[clang-tools-extra] [llvm] [llvm] add support for mustache templating language (PR #105893)

[Paul Kirth via cfe-commits]

================
@@ -0,0 +1,567 @@
+//===-- Mustache.cpp ------------------------------------------------------===//
+//
+// 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 "llvm/Support/Mustache.h"
+#include "llvm/Support/Error.h"
+#include <sstream>
+
+using namespace llvm;
+using namespace llvm::json;
+using namespace llvm::mustache;
+
+SmallString<0> escapeString(StringRef Input,
+                            DenseMap<char, StringRef> &Escape) {
+  SmallString<0> Output;
+  Output.reserve(Input.size());
+  for (char C : Input) {
+    auto It = Escape.find(C);
+    if (It != Escape.end())
+      Output += It->getSecond();
+    else
+      Output += C;
+  }
+  return Output;
+}
+
+Accessor split(StringRef Str, char Delimiter) {
+  Accessor Tokens;
+  if (Str == ".") {
+    Tokens.emplace_back(Str);
+    return Tokens;
+  }
+  StringRef Ref(Str);
+  while (!Ref.empty()) {
+    StringRef Part;
+    std::tie(Part, Ref) = Ref.split(Delimiter);
+    Tokens.emplace_back(Part.trim());
+  }
+  return Tokens;
+}
+
+void addIndentation(llvm::SmallString<0> &PartialStr, size_t IndentationSize) {
+  std::string Indent(IndentationSize, ' ');
+  llvm::SmallString<0> Result;
+  for (size_t I = 0; I < PartialStr.size(); ++I) {
+    Result.push_back(PartialStr[I]);
+    if (PartialStr[I] == '\n' && I < PartialStr.size() - 1)
+      Result.append(Indent);
+  }
+  PartialStr = Result;
+}
+
+Token::Token(StringRef RawBody, StringRef InnerBody, char Identifier)
+    : RawBody(RawBody), TokenBody(InnerBody), Indentation(0) {
+  TokenType = getTokenType(Identifier);
+  if (TokenType == Type::Comment)
+    return;
+
+  StringRef AccessorStr =
+      TokenType == Type::Variable ? InnerBody : InnerBody.substr(1);
+
+  Accessor = split(AccessorStr.trim(), '.');
+}
+
+Token::Token(StringRef Str)
+    : TokenType(Type::Text), RawBody(Str), Accessor({}), TokenBody(Str),
+      Indentation(0) {}
+
+Token::Type Token::getTokenType(char Identifier) {
+  switch (Identifier) {
+  case '#':
+    return Type::SectionOpen;
+  case '/':
+    return Type::SectionClose;
+  case '^':
+    return Type::InvertSectionOpen;
+  case '!':
+    return Type::Comment;
+  case '>':
+    return Type::Partial;
+  case '&':
+    return Type::UnescapeVariable;
+  default:
+    return Type::Variable;
+  }
+}
+
+// Function to check if there's no meaningful text behind
+bool noTextBehind(size_t Idx, const SmallVector<Token, 0> &Tokens) {
----------------
ilovepi wrote:

Looking though, the vector parameter should probably be an `ArrayRef`. For APIs where you need to modify the vector, you can use `SmallVectorImpl` as an in/out parameter.

There's a note in the programmer's manual about this. https://llvm.org/docs/ProgrammersManual.html#llvm-adt-smallvector-h

https://github.com/llvm/llvm-project/pull/105893