def try_parsing(self, parser: ParserPython, text_to_parse: str):
# noinspection PyBroadException
try:
parser.parse(text_to_parse)
except:
self.fail("Could not parse {0} using the {1} parser!".format(
text_to_parse, parser.__class__))
def convert_trog(inf: str, outf: TextIO) -> None:
pp = ParserPython(trogfile, skipws=False)
try:
tree = pp.parse_file(inf)
print(tree)
except Exception as e:
print(e)
def test_memoization_nomatch(capsys):
'''
Test that already failed match is found in the cache on
subsequent matches.
'''
def grammar():
return [(rule1, ruleb), [rule1, rulec]]
def rule1():
return rulea, ruleb
def rulea():
return "a"
def ruleb():
return "b"
def rulec():
return "c"
parser = ParserPython(grammar, memoization=True, debug=True)
parse_tree = parser.parse("c")
assert "Cache hit for [rule1=Sequence, 0] = '0'" in capsys.readouterr()[0]
assert parser.cache_hits == 1
assert parser.cache_misses == 4
def test_with_some_words_2(self):
parser = ParserPython(document, skipws=False)
text = f"{self.words1}\n\n"
parsed = parser.parse(text)
# print('\n: parsed') ; pp(parsed)
expect = NonTerminal(document(), [
NonTerminal(body(), [
NonTerminal(element(), [
Terminal(words(), 0, self.words1),
]),
NonTerminal(element(), [
Terminal(newline(), 0, '\n'),
]),
NonTerminal(element(), [
Terminal(blank_line(), 0, '\n'),
]),
]),
Terminal(EOF(), 0, ''),
])
# print('\n: expect') ; pp(expect)
assert parsed == expect, (
f"text = '{text}' :\n"
f"[expect]\n{pp_str(expect)}\n[parsed]\n{pp_str(parsed)}")
def test_direct_rule_call():
'''
Test regression where in direct rule call semantic action is
erroneously attached to both caller and callee.
'''
def grammar(): return rule1, rule2
def rule1(): return "a"
def rule2(): return rule1
call_count = [0]
class DummySemAction(SemanticAction):
def first_pass(self, parser, node, nodes):
call_count[0] += 1
return SemanticAction.first_pass(self, parser, node, nodes)
# Sem action is attached to rule2 only but
# this bug will attach it to rule1 also resulting in
# wrong call count.
rule2.sem = DummySemAction()
parser = ParserPython(grammar)
parse_tree = parser.parse("aa")
parser.getASG()
assert call_count[0] == 1, "Semantic action should be called once!"
def test_02_text_line_pair (self) :
def body():
return OneOrMore ( [ text_line ], rule_name='body' )
def document():
return Sequence( ( body, EOF ), rule_name='document' )
# print('\n: document') ; pp(document())
parser = ParserPython( document, skipws=False )
text = self.line1 + self.line2
parsed = parser.parse(text)
# print('\n: parsed') ; pp(parsed)
#
# print('\n: flatten') ; pp(flatten(parsed))
p_newline = Terminal(newline(), 0, '\n')
p_l1_words = Terminal(words(), 0, self.line1[:-1])
p_l1_text_line = NonTerminal(text_line(), [ p_l1_words, p_newline ])
p_l2_words = Terminal(words(), 0, self.line2[:-1])
p_l2_text_line = NonTerminal(text_line(), [ p_l2_words, p_newline ])
p_body = NonTerminal(body(), [ p_l1_text_line, p_l2_text_line ])
p_eof = Terminal(EOF(), 0, '')
expect = NonTerminal(document(), [p_body, p_eof] )
# print('\n: expect') ; pp(expect)
assert parsed == expect, ( f"text = '{text}' :\n"
f"[expect]\n{pp_str(expect)}\n[parsed]\n{pp_str(parsed)}" )
def test_05_paragraph_multiple (self) :
def body():
return OneOrMore ( OrderedChoice( [ paragraph, newline ] ), rule_name='body' )
def document():
return Sequence( ( body, EOF ), rule_name='document' )
# print('\n: document') ; pp(document())
parser = ParserPython( document, skipws=False )
text = self.line1 + self.line2 + '\n'
text = text * 3
parsed = parser.parse(text)
# print('\n: parsed') ; pp(parsed)
p_newline = Terminal(newline(), 0, '\n')
p_l1_words = Terminal(words(), 0, self.line1[:-1])
p_l1_text_line = NonTerminal(text_line(), [ p_l1_words, p_newline ])
p_l2_words = Terminal(words(), 0, self.line2[:-1])
p_l2_text_line = NonTerminal(text_line(), [ p_l2_words, p_newline ])
p_paragraph = NonTerminal(paragraph(), [ p_l1_text_line, p_l2_text_line ])
p_body = NonTerminal(body(), [
p_paragraph, p_newline,
p_paragraph, p_newline,
p_paragraph, p_newline,
])
p_eof = Terminal(EOF(), 0, '')
expect = NonTerminal(document(), [ p_body, p_eof ] )
# print('\n: expect') ; pp(expect)
assert parsed == expect, ( f"text = '{text}' :\n"
f"[expect]\n{pp_str(expect)}\n[parsed]\n{pp_str(parsed)}" )
def test_memoization_positive(capsys):
'''
Test that already matched rule is found in the cache on
subsequent matches.
