1
0
forked from VimPlug/jedi

Rework the parser so we can use arbitrary start nodes of the syntax.

This also includes a rework for error recovery in the parser. This is now just possible for file_input parsing, which means for full files.
Includes also a refactoring of the tokenizer. No more do we have to add an additional newline, because it now works correctly (removes certain confusion.
This commit is contained in:
Dave Halter
2015-12-20 22:21:47 +01:00
parent 9a93d599da
commit c4906e0e3f
22 changed files with 246 additions and 198 deletions

View File

@@ -4,7 +4,7 @@ Python 2.X)
"""
import jedi
from jedi._compatibility import u
from jedi.parser import Parser, load_grammar
from jedi.parser import ParserWithRecovery, load_grammar
from .. import helpers
@@ -12,7 +12,7 @@ def test_explicit_absolute_imports():
"""
Detect modules with ``from __future__ import absolute_import``.
"""
parser = Parser(load_grammar(), u("from __future__ import absolute_import"), "test.py")
parser = ParserWithRecovery(load_grammar(), u("from __future__ import absolute_import"), "test.py")
assert parser.module.has_explicit_absolute_import
@@ -20,7 +20,7 @@ def test_no_explicit_absolute_imports():
"""
Detect modules without ``from __future__ import absolute_import``.
"""
parser = Parser(load_grammar(), u("1"), "test.py")
parser = ParserWithRecovery(load_grammar(), u("1"), "test.py")
assert not parser.module.has_explicit_absolute_import
@@ -30,7 +30,7 @@ def test_dont_break_imports_without_namespaces():
assume that all imports have non-``None`` namespaces.
"""
src = u("from __future__ import absolute_import\nimport xyzzy")
parser = Parser(load_grammar(), src, "test.py")
parser = ParserWithRecovery(load_grammar(), src, "test.py")
assert parser.module.has_explicit_absolute_import

View File

@@ -35,3 +35,20 @@ def test_simple_annotations():
annot('')""")
assert [d.name for d in jedi.Script(source, ).goto_definitions()] == ['int']
@pytest.mark.skipif('sys.version_info[0] < 3')
@pytest.mark.parametrize('reference', [
'assert 1',
'1',
'lambda: 3',
'def x(): pass',
'1, 2',
r'1\n'
])
def test_illegal_forward_references(reference):
source = """
def foo(bar: "%s"):
bar""" % reference
assert not jedi.Script(source).goto_definitions()

View File

@@ -7,7 +7,7 @@ from jedi.evaluate.sys_path import (_get_parent_dir_with_file,
sys_path_with_modifications,
_check_module)
from jedi.evaluate import Evaluator
from jedi.parser import Parser, load_grammar
from jedi.parser import ParserWithRecovery, load_grammar
from ..helpers import cwd_at
@@ -37,7 +37,7 @@ def test_append_on_non_sys_path():
d = Dummy()
d.path.append('foo')"""))
grammar = load_grammar()
p = Parser(grammar, SRC)
p = ParserWithRecovery(grammar, SRC)
paths = _check_module(Evaluator(grammar), p.module)
assert len(paths) > 0
assert 'foo' not in paths
@@ -48,7 +48,7 @@ def test_path_from_invalid_sys_path_assignment():
import sys
sys.path = 'invalid'"""))
grammar = load_grammar()
p = Parser(grammar, SRC)
p = ParserWithRecovery(grammar, SRC)
paths = _check_module(Evaluator(grammar), p.module)
assert len(paths) > 0
assert 'invalid' not in paths
@@ -60,7 +60,7 @@ def test_sys_path_with_modifications():
import os
"""))
grammar = load_grammar()
p = Parser(grammar, SRC)
p = ParserWithRecovery(grammar, SRC)
p.module.path = os.path.abspath(os.path.join(os.curdir, 'module_name.py'))
paths = sys_path_with_modifications(Evaluator(grammar), p.module)
assert '/tmp/.buildout/eggs/important_package.egg' in paths
@@ -83,7 +83,7 @@ def test_path_from_sys_path_assignment():
if __name__ == '__main__':
sys.exit(important_package.main())"""))
grammar = load_grammar()
p = Parser(grammar, SRC)
p = ParserWithRecovery(grammar, SRC)
paths = _check_module(Evaluator(grammar), p.module)
assert 1 not in paths
assert '/home/test/.buildout/eggs/important_package.egg' in paths

View File

@@ -5,14 +5,14 @@ import sys
import pytest
from jedi._compatibility import unicode
from jedi.parser import Parser, load_grammar
from jedi.parser import ParserWithRecovery, load_grammar
from jedi.evaluate import sys_path, Evaluator
def test_paths_from_assignment():
def paths(src):
grammar = load_grammar()
stmt = Parser(grammar, unicode(src)).module.statements[0]
stmt = ParserWithRecovery(grammar, unicode(src)).module.statements[0]
return set(sys_path._paths_from_assignment(Evaluator(grammar), stmt))
assert paths('sys.path[0:0] = ["a"]') == set(['a'])

