def raw(self, expected: []):
token = CommonToken(type=Token.EOF)
token.text = "<eof-token>"
self.tokens.append(token)
raw_types = self.tokens_to_types(self.tokens)
self.assertEqual(expected, raw_types, "raw tokens")
return self
def makeNumber(parent: ParserRuleContext, n: int):
"""This node represents a virtual ParseTree node that does not come from
the parser but rather is constructed via a rewriting rule. For example,
the expression "a-b" is parsed as "Plus[a, Times[-1, b]]" in exitPlusOp(),
so a node for "-1" needs to be created even though "-1" does not appear as
a token in the token stream.
Note that makeNumber sets the NumberContext's parent but does not add
anything to parent's children."""
# The hierarchy is:
# CommonToken->TerminalNodeImpl->NumberLiteralContext
# ->NumberBaseTenContext->NumberContext->parent
digits_token = CommonToken(type=FoxySheepParser.DIGITS)
digits_token._text = str(n)
number = FoxySheepParser.NumberContext(
None, FoxySheepParser.ExprContext(None, parent=parent))
number_literal = FoxySheepParser.NumberBaseTenContext(
None, FoxySheepParser.NumberLiteralContext(None, parent=number))
number_literal.addTokenNode(digits_token)
addChild(number, number_literal)
return number
def nextToken(self):
# consume pending token if it's there.
if self.pending_token is not None:
t = self.pending_token
self.pending_token = None
else:
t = super(YAMLLexerWrapper, self).nextToken()
if t.type == YAMLLexer.S_INDENT:
num_spaces = len(t.text)
if num_spaces == self.last_num_spaces:
# skip to next token
t = self.nextToken()
elif num_spaces < self.last_num_spaces:
self.pending_token = t
t = CommonToken(type=YAMLParser.DEDENT)
elif num_spaces > self.last_num_spaces:
t = CommonToken(type=YAMLParser.INDENT)
else:
assert False
self.last_num_spaces = num_spaces
elif t.type == Token.EOF:
if self.last_token.type == YAMLLexer.B_BREAK:
if self.last_num_spaces > 0:
self.pending_token = t
t = CommonToken(type=YAMLParser.DEDENT)
self.last_token = t
return t
def createToken(self, type_, text="", length=0):
start = self._tokenStartCharIndex
stop = start + length
t = CommonToken(self._tokenFactorySourcePair, type_,
self.DEFAULT_TOKEN_CHANNEL, start, stop)
t.text = text
return t
def addToken(self, prefix: str, s: str, token_type: TT):
token = CommonToken(type=token_type.value)
token.text = prefix + s
token.column = self.pos
token.line = self.lineNo
self.pos += len(s)
self.builder.append(token)
def create(self, source, type, text, channel, start, stop, line, column):
t = CommonToken(source, type, channel, start, stop)
t.line = line
t.column = column
if text is not None:
t.text = text
elif self.copyText and source[1] is not None:
t.text = source[1].getText(start,stop)
return t
def emit_type(self, token_type, channel=Token.DEFAULT_CHANNEL, text=""):
char_index = self.getCharIndex()
token = CommonToken(self._tokenFactorySourcePair, token_type, channel,
char_index - len(text), char_index)
token.line = self.line
token.column = self.column
token.text = text
self.emitToken(token)
return token
def syntaxError(self, recognizer, offendingSymbol, line, column, msg, e):
t = CommonToken(source=(recognizer, recognizer._input),
type=Token.INVALID_TYPE,
channel=Token.DEFAULT_CHANNEL,
start=recognizer._tokenStartCharIndex,
stop=recognizer._tokenStartCharIndex)
t.line = recognizer._tokenStartLine
t.column = recognizer._tokenStartColumn
recognizer._type = Token.MIN_USER_TOKEN_TYPE
recognizer.emitToken(t)
def commonToken( self, _type, text ):
stop = self.getCharIndex() - 1
if len(text) == 0:
start = stop
else:
start = stop - len(text) + 1
return CommonToken( self._tokenFactorySourcePair, _type, self.DEFAULT_TOKEN_CHANNEL, start, stop )
def common_token(self, _type, text):
from antlr4.Token import CommonToken
stop = self.getCharIndex() - 1
if len(self.text) == 0:
start = stop
else:
start = stop - len(self.text) + 1
return CommonToken(self._tokenFactorySourcePair, _type, Lexer.DEFAULT_TOKEN_CHANNEL, start, stop)
def commonToken(self, type: int, text: str):
stop: int = self.getCharIndex() - 1
start: int = 0
if text:
start = stop - len(text) + 1
ct = CommonToken(self._tokenFactorySourcePair, type,
self.DEFAULT_TOKEN_CHANNEL, start, stop)
return ct
def exitPlusOp(self, ctx: FoxySheepParser.PlusOpContext):
"""PlusOp[expr1,expr2]
We have to treat PlusOp special, because we have to keep the
operators intact, and only plus and minus (not PlusMinus or
MinusPlus) are flat. The situation is complicated by the fact
that Mathematica parses "a-b" as "Plus[a, Times[-1, b]]". We
Rewrite the parse tree, inserting the Times context and
changing BINARYMINUS to BINARYPLUS."""