Args:
capsys - pytest fixture for output capture
'''
def grammar(): return [(rule1, ruleb), (rule1, rulec)]
def rule1(): return rulea, ruleb
def rulea(): return "a"
def ruleb(): return "b"
def rulec(): return "c"
parser = ParserPython(grammar, memoization=True, debug=True)
# Parse input where a rule1 will match but ruleb will fail
# Second sequence will try rule1 again on the same location
# and result should be found in the cache.
parse_tree = parser.parse("a b c")
# Assert that cached result is used
assert "Cache hit" in capsys.readouterr()[0]
assert parser.cache_hits == 1
assert parser.cache_misses == 4
def test_memoization_positive(capsys):
'''
Test that already matched rule is found in the cache on
subsequent matches.
Args:
capsys - pytest fixture for output capture
'''
def grammar():
return [(rule1, ruleb), (rule1, rulec)]
def rule1():
return rulea, ruleb
def rulea():
return "a"
def ruleb():
return "b"
def rulec():
return "c"
parser = ParserPython(grammar, memoization=True, debug=True)
# Parse input where a rule1 will match but ruleb will fail
# Second sequence will try rule1 again on the same location
# and result should be found in the cache.
parse_tree = parser.parse("a b c")
# Assert that cached result is used
assert "Cache hit" in capsys.readouterr()[0]
assert parser.cache_hits == 1
assert parser.cache_misses == 4
def parse_string(self, src, grammar=program, filename=None):
oldsrcs = self.input_sources
self.context.optimization_level = self.optimization_level
self.input_sources = src
parser = ParserPython(
grammar,
comment_def=comment,
skipws=True,
reduce_tree=False,
memoization=True,
debug=False,
)
self.context.parsers.append(parser)
self.context.filenames.append(filename)
try:
parse_tree = parser.parse(self.input_sources)
visitor = MuvVisitor(debug=False)
visitor.muvparser = self
parse_tree = visit_parse_tree(parse_tree, visitor)
out = parse_tree.generate_code(self.context)
if self.error_found:
return False
if len(self.context.filenames) == 1:
if self.context.filenames[-1]:
filetext = " from {0}".format(
self.context.filenames[-1]
)
else:
filetext = ''
self.output = (
"( Generated{0} by the MUV compiler. )\n"
"( https://github.com/revarbat/pymuv )\n"
"{1}\n"
).format(filetext, self.output)
self.output += out
if not self.error_found and len(self.context.filenames) == 1:
if self.wrapper_program:
self.output = (
"@program {0}\n"
"1 99999 d\n"
"1 i\n"
"{1}\n"
".\n"
"c\n"
"q\n"
).format(self.wrapper_program, self.output)
return True
except MuvError as e:
line, col = parser.pos_to_linecol(e.position)
self.print_error(filename, line, col, str(e))
return False
except NoMatch as e:
line, col = parser.pos_to_linecol(e.position)
expected = self.simplify_parse_error(e)
self.print_error(filename, line, col, "Expected %s" % expected)
return False
finally:
self.input_sources = oldsrcs
self.context.parsers.pop()
self.context.filenames.pop()
def parse_tree():
def grammar():
return ("first", "second", "third")
parser = ParserPython(grammar)
return parser.parse(" first \n\n second third")
def parse(file, enc):
with codecs.open(file, "r", encoding=enc) as opened_file:
opened_file_content = opened_file.read()
parser = ParserPython(segnetics_file, reduce_tree=True)
parse_tree = visit_parse_tree(parser.parse(opened_file_content),
SegneticsVisitor())
