def test_self(self):
import os
filename = os.path.join(os.path.dirname(parse_pgen.__file__), "..",
"pgen.pgen")
grammar = parse_pgen.load_grammar(filename)
self.maxDiff = None
pgen_grammar = parse_pgen.pgen_grammar
self.assertEqual(pgen_grammar.nonterminals, grammar.nonterminals)
self.assertEqual(pgen_grammar.variable_terminals,
grammar.variable_terminals)
self.assertEqual(pgen_grammar.goals(), grammar.goals())
with open(parse_pgen_generated.__file__) as f:
pre_generated = f.read()
import io
out = io.StringIO()
jsparagus.gen.generate_parser(out, grammar)
generated_from_file = out.getvalue()
self.maxDiff = None
self.assertEqual(pre_generated, generated_from_file)
# prose not wrapped in square brackets
# To avoid conflict with the `>` token, this is recognized only after a space.
PROSE=r'(?<= )>[^\n]*',
# prose wrapped in square brackets
WPROSE=r'\[>[^]]*\]',
# expression denoting a matched terminal or nonterminal
MATCH_REF=r'\$(?:0|[1-9][0-9]*)',
# the spec also gives a few productions names
RUSTCOMMENT=r'//.*\n',
)
ESGrammarParser = gen.compile(
parse_pgen.load_grammar(
os.path.join(os.path.dirname(__file__), "esgrammar.pgen")))
SIGIL_FALSE = '~'
SIGIL_TRUE = '+'
# Abbreviations for single-character terminals, used in the lexical grammar.
ECMASCRIPT_CODE_POINTS = {
# From <https://tc39.es/ecma262/#table-31>
'<ZWNJ>': grammar.Literal('\u200c'),
'<ZWJ>': grammar.Literal('\u200d'),
'<ZWNBSP>': grammar.Literal('\ufeff'),
# From <https://tc39.es/ecma262/#table-32>
'<TAB>': grammar.Literal('\t'),
'<VT>': grammar.Literal('\u000b'),
'<FF>': grammar.Literal('\u000c'),