[llvm-rc] Add basic RC scripts parsing ability.

As for now, the parser supports a limited set of statements and
resources. This will be extended in the following patches.

Thanks to Nico Weber (thakis) for his original work in this area.

Differential Revision: https://reviews.llvm.org/D36340

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@311175 91177308-0d34-0410-b5e6-96231b3b80d8

1 files changed, 143 insertions, 0 deletions

diff --git a/tools/llvm-rc/ResourceScriptParser.h b/tools/llvm-rc/ResourceScriptParser.h
new file mode 100644
index 00000000000..017e22f273d
--- /dev/null
+++ b/tools/llvm-rc/ResourceScriptParser.h
@@ -0,0 +1,143 @@
+//===-- ResourceScriptParser.h ----------------------------------*- C++-*-===//
+//
+//                     The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===---------------------------------------------------------------------===//
+//
+// This defines the RC scripts parser. It takes a sequence of RC tokens
+// and then provides the method to parse the resources one by one.
+//
+//===---------------------------------------------------------------------===//
+
+#ifndef LLVM_TOOLS_LLVMRC_RESOURCESCRIPTPARSER_H
+#define LLVM_TOOLS_LLVMRC_RESOURCESCRIPTPARSER_H
+
+#include "ResourceScriptStmt.h"
+#include "ResourceScriptToken.h"
+
+#include "llvm/Support/Compiler.h"
+#include "llvm/Support/raw_ostream.h"
+
+#include <system_error>
+#include <vector>
+
+namespace llvm {
+namespace rc {
+
+class RCParser {
+public:
+  using LocIter = std::vector<RCToken>::iterator;
+  using ParseType = Expected<std::unique_ptr<RCResource>>;
+  using ParseOptionType = Expected<std::unique_ptr<OptionalStmt>>;
+
+  // Class describing a single failure of parser.
+  class ParserError : public ErrorInfo<ParserError> {
+  public:
+    ParserError(Twine Expected, const LocIter CurLoc, const LocIter End);
+
+    void log(raw_ostream &OS) const override { OS << CurMessage; }
+    std::error_code convertToErrorCode() const override {
+      return std::make_error_code(std::errc::invalid_argument);
+    }
+    const std::string &getMessage() const { return CurMessage; }
+
+    static char ID; // Keep llvm::Error happy.
+
+  private:
+    std::string CurMessage;
+    LocIter ErrorLoc, FileEnd;
+  };
+
+  RCParser(const std::vector<RCToken> &TokenList);
+  RCParser(std::vector<RCToken> &&TokenList);
+
+  // Reads and returns a single resource definition, or error message if any
+  // occurred.
+  ParseType parseSingleResource();
+
+  bool isEof() const;
+
+private:
+  using Kind = RCToken::Kind;
+
+  // Checks if the current parser state points to the token of type TokenKind.
+  bool isNextTokenKind(Kind TokenKind) const;
+
+  // These methods assume that the parser is not in EOF state.
+
+  // Take a look at the current token. Do not fetch it.
+  const RCToken &look() const;
+  // Read the current token and advance the state by one token.
+  const RCToken &read();
+  // Advance the state by one token, discarding the current token.
+  void consume();
+
+  // The following methods try to read a single token, check if it has the
+  // correct type and then parse it.
+  Expected<uint32_t> readInt();           // Parse an integer.
+  Expected<StringRef> readString();       // Parse a string.
+  Expected<StringRef> readIdentifier();   // Parse an identifier.
+  Expected<IntOrString> readTypeOrName(); // Parse an integer or an identifier.
+
+  // Advance the state by one, discarding the current token.
+  // If the discarded token had an incorrect type, fail.
+  Error consumeType(Kind TokenKind);
+
+  // Check the current token type. If it's TokenKind, discard it.
+  // Return true if the parser consumed this token successfully.
+  bool consumeOptionalType(Kind TokenKind);
+
+  // Read at least MinCount, and at most MaxCount integers separated by
+  // commas. The parser stops reading after fetching MaxCount integers
+  // or after an error occurs. Whenever the parser reads a comma, it
+  // expects an integer to follow.
+  Expected<SmallVector<uint32_t, 8>> readIntsWithCommas(size_t MinCount,
+                                                        size_t MaxCount);
+
+  // Reads a set of optional statements. These can change the behavior of
+  // a number of resource types (e.g. STRINGTABLE, MENU or DIALOG) if provided
+  // before the main block with the contents of the resource.
+  // Usually, resources use a basic set of optional statements:
+  //    CHARACTERISTICS, LANGUAGE, VERSION
+  // However, DIALOG and DIALOGEX extend this list by the following items:
+  //    CAPTION, CLASS, EXSTYLE, FONT, MENU, STYLE
+  // UseExtendedStatements flag (off by default) allows the parser to read
+  // the additional types of statements.
+  //
+  // Ref (to the list of all optional statements):
+  //    msdn.microsoft.com/en-us/library/windows/desktop/aa381002(v=vs.85).aspx
+  Expected<OptionalStmtList>
+  parseOptionalStatements(bool UseExtendedStatements = false);
+
+  // Read a single optional statement.
+  Expected<std::unique_ptr<OptionalStmt>>
+  parseSingleOptionalStatement(bool UseExtendedStatements = false);
+
+  // Top-level resource parsers.
+  ParseType parseLanguageResource();
+  ParseType parseIconResource();
+  ParseType parseStringTableResource();
+
+  // Optional statement parsers.
+  ParseOptionType parseLanguageStmt();
+  ParseOptionType parseCharacteristicsStmt();
+  ParseOptionType parseVersionStmt();
+
+  // Raises an error. If IsAlreadyRead = false (default), this complains about
+  // the token that couldn't be parsed. If the flag is on, this complains about
+  // the correctly read token that makes no sense (that is, the current parser
+  // state is beyond the erroneous token.)
+  Error getExpectedError(const Twine Message, bool IsAlreadyRead = false);
+
+  std::vector<RCToken> Tokens;
+  LocIter CurLoc;
+  const LocIter End;
+};
+
+} // namespace rc
+} // namespace llvm
+
+#endif