Merge branch 'issue-619'

This commit is contained in:
BJ Dierkes 2022-07-02 21:16:29 -05:00
commit a352fb3a53
52 changed files with 8602 additions and 6671 deletions

View File

@ -4,7 +4,8 @@
Bugs:
- None
- `[cement.cli]` Cement CLI Broken on Python 3.10
- [Issue #619](https://github.com/datafolklabs/cement/issues/619)
Features:

View File

@ -1,4 +1,4 @@
FROM python:3.8-alpine
FROM python:3.10-alpine
LABEL MAINTAINER="BJ Dierkes <derks@datafolklabs.com>"
ENV PS1="\[\e[0;33m\]|> cement <| \[\e[1;35m\]\W\[\e[0m\] \[\e[0m\]# "
WORKDIR /src

View File

@ -0,0 +1,28 @@
Copyright 2007 Pallets
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
1. Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
2. Redistributions in binary form must reproduce the above copyright
notice, this list of conditions and the following disclaimer in the
documentation and/or other materials provided with the distribution.
3. Neither the name of the copyright holder nor the names of its
contributors may be used to endorse or promote products derived from
this software without specific prior written permission.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED
TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

View File

@ -1,83 +1,37 @@
# -*- coding: utf-8 -*-
"""Jinja is a template engine written in pure Python. It provides a
non-XML syntax that supports inline expressions and an optional
sandboxed environment.
"""
jinja2
~~~~~~
from .bccache import BytecodeCache as BytecodeCache
from .bccache import FileSystemBytecodeCache as FileSystemBytecodeCache
from .bccache import MemcachedBytecodeCache as MemcachedBytecodeCache
from .environment import Environment as Environment
from .environment import Template as Template
from .exceptions import TemplateAssertionError as TemplateAssertionError
from .exceptions import TemplateError as TemplateError
from .exceptions import TemplateNotFound as TemplateNotFound
from .exceptions import TemplateRuntimeError as TemplateRuntimeError
from .exceptions import TemplatesNotFound as TemplatesNotFound
from .exceptions import TemplateSyntaxError as TemplateSyntaxError
from .exceptions import UndefinedError as UndefinedError
from .loaders import BaseLoader as BaseLoader
from .loaders import ChoiceLoader as ChoiceLoader
from .loaders import DictLoader as DictLoader
from .loaders import FileSystemLoader as FileSystemLoader
from .loaders import FunctionLoader as FunctionLoader
from .loaders import ModuleLoader as ModuleLoader
from .loaders import PackageLoader as PackageLoader
from .loaders import PrefixLoader as PrefixLoader
from .runtime import ChainableUndefined as ChainableUndefined
from .runtime import DebugUndefined as DebugUndefined
from .runtime import make_logging_undefined as make_logging_undefined
from .runtime import StrictUndefined as StrictUndefined
from .runtime import Undefined as Undefined
from .utils import clear_caches as clear_caches
from .utils import is_undefined as is_undefined
from .utils import pass_context as pass_context
from .utils import pass_environment as pass_environment
from .utils import pass_eval_context as pass_eval_context
from .utils import select_autoescape as select_autoescape
Jinja2 is a template engine written in pure Python. It provides a
Django inspired non-XML syntax but supports inline expressions and
an optional sandboxed environment.
Nutshell
--------
Here a small example of a Jinja2 template::
{% extends 'base.html' %}
{% block title %}Memberlist{% endblock %}
{% block content %}
<ul>
{% for user in users %}
<li><a href="{{ user.url }}">{{ user.username }}</a></li>
{% endfor %}
</ul>
{% endblock %}
:copyright: (c) 2017 by the Jinja Team.
:license: BSD, see LICENSE for more details.
"""
__docformat__ = 'restructuredtext en'
__version__ = '2.10'
# high level interface
from jinja2.environment import Environment, Template
# loaders
from jinja2.loaders import BaseLoader, FileSystemLoader, PackageLoader, \
