mirror of
https://github.com/xonsh/xonsh.git
synced 2025-03-05 17:00:58 +01:00
refactoring per suggestions
This commit is contained in:
adam j hartz
parent
667c3695e4
commit
66292854ec
2 changed files with 58 additions and 97 deletions
120
xonsh/lexer.py
120
xonsh/lexer.py
|
|
@ -43,6 +43,7 @@ for (op, type) in op_map.items():
|
|||
token_map[tokenize.NAME] = 'NAME'
|
||||
token_map[tokenize.NUMBER] = 'NUMBER'
|
||||
token_map[tokenize.STRING] = 'STRING'
|
||||
token_map[tokenize.NEWLINE] = 'NEWLINE'
|
||||
|
||||
|
||||
def handle_indent(state, token, stream):
|
||||
|
|
@ -55,28 +56,17 @@ def handle_indent(state, token, stream):
|
|||
state['indents'].append(level)
|
||||
yield _new_token('INDENT', token.string, token.start)
|
||||
|
||||
try:
|
||||
n = next(stream)
|
||||
except:
|
||||
n = None
|
||||
if n is not None:
|
||||
if n.type != tokenize.ENDMARKER:
|
||||
for i in handle_token(state, n, stream):
|
||||
yield i
|
||||
|
||||
|
||||
def handle_dollar(state, token, stream):
|
||||
try:
|
||||
n = next(stream)
|
||||
except:
|
||||
n = next(stream, None)
|
||||
|
||||
if n is None:
|
||||
m = "missing token after $"
|
||||
yield _new_token("ERRORTOKEN", m, token.start)
|
||||
|
||||
if n.start != token.end:
|
||||
elif n.start != token.end:
|
||||
m = "unexpected whitespace after $"
|
||||
yield _new_token("ERRORTOKEN", m, token.start)
|
||||
|
||||
if n.type == tokenize.NAME:
|
||||
elif n.type == tokenize.NAME:
|
||||
state['last'] = n
|
||||
yield _new_token('DOLLAR_NAME', '$' + n.string, token.start)
|
||||
elif n.type == tokenize.OP and n.string == '(':
|
||||
|
|
@ -98,46 +88,39 @@ def handle_dollar(state, token, stream):
|
|||
|
||||
|
||||
def handle_at(state, token, stream):
|
||||
try:
|
||||
n = next(stream)
|
||||
except:
|
||||
n = next(stream, None)
|
||||
|
||||
if n is None:
|
||||
state['last'] = token
|
||||
m = "missing token after @"
|
||||
yield _new_token("ERRORTOKEN", m, token.start)
|
||||
|
||||
if n.type == tokenize.OP and n.string == '(' and \
|
||||
elif n.type == tokenize.OP and n.string == '(' and \
|
||||
n.start == token.end:
|
||||
state['pymode'].append(True)
|
||||
yield _new_token('AT_LPAREN', '@(', token.start)
|
||||
state['last'] = n
|
||||
yield _new_token('AT_LPAREN', '@(', token.start)
|
||||
else:
|
||||
yield _new_token('AT', '@', token.start)
|
||||
state['last'] = token
|
||||
for i in handle_token(state, n, stream):
|
||||
yield i
|
||||
yield _new_token('AT', '@', token.start)
|
||||
yield from handle_token(state, n, stream)
|
||||
|
||||
|
||||
def handle_question(state, token, stream):
|
||||
try:
|
||||
n = next(stream)
|
||||
except:
|
||||
n = None
|
||||
n = next(stream, None)
|
||||
|
||||
if n.type == tokenize.ERRORTOKEN and n.string == '?' and \
|
||||
n.start == token.end:
|
||||
yield _new_token('DOUBLE_QUESTION', '??', token.start)
|
||||
if n is not None and n.type == tokenize.ERRORTOKEN and \
|
||||
n.string == '?' and n.start == token.end:
|
||||
state['last'] = n
|
||||
yield _new_token('DOUBLE_QUESTION', '??', token.start)
|
||||
else:
|
||||
yield _new_token('QUESTION', '?', token.start)
|
||||
state['last'] = token
|
||||
for i in handle_token(state, n, stream):
|
||||
yield i
|
||||
yield _new_token('QUESTION', '?', token.start)
|
||||
if n is not None:
|
||||
yield from handle_token(state, n, stream)
|
||||
|
||||
|
||||
def handle_backtick(state, token, stream):
|
||||
try:
|
||||
n = next(stream)
|
||||
except:
|
||||
n = None
|
||||
n = next(stream, None)
|
||||
|
||||
found_match = False
|
||||
sofar = '`'
|
||||
|
|
@ -151,29 +134,15 @@ def handle_backtick(state, token, stream):
|
|||
except:
|
||||
