def p_error(p):
"""
print(p.lexer.prev.lineno, p.lineno)
if p.lexer.prev.lineno < p.lineno or p.type == "RBRACKET":
yacc.errok()
return
"""
if p == None:
if not restricted() and glob.g_tried_semi == False:
t = LexToken()
t.type = "SEMI"
t.value = ";"
t.lexpos = -1
t.lineno = -1
glob.g_lexer.push(t)
glob.g_tried_semi = True
yacc.errok()
else:
sys.stderr.write(glob.g_file + ": error: unexpected end of file\n")
return
else:
glob.g_error_pre = p
if handle_semi_error(p):
t = LexToken()
t.type = "SEMI"
t.value = ";"
t.lexpos = p.lexpos
t.lineno = p.lineno
#glob.g_lexer.push(t)
#glob.g_tried_semi = True
yacc.errok()
glob.g_error = False
return
else:
glob.g_error = True
print_err(p)
return
if glob.g_error:
print_err(glob.g_error_pre)
glob.g_error_pre = p
glob.g_error = True
try:
line = int(p.lineno)
except:
line = p.lineno(1)
try:
lexdata = p.lexer.lexer.lexdata
sline = p.lexer.lexer.lexpos
except:
lexdata = p.lexer.lexdata
sline = p.lexer.lexpos
sline = lexdata[sline-40:sline+1]
def t_ID(t: LexToken):
r"""[_a-zA-Z][_a-zA-Z0-9]*"""
# upper the token because MyLang is case insensitive
t.value = t.value.upper()
# handle special condition
if t.value == "TRUE" or t.value == "FALSE":
t.type = "BOOL"
t.value = str2bool(t.value)
elif t.value in constants.reserved:
t.type = t.value
return t
def t_RSTRING(t: lex.LexToken) -> lex.LexToken: # noqa: N802
r"r(\"([^\\\"\n]|\\.)*\")|r(\'([^\\\'\n]|\\.)*\')"
t.value = t.value[2:-1]
lexer = t.lexer
end = lexer.lexpos - lexer.linestart + 1
(s, e) = lexer.lexmatch.span()
start = end - (e - s)
t.value = LocatableString(
t.value, Range(lexer.inmfile, lexer.lineno, start, lexer.lineno, end),
lexer.lexpos, lexer.namespace)
return t
def t_STRING(t: lex.LexToken) -> lex.LexToken: # noqa: N802
r"(\"([^\\\"]|\\.)*\")|(\'([^\\\']|\\.)*\')"
t.value = bytes(t.value[1:-1], "utf-8").decode("unicode_escape")
lexer = t.lexer
end = lexer.lexpos - lexer.linestart + 1
(s, e) = lexer.lexmatch.span()
start = end - (e - s)
t.value = LocatableString(
t.value, Range(lexer.inmfile, lexer.lineno, start, lexer.lineno, end),
lexer.lexpos, lexer.namespace)
return t
def t_UINT_TYPE(self, t: LexToken) -> LexToken:
r"\buint[0-9]+\b"
cap: int = int(t.value[4:]) # uint{n}
t.value = Uint(
cap=cap, token=t.value, lineno=t.lineno, filepath=self.current_filepath()
)
return t
def _new_token(type, token):
tok = LexToken()
tok.type = type
tok.value = token.value
tok.lineno = token.lineno
tok.lexpos = token.lexpos
return tok
def token(self, value, ty=None):
t = LexToken()
t.type = ty if ty != None else value
t.value = value
t.lineno = -1
t.lexpos = -1
return t
def _lextoken_from_html(self, html_token):
token = LexToken()
token.type = {
0 : 'HTML_DOCTYPE',
1 : 'HTML_CHARS',
2 : 'HTML_WS',
3 : 'HTML_STARTTAG',
4 : 'HTML_ENDTAG',
5 : 'HTML_EMPTYTAG',
6 : 'HTML_COMMENT',
7 : 'HTML_PARSEERROR',
}[html_token['type']]
# TODO: fix lineno/lexpos
token.lineno = self.lineno
token.lexpos = self.lexpos
token.value = {
'self_closing' : html_token.get('selfClosing', False),
'name' : html_token.get('name', None),
}
if isinstance(html_token['data'], (list, tuple)):
token.value['attrs'] = html_token['data']
token.value['data'] = ''
if token.value['name'].lower() in voidElements:
token.type = 'HTML_VOID_TAG'
else:
token.value['data'] = html_token['data']
if token.type == tokenTypes['ParseError']:
raise SyntaxError("Got HTML Parse Error for token {}".format(html_token))
return token
def emit_autoend(self):
tok = LexToken()
tok.type = "AUTO_END"
tok.value = ""
tok.lineno = self.lineno
tok.lexpos = self.lexpos
return tok
def t_c8_DEPS(token: LexToken) -> LexToken:
# pylint: disable=missing-docstring
token.value = None if token.value == '_' else tuple(
(int(x[:x.index(':')]), x[x.index(':') + 1:])
for x in token.value.split('|'))
token.lexer.begin('v8')
return token
def new_dedent(amount, token):
tok = LexToken()
tok.type = "DEDENT"
tok.value = amount
tok.lineno = token.lineno
tok.lexpos = token.lexpos
return tok
def t_c5_FEATS(token: LexToken) -> LexToken:
# pylint: disable=missing-docstring
token.value = None if token.value == '_' else tuple(
(x[:x.index('=')], tuple(x[x.index('=') + 1:].split(',')))
for x in token.value.split('|'))
token.lexer.begin('v5')
return token
def token(self):
t = LexToken()
c = self.cur
if c >= len(self.str):
return None
c = self.str[c]
if c == "\\": t.type = "BACKSLASH"
elif c == "/": t.type = "DIVIDE"
elif c == "[": t.type = "LSBRACKET"
elif c == "]": t.type = "RSBRACKET"
elif c == "*": t.type = "STAR"
elif c == "\n" or c == "\r": t.type = "LT"
elif re.match(r"[a-zA-Z0-9_$]+", c) != None:
t.type = "ID_PART"
else:
t.type = "UCHAR"
t.value = c
t.lineno = 0
t.lexpos = self.cur
self.cur += 1
print(t)
return t
def token(self):
t = LexToken()
c = self.cur
if c >= len(self.str):
return None
c = self.str[c]
if c == "\\": t.type = "BACKSLASH"
elif c == "/": t.type = "DIVIDE"
elif c == "[": t.type = "LSBRACKET"
elif c == "]": t.type = "RSBRACKET"
elif c == "*": t.type = "STAR"
elif c == "\n" or c == "\r": t.type = "LT"
elif re.match(r"[a-zA-Z0-9_$]+", c) != None:
t.type = "ID_PART"
else: t.type = "UCHAR"
t.value = c
t.lineno = 0
t.lexpos = self.cur
self.cur += 1
print(t)
return t
def _lextoken( self, type_, value ) :
tok = LexToken()
tok.type = type_
tok.value = value
tok.lineno = self.lexer.lineno
tok.lexpos = self.lexer.lexpos
return tok
def _to_yacc(self, token_type, token_data):
token = LexToken()
token.type = token_type
token.value = (token_type, token_data)
token.lineno = 0 # TODO: file offset
token.lexpos = 0
self.__to_yacc(token)
def token(self, value, ty=None):
t = LexToken()
t.type = ty if ty != None else value
t.value = value
t.lineno = -1
t.lexpos = -1
return t
def t_provides_item(self, t: LexToken):
r'[^ \n]+'
t.type = 'PROVIDES_ITEM'
version = re.findall(r'([a-zA-Z-_\d.]+)([>=<]+)([\d.]+)', t.value)
if len(version): # version syntax
t.value = version[0]
return t
def _new_token(type, token):
tok = LexToken()
tok.type = type
tok.value = token.value
tok.lineno = token.lineno
tok.lexpos = token.lexpos
return tok
def _to_yacc(self, token_type, token_data):
token = LexToken()
token.type = token_type
token.value = (token_type, token_data)
token.lineno = 0 # TODO: file offset
token.lexpos = 0
self.__to_yacc(token)
def clone_token(old_token, new_type): token = LexToken() token.type = new_type token.value = old_token.value token.lineno = old_token.lineno token.lexpos = old_token.lexpos return token