DictLoader, FunctionLoader, PrefixLoader, ChoiceLoader, \
ModuleLoader
# bytecode caches
from jinja2.bccache import BytecodeCache, FileSystemBytecodeCache, \
MemcachedBytecodeCache
# undefined types
from jinja2.runtime import Undefined, DebugUndefined, StrictUndefined, \
make_logging_undefined
# exceptions
from jinja2.exceptions import TemplateError, UndefinedError, \
TemplateNotFound, TemplatesNotFound, TemplateSyntaxError, \
TemplateAssertionError, TemplateRuntimeError
# decorators and public utilities
from jinja2.filters import environmentfilter, contextfilter, \
evalcontextfilter
from jinja2.utils import Markup, escape, clear_caches, \
environmentfunction, evalcontextfunction, contextfunction, \
is_undefined, select_autoescape
__all__ = [
'Environment', 'Template', 'BaseLoader', 'FileSystemLoader',
'PackageLoader', 'DictLoader', 'FunctionLoader', 'PrefixLoader',
'ChoiceLoader', 'BytecodeCache', 'FileSystemBytecodeCache',
'MemcachedBytecodeCache', 'Undefined', 'DebugUndefined',
'StrictUndefined', 'TemplateError', 'UndefinedError', 'TemplateNotFound',
'TemplatesNotFound', 'TemplateSyntaxError', 'TemplateAssertionError',
'TemplateRuntimeError',
'ModuleLoader', 'environmentfilter', 'contextfilter', 'Markup', 'escape',
'environmentfunction', 'contextfunction', 'clear_caches', 'is_undefined',
'evalcontextfilter', 'evalcontextfunction', 'make_logging_undefined',
'select_autoescape',
]
def _patch_async():
from jinja2.utils import have_async_gen
if have_async_gen:
from jinja2.asyncsupport import patch_all
patch_all()
_patch_async()
del _patch_async
__version__ = "3.1.2"

View File

@ -1,99 +0,0 @@
# -*- coding: utf-8 -*-
"""
jinja2._compat
~~~~~~~~~~~~~~
Some py2/py3 compatibility support based on a stripped down
version of six so we don't have to depend on a specific version
of it.
:copyright: Copyright 2013 by the Jinja team, see AUTHORS.
:license: BSD, see LICENSE for details.
"""
import sys
PY2 = sys.version_info[0] == 2
PYPY = hasattr(sys, 'pypy_translation_info')
_identity = lambda x: x
if not PY2:
unichr = chr
range_type = range
text_type = str
string_types = (str,)
integer_types = (int,)
iterkeys = lambda d: iter(d.keys())
itervalues = lambda d: iter(d.values())
iteritems = lambda d: iter(d.items())
import pickle
from io import BytesIO, StringIO
NativeStringIO = StringIO
def reraise(tp, value, tb=None):
if value.__traceback__ is not tb:
raise value.with_traceback(tb)
raise value
ifilter = filter
imap = map
izip = zip
intern = sys.intern
implements_iterator = _identity
implements_to_string = _identity
encode_filename = _identity
else:
unichr = unichr
text_type = unicode
range_type = xrange
string_types = (str, unicode)
integer_types = (int, long)
iterkeys = lambda d: d.iterkeys()
itervalues = lambda d: d.itervalues()
iteritems = lambda d: d.iteritems()
import cPickle as pickle
from cStringIO import StringIO as BytesIO, StringIO
NativeStringIO = BytesIO
exec('def reraise(tp, value, tb=None):\n raise tp, value, tb')
from itertools import imap, izip, ifilter
intern = intern
def implements_iterator(cls):
cls.next = cls.__next__
del cls.__next__
return cls
def implements_to_string(cls):
cls.__unicode__ = cls.__str__
cls.__str__ = lambda x: x.__unicode__().encode('utf-8')
return cls
def encode_filename(filename):
if isinstance(filename, unicode):
return filename.encode('utf-8')
return filename
def with_metaclass(meta, *bases):
"""Create a base class with a metaclass."""
# This requires a bit of explanation: the basic idea is to make a
# dummy metaclass for one level of class instantiation that replaces
# itself with the actual metaclass.
class metaclass(type):
def __new__(cls, name, this_bases, d):
return meta(name, bases, d)
return type.__new__(metaclass, 'temporary_class', (), {})
try:
from urllib.parse import quote_from_bytes as url_quote
except ImportError:
from urllib import quote as url_quote

