================
@@ -0,0 +1,546 @@
+//===-- 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;
+ for (char C : Input) {
+ if (Escape.find(C) != Escape.end())
+ Output += Escape[C];
+ 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) {
+ if (Idx == 0 || Tokens[Idx - 1].getType() != Token::Type::Text)
+ return false;
+ const Token &PrevToken = Tokens[Idx - 1];
+ StringRef TokenBody = PrevToken.getRawBody().rtrim(" \t\v\t");
+ // We make a special exception for when previous token is empty
+ // and the current token is the second token
+ // ex.
+ // " {{#section}}A{{/section}}"
+ // that's why we check if the token body is empty and the index is 1
+ return TokenBody.ends_with("\n") || (TokenBody.empty() && Idx == 1);
+}
+
+// Function to check if there's no meaningful text ahead
+bool noTextAhead(size_t Idx, const SmallVector<Token, 0> &Tokens) {
+ if (Idx >= Tokens.size() - 1 ||
+ Tokens[Idx + 1].getType() != Token::Type::Text)
+ return false;
+
+ const Token &NextToken = Tokens[Idx + 1];
+ StringRef TokenBody = NextToken.getRawBody().ltrim(" ");
+ return TokenBody.starts_with("\r\n") || TokenBody.starts_with("\n");
+}
+
+bool requiresCleanUp(Token::Type T) {
+ // We must clean up all the tokens that could contain child nodes
+ return T == Token::Type::SectionOpen || T == Token::Type::InvertSectionOpen
||
+ T == Token::Type::SectionClose || T == Token::Type::Comment ||
+ T == Token::Type::Partial;
+}
+
+// Simple tokenizer that splits the template into tokens
+// the mustache spec allows {{{ }}} to unescape variables
+// but we don't support that here unescape variable
+// is represented only by {{& variable}}
+SmallVector<Token, 0> tokenize(StringRef Template) {
+ SmallVector<Token, 0> Tokens;
+ StringRef Open("{{");
+ StringRef Close("}}");
+ size_t Start = 0;
+ size_t DelimiterStart = Template.find(Open);
+ if (DelimiterStart == StringRef::npos) {
+ Tokens.emplace_back(Template);
+ return Tokens;
+ }
+ while (DelimiterStart != StringRef::npos) {
+ if (DelimiterStart != Start) {
+ Tokens.emplace_back(Template.substr(Start, DelimiterStart - Start));
+ }
+
+ size_t DelimiterEnd = Template.find(Close, DelimiterStart);
+ if (DelimiterEnd == StringRef::npos) {
+ break;
+ }
+
+ // Extract the Interpolated variable without {{ and }}
+ size_t InterpolatedStart = DelimiterStart + Open.size();
+ size_t InterpolatedEnd = DelimiterEnd - DelimiterStart - Close.size();
+ SmallString<0> Interpolated =
+ Template.substr(InterpolatedStart, InterpolatedEnd);
+ SmallString<0> RawBody({Open, Interpolated, Close});
+ Tokens.emplace_back(RawBody, Interpolated, Interpolated[0]);
+ Start = DelimiterEnd + Close.size();
+ DelimiterStart = Template.find(Open, Start);
+ }
+
+ if (Start < Template.size())
+ Tokens.emplace_back(Template.substr(Start));
+
+ // Fix up white spaces for:
+ // open sections/inverted sections/close section/comment
+ // This loop attempts to find standalone tokens and tries to trim out
+ // the surrounding whitespace.
+ // For example:
+ // if you have the template string
+ // "Line 1\n {{#section}} \n Line 2 \n {{/section}} \n Line 3"
+ // The output would be
+ // "Line 1\n Line 2\n Line 3"
----------------
ilovepi wrote:
```suggestion
// Fix up white spaces for:
// - open sections
// - inverted sections
// - close sections
// - comments
//
// This loop attempts to find standalone tokens and tries to trim out
// the surrounding whitespace.
// For example:
// if you have the template string
// "Line 1\n {{#section}} \n Line 2 \n {{/section}} \n Line 3"
// The output would be
// "Line 1\n Line 2\n Line 3"
```
https://github.com/llvm/llvm-project/pull/105893
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