return parse_tree
def parse_tree():
def grammar(): return ("first", "second", "third")
parser = ParserPython(grammar)
return parser.parse(" first \n\n second third")
def setUp(self):
# first get defaults, should all be False for boolean flags
super().setUp()
global parse_debug, record, analyzing
self.parse_debug = parse_debug
self.record = record
self.analyzing = analyzing
# quiet, no parse trees displayeda
# self.debug = False
# show parse tree for pass >= self.debug
# self.debug = 2
# Show text being parsed
# self.show = True
# and again, to apply behavior per altered settings
super().setUp()
self.grammar = document
self.parser = ParserPython ( language_def = self.grammar,
reduce_tree = False,
debug = self.parse_debug, )
if self.record :
write_scratch ( _clean = True )
def main(text):
from arpeggio import Optional, ZeroOrMore, OneOrMore, EOF
from arpeggio import RegExMatch as _
def number():
return _(r'\d*\.\d*|\d+')
def factor():
return Optional(["+", "-"]), [number, ("(", expression, ")")]
def term():
return factor, ZeroOrMore(["*", "/"], factor)
def expression():
return term, ZeroOrMore(["+", "-"], term)
def calc():
return OneOrMore(expression), EOF
from arpeggio import ParserPython
parser = ParserPython(calc) # calc is the root rule of your grammar
# Use param debug=True for verbose debugging
# messages and grammar and parse tree visualization
# using graphviz and dot
parse_tree = parser.parse(text)
pass
def test_memoization_nomatch(capsys: FixtureRequest) -> None:
"""
Test that already failed match is found in the cache on
subsequent matches.
"""
def grammar() -> List[Any]:
return [(rule1, ruleb), [rule1, rulec]]
def rule1() -> Tuple[Any, ...]:
return rulea, ruleb
def rulea() -> str:
return "a"
def ruleb() -> str:
return "b"
def rulec() -> str:
return "c"
parser = ParserPython(grammar, memoization=True, debug=True)
parse_tree = parser.parse("c")
assert "Cache hit for [rule1=Sequence, 0] = '0'" in capsys.readouterr()[0]
assert parser.cache_hits == 1
assert parser.cache_misses == 4
def test_combine_python():
# This will result in NonTerminal node
def root():
return my_rule(), "."
# This will result in Terminal node
def my_rule():
return Combine(ZeroOrMore("a"), OneOrMore("b"))
parser = ParserPython(root)
input1 = "abbb."
# Whitespaces are preserved in lexical rules so the following input
# should not be recognized.
input2 = "a b bb."
ptree1 = parser.parse(input1)
with pytest.raises(NoMatch):
parser.parse(input2)
assert isinstance(ptree1, NonTerminal)
assert isinstance(ptree1[0], Terminal)
assert ptree1[0].value == "abbb"
def test_direct_rule_call() -> None:
"""
Test regression where in direct rule call semantic action is
erroneously attached to both caller and callee.
"""
def grammar(): return rule1, rule2
def rule1(): return "a"
def rule2(): return rule1
call_count = [0]
class DummySemAction(SemanticAction):
def first_pass(self, parser, node, nodes):
call_count[0] += 1
return SemanticAction.first_pass(self, parser, node, nodes)
# Sem action is attached to rule2 only but
# this bug will attach it to rule1 also resulting in
# wrong call count.
rule2.sem = DummySemAction() # type: ignore
parser = ParserPython(grammar)
parse_tree = parser.parse("aa")
parser.getASG()
assert call_count[0] == 1, "Semantic action should be called once!"