View File

@ -1,2 +1,6 @@
import re
# generated by scripts/generate_identifier_pattern.py
pattern = '·̀-ͯ·҃-֑҇-ׇֽֿׁׂׅׄؐ-ًؚ-ٰٟۖ-ۜ۟-۪ۤۧۨ-ܑۭܰ-݊ަ-ް߫-߳ࠖ-࠙ࠛ-ࠣࠥ-ࠧࠩ-࡙࠭-࡛ࣔ-ࣣ࣡-ःऺ-़ा-ॏ॑-ॗॢॣঁ-ঃ়া-ৄেৈো-্ৗৢৣਁ-ਃ਼ਾ-ੂੇੈੋ-੍ੑੰੱੵઁ-ઃ઼ા-ૅે-ૉો-્ૢૣଁ-ଃ଼ା-ୄେୈୋ-୍ୖୗୢୣஂா-ூெ-ைொ-்ௗఀ-ఃా-ౄె-ైొ-్ౕౖౢౣಁ-ಃ಼ಾ-ೄೆ-ೈೊ-್ೕೖೢೣഁ-ഃാ-ൄെ-ൈൊ-്ൗൢൣංඃ්ා-ුූෘ-ෟෲෳัิ-ฺ็-๎ັິ-ູົຼ່-ໍ༹༘༙༵༷༾༿ཱ-྄྆྇ྍ-ྗྙ-ྼ࿆ါ-ှၖ-ၙၞ-ၠၢ-ၤၧ-ၭၱ-ၴႂ-ႍႏႚ-ႝ፝-፟ᜒ-᜔ᜲ-᜴ᝒᝓᝲᝳ឴-៓៝᠋-᠍ᢅᢆᢩᤠ-ᤫᤰ-᤻ᨗ-ᨛᩕ-ᩞ᩠-᩿᩼᪰-᪽ᬀ-ᬄ᬴-᭄᭫-᭳ᮀ-ᮂᮡ-ᮭ᯦-᯳ᰤ-᰷᳐-᳔᳒-᳨᳭ᳲ-᳴᳸᳹᷀-᷵᷻-᷿‿⁀⁔⃐-⃥⃜⃡-⃰℘℮⳯-⵿⳱ⷠ-〪ⷿ-゙゚〯꙯ꙴ-꙽ꚞꚟ꛰꛱ꠂ꠆ꠋꠣ-ꠧꢀꢁꢴ-ꣅ꣠-꣱ꤦ-꤭ꥇ-꥓ꦀ-ꦃ꦳-꧀ꧥꨩ-ꨶꩃꩌꩍꩻ-ꩽꪰꪲ-ꪴꪷꪸꪾ꪿꫁ꫫ-ꫯꫵ꫶ꯣ-ꯪ꯬꯭ﬞ︀-️︠-︯︳︴﹍-﹏_𐇽𐋠𐍶-𐍺𐨁-𐨃𐨅𐨆𐨌-𐨏𐨸-𐨿𐨺𐫦𐫥𑀀-𑀂𑀸-𑁆𑁿-𑂂𑂰-𑂺𑄀-𑄂𑄧-𑅳𑄴𑆀-𑆂𑆳-𑇊𑇀-𑇌𑈬-𑈷𑈾𑋟-𑋪𑌀-𑌃𑌼𑌾-𑍄𑍇𑍈𑍋-𑍍𑍗𑍢𑍣𑍦-𑍬𑍰-𑍴𑐵-𑑆𑒰-𑓃𑖯-𑖵𑖸-𑗀𑗜𑗝𑘰-𑙀𑚫-𑚷𑜝-𑜫𑰯-𑰶𑰸-𑰿𑲒-𑲧𑲩-𑲶𖫰-𖫴𖬰-𖬶𖽑-𖽾𖾏-𖾒𛲝𛲞𝅥-𝅩𝅭-𝅲𝅻-𝆂𝆅-𝆋𝆪-𝆭𝉂-𝉄𝨀-𝨶𝨻-𝩬𝩵𝪄𝪛-𝪟𝪡-𝪯𞀀-𞀆𞀈-𞀘𞀛-𞀡𞀣𞀤𞀦-𞣐𞀪-𞣖𞥄-𞥊󠄀-󠇯'