def t_ANY_STRING(self, t: LexToken) -> LexToken:
for g in self.string_group_enc:
str_value = t.lexer.lexmatch.group(g)
if str_value:
t.value = str_value
break
return t
def make_tok(type_, value, lineno, lexpos):
token = LexToken()
token.type = type_
token.value = value
token.lineno = lineno
token.lexpos = lexpos
return token
def clone_token(old_token, new_type):
token = LexToken()
token.type = new_type
token.value = old_token.value
token.lineno = old_token.lineno
token.lexpos = old_token.lexpos
return token
def _create_token(self, type):
token = LexToken()
token.type = type
token.value = ''
token.lineno = 0
token.lexpos = 0
return token
def new_dedent(amount, token):
tok = LexToken()
tok.type = "DEDENT"
tok.value = amount
tok.lineno = token.lineno
tok.lexpos = token.lexpos
return tok
def p_class0(self, p):
""" class : CLASS TYPEID '{' feature_list '}' ';' """
tok = LexToken()
tok.type = "OBJECTID"
tok.value = "object"
tok.lineno = -1 #this is just inserted not present
tok.lexpos = -1
p[0] = class_(p[2], tok, p[4], self.filename)
def correct_tag_name(original):
token = LexToken()
token.type = original.type
token.value = original.value[:-1]
token.lineno = original.lineno
token.lexpos = original.lexpos
token.lexer = original.lexer
return token
def newtok(tok, ttype=None):
if tok.type != ttype and (ttype != None or tok.value != ""):
if tok.type != None:
push(tok)
tok = LexToken()
tok.type = ttype
tok.value = ""
return tok
def make_ws(tok):
lt = LexToken()
lt.type = "CPP_WS"
lt.value = "\n"
lt.lexer = lexer
lt.lineno = tok.lineno
lt.lexpos = tok.lexpos
return lt
def remove_first_and_last_char(original):
token = LexToken()
token.type = original.type
token.value = original.value[1:-1]
token.lineno = original.lineno
token.lexpos = original.lexpos
token.lexer = original.lexer
return token
def new_tok(lexpos, tok_type, lineno, value):
# Create a token for return
tok = LexToken()
tok.value = value
tok.lineno = lineno
tok.lexpos = lexpos
tok.type = tok_type
return tok
def createFunctionDefinition(self, def_token, var_token, params, val_node):
lamToken = LexToken()
lamToken.value = 'lambda'
lamToken.type = 'LAMBDA'
return LetNode(def_token, [
VariableNode(var_token),
LambdaNode(lamToken, [Node(None, None, nodes(params)), val_node]),
])
def gen(code): for line in code: for item in line: t = LexToken() t.type = item[1] t.value = item[0] yield t yield None
def createFunctionDefinition(self, def_token, var_token, params, val_node):
lamToken = LexToken()
lamToken.value = 'lambda'
lamToken.type = 'LAMBDA'
return LetNode(def_token, [
VariableNode(var_token),
LambdaNode(lamToken, [ Node(None, None, nodes(params)), val_node ]),
])
def t_config_help(self, t: LexToken):
r'(help|\-{3}help\-{3})\n'
t.lexer.help_indent = None
t.lexer.push_state('helpline')
t.type = "CONFIG_HELP"
t.value = t.value[0:-1]
t.lexer.lineno += 1
return t
def newtok(tok, ttype=None):
if tok.type != ttype and (ttype != None or tok.value != ""):
if tok.type != None:
push(tok)
tok = LexToken()
tok.type = ttype
tok.value = ""
return tok
def _parse_chars(self, data):
m = js_start_rx.match(data)
if m is None:
return None
pretext = m.group(1)
start_type = m.group(2)
