[clang-tools-extra] [clang-tidy] Add `llvm-formatv-string` (PR #195974)

[Victor Chernyakin via cfe-commits]

================
@@ -0,0 +1,198 @@
+//===----------------------------------------------------------------------===//
+//
+// 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 "FormatvStringCheck.h"
+#include "clang/AST/DeclTemplate.h"
+#include "clang/AST/Expr.h"
+#include "clang/ASTMatchers/ASTMatchers.h"
+#include "llvm/ADT/STLExtras.h"
+#include "llvm/ADT/SmallBitVector.h"
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/Support/Error.h"
+
+using namespace clang::ast_matchers;
+
+namespace clang::tidy::llvm_check {
+
+namespace {
+
+struct ParseResult {
+  llvm::SmallVector<unsigned, 4> Indices;
+  unsigned MaxIndex = 0;
+};
+
+} // namespace
+
+static llvm::Expected<ParseResult> parseFormatvString(llvm::StringRef Fmt) {
+  ParseResult Result;
+  unsigned NextAutoIndex = 0;
+  bool HasAutomatic = false;
+  bool HasExplicit = false;
+
+  while (!Fmt.empty()) {
+    const size_t OpenBrace = Fmt.find('{');
+    if (OpenBrace == llvm::StringRef::npos)
+      break;
+
+    Fmt = Fmt.drop_front(OpenBrace);
+
+    // Handle escaped braces '{{'.
+    if (Fmt.size() > 1 && Fmt[1] == '{') {
+      Fmt = Fmt.drop_front(2);
+      continue;
+    }
+
+    // Find the closing '}'.
+    const size_t CloseBrace = Fmt.find('}');
+    if (CloseBrace == llvm::StringRef::npos)
+      return llvm::createStringError("unterminated brace in format string");
+
+    // Extract the content between braces.
+    const llvm::StringRef Content = Fmt.substr(1, CloseBrace - 1);
+    Fmt = Fmt.drop_front(CloseBrace + 1);
+
+    // Parse the replacement field: [index] ["," layout] [":" format]
+    llvm::StringRef IndexStr = Content;
+
+    // Strip layout and format parts for index parsing.
+    const size_t CommaPos = Content.find(',');
+    const size_t ColonPos = Content.find(':');
+    if (CommaPos != llvm::StringRef::npos)
+      IndexStr = Content.substr(0, CommaPos);
+    else if (ColonPos != llvm::StringRef::npos)
+      IndexStr = Content.substr(0, ColonPos);
+
+    IndexStr = IndexStr.trim();
+
+    unsigned Index = 0;
+    if (IndexStr.empty()) {
+      Index = NextAutoIndex++;
+      HasAutomatic = true;
+    } else {
+      if (IndexStr.getAsInteger(10, Index))
+        return llvm::createStringError(
+            "invalid replacement index in format string");
+      HasExplicit = true;
+    }
+
+    Result.Indices.push_back(Index);
+    Result.MaxIndex = std::max(Result.MaxIndex, Index);
+  }
+
+  if (HasAutomatic && HasExplicit)
+    return llvm::createStringError(
+        "format string mixes automatic and explicit indices");
+
+  return Result;
+}
+
+FormatvStringCheck::FormatvStringCheck(StringRef Name,
+                                       ClangTidyContext *Context)
+    : ClangTidyCheck(Name, Context),
+      AdditionalFunctions(Options.get("AdditionalFunctions", "")) {
+  Functions.insert("llvm::formatv");
+  Functions.insert("llvm::createStringErrorV");
+
+  // Parse semicolon-separated function names from AdditionalFunctions.
+  const llvm::StringRef Input(AdditionalFunctions);
+  llvm::SmallVector<llvm::StringRef, 8> Entries;
+  Input.split(Entries, ';', -1, false);
+  for (llvm::StringRef Entry : Entries) {
+    Entry = Entry.trim();
+    if (!Entry.empty())
+      Functions.insert(Entry);
+  }
----------------
localspook wrote:

There's a function for this, `utils::options::parseStringList` (it's in `"../utils/OptionsUtils.h"`).

https://github.com/llvm/llvm-project/pull/195974