AST/ASTContext.h"
+#include "clang/ASTMatchers/ASTMatchFinder.h"
+#include "clang/Lex/Lexer.h"
+#include "llvm/Support/FormatVariadic.h"
+
+using namespace clang::ast_matchers;
+
+namespace {
+std::string
+getFormatedScientificFloatString(const llvm::StringRef OriginalLiteralString);
+
+std::vector<std::basic_string<char>>
+splitStringByGroupSize(const std::basic_string<char> &String,
+                       size_t GroupSize) {
+  std::vector<std::basic_string<char>> Result;
+  std::basic_string<char> ReversedString(String.rbegin(), String.rend());
+
+  for (size_t I = 0; I < ReversedString.size(); I += GroupSize) {
+    Result.push_back(ReversedString.substr(I, GroupSize));
+  }
+
+  std::reverse(Result.begin(), Result.end());
+  std::for_each(Result.begin(), Result.end(), [](std::basic_string<char> &Str) {
+    return std::reverse(Str.begin(), Str.end());
+  });
+
+  return Result;
+}
+
+std::string
+getFormatedIntegerString(const llvm::StringRef OriginalLiteralString,
+                         const llvm::APInt IntegerValue) {
+  // Configure formatting
+  unsigned int Radix;
+  size_t GroupSize;
+  std::string Prefix;
+  std::string Postfix;
+  if (OriginalLiteralString.starts_with("0b")) {
+    Radix = 2;
+    GroupSize = 4;
+    Prefix = "0b";
+  } else if (OriginalLiteralString.starts_with("0x")) {
+    Radix = 16;
+    GroupSize = 4;
+    Prefix = "0x";
+  } else if (OriginalLiteralString.starts_with("0") &&
+             OriginalLiteralString != "0") {
+    Radix = 8;
+    GroupSize = 3;
+    Prefix = "0";
+  } else {
+    Radix = 10;
+    GroupSize = 3;
+  }
+
+  for (const char &Character : OriginalLiteralString.substr(
+           Prefix.size(), OriginalLiteralString.size())) {
+    if (((Prefix == "0x" && !std::isxdigit(Character)) ||
+         (Prefix != "0x" && !std::isdigit(Character))) &&
+        Character != '\'') {
+      Postfix += Character;
+    }
+  }
+
+  // Get formatting literal text
+  const std::vector<std::string> SplittedIntegerLiteral =
+      splitStringByGroupSize(toString(IntegerValue, Radix, true), GroupSize);
+  const std::string FormatedLiteralString =
+      Prefix +
+      std::accumulate(SplittedIntegerLiteral.begin(),
+                      SplittedIntegerLiteral.end(), std::string(""),
+                      [](std::basic_string<char> S1,
+                         std::basic_string<char> S2) { return S1 + "\'" + S2; })
+          .erase(0, 1) +
+      Postfix;
+
+  return FormatedLiteralString;
+}
+
+std::string getFormatedFloatString(const llvm::StringRef OriginalLiteralString,
+                                   const llvm::APFloat FloatValue) {
+  if (OriginalLiteralString.contains('E') ||
+      OriginalLiteralString.contains('e')) {
+    return getFormatedScientificFloatString(OriginalLiteralString);
+  }
+
+  // Configure formatting
+  std::string Postfix;
+  for (const char &Character : OriginalLiteralString) {
+    if (!std::isdigit(Character) && Character != '.' && Character != '\'') {
+      Postfix += Character;
+    }
+  }
+
+  // Get precision
+  const std::string::size_type OriginalDotPosition =
----------------
PiotrZSL wrote:

mising std::string with llvm::StringRef, use just llvm::StringRef::size_type

https://github.com/llvm/llvm-project/pull/76153