self.lexpos -= len(data)
if len(pretext):
pretext_tok = LexToken()
pretext_tok.type = 'HTML_CHARS'
pretext_tok.value = pretext
pretext_tok.lineno = self.lineno - pretext.count("\n")
pretext_tok.lexpos = self.lexpos
self.next_tokens.append(pretext_tok)
self.lexpos += len(pretext)
start_tok = LexToken()
start_tok.type = self.tbtype[start_type]
start_tok.value = start_type
start_tok.lineno = self.lineno
start_tok.lexpos = self.lexpos
self.next_tokens.append(start_tok)
self.lexpos += len(start_type)
js_lexer = JSLexer()
js_lexer.input(data[m.end(2):])
for t in js_lexer:
t.lineno += self.lineno - 1
t.lexpos = self.lexpos
self.lexpos += js_lexer.lexer.lexpos
if t.type in ('EXPRESSION_TERMINATOR', 'ESCAPED_TERMINATOR', 'JS_TERMINATOR'):
if t.type != self.ttype[start_type]:
raise SyntaxError("Expected {} but got {} in char data `{}`".format(self.ttype[start_type], t.type, data))
self.next_tokens.append(t)
break
self.next_tokens.append(t)
remaining_text = data[m.end(2) + js_lexer.lexer.lexpos:]
self.lexpos += len(remaining_text)
return remaining_text
def p_expr5(self, p):
""" expr : OBJECTID '(' ')' """
tok = LexToken()
tok.type = "OBJECTID"
tok.value = "self"
tok.lineno = -1 #this is just inserted not present
tok.lexpos = -1
p[0] = dispatch(object_(tok), p[1], nil_Expressions())
p.lineno = p.lineno(1)
def gen_token(value, type, line, lexpos):
t = LexToken()
t.value = value
t.type = type
t.line = line
t.lexpos = lexpos
t.lexer = self
return t
def p_expr4(self, p):
""" expr : OBJECTID '(' expr_arg_list ')' """
tok = LexToken()
tok.type = "OBJECTID"
tok.value = "self"
tok.lineno = -1 #this is just inserted not present
tok.lexpos = -1
p[0] = dispatch(object_(tok), p[1], p[3])
p[0].lineno = p.lineno(1)
def gen_token(value, type, line, lexpos):
t = LexToken()
t.value = value
t.type = type
t.line = line
t.lexpos = lexpos
t.lexer = self
return t
def _new_token(self, type=None, value=None, lexpos=None, lineno=None) -> LexToken:
"""
Creates a new lexer token with the given properties.
:return: a new lexer token with the given properties.
"""
token = LexToken()
token.type = type
token.value = value
token.lexpos = lexpos
token.lineno = lineno
def t_INT_TYPE(self, t: LexToken) -> LexToken:
r"\bint[0-9]+\b"
# Why use the regexp `int[0-9]+` instead of `int(8|16|32|64)`?
# We want the unsupported cases like `int0`, `int3` to raise an error instead of
# lexing into identifiers.
cap: int = int(t.value[3:])
t.value = Int(
cap=cap, token=t.value, lineno=t.lineno, filepath=self.current_filepath()
)
return t
def _new_token(self, new_type, new_value, lineno: int, lexpos: int):
"""
Creates a new token with the given data.
:return: new token with the given data.
"""
token = LexToken()
token.type = new_type
token.value = new_value
token.lineno = lineno
token.lexpos = lexpos
return token
def to_tokens(self, token_list):
result = []
for values in token_list:
token = LexToken()
token.type = values[0]
token.value = values[1]
token.lineno = values[2]
token.lexpos = values[3]
token.lexer = self.lexer
result.append(token)
return result
def _gen_token(self, type, value='', lnum=None, position=0, lexpos=None):
"""
Generates a LexToken with the paramaters given.