# If the op isn't Plus or Minus, nothing to do.
if ctx.BINARYMINUS() is None and ctx.BINARYPLUS() is None:
return
# Since ANTLR4 parses this operator as left associative, we only
# need to check the left hand side expr.
rhs = ctx.getChild(2)
# If the operator of the PlusOp is BINARYMINUS, we rewrite the tree as
# "Plus[lhs, Times[-1, rhs]]". Note that if rhs is TIMES, we have to
# keep that TIMES flat.
if ctx.BINARYMINUS() is not None:
# Construct Times, or obtain it from the rhs.
times = None
if isinstance(rhs, FoxySheepParser.TimesContext):
times = rhs
else:
# If rhs is already a times, keep it flat.
times = FoxySheepParser.TimesContext(
None, FoxySheepParser.ExprContext(None))
ctx.children.remove(rhs)
adopt(ctx, times)
adopt(times, rhs)
# Add "-1" as the first child of Times.
addChild(times, makeNumber(times, -1), 0)
# Finally, we have to change operator to BINARYPLUS.
plustoken = CommonToken(type=FoxySheepParser.BINARYPLUS)
plustoken.text = '+'
plus = TerminalNodeImpl(plustoken)
# Replace minus token with plus.
ctx.children[1] = plus
plus.parentCtx = ctx
# Flatten
flatten(ctx)
def __emit_token_type_on_channel(self, token_type: int, channel: int,
text: str) -> None:
char_index: int = self.getCharIndex()
token: CommonToken = CommonToken(self._tokenFactorySourcePair,
token_type, channel,
char_index - len(text),
char_index - 1)
token.line = self.line
token.column = self.column
token.text = text
self.emitToken(token)
def validate_common_token(py_tok:CommonToken, cpp_tok:CommonToken):
assert type(py_tok) == type(cpp_tok)
if py_tok is None:
return
assert py_tok.type == cpp_tok.type
assert py_tok.channel == cpp_tok.channel
assert py_tok.start == cpp_tok.start
assert py_tok.stop == cpp_tok.stop
assert py_tok.tokenIndex == cpp_tok.tokenIndex
assert py_tok.line == cpp_tok.line
assert py_tok.column == cpp_tok.column
assert py_tok.text == cpp_tok.text
assert isinstance(cpp_tok.getInputStream(), InputStream)
def main(argv):
sai = StaticAnalysisInfo()
input = FileStream(argv[1])
lexer = CPP14Lexer(input)
stream = CommonTokenStream(lexer)
parser = CPP14Parser(stream)
tree = parser.translationunit()
tracker = Tracker1(sai, stream)
walker = ParseTreeWalker()
walker.walk(tracker, tree)
print(stream.getText())
stream.tokens[0].text = "test"
tmp_tok = stream.getTokens(0,3)
space = CommonToken()
space.text = ' '
tmp_tok.append(space)
stream.tokens[2:2] = tmp_tok
#help(stream.tokens[0])
#tmp_tok = stream.tokens[2].clone()
#stream.tokens.insert(2, tmp_tok)
#ttt = CommonToken()
#ttt.text = '???'
#stream.tokens.insert(2, ttt)
print(stream.getText())
print("---------")
#lll = stream.getTokens(0,5)
#print(lll)
#print(stream.tokens[0].getTokenSource().text)
#print(tracker.rewrite.getTokenStream().getText())
'''
def insertLeft(self, type, text):
token = CommonToken()
token.type = type.value
token.text = text
token.line = self._dllTokens.value.line
token.tokenIndex = self._dllTokens.value.tokenIndex
inserted = self._dllAll.insert(token, self._dllTokens)
return TokenEditor(inserted, self._dllAll)
def create(self, source, type, text, channel, start, stop, line, column):
t = CommonToken(source, type, channel, start, stop)
t.line = line
t.column = column
if text is not None:
t.text = text
elif self.copyText and source[1] is not None:
t.text = source[1].getText(start, stop)
return t
def nextToken(self):
token = super(PhpBaseLexer, self).nextToken()
if token.type == self.PHPEnd or token.type == self.PHPEndSingleLineComment:
if self._mode == self.SingleLineCommentMode:
# SingleLineCommentMode for such allowed syntax:
# // <?php echo "Hello world"; // comment ?>
self.popMode()
self.popMode()
if token.text == "</script>":
self._phpScript = False
token.type = self.ScriptClose
else:
