def test_simple_compose (self):
testpath = os.path.split(__file__)[0]
gobj1 = PyPgen.PyPgen().handleStart(
PgenParser.parseFile(os.path.join(testpath, 'test.pgen')))
gobj2 = PyPgen.PyPgen().handleStart(
PgenParser.parseFile(os.path.join(testpath, 'ext.pgen')))
gobj3 = nfa.compose_nfas(gobj1, gobj2)
self.assertTrue(gobj3)
self.assertEqual(len(gobj3[0]), len(gobj1[0]))
self.assertNotEqual(gobj3[0], gobj1[0])
def main (fileName = None):
"""main
Read a pgen input grammar from the given file name. If no file name is
given, read the grammar from stdin. Print a XML document for the
resulting grammar model translation.
"""
if None == fileName:
parseTree = PgenParser.parseString(sys.stdin.read())
else:
parseTree = PgenParser.parseFile(fileName)
grammarFactoryClass = BasilGrammarModel.getModelFactory()
grammarFactory = grammarFactoryClass()
internalizer = PgenInternalizer(grammarFactory)
model = internalizer(parseTree)
print grammarFactory.externalizeXML(model)
def main ():
import sys, PgenParser, pprint
# ____________________________________________________________
# Generate a test parser
if len(sys.argv) > 1:
grammarFile = sys.argv[1]
else:
grammarFile = "tests/test.pgen"
grammarST = PgenParser.parseFile(grammarFile)
parser = buildParser(grammarST)
pprint.pprint(parser.toTuple())
# ____________________________________________________________
# Parse some input
if len(sys.argv) > 2:
inputFile = sys.argv[2]
fileObj = open(inputFile)
else:
inputFile = "<stdin>"
fileObj = sys.stdin
tokenizer = StdTokenizer.StdTokenizer.tokenize(fileObj)
parser.setStart(257)
parseTree = parser.parseTokens(tokenizer)
fileObj.close()
# ____________________________________________________________
# Show the result
from basil.visuals.TreeBox import showTree
showTree(parseTree).mainloop()
def main (*args):
"""Main routine for the basil.parsing.nfa module.
Composes one or more grammars (represented as a collections of
pgen non-deterministic finite state automata, or NFA's) into a single
NFA. By default, main() pretty-prints the resulting NFA to stdout.
Flags:
-n <symbol> Output a Graphviz DOT file for the NFA that recognizes the
given symbol (can specify more than one).
-o <filename> Output the pretty-printed NFA collection to the given file.
-q Quiet. Suppress output of the resulting NFA. Overrides -o.
Non-flag arguments should be the file names of a pgen input files
that will be composed into a single NFA collection.
"""
import getopt
import pprint
from basil.parsing import PgenParser, PyPgen
nfas_to_output = []
output_file = None
quiet = False
opts, args = getopt.getopt(args, "n:o:q")
for (opt_flag, opt_arg) in opts:
if opt_flag == "-q":
quiet = True
elif opt_flag == "-o":
output_file = opt_arg
elif opt_flag == "-n":
nfas_to_output.append(opt_arg)
if len(args) < 1:
print "Usage: nfa.py [flags] <pgenfile1> [<pgenfile2> ...]"
return
index = 0
nfas_tups = []
out_nfas = [[],[]]
for arg in args:
grammar_st = PgenParser.parseFile(arg)
crnt_nfas = PyPgen.PyPgen().handleStart(grammar_st)
nfas_tups.append((crnt_nfas, "g%d_" % index))
out_nfas = compose_nfas(out_nfas, crnt_nfas)
index += 1
nfas_tups.append((out_nfas, "res"))
if not quiet:
out_nfas_str = pprint.pformat(out_nfas)
if output_file:
open(output_file, "w").write(out_nfas_str)
else:
print(out_nfas_str)
if nfas_to_output:
for nfa_name in nfas_to_output:
nfas = [(find_nfa(nfas, nfa_name), nfas, nfas_pre)
for nfas, nfas_pre in nfas_tups]
nfas_code = ["" if nfa is None else
nfa_to_dot(nfa, nfas[1], nfas_pre)
for (nfa, nfas, nfas_pre) in nfas]
dotfile = open(nfa_name + ".dot", "w")
dotfile.write("digraph G {\n%s}\n" % ("\n".join(nfas_code)))
dotfile.close()
def pgen_to_grammar_obj (source):
pgen = PyPgen.PyPgen()
nfa_grammar = pgen.handleStart(PgenParser.parseString(source))
dfa_grammar = pgen.generateDfaGrammar(nfa_grammar)
pgen.translateLabels(dfa_grammar)
pgen.generateFirstSets(dfa_grammar)
dfa_grammar = DFAParser.addAccelerators(dfa_grammar)
return dfa_grammar
def pgen_to_grammar_obj (source):
'''XXX Stolen from PyCon 2010 sprint sandbox. Move into Basil proper.'''