View File

@@ -1,11 +1,11 @@
from jedi._compatibility import u
from jedi.parser import Parser, load_grammar
from jedi.parser import ParserWithRecovery, load_grammar
def test_basic_parsing():
def compare(string):
"""Generates the AST object and then regenerates the code."""
assert Parser(load_grammar(), string).module.get_code() == string
assert ParserWithRecovery(load_grammar(), string).module.get_code() == string
compare(u('\na #pass\n'))
compare(u('wblabla* 1\t\n'))

View File

@@ -3,7 +3,7 @@ import difflib
import pytest
from jedi._compatibility import u
from jedi.parser import Parser, load_grammar
from jedi.parser import ParserWithRecovery, load_grammar
code_basic_features = u('''
"""A mod docstring"""
@@ -44,7 +44,7 @@ def diff_code_assert(a, b, n=4):
def test_basic_parsing():
"""Validate the parsing features"""
prs = Parser(load_grammar(), code_basic_features)
prs = ParserWithRecovery(load_grammar(), code_basic_features)
diff_code_assert(
code_basic_features,
prs.module.get_code()
@@ -53,7 +53,7 @@ def test_basic_parsing():
def test_operators():
src = u('5 * 3')
prs = Parser(load_grammar(), src)
prs = ParserWithRecovery(load_grammar(), src)
diff_code_assert(src, prs.module.get_code())
@@ -82,7 +82,7 @@ def method_with_docstring():
"""class docstr"""
pass
''')
assert Parser(load_grammar(), s).module.get_code() == s
assert ParserWithRecovery(load_grammar(), s).module.get_code() == s
def test_end_newlines():
@@ -92,7 +92,7 @@ def test_end_newlines():
line the parser needs.
"""
def test(source, end_pos):
module = Parser(load_grammar(), u(source)).module
module = ParserWithRecovery(load_grammar(), u(source)).module
assert module.get_code() == source
assert module.end_pos == end_pos

View File

@@ -3,7 +3,7 @@ import sys
import jedi
from jedi._compatibility import u, is_py3
from jedi.parser import Parser, load_grammar
from jedi.parser import ParserWithRecovery, load_grammar
from jedi.parser.user_context import UserContextParser
from jedi.parser import tree as pt
from textwrap import dedent
@@ -23,7 +23,7 @@ def test_user_statement_on_import():
class TestCallAndName():
def get_call(self, source):
# Get the simple_stmt and then the first one.
simple_stmt = Parser(load_grammar(), u(source)).module.children[0]
simple_stmt = ParserWithRecovery(load_grammar(), u(source)).module.children[0]
return simple_stmt.children[0]
def test_name_and_call_positions(self):
@@ -58,7 +58,7 @@ class TestCallAndName():
class TestSubscopes():
def get_sub(self, source):
return Parser(load_grammar(), u(source)).module.subscopes[0]
return ParserWithRecovery(load_grammar(), u(source)).module.subscopes[0]
def test_subscope_names(self):
name = self.get_sub('class Foo: pass').name
@@ -74,7 +74,7 @@ class TestSubscopes():
class TestImports():
def get_import(self, source):
return Parser(load_grammar(), source).module.imports[0]
return ParserWithRecovery(load_grammar(), source).module.imports[0]
def test_import_names(self):
imp = self.get_import(u('import math\n'))
@@ -89,13 +89,13 @@ class TestImports():
def test_module():
module = Parser(load_grammar(), u('asdf'), 'example.py').module
module = ParserWithRecovery(load_grammar(), u('asdf'), 'example.py').module
name = module.name
assert str(name) == 'example'
assert name.start_pos == (1, 0)
assert name.end_pos == (1, 7)
module = Parser(load_grammar(), u('asdf')).module
module = ParserWithRecovery(load_grammar(), u('asdf')).module
name = module.name
assert str(name) == ''
assert name.start_pos == (1, 0)
@@ -108,7 +108,7 @@ def test_end_pos():
def func():
y = None
'''))
parser = Parser(load_grammar(), s)
parser = ParserWithRecovery(load_grammar(), s)
scope = parser.module.subscopes[0]
assert scope.start_pos == (3, 0)
assert scope.end_pos == (5, 0)
@@ -121,7 +121,7 @@ def test_carriage_return_statements():
# this is a namespace package
'''))
source = source.replace('\n', '\r\n')
stmt = Parser(load_grammar(), source).module.statements[0]
stmt = ParserWithRecovery(load_grammar(), source).module.statements[0]
assert '#' not in stmt.get_code()
@@ -129,7 +129,7 @@ def test_incomplete_list_comprehension():
""" Shouldn't raise an error, same bug as #418. """
# With the old parser this actually returned a statement. With the new
# parser only valid statements generate one.
assert Parser(load_grammar(), u('(1 for def')).module.statements == []
assert ParserWithRecovery(load_grammar(), u('(1 for def')).module.statements == []
def test_hex_values_in_docstring():
@@ -141,7 +141,7 @@ def test_hex_values_in_docstring():
return 1
'''
doc = Parser(load_grammar(), dedent(u(source))).module.subscopes[0].raw_doc
doc = ParserWithRecovery(load_grammar(), dedent(u(source))).module.subscopes[0].raw_doc
if is_py3:
assert doc == '\xff'
else:
@@ -160,7 +160,7 @@ def test_error_correction_with():
def test_newline_positions():
endmarker = Parser(load_grammar(), u('a\n')).module.children[-1]
endmarker = ParserWithRecovery(load_grammar(), u('a\n')).module.children[-1]
assert endmarker.end_pos == (2, 0)
new_line = endmarker.get_previous()
assert new_line.start_pos == (1, 1)
@@ -174,7 +174,7 @@ def test_end_pos_error_correction():
end_pos, even if something breaks in the parser (error correction).
"""
s = u('def x():\n .')
m = Parser(load_grammar(), s).module
m = ParserWithRecovery(load_grammar(), s).module
func = m.children[0]
assert func.type == 'funcdef'
# This is not exactly correct, but ok, because it doesn't make a difference
@@ -191,7 +191,7 @@ def test_param_splitting():
def check(src, result):
# Python 2 tuple params should be ignored for now.
grammar = load_grammar('grammar%s.%s' % sys.version_info[:2])
m = Parser(grammar, u(src)).module
m = ParserWithRecovery(grammar, u(src)).module
if is_py3:
assert not m.subscopes
else:
@@ -211,5 +211,5 @@ def test_unicode_string():
def test_backslash_dos_style():
grammar = load_grammar()
m = Parser(grammar, u('\\\r\n')).module
m = ParserWithRecovery(grammar, u('\\\r\n')).module
assert m

