def id_lt(self):
if self.peek() == '=':
self.advance(2)
return TokenZ(TypesZ.LTE, '<=')
else:
self.advance()
return TokenZ(TypesZ.LT, '<')
def id_bang(self):
if self.peek() == '=':
self.advance(2)
return TokenZ(TypesZ.NEQ, '!=')
else:
self.advance()
return TokenZ(TypesZ.BANG, '!')
def id_gt(self):
if self.peek() == '=':
self.advance(2)
return TokenZ(TypesZ.GTE, '>=')
else:
self.advance()
return TokenZ(TypesZ.GT, '>')
def id_eq(self):
if self.peek() == '=':
self.advance(2)
return TokenZ(TypesZ.EQEQ, '==')
else:
self.advance()
return TokenZ(TypesZ.EQ, '=')
def identify(self):
result = ''
if self.current_char == '/': # Endfor or Endif
result += self.current_char
self.advance()
while self.current_char is not None and (self.current_char.isalnum()
or self.current_char == '_'):
result += self.current_char
self.advance()
if result in ZephyrFuncs:
return TokenZ(TypesZ.FUNC, result)
else:
return ReservedKeywords.as_dict().get(result,
TokenZ(TypesZ.VAR, result))
def number(self):
result = ''
while self.current_char is not None and self.current_char.isnumeric():
result += self.current_char
self.advance()
if self.current_char == '.': # Floating point
result += self.current_char
self.advance()
while self.current_char is not None and self.current_char.isnumeric(
):
result += self.current_char
self.advance()
if result[-1] == '.':
raise Exception('Real val cannot terminate with \'.\'')
return TokenZ(TypesZ.REAL, float(result))
return TokenZ(TypesZ.INTEGER, int(result))
def string(self):
result = ''
self.advance() # Skip the quotation mark
while self.current_char is not None and (self.current_char != '\''
and self.current_char != '"'):
result += self.current_char
self.advance()
self.advance() # Skip the closing quotation mark
return TokenZ(TypesZ.STRING, result)
def html_or_text(self):
result = ''
while self.current_char is not None and self.current_char != '{':
result += self.current_char
self.advance()
if all(css_keyword in result
for css_keyword in ['text/css', '<style']):
# Good indicatio that this is CSS which should be skipped.
while result[-8:] != '</style>' and self.current_char is not None:
result += self.current_char
self.advance()
return TokenZ(TypesZ.HTML_OR_TEXT, result)
def tokenize(self):
if self.current_char is None:
return None
if self.current_char == '{':
self.zephyr_mode = True
self.advance()
return TokenZ(TypesZ.LBRACE, '{')
if self.current_char == '}':
self.zephyr_mode = False
self.advance()
return TokenZ(TypesZ.RBRACE, '}')
if not self.zephyr_mode:
return self.html_or_text()
if self.current_char.isspace():
self.skip_whitespace()
if self.current_char.isnumeric():
return self.number()
if self.current_char == '\'' or self.current_char == '"':
return self.string()
if self.current_char.isalpha():
return self.identify()
result = self.char_to_token_dict.get(self.current_char)
if type(result) is str:
result = getattr(self, result)
return result()
elif result is not None:
self.advance()
return result
else:
import pdb
pdb.set_trace()
raise Exception('Invalid character: %s' % self.current_char)
def id_div(self):
if self.lookback() == '{':
return self.identify()
else:
self.advance()
return TokenZ(TypesZ.DIV, '/')
def id_and(self):
if self.peek() == '&':
self.advance(2)
return TokenZ(TypesZ.AND, '&&')
raise NotImplemented('Binary operators haven\'t been implemented yet.')
class ScannerZ:
char_to_token_dict = {
'(': TokenZ(TypesZ.LPAREN, '('),
')': TokenZ(TypesZ.RPAREN, ')'),
'[': TokenZ(TypesZ.LBRACKET, '['),
']': TokenZ(TypesZ.RBRACKET, ']'),
'.': TokenZ(TypesZ.DOT, '.'),
',': TokenZ(TypesZ.COMMA, ','),
':': TokenZ(TypesZ.COLON, ':'),
'?': TokenZ(TypesZ.QUESTION, '?'),
'+': TokenZ(TypesZ.PLUS, '+'),
'-': TokenZ(TypesZ.MINUS, '-'),
'*': TokenZ(TypesZ.MUL, '*'),
'%': TokenZ(TypesZ.MOD, '%'),
'/': 'id_div',
'<': 'id_lt',
'>': 'id_gt',
'=': 'id_eq',
'!': 'id_bang',
'&': 'id_and',
'|': 'id_or'
}
def __init__(self, text):
self.text = text
self.pos = 0
self.current_char = self.text[self.pos]
self.zephyr_mode = False # indicates that we're parsing Zephyr code and not txt/html.