def parse(content: str) -> 'Atom':
from foil.language.grammar import atom
from foil.language.grammar import comment
parser = ParserPython(atom, comment_def=comment)
parse_tree = parser.parse(content)
return visit_parse_tree(parse_tree, FoilVisitor())
def parse(content: str) -> 'Literal':
from foil.language.grammar import literal
from foil.language.grammar import comment
parser = ParserPython(literal, comment_def=comment)
parse_tree = parser.parse(content)
return visit_parse_tree(parse_tree, FoilVisitor())
def test_nondeterministic_unordered_group():
def root():
return 'word1', UnorderedGroup(some_rule, 'word2', some_rule), EOF
def some_rule():
return Optional('word2'), Optional('word3')
content = '''word1 word2 '''
# If the 'word2' from unordered group in the `root` rule matches first
# the input parses, else it fails.
# We repeat parser construction and parsing many times to check
# if it fails every time. The current fix will iterate in order from left
# to right and repeat matching until all rules in a unordered group
# succeeds.
fail = 0
success = 0
for _ in range(100):
try:
parser = ParserPython(root)
parser.parse(content)
success += 1
except NoMatch:
fail += 1
assert fail == 100
def test_reduce_tree():
input = "34 a 3 3 b 3 b"
parser = ParserPython(grammar, reduce_tree=False)
result = parser.parse(input)
# PTDOTExporter().exportFile(result, 'test_reduce_tree_pt.dot')
assert result[0].rule_name == 'first'
assert isinstance(result[0], NonTerminal)
assert result[3].rule_name == 'first'
assert result[0][0].rule_name == 'fourth'
# Check reduction for direct OrderedChoice
assert result[2][0].rule_name == 'third'
parser = ParserPython(grammar, reduce_tree=True)
result = parser.parse(input)
# PTDOTExporter().exportFile(result, 'test_reduce_tree_pt.dot')
assert result[0].rule_name == 'fourth'
assert isinstance(result[0], Terminal)
assert result[3].rule_name == 'fourth'
# Check reduction for direct OrderedChoice
assert result[2][0].rule_name == 'third_str'
def main(argv):
# Parsing
#different alg relation next to each other i.e a*|b require brackets (a*)|b
parser = ParserPython(formula) #, debug=True) #, reduce_tree = True)
parse_tree = parser.parse(argv)
result = visit_parse_tree(parse_tree, formVisitor())
result.tostr()
return result
def load_from_str(self, content: str) -> 'Builder':
parser = ParserPython(cypher, comment_def=comment)
parsed = parser.parse(content)
visited = visit_parse_tree(parsed, KnowledgeVisitor())
base = RuleBase(visited['data'])
self.load_from_base(base)
return self
def language_from_str(language_def, metamodel):
"""
Constructs parser and initializes metamodel from language description
given in textX language.
Args:
language_def (str): A language description in textX.
metamodel (TextXMetaModel): A metamodel to initialize.
Returns:
Parser for the new language.
"""
if metamodel.debug:
metamodel.dprint("*** PARSING LANGUAGE DEFINITION ***")
# Check the cache for already conctructed textX parser
if metamodel.debug in textX_parsers:
parser = textX_parsers[metamodel.debug]
else:
# Create parser for TextX grammars using
# the arpeggio grammar specified in this module
parser = ParserPython(textx_model,
comment_def=comment,
ignore_case=False,
reduce_tree=False,
memoization=metamodel.memoization,
debug=metamodel.debug,
file=metamodel.file)
# Cache it for subsequent calls
textX_parsers[metamodel.debug] = parser
# Parse language description with textX parser
try:
parse_tree = parser.parse(language_def)
except NoMatch as e:
line, col = parser.pos_to_linecol(e.position)
raise TextXSyntaxError(text(e), line, col)
# Construct new parser and meta-model based on the given language
# description.
lang_parser = visit_parse_tree(parse_tree, TextXVisitor(parser, metamodel))
# Meta-model is constructed. Validate its semantics.
metamodel.validate()