View File

@@ -5,7 +5,7 @@ from textwrap import dedent
import pytest
from jedi._compatibility import u, unicode
from jedi.parser import Parser, load_grammar
from jedi.parser import ParserWithRecovery, load_grammar
from jedi.parser import tree as pt
@@ -27,7 +27,7 @@ class TestsFunctionAndLambdaParsing(object):
@pytest.fixture(params=FIXTURES)
def node(self, request):
parsed = Parser(load_grammar(), dedent(u(request.param[0])))
parsed = ParserWithRecovery(load_grammar(), dedent(u(request.param[0])))
request.keywords['expected'] = request.param[1]
return parsed.module.subscopes[0]

View File

@@ -7,7 +7,7 @@ import pytest
from jedi._compatibility import u, is_py3
from jedi.parser.token import NAME, OP, NEWLINE, STRING, INDENT
from jedi.parser import Parser, load_grammar, tokenize
from jedi.parser import ParserWithRecovery, load_grammar, tokenize
from ..helpers import unittest
@@ -15,7 +15,7 @@ from ..helpers import unittest
class TokenTest(unittest.TestCase):
def test_end_pos_one_line(self):
parsed = Parser(load_grammar(), dedent(u('''
parsed = ParserWithRecovery(load_grammar(), dedent(u('''
def testit():
a = "huhu"
''')))
@@ -23,7 +23,7 @@ class TokenTest(unittest.TestCase):
assert tok.end_pos == (3, 14)
def test_end_pos_multi_line(self):
parsed = Parser(load_grammar(), dedent(u('''
parsed = ParserWithRecovery(load_grammar(), dedent(u('''
def testit():
a = """huhu
asdfasdf""" + "h"
@@ -108,7 +108,7 @@ class TokenTest(unittest.TestCase):
]
for s in string_tokens:
parsed = Parser(load_grammar(), u('''a = %s\n''' % s))
parsed = ParserWithRecovery(load_grammar(), u('''a = %s\n''' % s))
simple_stmt = parsed.module.children[0]
expr_stmt = simple_stmt.children[0]
assert len(expr_stmt.children) == 3

View File

@@ -15,7 +15,7 @@ from jedi._compatibility import u
from jedi import Script
from jedi import api
from jedi.evaluate import imports
from jedi.parser import Parser, load_grammar
from jedi.parser import ParserWithRecovery, load_grammar
#jedi.set_debug_function()
@@ -102,7 +102,7 @@ class TestRegression(TestCase):
def test_end_pos_line(self):
# jedi issue #150
s = u("x()\nx( )\nx( )\nx ( )")
parser = Parser(load_grammar(), s)
parser = ParserWithRecovery(load_grammar(), s)
for i, s in enumerate(parser.module.statements):
assert s.end_pos == (i + 1, i + 3)