def advance(self, count=1):
self.pos += count
if self.pos > len(self.text) - 1:
self.current_char = None
else:
self.current_char = self.text[self.pos]
def peek(self):
peek_pos = self.pos + 1
if peek_pos < len(self.text) - 1:
return self.text[peek_pos]
else:
return None
def lookback(self):
lookback_pos = self.pos - 1
if lookback_pos >= 0:
return self.text[lookback_pos]
else:
return None
def skip_whitespace(self):
while self.current_char.isspace() and self.current_char is not None:
self.advance()
def identify(self):
result = ''
if self.current_char == '/': # Endfor or Endif
result += self.current_char
self.advance()
while self.current_char is not None and (self.current_char.isalnum()
or self.current_char == '_'):
result += self.current_char
self.advance()
if result in ZephyrFuncs:
return TokenZ(TypesZ.FUNC, result)
else:
return ReservedKeywords.as_dict().get(result,
TokenZ(TypesZ.VAR, result))
def id_and(self):
if self.peek() == '&':
self.advance(2)
return TokenZ(TypesZ.AND, '&&')
raise NotImplemented('Binary operators haven\'t been implemented yet.')
def id_bang(self):
if self.peek() == '=':
self.advance(2)
return TokenZ(TypesZ.NEQ, '!=')
else:
self.advance()
return TokenZ(TypesZ.BANG, '!')
def id_div(self):
if self.lookback() == '{':
return self.identify()
else:
self.advance()
return TokenZ(TypesZ.DIV, '/')
def id_lt(self):
if self.peek() == '=':
self.advance(2)
return TokenZ(TypesZ.LTE, '<=')
else:
self.advance()
return TokenZ(TypesZ.LT, '<')
def id_eq(self):
if self.peek() == '=':
self.advance(2)
return TokenZ(TypesZ.EQEQ, '==')
else:
self.advance()
return TokenZ(TypesZ.EQ, '=')
def id_gt(self):
if self.peek() == '=':
self.advance(2)
return TokenZ(TypesZ.GTE, '>=')
else:
self.advance()
return TokenZ(TypesZ.GT, '>')
def id_or(self):
if self.peek() == '|':
self.advance(2)
return TokenZ(TypesZ.OR, '||')
raise NotImplemented('Binary operators haven\'t been implemented yet.')
def number(self):
result = ''
while self.current_char is not None and self.current_char.isnumeric():
result += self.current_char
self.advance()
if self.current_char == '.': # Floating point
result += self.current_char
self.advance()
while self.current_char is not None and self.current_char.isnumeric(
):
result += self.current_char
self.advance()
if result[-1] == '.':
raise Exception('Real val cannot terminate with \'.\'')
return TokenZ(TypesZ.REAL, float(result))
return TokenZ(TypesZ.INTEGER, int(result))
def string(self):
result = ''
self.advance() # Skip the quotation mark
while self.current_char is not None and (self.current_char != '\''
and self.current_char != '"'):
result += self.current_char
self.advance()
self.advance() # Skip the closing quotation mark
return TokenZ(TypesZ.STRING, result)
def html_or_text(self):
result = ''
while self.current_char is not None and self.current_char != '{':
result += self.current_char
self.advance()
if all(css_keyword in result
for css_keyword in ['text/css', '<style']):
# Good indicatio that this is CSS which should be skipped.
while result[-8:] != '</style>' and self.current_char is not None:
result += self.current_char
self.advance()
return TokenZ(TypesZ.HTML_OR_TEXT, result)
def tokenize(self):
if self.current_char is None:
return None
if self.current_char == '{':
self.zephyr_mode = True
self.advance()
return TokenZ(TypesZ.LBRACE, '{')
if self.current_char == '}':
self.zephyr_mode = False
self.advance()
return TokenZ(TypesZ.RBRACE, '}')
if not self.zephyr_mode:
return self.html_or_text()
if self.current_char.isspace():
self.skip_whitespace()
if self.current_char.isnumeric():
return self.number()
if self.current_char == '\'' or self.current_char == '"':
return self.string()
if self.current_char.isalpha():
return self.identify()
result = self.char_to_token_dict.get(self.current_char)
if type(result) is str:
result = getattr(self, result)
return result()
elif result is not None:
self.advance()
return result
else:
import pdb
pdb.set_trace()
raise Exception('Invalid character: %s' % self.current_char)
def scan(self):
tokens = []
while self.current_char is not None:
tokens.append(self.tokenize())
return tokens
def id_or(self):
if self.peek() == '|':
self.advance(2)
return TokenZ(TypesZ.OR, '||')
raise NotImplemented('Binary operators haven\'t been implemented yet.')
def as_dict(cls):
return {
m.value: TokenZ(m.name, None)
for m in cls.__members__.values()
}