def generate_parse_tree(source):
"""Generates an ANTLR parse tree from a piece of source code."""
input_stream = FileStream(source)
lexer = JavaLexer(input_stream)
stream = CommonTokenStream(lexer)
parser = JavaParser(stream)
tree = parser.compilationUnit()
return tree
def parse(self, file_content):
inputS = InputStream.InputStream(file_content)
lexer = JavaLexer(inputS)
stream = CommonTokenStream(lexer)
parser = JavaParser(stream)
tree = parser.compilationUnit()
self.lookupTable = dict()
# tree.getText()
return tree
def parse_file(path):
file = open(path, 'r').read()
codeStream = InputStream(file)
lexer = JavaLexer(codeStream)
# First lexing way
tokens = CommonTokenStream(lexer)
parser = JavaParser(tokens)
parser.buildParseTrees = True
return parser
def main(argv):
if len(argv)>1:
input = FileStream(argv[1])
else:
input = InputStream(sample)
lexer = JavaLexer(input)
stream = CommonTokenStream(lexer)
parser = JavaParser(stream)
tree = parser.compilationUnit()
# print(Trees.toStringTree(tree, None, parser))
print "Grammar %s has %d rules, %d tokens" % (parser.grammarFileName, len(parser.ruleNames), len(lexer.ruleNames))
print ', '.join(CollectTokenFeatures.feature_names)
token_features = CollectTokenFeatures(stream)
walker = ParseTreeWalker()
walker.walk(token_features, tree)
print token_features.feature_names
def extract_data(code):
"""
Parse a code string and collect features with a CollectTokenFeatures.
Returns (tokens:list, inject_newlines:boolean[], indent:int[],
whitespace:int[], features:list<object[]>)
"""
input = InputStream(code)
lexer = JavaLexer(input)
stream = CommonTokenStream(lexer)
parser = JavaParser(stream)
tree = parser.compilationUnit()
collector = CollectTokenFeatures(stream)
walker = ParseTreeWalker()
walker.walk(collector, tree)
return (stream.tokens, collector.inject_newlines,
collector.indent, collector.whitespace, collector.features)
def main(argv):
if len(argv) > 1:
input = FileStream(argv[1])
else:
input = InputStream(sample)
lexer = JavaLexer(input)
stream = CommonTokenStream(lexer)
parser = JavaParser(stream)
tree = parser.compilationUnit()
# print(Trees.toStringTree(tree, None, parser))
print "Grammar %s has %d rules, %d tokens" % (
parser.grammarFileName, len(parser.ruleNames), len(lexer.ruleNames))
print ', '.join(CollectTokenFeatures.feature_names)
token_features = CollectTokenFeatures(stream)
walker = ParseTreeWalker()
walker.walk(token_features, tree)
print token_features.feature_names
def parse_for_methods(repo_path):
"""[this is the main mehtod where the actual antlr4 parsing happens]
Arguments:
repo_path {[str]} -- [holds repository path and only used for logging purpose]
"""
try:
istream = FileStream(repo_path, encoding='utf-8')
lexer = JavaLexer(istream)
stream = CommonTokenStream(lexer)
parser = JavaParser(stream)
tree = parser.compilationUnit()
# _using PatternListener to walk only through class and method Declarations
walker = ParseTreeWalker()
walker.walk(PatternListener(), tree)
except Exception as e:
print("Unexpected error: " + repo_path + " " + str(e))
def main(argv):
input = FileStream(argv[1])
lexer = JavaLexer(input)
stream = CommonTokenStream(lexer)
parser = JavaParser(stream)
tree = parser.compilationUnit()
walker = ParseTreeWalker()
listener = CompilationUnitListener()
walker.walk(listener, tree)
Classes = listener.Classes
Interfaces = listener.Interfaces
symbolTable = {}
error = InitalizeSymbolTable(symbolTable, Classes, Interfaces)
if (error < 0):
return False
graph = UMLGraph()
for Interface in Interfaces:
graph.addInterface(Interface)
for Class in Classes:
graph.addClass(Class)
if Class.Extends in symbolTable:
graph.addExtendsRelation(Class.ClassName, Class.Extends)
for interface in Class.ImplementList:
if interface["Type"] in symbolTable:
graph.addImplementsRelation(Class.ClassName, interface["Type"])
print(str(graph))
#if __name__ == '__main__':
#main(sys.argv)
#MainClass = Java2UML()
#result = MainClass.JavaCode2UML(sys.argv[1])
#print(result)
def standardizeJava(javaFiles: List[HookFile]) -> List[StandardizedFile]:
standardizedFiles: List[StandardizedFile] = []
for javaFile in javaFiles:
lexer = JavaLexer(InputStream(javaFile.fileData))
tokenStream = CommonTokenStream(lexer)
parser = JavaParser(tokenStream)
tree = parser.compilationUnit()
visitor = JavaParserVisitor()
visitor.visit(tree)
var, funcs, classes, types = JavaSyntax.getLists()
words = javaFile.fileData.split(' ')
for i in range(len(words)):
for v in var:
words[i] = replaceIfContains(v, words[i], "V")
for f in funcs:
words[i] = replaceIfContains(f, words[i], "F")
for c in classes:
words[i] = replaceIfContains(c, words[i], "C")
for t in types:
words[i] = replaceIfContains(t, words[i], "T")
stripped = ''.join(words).replace('\t', '').lower()
standardizedFiles.append(StandardizedFile(javaFile, stripped))
JavaSyntax.clearLists()
return standardizedFiles
def JavaCode2UML(self, string):
#input = FileStream(string)
#lexer = JavaLexer(input)
lexer = JavaLexer(InputStream(string))
stream = CommonTokenStream(lexer)
parser = JavaParser(stream)
tree = parser.compilationUnit()
walker = ParseTreeWalker()
listener = CompilationUnitListener()
walker.walk(listener, tree)
Classes = listener.Classes
Interfaces = listener.Interfaces
symbolTable = {}
error = InitalizeSymbolTable(symbolTable, Classes, Interfaces)
if (error < 0):
return False
graph = UMLGraph()
for Interface in Interfaces:
graph.addInterface(Interface)
for Class in Classes:
graph.addClass(Class)
if Class.Extends in symbolTable:
graph.addExtendsRelation(Class.ClassName, Class.Extends)
for interface in Class.ImplementList:
if interface["Type"] in symbolTable:
graph.addImplementsRelation(Class.ClassName,
interface["Type"])
return str(graph)
def format_code(newline_forest, indent_forest, whitespace_forest, vec, code):
"""
Tokenize code and then, one token at a time, predict newline or not.