pattern = re.compile(
r"[\w·̀-ͯ·҃-֑҇-ׇֽֿׁׂׅׄؐ-ًؚ-ٰٟۖ-ۜ۟-۪ۤۧۨ-ܑۭܰ-݊ަ-ް߫-߽߳ࠖ-࠙ࠛ-ࠣࠥ-ࠧࠩ-࡙࠭-࡛࣓-ࣣ࣡-ःऺ-़ा-ॏ॑-ॗॢॣঁ-ঃ়া-ৄেৈো-্ৗৢৣ৾ਁ-ਃ਼ਾ-ੂੇੈੋ-੍ੑੰੱੵઁ-ઃ઼ા-ૅે-ૉો-્ૢૣૺ-૿ଁ-ଃ଼ା-ୄେୈୋ-୍ୖୗୢୣஂா-ூெ-ைொ-்ௗఀ-ఄా-ౄె-ైొ-్ౕౖౢౣಁ-ಃ಼ಾ-ೄೆ-ೈೊ-್ೕೖೢೣഀ-ഃ഻഼ാ-ൄെ-ൈൊ-്ൗൢൣංඃ්ා-ුූෘ-ෟෲෳัิ-ฺ็-๎ັິ-ູົຼ່-ໍ༹༘༙༵༷༾༿ཱ-྄྆྇ྍ-ྗྙ-ྼ࿆ါ-ှၖ-ၙၞ-ၠၢ-ၤၧ-ၭၱ-ၴႂ-ႍႏႚ-ႝ፝-፟ᜒ-᜔ᜲ-᜴ᝒᝓᝲᝳ឴-៓៝᠋-᠍ᢅᢆᢩᤠ-ᤫᤰ-᤻ᨗ-ᨛᩕ-ᩞ᩠-᩿᩼᪰-᪽ᬀ-ᬄ᬴-᭄᭫-᭳ᮀ-ᮂᮡ-ᮭ᯦-᯳ᰤ-᰷᳐-᳔᳒-᳨᳭ᳲ-᳴᳷-᳹᷀-᷹᷻-᷿‿⁀⁔⃐-⃥⃜⃡-⃰℘℮⳯-⵿⳱ⷠ-〪ⷿ-゙゚〯꙯ꙴ-꙽ꚞꚟ꛰꛱ꠂ꠆ꠋꠣ-ꠧꢀꢁꢴ-ꣅ꣠-꣱ꣿꤦ-꤭ꥇ-꥓ꦀ-ꦃ꦳-꧀ꧥꨩ-ꨶꩃꩌꩍꩻ-ꩽꪰꪲ-ꪴꪷꪸꪾ꪿꫁ꫫ-ꫯꫵ꫶ꯣ-ꯪ꯬꯭ﬞ︀-️︠-︯︳︴﹍-﹏_𐇽𐋠𐍶-𐍺𐨁-𐨃𐨅𐨆𐨌-𐨏𐨸-𐨿𐨺𐫦𐫥𐴤-𐽆𐴧-𐽐𑀀-𑀂𑀸-𑁆𑁿-𑂂𑂰-𑂺𑄀-𑄂𑄧-𑄴𑅅𑅆𑅳𑆀-𑆂𑆳-𑇀𑇉-𑇌𑈬-𑈷𑈾𑋟-𑋪𑌀-𑌃𑌻𑌼𑌾-𑍄𑍇𑍈𑍋-𑍍𑍗𑍢𑍣𑍦-𑍬𑍰-𑍴𑐵-𑑆𑑞𑒰-𑓃𑖯-𑖵𑖸-𑗀𑗜𑗝𑘰-𑙀𑚫-𑚷𑜝-𑜫𑠬-𑠺𑨁-𑨊𑨳-𑨹𑨻-𑨾𑩇𑩑-𑩛𑪊-𑪙𑰯-𑰶𑰸-𑰿𑲒-𑲧𑲩-𑲶𑴱-𑴶𑴺𑴼𑴽𑴿-𑵅𑵇𑶊-𑶎𑶐𑶑𑶓-𑶗𑻳-𑻶𖫰-𖫴𖬰-𖬶𖽑-𖽾𖾏-𖾒𛲝𛲞𝅥-𝅩𝅭-𝅲𝅻-𝆂𝆅-𝆋𝆪-𝆭𝉂-𝉄𝨀-𝨶𝨻-𝩬𝩵𝪄𝪛-𝪟𝪡-𝪯𞀀-𞀆𞀈-𞀘𞀛-𞀡𞀣𞀤𞀦-𞣐𞀪-𞣖𞥄-𞥊󠄀-󠇯]+" # noqa: B950
)