# Here we connect meta-model and language parser for convenience.
lang_parser.metamodel = metamodel
metamodel._parser_blueprint = lang_parser
if metamodel.debug:
# Create dot file for debuging purposes
PMDOTExporter().exportFile(
lang_parser.parser_model,
"{}_parser_model.dot".format(metamodel.rootcls.__name__))
return lang_parser
def parse(text):
parser = ParserPython(mspaText,
debug=False,
reduce_tree=False,
skipws=False,
ws="\t ",
memoization=True)
return parser.parse(text)
def __init__(self, graph, platform, cfg, mappingDict={}, debug=False):
self.__graph = graph
self.__platform = platform
self.__mappingDict = mappingDict
self.__parser = ParserPython(
Grammar.logicLanguage, reduce_tree=True, debug=debug
)
self.__debug = debug
self.__cfg = cfg
def test_parse_input():
parser = ParserPython(calc)
input = "4+5*7/3.45*-45*(2.56+32)/-56*(2-1.34)"
result = parser.parse(input)
assert isinstance(result, NonTerminal)
assert str(result) == "4 | + | 5 | * | 7 | / | 3.45 | * | - | 45 | * | ( | 2.56 | + | 32 | ) | / | - | 56 | * | ( | 2 | - | 1.34 | ) | "
assert repr(result) == "[ [ [ [ number '4' [0] ] ], '+' [1], [ [ number '5' [2] ], '*' [3], [ number '7' [4] ], '/' [5], [ number '3.45' [6] ], '*' [10], [ '-' [11], number '45' [12] ], '*' [14], [ '(' [15], [ [ [ number '2.56' [16] ] ], '+' [20], [ [ number '32' [21] ] ] ], ')' [23] ], '/' [24], [ '-' [25], number '56' [26] ], '*' [28], [ '(' [29], [ [ [ number '2' [30] ] ], '-' [31], [ [ number '1.34' [32] ] ] ], ')' [36] ] ] ], EOF [37] ]"
def test_parse_input():
parser = ParserPython(calc)
input = "4+5*7/3.45*-45*(2.56+32)/-56*(2-1.34)"
result = parser.parse(input)
assert isinstance(result, NonTerminal)
assert str(result) == "4 | + | 5 | * | 7 | / | 3.45 | * | - | 45 | * | ( | 2.56 | + | 32 | ) | / | - | 56 | * | ( | 2 | - | 1.34 | ) | "
assert repr(result) == "[ [ [ [ number '4' [0] ] ], '+' [1], [ [ number '5' [2] ], '*' [3], [ number '7' [4] ], '/' [5], [ number '3.45' [6] ], '*' [10], [ '-' [11], number '45' [12] ], '*' [14], [ '(' [15], [ [ [ number '2.56' [16] ] ], '+' [20], [ [ number '32' [21] ] ] ], ')' [23] ], '/' [24], [ '-' [25], number '56' [26] ], '*' [28], [ '(' [29], [ [ [ number '2' [30] ] ], '-' [31], [ [ number '1.34' [32] ] ] ], ')' [36] ] ] ], EOF [37] ]"
def test_sequence_suppress():
"""
"""
def grammar():
return Sequence("one", "two", "three", suppress=True), "four"
parser = ParserPython(grammar)
result = parser.parse("one two three four")
assert result[0] == "four"
def _from_peg(self, language_def):
parser = ParserPython(peggrammar, comment, reduce_tree=False,
debug=self.debug)
parser.root_rule_name = self.root_rule_name
parse_tree = parser.parse(language_def)
return visit_parse_tree(parse_tree, PEGVisitor(self.root_rule_name,
self.comment_rule_name,
self.ignore_case,
debug=self.debug))
def test_sequence():
def grammar():
return ("a", "b", "c")
parser = ParserPython(grammar)
parsed = parser.parse("a b c")
assert str(parsed) == "a | b | c"
assert repr(parsed) == "[ 'a' [0], 'b' [2], 'c' [4] ]"
def test_zeroormore_with_separator():
def grammar():
return ZeroOrMore(['a', 'b'], sep=','), EOF
parser = ParserPython(grammar, reduce_tree=False)
result = parser.parse('a, b, b, b, a')
assert result
with pytest.raises(NoMatch):
parser.parse('a, b a')
def test_sequence_suppress():
"""
"""
def grammar(): return Sequence("one", "two", "three", suppress=True), "four"
parser = ParserPython(grammar)
result = parser.parse("one two three four")
assert result[0] == "four"
def _from_peg(self, language_def):
parser = ParserPython(peggrammar, comment, reduce_tree=False,
debug=self.debug)
parser.root_rule_name = self.root_rule_name
parse_tree = parser.parse(language_def)
return visit_parse_tree(parse_tree, PEGVisitor(self.root_rule_name,
self.comment_rule_name,
