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

Refactored Parsers. Introduced BaseCustomGrammarParser. Renamed DefaultParser into DefConceptParser

Browse Source
This commit is contained in:
Kodjo Sossouvi
2020-10-02 04:45:47 +02:00
parent d100b7e8b3
commit e8f2705dcf
28 changed files with 1411 additions and 872 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/parsers/FunctionParser.py
+12 -9
View File
@@ -7,7 +7,7 @@ from core.sheerka.services.SheerkaExecute import ParserInput
from core.tokenizer import TokenKind, Token
from core.utils import get_n_clones
from parsers.BaseNodeParser import SourceCodeNode, SourceCodeWithConceptNode, UnrecognizedTokensNode
from parsers.BaseParser import BaseParser, UnexpectedTokenErrorNode, UnexpectedEof, Node
from parsers.BaseParser import BaseParser, UnexpectedTokenErrorNode, UnexpectedEofNode, Node
from parsers.PythonWithConceptsParser import PythonWithConceptsParser
# No need to check for Python code as the source code node will resolve to python code anyway
@@ -143,7 +143,7 @@ class FunctionParser(BaseParser):
so 'twenty one' will resolve to [[c:twenty one:]], not [[c:twenty one:], [c:twenty:, c:one:]]
:param kwargs:
"""
super().__init__("Function", 55, True)
super().__init__("Function", 55)
self.sep = sep
self.longest_concepts_only = longest_concepts_only
self.record_errors = True
@@ -179,6 +179,7 @@ class FunctionParser(BaseParser):
False,
context.sheerka.new(BuiltinConcepts.ERROR, body=self.error_sink))
self.parser_input.next_token()
node = self.parse_function()
if self.parser_input.next_token():
@@ -219,7 +220,7 @@ class FunctionParser(BaseParser):
return None
if not self.parser_input.next_token():
self.add_error(UnexpectedEof(f"Unexpected EOF while parsing left parenthesis"))
self.add_error(UnexpectedEofNode(f"Unexpected EOF while parsing left parenthesis"))
return None
token = self.parser_input.token
@@ -231,7 +232,7 @@ class FunctionParser(BaseParser):
start_node = NamesNode(start, start + 1, self.parser_input.tokens[start:start + 2])
if not self.parser_input.next_token():
self.add_error(UnexpectedEof(f"Unexpected EOF after left parenthesis"))
self.add_error(UnexpectedEofNode(f"Unexpected EOF after left parenthesis"))
return FunctionNode(start_node, None, None)
params = self.parse_parameters()
@@ -239,7 +240,7 @@ class FunctionParser(BaseParser):
return FunctionNode(start_node, None, params)
token = self.parser_input.token
if token.type != TokenKind.RPAR:
if not token or token.type != TokenKind.RPAR:
self.add_error(UnexpectedTokenErrorNode(f"Right parenthesis not found",
token,
[TokenKind.RPAR]))
@@ -261,7 +262,7 @@ class FunctionParser(BaseParser):
token = self.parser_input.token
if token.type == TokenKind.EOF:
self.add_error(UnexpectedEof(f"Unexpected EOF while parsing parameters"))
self.add_error(UnexpectedEofNode(f"Unexpected EOF while parsing parameters"))
return None
if token.type == TokenKind.RPAR:
@@ -269,10 +270,12 @@ class FunctionParser(BaseParser):
if token.value == self.sep:
sep_pos = self.parser_input.pos
self.parser_input.next_token()
has_next = self.parser_input.next_token() # it's before add_sep() to capture trailing whitespace
function_parameter.add_sep(sep_pos,
self.parser_input.pos - 1,
self.parser_input.tokens[sep_pos: self.parser_input.pos])
if not has_next:
break
return nodes
@@ -292,8 +295,8 @@ class FunctionParser(BaseParser):
tokens = []
while True:
token = self.parser_input.token
# if token is None:
# break
if token is None:
break
if token.value == self.sep or token.type == TokenKind.RPAR:
break