# Add semicolon to the end of statement if it is absent.
# For example: <?php echo "Hello world" ?>
if self._prevTokenType == self.SemiColon or \
self._prevTokenType == self.Colon or \
self._prevTokenType == self.OpenCurlyBracket or \
self._prevTokenType == self.CloseCurlyBracket:
token = super(PhpBaseLexer, self).nextToken()
else:
token = CommonToken(type=self.SemiColon)
token.text = ';'
elif token.type == self.HtmlName:
self._htmlNameText = token.text
elif token.type == self.HtmlDoubleQuoteString:
if token.text == "php" and self._htmlNameText == "language":
self._phpScript = True
elif self._mode == self.HereDoc:
# Heredoc and Nowdoc syntax support: http://php.net/manual/en/language.types.string.php#language.types.string.syntax.heredoc
if token.type == self.StartHereDoc or token.type == self.StartNowDoc:
self._heredocIdentifier = token.text[3:].strip().replace(
"'", "")
if token.type == self.HereDocText:
if self.CheckHeredocEnd(token.text):
self.popMode()
heredoc_identifier = self.GetHeredocEnd(token.text)
if token.text.strip().endswith(';'):
text = heredoc_identifier + ";\n"
token = CommonToken(type=self.SemiColon)
token.text = text
else:
token = super(PhpBaseLexer, self).nextToken()
token.text = heredoc_identifier + "\n;"
elif self._mode == self.PHP:
if self._channel == self.HIDDEN:
self._prevTokenType = token.type
return token
def insertRight(self, type, text):
token = CommonToken()
token.type = type.value
token.text = text
token.line = self._dllTokens.value.line
token.tokenIndex = self._dllTokens.value.tokenIndex
next = self._dllTokens.next
if next:
inserted = self._dllAll.insert(token, next)
else:
inserted = self._dllAll.append(token)
return TokenEditor(inserted, self._dllAll)
def nextToken(self) -> Token:
if not self.queue.empty():
return self.queue.get(block=False)
else:
next_: Token = super(TurkishLexer, self).nextToken()
if next_.type != 16:
return next_
else:
next2: Token = super(TurkishLexer, self).nextToken()
if next2.type == 19 and next2.text == ".":
abbrev = next_.text + "."
if self.abbreviations and abbrev in self.abbreviations:
common_token: CommonToken = CommonToken(type=1)
common_token.text = abbrev
common_token.start = next_.start
common_token.stop = next2.stop
common_token.tokenIndex = next_.tokenIndex
common_token.column = next_.column
common_token.line = next_.line
return common_token
self.queue.put_nowait(next2)
return next_
def get_injected_token(self, copy_from: CommonToken, token_type_str):
new_token = copy_from.clone()
new_token.text = token_type_str
return new_token
def commonToken(self, type, text, indent=0):
stop = self.getCharIndex() - 1 - indent
start = (stop - len(text) + 1) if text else stop
return CommonToken(self._tokenFactorySourcePair, type,
super().DEFAULT_TOKEN_CHANNEL, start, stop)
def commonToken(self, mytype, text):
stop = self.getCharIndex() - 1
start = stop if len(text) == 0 else (stop - len(text) + 1)
return CommonToken(self._tokenFactorySourcePair, mytype,
super().DEFAULT_TOKEN_CHANNEL, start, stop)
def createThin(self, type, text):
t = CommonToken(type=type)
t.text = text
return t
def createThin(self, type, text):
t = CommonToken(type=type)
t.text = text
return t
def createThin(self, type: int, text: str):
t = CommonToken(type=type)
t.text = text
return t
def deriveToken(self, token: CommonToken, type: int):
res = token.clone()
res.type = type
return res
def createThin(self, type:int, text:str):
t = CommonToken(type=type)
t.text = text
return t
def commonToken(self, _type, text):
stop = self.getCharIndex() - 1
start = stop if text == "" else stop - len(text) + 1
return CommonToken(self._tokenFactorySourcePair, _type, start=start, stop=stop)