n = None
|
||||
if found_match:
|
||||
yield _new_token('REGEXPATH', sofar, token.start)
|
||||
state['last'] = n
|
||||
yield _new_token('REGEXPATH', sofar, token.start)
|
||||
else:
|
||||
state['last'] = token
|
||||
e = "Could not find matching backtick for regex on line {0}"
|
||||
m = e.format(token.start[0])
|
||||
yield _new_token("ERRORTOKEN", m, token.start)
|
||||
|
||||
|
||||
def handle_newline(state, token, stream):
|
||||
try:
|
||||
n = next(stream)
|
||||
except:
|
||||
n = None
|
||||
|
||||
yield _new_token('NEWLINE', '\n', token.start)
|
||||
state['last'] = token
|
||||
|
||||
if n is not None:
|
||||
if n.type != tokenize.ENDMARKER:
|
||||
for i in handle_token(state, n, stream):
|
||||
yield i
|
||||
|
||||
|
||||
def handle_lparen(state, token, stream):
|
||||
state['pymode'].append(True)
|
||||
state['last'] = token
|
||||
|
|
@ -213,15 +182,19 @@ def handle_rbracket(state, token, stream):
|
|||
def handle_error_space(state, token, stream):
|
||||
if not state['pymode'][-1]:
|
||||
state['last'] = token
|
||||
yield _new_token('WS', ' ', token.start)
|
||||
yield _new_token('WS', token.string, token.start)
|
||||
else:
|
||||
yield from []
|
||||
|
||||
|
||||
def handle_ignore(state, token, stream):
|
||||
yield from []
|
||||
|
||||
|
||||
special_handlers = {
|
||||
tokenize.ENCODING: lambda s, t, st: [],
|
||||
tokenize.COMMENT: lambda s, t, st: [],
|
||||
tokenize.ENDMARKER: lambda s, t, st: [],
|
||||
tokenize.NEWLINE: handle_newline,
|
||||
tokenize.COMMENT: handle_ignore,
|
||||
tokenize.ENCODING: handle_ignore,
|
||||
tokenize.ENDMARKER: handle_ignore,
|
||||
(tokenize.OP, '('): handle_lparen,
|
||||
(tokenize.OP, ')'): handle_rparen,
|
||||
(tokenize.OP, '['): handle_lbracket,
|
||||
|
|
@ -253,25 +226,22 @@ def handle_token(state, token, stream):
|
|||
state['last'] = token
|
||||
yield _new_token(token_map[typ], st, token.start)
|
||||
elif (typ, st) in special_handlers:
|
||||
for i in special_handlers[(typ, st)](state, token, stream):
|
||||
yield i
|
||||
yield from special_handlers[(typ, st)](state, token, stream)
|
||||
elif typ in special_handlers:
|
||||
for i in special_handlers[typ](state, token, stream):
|
||||
yield i
|
||||
yield from special_handlers[typ](state, token, stream)
|
||||
else:
|
||||
m = "Unexpected token: {0}".format(token)
|
||||
yield _new_token("ERRORTOKEN", m, token.start)
|
||||
|
||||
|
||||
def preprocess_tokens(tokstream):
|
||||
tokstream = clear_NL(tokstream)
|
||||
tokstream = clear_nl(tokstream)
|
||||
state = {'indents': [0], 'pymode': [True], 'last': None}
|
||||
for token in tokstream:
|
||||
for i in handle_token(state, token, tokstream):
|
||||
yield i
|
||||
yield from handle_token(state, token, tokstream)
|
||||
|
||||
|
||||
def clear_NL(tokstream):
|
||||
def clear_nl(tokstream):
|
||||
for i in tokstream:
|
||||
if i.type != tokenize.NL:
|
||||
yield i
|
||||
|
|
@ -284,9 +254,9 @@ def single_error(exc):
|
|||
|
||||
def tok(s):
|
||||
try:
|
||||
return iter(tokenize.tokenize(BytesIO(s.encode('utf-8')).readline))
|
||||
return tokenize.tokenize(BytesIO(s.encode('utf-8')).readline)
|
||||
except Exception as e:
|
||||
return iter(single_error(e))
|
||||
return single_error(e)
|
||||
|
||||
|
||||
# synthesize a new PLY token
|
||||
|
|
@ -298,8 +268,7 @@ def _new_token(type, value, pos):
|
|||
return o
|
||||
|
||||
|
||||
def anyof(*regexes):
|
||||
return '(' + '|'.join(regexes) + ')'
|
||||
COMMENT_REGEX = re.compile(r'#.*')
|
||||
|
||||
|
||||
class Lexer(object):
|
||||
|
|
@ -337,16 +306,13 @@ class Lexer(object):
|
|||
|
||||
def input(self, s):
|
||||
"""Calls the lexer on the string s."""