self.ignore_case,
debug=self.debug))
def main():
parser = ParserPython(calc) # calc is the root rule of your grammar
# Use param debug=True for verbose debugging
# messages and grammar and parse tree visualization
# using graphviz and dot
parse_tree = parser.parse("-(4-1)*5+(2+4.67)+5.89/(.2+7)")
print(parse_tree)
result = visit_parse_tree(parse_tree, CalcVisitor(debug=False))
print(result)
def test_sequence():
def grammar():
return ("a", "b", "c")
parser = ParserPython(grammar)
parsed = parser.parse("a b c")
assert str(parsed) == "a | b | c"
assert repr(parsed) == "[ 'a' [0], 'b' [2], 'c' [4] ]"
def test_rrel_basic_parser1():
parser = ParserPython(rrel_standalone)
parse_tree = parser.parse("^pkg*.cls")
assert len(parse_tree) == 2 # always true (one path, one EOF)
parse_tree = parser.parse("obj.ref.~extension *.methods")
assert len(parse_tree) == 2 # always true (one path, one EOF)
parse_tree = parser.parse("instance.(type.vals)*")
assert len(parse_tree) == 2 # always true (one path, one EOF)
def parse(content: str) -> 'Literal':
from arpeggio import ParserPython
from arpeggio import visit_parse_tree
from depysible.language.grammar import literal, comment
from depysible.language.visitor import DefeasibleVisitor
parser = ParserPython(literal, comment_def=comment)
parse_tree = parser.parse(content)
return visit_parse_tree(parse_tree, DefeasibleVisitor())
def test_zeroormore_with_separator():
def grammar():
return ZeroOrMore(['a', 'b'], sep=','), EOF
parser = ParserPython(grammar, reduce_tree=False)
result = parser.parse('a, b, b, b, a')
assert result
with pytest.raises(NoMatch):
parser.parse('a, b a')
def main(debug=False):
parser = ParserPython(initial, debug=debug)
file_input = open("input.txt", 'r')
input_expr = file_input.read()
parse_tree = parser.parse(input_expr)
result = parser.getASG()
def test_eolterm():
# first rule should match only first line
# so that second rule will match "a" on the new line
input = """a a b a b b
a"""
parser = ParserPython(grammar, reduce_tree=False)
result = parser.parse(input)
assert result
def test_default_action_disabled():
parser = ParserPython(grammar)
parser.parse('(-34) strmatch')
parser.getASG(defaults=False)
assert not p_removed
assert not number_str
assert parse_tree_node
def test_empty_nested_parse():
def grammar(): return [first]
def first(): return ZeroOrMore("second")
parser = ParserPython(grammar)
# Parse tree will be empty
# as nothing will be parsed
tree = parser.parse("something")
assert not tree
def test_not_match_as_alternative():
"""
Test that Not is not reported if a part of OrderedChoice.
"""
def grammar(): return ['one', Not('two')], _(r'\w+')
parser = ParserPython(grammar)
with pytest.raises(NoMatch) as e:
parser.parse(' three ident')
assert "Expected 'one' at " in str(e)
def parse_bibtex(file_name, debug=False):
global parser
if parser is None:
parser = ParserPython(bibfile, debug=debug)
with codecs.open(file_name, "r", encoding="utf-8") as bibtexfile:
bibtexfile_content = bibtexfile.read()
parse_tree = parser.parse(bibtexfile_content)
return visit_parse_tree(parse_tree,
BibtexVisitor(debug=debug))
def test_optional_no_error():
"""
"""
def grammar(): return Optional('a'), 'b'
parser = ParserPython(grammar)
try:
parser.parse('c')
assert False
except NoMatch as e:
assert "Expected 'b'" in str(e)
def test_file_name_reporting():
"""
Test that if parser has file name set it will be reported.
"""
def grammar(): return Optional('a'), 'b', EOF
parser = ParserPython(grammar)
with pytest.raises(NoMatch) as e:
parser.parse("\n\n a c", file_name="test_file.peg")
assert "Expected 'b' at position test_file.peg:(3, 6)" in str(e)
assert (e.value.line, e.value.col) == (3, 6)
def test_sequence_of_nots():
"""
Test that sequence of Not rules is handled properly.
