Lexer.cpp

Go to the documentation of this file.

//===- Lexer.cpp - MLIR Lexer Implementation ------------------------------===//
//
// 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
//
//===----------------------------------------------------------------------===//
//
// This file implements the lexer for the MLIR textual form.
//
//===----------------------------------------------------------------------===//
 
#include "Lexer.h"
#include "Token.h"
#include "mlir/AsmParser/CodeComplete.h"
#include "mlir/IR/Diagnostics.h"
#include "mlir/IR/Location.h"
#include "mlir/IR/MLIRContext.h"
#include "mlir/Support/LLVM.h"
#include "llvm/ADT/STLExtras.h"
#include "llvm/ADT/StringExtras.h"
#include "llvm/ADT/StringSwitch.h"
#include "llvm/Support/ErrorHandling.h"
#include "llvm/Support/SourceMgr.h"
#include <cassert>
#include <cctype>
 
using namespace mlir;
 
// Returns true if 'c' is an allowable punctuation character: [$._-]
// Returns false otherwise.
static bool isPunct(char c) {
  return c == '$' || c == '.' || c == '_' || c == '-';
}
 
Lexer::Lexer(const llvm::SourceMgr &sourceMgr, MLIRContext *context,
             AsmParserCodeCompleteContext *codeCompleteContext)
    : sourceMgr(sourceMgr), context(context), codeCompleteLoc(nullptr) {
  auto bufferID = sourceMgr.getMainFileID();
 
  // Check to see if the main buffer contains the last buffer, and if so the
  // last buffer should be used as main file for parsing.
  if (sourceMgr.getNumBuffers() > 1) {
    unsigned lastFileID = sourceMgr.getNumBuffers();
    const llvm::MemoryBuffer *main = sourceMgr.getMemoryBuffer(bufferID);
    const llvm::MemoryBuffer *last = sourceMgr.getMemoryBuffer(lastFileID);
    if (main->getBufferStart() <= last->getBufferStart() &&
        main->getBufferEnd() >= last->getBufferEnd()) {
      bufferID = lastFileID;
    }
  }
  curBuffer = sourceMgr.getMemoryBuffer(bufferID)->getBuffer();
  curPtr = curBuffer.begin();
 
  // Set the code completion location if it was provided.
  if (codeCompleteContext)
    codeCompleteLoc = codeCompleteContext->getCodeCompleteLoc().getPointer();
}
 
/// Encode the specified source location information into an attribute for
/// attachment to the IR.
Location Lexer::getEncodedSourceLocation(SMLoc loc) {
  auto &sourceMgr = getSourceMgr();
  unsigned mainFileID = sourceMgr.getMainFileID();
 
  auto [lineNo, column] = sourceMgr.getLineAndColumn(loc);
  auto *buffer = sourceMgr.getMemoryBuffer(mainFileID);
 
  return FileLineColLoc::get(context, buffer->getBufferIdentifier(), lineNo,
                             column);
}
 
/// emitError - Emit an error message and return an Token::error token.
Token Lexer::emitError(const char *loc, const Twine &message) {
  mlir::emitError(getEncodedSourceLocation(SMLoc::getFromPointer(loc)),
                  message);
  return formToken(Token::error, loc);
}
 
Token Lexer::lexToken() {
  const char *curBufferEnd = curBuffer.end();
  while (true) {
    const char *tokStart = curPtr;
 
    // Check to see if the current token is at the code completion location.
    if (tokStart == codeCompleteLoc)
      return formToken(Token::code_complete, tokStart);
 
    if (tokStart == curBufferEnd)
      return formToken(Token::eof, tokStart);
 
    // Lex the next token.
    switch (*curPtr++) {
    default:
      // Handle bare identifiers.
      if (isalpha(curPtr[-1]))
        return lexBareIdentifierOrKeyword(tokStart);
 
      // Unknown character, emit an error.
      return emitError(tokStart, "unexpected character");
 
    case ' ':
    case '\t':
    case '\n':
    case '\r':
      // Handle whitespace.
      continue;
 
    case '_':
      // Handle bare identifiers.
      return lexBareIdentifierOrKeyword(tokStart);
 
    case 0:
      // A nul character in the source is skipped like whitespace.
      continue;
 
    case ':':
      return formToken(Token::colon, tokStart);
    case ',':
      return formToken(Token::comma, tokStart);
    case '.':
      return lexEllipsis(tokStart);
    case '(':
      return formToken(Token::l_paren, tokStart);
    case ')':
      return formToken(Token::r_paren, tokStart);
    case '{':
      if (peek() == '-' && peek(1) == '#') {
        curPtr += 2;
        return formToken(Token::file_metadata_begin, tokStart);
      }
      return formToken(Token::l_brace, tokStart);
    case '}':
      return formToken(Token::r_brace, tokStart);
    case '[':
      return formToken(Token::l_square, tokStart);
    case ']':
      return formToken(Token::r_square, tokStart);
    case '<':
      return formToken(Token::less, tokStart);
    case '>':
      return formToken(Token::greater, tokStart);
    case '=':
      return formToken(Token::equal, tokStart);
 