"""
tok = LexToken()
tok.lexer = self.lex
tok.type = type
tok.value = value
tok.line_position = position
# I think this will work...
tok.lineno = self.lex.lineno if lnum is None else lnum
tok.lexpos = self.lex.lexpos if lexpos is None else lexpos
return tok
def p_error(self, p):
# TODO
if p:
self._errors.append(p)
pass # self._parser.errok()
else:
# hack handle eof, don't know why ply behaves this way
from ply.lex import LexToken
tok = LexToken()
tok.value = self.lexer.lexdata[self.lexer.lexpos:]
tok.lineno = self.lexer.lineno
tok.type = 'error'
tok.lexpos = self.lexer.lexpos
self._parser.errok()
return tok
def handle_semi_error(p):
tok = p.lexer.peek()
if len(p.lexer.peeks) > 1:
prev = p.lexer.peeks[-2]
else:
prev = p.lexer.prev
cur = p.lexer.cur
if prev == None:
prev = tok
if cur == None:
cur = tok
if type(prev) == list: prev = prev[0]
if type(cur) == list: cur = cur[0]
if type(tok) == list: tok = tok[0]
ret = tok == None or cur == None or prev.lineno < tok.lineno
ret = ret or tok.type == "RBRACKET" or prev.type == "RBRACKET"
ret = ret or cur.type == "RBRACKET"
p2 = restricted()
if p2 != None and not (prev.type in ["RSBRACKET", "RPAREN"] and restrict_prev() == None):
ret = False
p = p2
glob.g_line = p.lineno
glob.g_lexpos = p.lexpos
if ret and not glob.g_tried_semi:
t = LexToken()
t.type = "SEMI"
t.value = ";"
t.lineno = cur.lineno
t.lexpos = cur.lexpos
p.lexer.push(p.lexer.cur)
p.lexer.push(t)
yacc.errok()
glob.g_error = False
glob.g_tried_semi = True
else:
ret = False
glob.g_error = True
glob.g_error_pre = p
return ret
def _new_token(type, value, pos):
o = LexToken()
o.type = type
o.value = value
o.lineno, o.lexpos = pos
return o
def t_ANY_IRI(self, t: LexToken) -> LexToken:
lexmatch = t.lexer.lexmatch
t.value = (lexmatch.group('prefix_name'), lexmatch.group('prefix_ind'), lexmatch.group('iri_value'))
return t
def t_NEWLINE(self, t: LexToken) -> LexToken:
t.lexer.lineno += len(t.value)
t.value = t.value[0]
return t
def t_ANY_NUMBER(self, t: LexToken) -> LexToken:
t.value = int(t.value)
return t
# elif tokens[i].value < dentstack[-1]:
# newtok.type = 'DEDENT'
# dentstack += [tokens[i].value]
# else:
# newtok.type = 'NEWLINE'
#
# fixedtokens += [newtok]
# else:
# fixedtokens += [tokens[i]]
for i in range(len(tokens)-1):
print ">>>", tokens[i].type, tokens[i].value
fixedtokens += [tokens[i]]
if tokens[i].type == 'NEWLINE':
newtok = LexToken()
newtok.value = None
newtok.lineno = tokens[i].lineno
newtok.lexpos = tokens[i].lexpos
if tokens[i].value > dentstack[-1]:
newtok.type = 'INDENT'
dentstack += [tokens[i].value]
elif tokens[i].value < dentstack[-1]:
newtok.type = 'DEDENT'
dentstack = dentstack[:-1]
if 'type' in newtok.__dict__:
fixedtokens += [newtok]
fixedtokens += [tokens[-1]]
dedent = LexToken()
dedent.value = None
def handle_semi_error(p):
if glob.g_production_debug:
print("in handle_semi_error")