Do prediction of newline on the fly, adjusting token line/column.
"""
# tokenize and wack location info in tokens
input = InputStream(code)
lexer = JavaLexer(input)
stream = CommonTokenStream(lexer)
stream.fill()
# wipe out token location information in sample
for t in stream.tokens:
t.line = 0
t.column = 0
# parse to get parse tree
parser = JavaParser(stream)
tree = parser.compilationUnit()
# compute feature vector for each token and adjust line/column as we walk tree
collector = ProcessTokens(newline_forest, indent_forest, whitespace_forest, vec, stream)
walker = ParseTreeWalker()
walker.walk(collector, tree)
__author__ = 'jszheng'
import sys
from antlr4 import *
from antlr4.InputStream import InputStream
from JavaLexer import JavaLexer
from JavaParser import JavaParser
from ExtractInterfaceListener import ExtractInterfaceListener
if __name__ == '__main__':
if len(sys.argv) > 1:
input_stream = FileStream(sys.argv[1])
else:
input_stream = InputStream(sys.stdin.readline())
lexer = JavaLexer(input_stream)
token_stream = CommonTokenStream(lexer)
parser = JavaParser(token_stream)
tree = parser.compilationUnit()
listener = ExtractInterfaceListener(parser)
walker = ParseTreeWalker()
walker.walk(listener, tree)
def parse_java_deep(file):
Command = "ctags -f - --kinds-java=* --fields=neK " + file
global delimiter
delimiter = "\r\0?\r?\0\r"
try:
astString = subprocess.check_output(Command,
stderr=subprocess.STDOUT,
shell=True).decode()
except subprocess.CalledProcessError as e:
print("Parser Error:", e)
astString = ""
f = open(file, 'r')
lines = f.readlines()
methodList = astString.split('\n')
method = re.compile(r'(method)')
number = re.compile(r'(\d+)')
funcBody = re.compile(r'{([\S\s]*)}')
string = ""
funcId = 1
methodInstanceList = []
for i in methodList:
elemList = re.sub(r'[\t\s ]{2,}', '', i)
elemList = elemList.split("\t")
methodInstance = function(file)
methodInstance.funcBody = ''
if i != '' and method.match(elemList[3]) and len(elemList) >= 6:
methodInstance.name = elemList[0]
methodInstance.parentFile = elemList[1]
methodInstance.lines = (int(number.search(elemList[4]).group(0)),
int(number.search(elemList[5]).group(0)))
methodInstance.parentNumLoc = len(lines)
string = ""
string = string.join(lines[methodInstance.lines[0] -
1:methodInstance.lines[1]])
if funcBody.search(string):
methodInstance.funcBody = methodInstance.funcBody + funcBody.search(
string).group(0)
lexer = JavaLexer(InputStream(string))
tokens = CommonTokenStream(lexer)
parser = JavaParser(tokens)
tree = parser.memberDeclaration()
walker = ParseTreeWalker()
listener = JavaParserListener()
listener.variables = []
listener.parameters = []
listener.dataTypes = []
listener.methodCalls = []
walker.walk(listener, tree)
methodInstance.variableList = listener.variables
methodInstance.dataTypeList = listener.dataTypes
methodInstance.funcCalleeList = listener.methodCalls
methodInstance.parameterList = listener.parameters
else:
methodInstance.funcBody = " "
methodInstance.funcId = funcId
funcId += 1
methodInstanceList.append(methodInstance)
print(methodInstance.funcBody)
# print(tree.toStringTree(recog=parser))
#Not finished
return methodInstanceList
import sys import antlr3 from JavaLexer import JavaLexer from JavaParser import JavaParser cStream = antlr3.StringStream(open(sys.argv[1]).read()) lexer = JavaLexer(cStream) tStream = antlr3.CommonTokenStream(lexer) parser = JavaParser(tStream) parser.compilationUnit() print "finished parsing OK"