|
||||
s = re.sub(r'#.*', '', s)
|
||||
s = re.sub(COMMENT_REGEX, '', s)
|
||||
self.token_stream = preprocess_tokens(tok(s))
|
||||
|
||||
def token(self):
|
||||
"""Retrieves the next token."""
|
||||
try:
|
||||
self.last = next(self.token_stream)
|
||||
self.last = next(self.token_stream, None)
|
||||
return self.last
|
||||
except StopIteration:
|
||||
return None
|
||||
|
||||
def __iter__(self):
|
||||
t = self.token()
|
||||
|
|
|
|||
|
|
@ -335,14 +335,10 @@ class Parser(object):
|
|||
|
||||
@property
|
||||
def lineno(self):
|
||||
try:
|
||||
return self.lexer.last.lineno
|
||||
except:
|
||||
if self.lexer.last is None:
|
||||
return 0
|
||||
|
||||
@lineno.setter
|
||||
def lineno(self, value):
|
||||
pass
|
||||
else:
|
||||
return self.lexer.last.lineno
|
||||
|
||||
@property
|
||||
def col(self):
|
||||
|
|
@ -412,15 +408,12 @@ class Parser(object):
|
|||
"""newline_or_stmt : NEWLINE
|
||||
| stmt
|
||||
"""
|
||||
if p[1] == '\n':
|
||||
self.lineno += 1
|
||||
p[0] = p[1]
|
||||
|
||||
def p_newlines(self, p):
|
||||
"""newlines : NEWLINE
|
||||
| newlines NEWLINE
|
||||
"""
|
||||
self.lineno += 1
|
||||
p[0] = p[1] if len(p) == 2 else p[1] + p[2]
|
||||
|
||||
def p_eval_input(self, p):
|
||||
|
|
@ -470,7 +463,6 @@ class Parser(object):
|
|||
else:
|
||||
p0 = ast.Call(func=name, lineno=self.lineno, col_offset=self.col,
|
||||
**p3)
|
||||
self.lineno += 1 # needs to be at the end
|
||||
p[0] = p0
|
||||
|
||||
def p_decorators(self, p):
|
||||
|
|
@ -1208,8 +1200,6 @@ class Parser(object):
|
|||
| NEWLINE INDENT stmt_list DEDENT
|
||||
"""
|
||||
p[0] = p[1] if len(p) == 2 else p[3]
|
||||
if len(p) < 4:
|
||||
self.lineno += 1 # needs to be at the end
|
||||
|
||||
def p_test(self, p):
|
||||
"""test : or_test
|
||||
|
|
@ -1520,8 +1510,14 @@ class Parser(object):
|
|||
bt = '`'
|
||||
if isinstance(p1, (ast.Num, ast.Str, ast.Bytes)):
|
||||
pass
|
||||
elif (p1 == 'True') or (p1 == 'False') or (p1 == 'None'):
|
||||
p1 = ast.NameConstant(value=eval(p1), lineno=self.lineno,
|
||||
elif p1 == 'True':
|
||||
p1 = ast.NameConstant(value=True, lineno=self.lineno,
|
||||
col_offset=self.col)
|
||||
elif p1 == 'False':
|
||||
p1 = ast.NameConstant(value=False, lineno=self.lineno,
|
||||
col_offset=self.col)
|
||||
elif p1 == 'None':
|
||||
p1 = ast.NameConstant(value=None, lineno=self.lineno,
|
||||
col_offset=self.col)
|
||||
elif p1 == '...':
|
||||
p1 = ast.Ellipsis(lineno=self.lineno, col_offset=self.col)
|
||||
|
|
@ -2066,7 +2062,7 @@ class Parser(object):
|
|||
lineno=self.lineno, col=self.col)
|
||||
p0._cliarg_action = 'extend'
|
||||
elif p1.startswith('$'):
|
||||
p0 = self._envvar_by_name(p[1][1:], lineno=self.lineno, col=self.col)
|
||||
p0 = self._envvar_by_name(p1[1:], lineno=self.lineno, col=self.col)
|
||||
p0._cliarg_action = 'ensure_list'
|
||||
else:
|
||||
p0._cliarg_action = 'append'
|
||||
|
|
@ -2149,8 +2145,7 @@ class Parser(object):
|
|||
self._parse_error('no further code', None)
|
||||
elif p.type == 'ERRORTOKEN':
|
||||
self._parse_error(p.value,
|
||||
self.currloc(lineno=p.lineno,
|
||||
column=p.lexpos))
|
||||
self.currloc(lineno=p.lineno, column=p.lexpos))
|
||||
else:
|
||||
msg = 'code: {0}'.format(p.value),
|
||||
self._parse_error(msg, self.currloc(lineno=p.lineno,
|
||||
|
|
|
|||
Loading…
Add table
Reference in a new issue