pgen = PyPgen.PyPgen()
nfa_grammar = pgen.handleStart(PgenParser.parseString(source))
dfa_grammar = pgen.generateDfaGrammar(nfa_grammar)
pgen.translateLabels(dfa_grammar)
pgen.generateFirstSets(dfa_grammar)
dfa_grammar = DFAParser.addAccelerators(dfa_grammar)
return dfa_grammar
def internalizePGEN (self, stream):
"""BasilGrammarModelFactory.internalizePgen
"""
from basil.parsing import PgenParser
from basil.models.grammar.InternalizePgen import PgenInternalizer
if type(stream) == types.StringType:
text = stream
else:
text = stream.read()
parse = PgenParser.parseString(text)
internalizer = PgenInternalizer(self)
return internalizer(parse)
def setUp (self):
# Parse the MyFront grammar, create a set of automata for it (like
# pgen does), and then convert the automata to generators for the
# treepoline.
grammar_st = PgenParser.parseFile(basil.lang.python.__path__[0] +
"/python26/Grammar")
grammar_obj = PyPgen.buildParser(grammar_st)
grammar_obj.setStart(grammar_obj.stringToSymbolMap()['file_input'])
self.parser1 = grammar_obj
gram_tup0 = grammar_obj.toTuple()
gram_tup1 = DFAParser.addAccelerators(gram_tup0)
handlers = trampoline.pgen_grammar_to_handlers(gram_tup1, {})
# Override the start special nonterminal to just do what it is
# supposed to:
def parse_start (instream, outtree):
yield 'file_input'
handlers['start'] = parse_start
self.handlers = handlers
def main (*args):
"""main() - Unit test routine for the PushdownAutomaton module."""
from basil.lang.python import DFAParser
from basil.parsing import PgenParser, PyPgen
import sys, getopt
# ____________________________________________________________
opts, args = getopt.getopt(args, "o:")
outFile = sys.stdout
for optFlag, optArg in opts:
if optFlag == "-o":
outFile = open(optArg, "w")
argc = len(args)
if argc > 1:
print "Usage:\n\tPushdownAutomaton.py [opts] <file.pgen>\n"
sys.exit(-1)
elif argc == 1:
srcFile = args[0]
else:
srcFile = "./tests/test.pgen"
grammarST = PgenParser.parseFile(srcFile)
parser = PyPgen.buildParser(grammarST)
grammarTup = parser.grammarObj
# ____________________________________________________________
# Round trip test
myGrammarObj = PushdownGrammar(grammarTup)
myGrammarObj.compareToTuple(grammarTup)
grammarTup2 = myGrammarObj.toTuple()
myGrammarObj.compareToTuple(grammarTup2)
# Now with accelerators...
grammarTup3 = DFAParser.addAccelerators(grammarTup)
myGrammarObj2 = PushdownGrammar(grammarTup3)
myGrammarObj2.compareToTuple(grammarTup3)
grammarTup4 = myGrammarObj2.toTuple()
myGrammarObj2.compareToTuple(grammarTup4)
# ____________________________________________________________
# Output tests.
outFile.write("%s\n" % myGrammarObj2.toString("graphviz"))
if outFile != sys.stdout:
outFile.close()
from basil.lang.mython.MyFrontExceptions import MyFrontSyntaxError
# ______________________________________________________________________
# Module data
MY_GRAMMAR_EXT = """
compound_stmt: quotedef
quotedef: 'quote' ['[' expr ']'] [NAME] qsuite
qsuite: ':' (QUOTED NEWLINE | NEWLINE QUOTED)
"""
MY_START_SYMBOL = 'file_input'
pgen = PyPgen.PyPgen()
py_grammar_path = os.path.split(basil.lang.python.__file__)[0]
py_nfa_grammar = pgen.handleStart(PgenParser.parseFile(
os.path.join(py_grammar_path, 'python26/Grammar')))
ext_nfa_grammar = pgen.handleStart(PgenParser.parseString(MY_GRAMMAR_EXT))
my_nfa_grammar = nfa.compose_nfas(py_nfa_grammar, ext_nfa_grammar)
my_grammar0 = pgen.generateDfaGrammar(my_nfa_grammar, MY_START_SYMBOL)
pgen.translateLabels(my_grammar0, {'QUOTED' : mylexer.QUOTED})
pgen.generateFirstSets(my_grammar0)
my_grammar0[0] = map(tuple, my_grammar0[0])
my_grammar0 = tuple(my_grammar0)
my_grammar = DFAParser.addAccelerators(my_grammar0)
del my_grammar0
__DEBUG__ = False
if __DEBUG__:
import pprint