"""
def grammar():
return Not('one'), Not('two'), _(r'\w+')
parser = ParserPython(grammar)
with pytest.raises(NoMatch) as e:
parser.parse(' two ident')
assert "Not expected input" in str(e)
def test_file_name_reporting():
"""
Test that if parser has file name set it will be reported.
"""
def grammar(): return Optional('a'), 'b', EOF
parser = ParserPython(grammar)
try:
parser.parse("\n\n a c", file_name="test_file.peg")
except NoMatch as e:
assert "Expected 'b' at test_file.peg:(3, 6)" in str(e)
def test_not_match_at_beginning():
"""
Test that matching of Not ParsingExpression is not reported in the
error message.
"""
def grammar():
return Not('one'), _(r'\w+')
parser = ParserPython(grammar)
with pytest.raises(NoMatch) as e:
parser.parse(' one ident')
assert "Not expected input" in str(e)
def test_pos_to_linecol():
def grammar(): return ("a", "b", "c")
parser = ParserPython(grammar)
parse_tree = parser.parse("a\n\n\n b\nc")
a_pos = parse_tree[0].position
assert parser.pos_to_linecol(a_pos) == (1, 1)
b_pos = parse_tree[1].position
assert parser.pos_to_linecol(b_pos) == (4, 2)
c_pos = parse_tree[2].position
assert parser.pos_to_linecol(c_pos) == (5, 1)
def test_alternative_added():
"""
Test that matches from alternative branches at the same positiona are
reported.
"""
def grammar(): return ['one', 'two'], _(r'\w+')
parser = ParserPython(grammar)
with pytest.raises(NoMatch) as e:
parser.parse(' three ident')
assert "Expected 'one' or 'two'" in str(e)
assert (e.value.line, e.value.col) == (1, 4)
def test_optional_no_error():
"""
Test that optional match failure does not show up in the NoMatch errors.
"""
def grammar(): return Optional('a'), 'b'
parser = ParserPython(grammar)
try:
parser.parse('c')
assert False
except NoMatch as e:
assert "Expected 'b'" in str(e)
def test_comment_matching_not_reported():
"""
Test that matching of comments is not reported.
"""
def grammar(): return Optional('a'), 'b', EOF
def comments(): return _('\/\/.*$')
parser = ParserPython(grammar, comments)
try:
parser.parse('\n\n a // This is a comment \n c')
except NoMatch as e:
assert "Expected 'b' at position (4, 2)" in str(e)
def test_skipws():
"""
skipws may be defined per Sequence.
"""
def grammar(): return Sequence("one", "two", "three"), "four"
parser = ParserPython(grammar)
# By default, skipws is True and whitespaces will be skipped.
parser.parse("one two three four")
def grammar(): return Sequence("one", "two", "three",
skipws=False), "four"
parser = ParserPython(grammar)
# If we disable skipws for sequence only then whitespace
# skipping should not be done inside sequence.
with pytest.raises(NoMatch):
parser.parse("one two three four")
# But it will be done outside of it
parser.parse("onetwothree four")
def test_comment_matching_not_reported():
"""
Test that matching of comments is not reported.
"""
def grammar(): return Optional('a'), 'b', EOF
def comments(): return _('\/\/.*$')
parser = ParserPython(grammar, comments)
with pytest.raises(NoMatch) as e:
parser.parse('\n\n a // This is a comment \n c')
assert "Expected 'b' at position (4, 2)" in str(e)
assert (e.value.line, e.value.col) == (4, 2)
def test_optional_inside_zeroormore():
"""
Test optional match inside a zero or more.
Optional should always succeed thus inducing ZeroOrMore
to try the match again.
Arpeggio handle this using soft failures.
"""
def grammar(): return ZeroOrMore(Optional('a'))
parser = ParserPython(grammar)
with pytest.raises(NoMatch):
# This could lead to infinite loop
parser.parse('b')
def test_ws():
"""
ws consists of chars that will be skipped if skipws is enables.
By default it consists of space, tab and newline.
"""
def grammar():
return ("one", "two", "three")
parser = ParserPython(grammar)
# With default ws this should parse without error
parser.parse("""one
two three""")
# If we make only a space char to be ws than the
# same input will raise exception.
parser = ParserPython(grammar, ws=" ")
with pytest.raises(NoMatch):
parser.parse("""one
two three""")
# But if only spaces are between words than it will
# parse.
parser.parse("one two three")