tok = p.lexer.peek()
if len(p.lexer.peeks) > 1:
prev = p.lexer.peeks[-2]
else:
prev = p.lexer.prev
cur = p.lexer.cur
if prev == None:
prev = tok
if cur == None:
cur = tok
#print("p", prev)
#print("c", cur)
#print("t", tok)
if type(prev) == list: prev = prev[0]
if type(cur) == list: cur = cur[0]
if type(tok) == list: tok = tok[0]
if p != None and type(p) != LexToken:
print(list(p))
ret = tok == None or cur == None or prev.lineno < tok.lineno
ret = ret or tok.type == "RBRACKET" or prev.type == "RBRACKET"
ret = ret or cur.type == "RBRACKET"
p2 = restricted()
if p2 != None and not (prev.type in ["RSBRACKET", "RPAREN"] and restrict_prev() == None):
ret = False
p = p2
print(prev.type, cur.type, p2, restrict_prev())
print("didn't handle semi error")
glob.g_line = p.lineno
glob.g_lexpos = p.lexpos
#print_err(p)
if ret and not glob.g_tried_semi:
#"""
t = LexToken()
t.type = "SEMI"
t.value = ";"
t.lineno = cur.lineno
t.lexpos = cur.lexpos
#"""
p.lexer.push(p.lexer.cur)
p.lexer.push(t)
yacc.errok()
glob.g_error = False
glob.g_tried_semi = True
else:
ret = False
glob.g_error = True
glob.g_error_pre = p
#for l in prodname_log[-5:-1]:
# print(l)
#print("a real error occurred 2!?")
#print_err(p)
return ret
def p_error(p):
"""
print(p.lexer.prev.lineno, p.lineno)
if p.lexer.prev.lineno < p.lineno or p.type == "RBRACKET":
yacc.errok()
return
"""
if glob.g_production_debug:
if p == None:
print("in p_error")
else:
print("in p_error", p.type, p.value)
if p == None:
if not restricted() and glob.g_tried_semi == False:
t = LexToken()
t.type = "SEMI"
t.value = ";"
t.lexpos = -1
t.lineno = -1
glob.g_lexer.push(t)
glob.g_tried_semi = True
yacc.errok()
else:
sys.stderr.write(glob.g_file + ": error: unexpected end of file\n")
return
else:
glob.g_error_pre = p
if handle_semi_error(p):
t = LexToken()
t.type = "SEMI"
t.value = ";"
t.lexpos = p.lexpos
t.lineno = p.lineno
#glob.g_lexer.push(t)
#glob.g_tried_semi = True
yacc.errok()
glob.g_error = False
if glob.g_production_debug or glob.g_semi_debug:
linestr, colstr = err_find_line(p.lexer, p.lexpos);
lineno = p.lineno if type(p.lineno) == int else p.lineno(0)
sys.stdout.write("handled semicolon error : %d\n" % lineno)
sys.stdout.write(linestr+"\n")
sys.stdout.write(colstr+"\n")
return
else:
glob.g_error = True
print_err(p)
return
if glob.g_error:
print_err(glob.g_error_pre)
glob.g_error_pre = p
glob.g_error = True
try:
line = int(p.lineno)
except:
line = p.lineno(1)
try:
lexdata = p.lexer.lexer.lexdata
sline = p.lexer.lexer.lexpos
except:
lexdata = p.lexer.lexdata
sline = p.lexer.lexpos
sline = lexdata[sline-40:sline+1]
def indent_generator(toks):
"""Post process the given stream of tokens to generate INDENT/DEDENT
tokens.
Note
----
Each generated token's value is the total amount of spaces from the
beginning of the line.
The way indentation tokens are generated is similar to how it works in
python."""
