kodjo/Sheerka-Old
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Added ZeroAndMore and OneAndMore to BNF. BNF expressions can now be captured

Browse Source
This commit is contained in:
Kodjo Sossouvi
2019-12-18 12:01:51 +01:00
parent 88cd3162be
commit 8dbe2e1b20
9 changed files with 425 additions and 91 deletions
Download Patch File Download Diff File Expand all files Collapse all files
parsers/BnfParser.py
+41 -12
View File
@@ -115,7 +115,11 @@ class BnfParser:
def parse(self, context: ExecutionContext, text):
self.reset_parser(context, text)
tree = self.parse_choice()
tree = self.parser_outer_rule_name()
token = self.get_token()
if token and token.type != TokenKind.EOF:
self.add_error(UnexpectedTokenErrorNode(f"Unexpected token '{token}'", []))
ret = self.sheerka.ret(
self.name,
@@ -129,12 +133,15 @@ class BnfParser:
return ret
def parser_outer_rule_name(self):
return self.parser_rule_name(self.parse_choice)
def parse_choice(self):
sequence = self.parse_sequence()
self.eat_white_space()
token = self.get_token()
if token is None or token.type == TokenKind.EOF or token.type != TokenKind.VBAR:
if token is None or token.type != TokenKind.VBAR:
return sequence
elements = [sequence]
@@ -142,7 +149,7 @@ class BnfParser:
# maybe eat the vertical bar
self.eat_white_space()
token = self.get_token()
if token is None or token.type == TokenKind.EOF or token.type != TokenKind.VBAR:
if token is None or token.type != TokenKind.VBAR:
break
self.next_token(skip_whitespace=True)
@@ -152,30 +159,33 @@ class BnfParser:
return OrderedChoice(*elements)
def parse_sequence(self):
expr_and_modifier = self.parse_expression_and_modifier()
expr_and_modifier = self.parse_modifier()
token = self.get_token()
if token is None or token.type == TokenKind.EOF or \
if token is None or \
token.type == TokenKind.EOF or \
token.type == TokenKind.EQUALS or \
self.maybe_sequence(TokenKind.WHITESPACE, TokenKind.VBAR) or \
self.nb_open_par > 0 and self.maybe_sequence(TokenKind.WHITESPACE, TokenKind.RPAR):
return expr_and_modifier
elements = [expr_and_modifier]
while True:
# maybe eat the comma
token = self.get_token()
if token is None or token.type == TokenKind.EOF or \
if token is None or \
token.type == TokenKind.EOF or \
token.type == TokenKind.EQUALS or \
self.maybe_sequence(TokenKind.WHITESPACE, TokenKind.VBAR) or \
self.nb_open_par > 0 and self.maybe_sequence(TokenKind.WHITESPACE, TokenKind.RPAR):
break
self.eat_white_space()
sequence = self.parse_expression_and_modifier()
sequence = self.parse_modifier()
elements.append(sequence)
return Sequence(*elements)
def parse_expression_and_modifier(self):
expression = self.parse_expression()
def parse_modifier(self):
expression = self.parser_inner_rule_name()
token = self.get_token()
@@ -193,6 +203,9 @@ class BnfParser:
return expression
def parser_inner_rule_name(self):
return self.parser_rule_name(self.parse_expression)
def parse_expression(self):
token = self.get_token()
if token.type == TokenKind.EOF:
@@ -207,7 +220,7 @@ class BnfParser:
self.next_token()
return expression
else:
self.add_error(UnexpectedTokenErrorNode(f"Unexpected token '{token.type}'", [TokenKind.RPAR]))
self.add_error(UnexpectedTokenErrorNode(f"Unexpected token '{token}'", [TokenKind.RPAR]))
return expression
if token.type == TokenKind.IDENTIFIER:
@@ -224,4 +237,20 @@ class BnfParser:
ret = StrMatch(core.utils.strip_quotes(token.value))
self.next_token()
return ret
return ret
def parser_rule_name(self, next_to_parse):
expression = next_to_parse()
token = self.get_token()
if token is None or token.type != TokenKind.EQUALS:
return expression
self.next_token() # eat equals
token = self.get_token()
if token is None or token.type != TokenKind.IDENTIFIER:
return self.add_error(UnexpectedTokenErrorNode(f"Unexpected token '{token}'", [TokenKind.IDENTIFIER]))
expression.rule_name = token.value
self.next_token()
return expression