View File

@ -0,0 +1,84 @@
import inspect
import typing as t
from functools import WRAPPER_ASSIGNMENTS
from functools import wraps
from .utils import _PassArg
from .utils import pass_eval_context
V = t.TypeVar("V")
def async_variant(normal_func): # type: ignore
def decorator(async_func): # type: ignore
pass_arg = _PassArg.from_obj(normal_func)
need_eval_context = pass_arg is None
if pass_arg is _PassArg.environment:
def is_async(args: t.Any) -> bool:
return t.cast(bool, args[0].is_async)
else:
def is_async(args: t.Any) -> bool:
return t.cast(bool, args[0].environment.is_async)
# Take the doc and annotations from the sync function, but the
# name from the async function. Pallets-Sphinx-Themes
# build_function_directive expects __wrapped__ to point to the
# sync function.
async_func_attrs = ("__module__", "__name__", "__qualname__")
normal_func_attrs = tuple(set(WRAPPER_ASSIGNMENTS).difference(async_func_attrs))
@wraps(normal_func, assigned=normal_func_attrs)
@wraps(async_func, assigned=async_func_attrs, updated=())
def wrapper(*args, **kwargs): # type: ignore
b = is_async(args)
if need_eval_context:
args = args[1:]
if b:
return async_func(*args, **kwargs)
return normal_func(*args, **kwargs)
if need_eval_context:
wrapper = pass_eval_context(wrapper)
wrapper.jinja_async_variant = True
return wrapper
return decorator
_common_primitives = {int, float, bool, str, list, dict, tuple, type(None)}
async def auto_await(value: t.Union[t.Awaitable["V"], "V"]) -> "V":
# Avoid a costly call to isawaitable
if type(value) in _common_primitives:
return t.cast("V", value)
if inspect.isawaitable(value):
return await t.cast("t.Awaitable[V]", value)
return t.cast("V", value)
async def auto_aiter(
iterable: "t.Union[t.AsyncIterable[V], t.Iterable[V]]",
) -> "t.AsyncIterator[V]":
if hasattr(iterable, "__aiter__"):
async for item in t.cast("t.AsyncIterable[V]", iterable):
yield item
else:
for item in t.cast("t.Iterable[V]", iterable):
yield item
async def auto_to_list(
value: "t.Union[t.AsyncIterable[V], t.Iterable[V]]",
) -> t.List["V"]:
return [x async for x in auto_aiter(value)]