stack = [0]
# Dummy token to track the token just before the current one
former = LexToken()
former.type = "NEWLINE"
former.value = "dummy"
former.lineno = 0
former.lexpos = -1
def generate_dedent(stck, tok):
amount = stck.pop(0)
return new_dedent(amount, tok)
for token in toks:
if former.type == "NEWLINE":
if token.type == "WS":
indent = len(token.value)
else:
indent = 0
if indent == stack[0]:
former = token
if indent > 0:
token = six.advance_iterator(toks)
former = token
yield token
else:
yield former
elif indent > stack[0]:
stack.insert(0, indent)
ind = new_indent(indent, token)
former = ind
yield ind
elif indent < stack[0]:
if not indent in stack:
raise ValueError("Wrong indent at line %d" % token.lineno)
while stack[0] > indent:
former = generate_dedent(stack, token)
yield former
if stack[0] > 0:
former = six.advance_iterator(toks)
yield former
else:
former = token
yield token
else:
former = token
yield token
# Generate additional DEDENT so that the number of INDENT/DEDENT always
# match
while len(stack) > 1:
former = generate_dedent(stack, token)
yield former
def parse(self):
i = 0
data = self.lexdata
states = self.bracketstates
prev = self.get_prev
next = self.get_next
toks = []
tok = LexToken()
tok.type = None
tok.value = ""
stack = []
def escape(i1):
if i1 == None:
i1 = i
return prev(i1) == "\\" and prev(i1, 2) != "\\"
def inc_i(i, off=1):
for j in range(abs(off)):
if i < 0 or i >= len(data):
break
if data[i] in ["\n", "\r"]:
self.lineno += 1
self.lexpos += 1
if off < 0:
i -= 1
else:
i += 1
return i
def push(tok):
if tok.type == None:
traceback.print_stack()
print("ERROR: None token!")
return
tok.lineno = self.lineno
tok.lexpos = self.lexpos
tok.lexer = self
toks.append(tok)
self.tokens.append(tok)
# print(tok)
def newtok(tok, ttype=None):
if tok.type != ttype and (ttype != None or tok.value != ""):
if tok.type != None:
push(tok)
tok = LexToken()
tok.type = ttype
tok.value = ""
return tok
in_set = 0
while i < len(data):
cp = prev(i)
cc = data[i]
cn = next(i)
handled = False
if not escape(i):
if cc == "$":
tok = newtok(tok)
if cn == "{":
tok.type = "LBK"
tok.value = "${"
i = inc_i(i)
in_set += 1
for k in states.keys():
states[k].append(0)
else:
tok.type = "SPECIAL"
tok.value = "$"
handled = True
elif cc == "}" and cn == "$":
tok = newtok(tok)
tok.type = "RBK"
tok.value = "$}"
i = inc_i(i)
in_set -= 1
for k in states.keys():
states[k].pop(-1)
handled = True
elif cp == "*" and cn == "$":
tok = newtok(tok)
tok.type = "STAR"
tok.value = "*"
i = inc_i(i)
handled = True
elif cp == "^" and cn == "$":
tok = newtok(tok)
tok.type = "NOT"
tok.value = "^"
i = inc_i(i)
handled = True
elif cp == "|" and cn == "$":
tok = newtok(tok)
tok.type = "OR"
tok.value = "|"
i = inc_i(i)
handled = True
elif cc == "," and in_set:
k = 0
for t in self.bracketstates.keys():
s = self.bracketstates[t]
if s[-1] < 0:
# print(t, prev(i, 2), cp, cc, cn, "end")
pass
k += s[-1]
# print(k)
if k == 0:
tok = newtok(tok)
tok.type = "COMMA"
tok.value = ","
handled = True
if not handled and in_set > 0:
if cc in self.bracketstates:
states[cc][-1] += 1
elif cc in self.bracket_endchars:
states[self.bracket_endchars[cc]][-1] -= 1
def is_word_char(cc):
return re_word_pat.match(cc) != None
if not handled:
cp = prev(i)
if cp == "$" and tok.type not in ["WORD", "CODE"] and is_word_char(cc):
tok = newtok(tok)
tok.type = "WORD"
tok.value = cc
while i < len(data) and re_word_pat.match(tok.value).span() == (0, len(tok.value)):
i = inc_i(i)
cc = data[i]
tok.value += cc
i = inc_i(i, -1)
tok.value = tok.value[:-1]
tok = newtok(tok)
else:
tok = newtok(tok, "CODE")
tok.value += cc
i = inc_i(i)
if tok.type != None:
push(tok)