    case '+':
      return formToken(Token::plus, tokStart);
    case '*':
      return formToken(Token::star, tokStart);
    case '-':
      if (peek() == '>') {
        ++curPtr;
        return formToken(Token::arrow, tokStart);
      }
      return formToken(Token::minus, tokStart);
 
    case '?':
      return formToken(Token::question, tokStart);
 
    case '|':
      return formToken(Token::vertical_bar, tokStart);
 
    case '/':
      if (peek() == '/') {
        skipComment();
        continue;
      }
      return formToken(Token::slash, tokStart);
 
    case '@':
      return lexAtIdentifier(tokStart);
 
    case '#':
      if (peek() == '-' && peek(1) == '}') {
        curPtr += 2;
        return formToken(Token::file_metadata_end, tokStart);
      }
      [[fallthrough]];
    case '!':
    case '^':
    case '%':
      return lexPrefixedIdentifier(tokStart);
    case '"':
      return lexString(tokStart);
 
    case '0':
    case '1':
    case '2':
    case '3':
    case '4':
    case '5':
    case '6':
    case '7':
    case '8':
    case '9':
      return lexNumber(tokStart);
    }
  }
}
 
/// Lex an '@foo' identifier.
///
///   symbol-ref-id ::= `@` (bare-id | string-literal)
///
Token Lexer::lexAtIdentifier(const char *tokStart) {
  char cur = peek();
 
  // Try to parse a string literal, if present.
  if (cur == '"') {
    ++curPtr;
    Token stringIdentifier = lexString(curPtr);
    if (stringIdentifier.is(Token::error))
      return stringIdentifier;
    return formToken(Token::at_identifier, tokStart);
  }
 
  // Otherwise, these always start with a letter or underscore.
  if (!isalpha(cur) && cur != '_')
    return emitError(curPtr,
                     "@ identifier expected to start with letter or '_'");
 
  ++curPtr;
  while (isalpha(peek()) || isdigit(peek()) || peek() == '_' || peek() == '$' ||
         peek() == '.')
    ++curPtr;
  return formToken(Token::at_identifier, tokStart);
}
 
/// Lex a bare identifier or keyword that starts with a letter.
///
///   bare-id ::= (letter|[_]) (letter|digit|[_$.])*
///   integer-type ::= `[su]?i[1-9][0-9]*`
///
Token Lexer::lexBareIdentifierOrKeyword(const char *tokStart) {
  // Match the rest of the identifier regex: [0-9a-zA-Z_.$]*
  while (isalpha(peek()) || isdigit(peek()) || peek() == '_' || peek() == '$' ||
         peek() == '.')
    ++curPtr;
 
  // Check to see if this identifier is a keyword.
  StringRef spelling(tokStart, curPtr - tokStart);
 
  auto isAllDigit = [](StringRef str) {
    return llvm::all_of(str, llvm::isDigit);
  };
 
  // Check for i123, si456, ui789.
  if ((spelling.size() > 1 && tokStart[0] == 'i' &&
       isAllDigit(spelling.drop_front())) ||
      ((spelling.size() > 2 && tokStart[1] == 'i' &&
        (tokStart[0] == 's' || tokStart[0] == 'u')) &&
       isAllDigit(spelling.drop_front(2))))
    return Token(Token::inttype, spelling);
 
  Token::Kind kind = StringSwitch<Token::Kind>(spelling)
#define TOK_KEYWORD(SPELLING) .Case(#SPELLING, Token::kw_##SPELLING)
#include "TokenKinds.def"
                         .Default(Token::bare_identifier);
 
  return Token(kind, spelling);
}
 
/// Skip a comment line, starting with a '//'.
///
///   TODO: add a regex for comments here and to the spec.
///
void Lexer::skipComment() {
  // Advance over the second '/' in a '//' comment.
  assert(peek() == '/');
  ++curPtr;
 
  const char *curBufferEnd = curBuffer.end();
  while (true) {
    if (curPtr == curBufferEnd)
      return;
 
    switch (*curPtr++) {
    case '\n':
    case '\r':
      // Newline is end of comment.
      return;
    default:
      // Skip over other characters.
      break;
    }
  }
}
 
/// Lex an ellipsis.
///
///   ellipsis ::= '...'
///
Token Lexer::lexEllipsis(const char *tokStart) {
  assert(curPtr[-1] == '.');
 
  if (peek() != '.' || peek(1) != '.')
    return emitError(curPtr, "expected three consecutive dots for an ellipsis");
 
  curPtr += 2;
  return formToken(Token::ellipsis, tokStart);
}
 
/// Lex a number literal.
///
///   integer-literal ::= digit+ | `0x` hex_digit+
///   float-literal ::= [-+]?[0-9]+[.][0-9]*([eE][-+]?[0-9]+)?
///
Token Lexer::lexNumber(const char *tokStart) {
  assert(isdigit(curPtr[-1]));
 