View File

@ -1,146 +0,0 @@
from functools import wraps
from jinja2.asyncsupport import auto_aiter
from jinja2 import filters
async def auto_to_seq(value):
seq = []
if hasattr(value, '__aiter__'):
async for item in value:
seq.append(item)
else:
for item in value:
seq.append(item)
return seq
async def async_select_or_reject(args, kwargs, modfunc, lookup_attr):
seq, func = filters.prepare_select_or_reject(
args, kwargs, modfunc, lookup_attr)
if seq:
async for item in auto_aiter(seq):
if func(item):
yield item
def dualfilter(normal_filter, async_filter):
wrap_evalctx = False
if getattr(normal_filter, 'environmentfilter', False):
is_async = lambda args: args[0].is_async
wrap_evalctx = False
else:
if not getattr(normal_filter, 'evalcontextfilter', False) and \
not getattr(normal_filter, 'contextfilter', False):
wrap_evalctx = True
is_async = lambda args: args[0].environment.is_async
@wraps(normal_filter)
def wrapper(*args, **kwargs):
b = is_async(args)
if wrap_evalctx:
args = args[1:]
if b:
return async_filter(*args, **kwargs)
return normal_filter(*args, **kwargs)
if wrap_evalctx:
wrapper.evalcontextfilter = True
wrapper.asyncfiltervariant = True
return wrapper
def asyncfiltervariant(original):
def decorator(f):
return dualfilter(original, f)
return decorator
@asyncfiltervariant(filters.do_first)
async def do_first(environment, seq):
try:
return await auto_aiter(seq).__anext__()
except StopAsyncIteration:
return environment.undefined('No first item, sequence was empty.')
@asyncfiltervariant(filters.do_groupby)
async def do_groupby(environment, value, attribute):
expr = filters.make_attrgetter(environment, attribute)
return [filters._GroupTuple(key, await auto_to_seq(values))
for key, values in filters.groupby(sorted(
await auto_to_seq(value), key=expr), expr)]
@asyncfiltervariant(filters.do_join)
async def do_join(eval_ctx, value, d=u'', attribute=None):
return filters.do_join(eval_ctx, await auto_to_seq(value), d, attribute)
@asyncfiltervariant(filters.do_list)
async def do_list(value):
return await auto_to_seq(value)
@asyncfiltervariant(filters.do_reject)
async def do_reject(*args, **kwargs):
return async_select_or_reject(args, kwargs, lambda x: not x, False)
@asyncfiltervariant(filters.do_rejectattr)
async def do_rejectattr(*args, **kwargs):
return async_select_or_reject(args, kwargs, lambda x: not x, True)
@asyncfiltervariant(filters.do_select)
async def do_select(*args, **kwargs):
return async_select_or_reject(args, kwargs, lambda x: x, False)
@asyncfiltervariant(filters.do_selectattr)
async def do_selectattr(*args, **kwargs):
return async_select_or_reject(args, kwargs, lambda x: x, True)
@asyncfiltervariant(filters.do_map)
async def do_map(*args, **kwargs):
seq, func = filters.prepare_map(args, kwargs)
if seq:
async for item in auto_aiter(seq):
yield func(item)
@asyncfiltervariant(filters.do_sum)
async def do_sum(environment, iterable, attribute=None, start=0):
rv = start
if attribute is not None:
func = filters.make_attrgetter(environment, attribute)
else:
func = lambda x: x
async for item in auto_aiter(iterable):
rv += func(item)
return rv
@asyncfiltervariant(filters.do_slice)
async def do_slice(value, slices, fill_with=None):
return filters.do_slice(await auto_to_seq(value), slices, fill_with)
ASYNC_FILTERS = {
'first': do_first,
'groupby': do_groupby,
'join': do_join,
'list': do_list,
# we intentionally do not support do_last because that would be
# ridiculous
'reject': do_reject,
'rejectattr': do_rejectattr,
'map': do_map,
'select': do_select,
'selectattr': do_selectattr,
'sum': do_sum,
'slice': do_slice,
}