  // Handle the hexadecimal case.
  if (curPtr[-1] == '0' && peek() == 'x') {
    // If we see stuff like 0xi32, this is a literal `0` followed by an
    // identifier `xi32`, stop after `0`.
    if (!isxdigit(peek(1)))
      return formToken(Token::integer, tokStart);
 
    curPtr += 2;
    while (isxdigit(peek()))
      ++curPtr;
 
    return formToken(Token::integer, tokStart);
  }
 
  // Handle the normal decimal case.
  while (isdigit(peek()))
    ++curPtr;
 
  if (peek() != '.')
    return formToken(Token::integer, tokStart);
  ++curPtr;
 
  // Skip over [0-9]*([eE][-+]?[0-9]+)?
  while (isdigit(peek()))
    ++curPtr;
 
  if (peek() == 'e' || peek() == 'E') {
    if (isdigit(static_cast<unsigned char>(peek(1))) ||
        ((peek(1) == '-' || peek(1) == '+') &&
         isdigit(static_cast<unsigned char>(peek(2))))) {
      curPtr += 2;
      while (isdigit(peek()))
        ++curPtr;
    }
  }
  return formToken(Token::floatliteral, tokStart);
}
 
/// Lex an identifier that starts with a prefix followed by suffix-id.
///
///   attribute-id  ::= `#` suffix-id
///   ssa-id        ::= '%' suffix-id
///   block-id      ::= '^' suffix-id
///   type-id       ::= '!' suffix-id
///   suffix-id     ::= digit+ | (letter|id-punct) (letter|id-punct|digit)*
///   id-punct      ::= `$` | `.` | `_` | `-`
///
Token Lexer::lexPrefixedIdentifier(const char *tokStart) {
  Token::Kind kind;
  StringRef errorKind;
  switch (*tokStart) {
  case '#':
    kind = Token::hash_identifier;
    errorKind = "invalid attribute name";
    break;
  case '%':
    kind = Token::percent_identifier;
    errorKind = "invalid SSA name";
    break;
  case '^':
    kind = Token::caret_identifier;
    errorKind = "invalid block name";
    break;
  case '!':
    kind = Token::exclamation_identifier;
    errorKind = "invalid type identifier";
    break;
  default:
    llvm_unreachable("invalid caller");
  }
 
  // Parse suffix-id.
  if (isdigit(peek())) {
    // If suffix-id starts with a digit, the rest must be digits.
    while (isdigit(peek()))
      ++curPtr;
  } else if (isalpha(peek()) || isPunct(peek())) {
    do {
      ++curPtr;
    } while (isalpha(peek()) || isdigit(peek()) || isPunct(peek()));
  } else if (curPtr == codeCompleteLoc) {
    return formToken(Token::code_complete, tokStart);
  } else {
    return emitError(curPtr - 1, errorKind);
  }
 
  // Check for a code completion within the identifier.
  if (codeCompleteLoc && codeCompleteLoc >= tokStart &&
      codeCompleteLoc <= curPtr) {
    return Token(Token::code_complete,
                 StringRef(tokStart, codeCompleteLoc - tokStart));
  }
 
  return formToken(kind, tokStart);
}
 
/// Lex a string literal.
///
///   string-literal ::= '"' [^"\n\f\v\r]* '"'
///
/// TODO: define escaping rules.
Token Lexer::lexString(const char *tokStart) {
  assert(curPtr[-1] == '"');
 
  const char *curBufferEnd = curBuffer.end();
  while (true) {
    // Check to see if there is a code completion location within the string. In
    // these cases we generate a completion location and place the currently
    // lexed string within the token. This allows for the parser to use the
    // partially lexed string when computing the completion results.
    if (curPtr == codeCompleteLoc)
      return formToken(Token::code_complete, tokStart);
 
    if (curPtr == curBufferEnd)
      return emitError(curPtr, "expected '\"' in string literal");
 
    // A nul character in the string is included like any other character.
    switch (*curPtr++) {
    case '"':
      return formToken(Token::string, tokStart);
    case '\n':
    case '\v':
    case '\f':
      return emitError(curPtr - 1, "expected '\"' in string literal");
    case '\\':
      // Handle explicitly a few escapes.
      if (peek() == '"' || peek() == '\\' || peek() == 'n' || peek() == 't')
        ++curPtr;
      else if (llvm::isHexDigit(peek()) && llvm::isHexDigit(peek(1)))
        // Support \xx for two hex digits.
        curPtr += 2;
      else
        return emitError(curPtr - 1, "unknown escape in string literal");
      continue;
 
    default:
      continue;
    }
  }
}