View File

@ -1,256 +0,0 @@
# -*- coding: utf-8 -*-
"""
jinja2.asyncsupport
~~~~~~~~~~~~~~~~~~~
Has all the code for async support which is implemented as a patch
for supported Python versions.
:copyright: (c) 2017 by the Jinja Team.
:license: BSD, see LICENSE for more details.
"""
import sys
import asyncio
import inspect
from functools import update_wrapper
from jinja2.utils import concat, internalcode, Markup
from jinja2.environment import TemplateModule
from jinja2.runtime import LoopContextBase, _last_iteration
async def concat_async(async_gen):
rv = []
async def collect():
async for event in async_gen:
rv.append(event)
await collect()
return concat(rv)
async def generate_async(self, *args, **kwargs):
vars = dict(*args, **kwargs)
try:
async for event in self.root_render_func(self.new_context(vars)):
yield event
except Exception:
exc_info = sys.exc_info()
else:
return
yield self.environment.handle_exception(exc_info, True)
def wrap_generate_func(original_generate):
def _convert_generator(self, loop, args, kwargs):
async_gen = self.generate_async(*args, **kwargs)
try:
while 1:
yield loop.run_until_complete(async_gen.__anext__())
except StopAsyncIteration:
pass
def generate(self, *args, **kwargs):
if not self.environment.is_async:
return original_generate(self, *args, **kwargs)
return _convert_generator(self, asyncio.get_event_loop(), args, kwargs)
return update_wrapper(generate, original_generate)
async def render_async(self, *args, **kwargs):
if not self.environment.is_async:
raise RuntimeError('The environment was not created with async mode '
'enabled.')
vars = dict(*args, **kwargs)
ctx = self.new_context(vars)
try:
return await concat_async(self.root_render_func(ctx))
except Exception:
exc_info = sys.exc_info()
return self.environment.handle_exception(exc_info, True)
def wrap_render_func(original_render):
def render(self, *args, **kwargs):
if not self.environment.is_async:
return original_render(self, *args, **kwargs)
loop = asyncio.get_event_loop()
return loop.run_until_complete(self.render_async(*args, **kwargs))
return update_wrapper(render, original_render)
def wrap_block_reference_call(original_call):
@internalcode
async def async_call(self):
rv = await concat_async(self._stack[self._depth](self._context))
if self._context.eval_ctx.autoescape:
rv = Markup(rv)
return rv
@internalcode
def __call__(self):
if not self._context.environment.is_async:
return original_call(self)
return async_call(self)
return update_wrapper(__call__, original_call)
def wrap_macro_invoke(original_invoke):
@internalcode
async def async_invoke(self, arguments, autoescape):
rv = await self._func(*arguments)
if autoescape:
rv = Markup(rv)
return rv
@internalcode
def _invoke(self, arguments, autoescape):
if not self._environment.is_async:
return original_invoke(self, arguments, autoescape)
return async_invoke(self, arguments, autoescape)
return update_wrapper(_invoke, original_invoke)
@internalcode
async def get_default_module_async(self):
if self._module is not None:
return self._module
self._module = rv = await self.make_module_async()
return rv
def wrap_default_module(original_default_module):
@internalcode
def _get_default_module(self):
if self.environment.is_async:
raise RuntimeError('Template module attribute is unavailable '
'in async mode')
return original_default_module(self)
return _get_default_module
async def make_module_async(self, vars=None, shared=False, locals=None):
context = self.new_context(vars, shared, locals)
body_stream = []
async for item in self.root_render_func(context):
body_stream.append(item)
return TemplateModule(self, context, body_stream)
def patch_template():
from jinja2 import Template
Template.generate = wrap_generate_func(Template.generate)
Template.generate_async = update_wrapper(
generate_async, Template.generate_async)
Template.render_async = update_wrapper(
render_async, Template.render_async)
Template.render = wrap_render_func(Template.render)
Template._get_default_module = wrap_default_module(
Template._get_default_module)
Template._get_default_module_async = get_default_module_async
Template.make_module_async = update_wrapper(
make_module_async, Template.make_module_async)
def patch_runtime():
from jinja2.runtime import BlockReference, Macro
BlockReference.__call__ = wrap_block_reference_call(
BlockReference.__call__)
Macro._invoke = wrap_macro_invoke(Macro._invoke)
def patch_filters():
from jinja2.filters import FILTERS
from jinja2.asyncfilters import ASYNC_FILTERS
FILTERS.update(ASYNC_FILTERS)
def patch_all():
patch_template()
patch_runtime()
patch_filters()
async def auto_await(value):
if inspect.isawaitable(value):
return await value
return value
async def auto_aiter(iterable):
if hasattr(iterable, '__aiter__'):
async for item in iterable:
yield item
return
for item in iterable:
yield item
class AsyncLoopContext(LoopContextBase):
def __init__(self, async_iterator, undefined, after, length, recurse=None,
depth0=0):
LoopContextBase.__init__(self, undefined, recurse, depth0)
self._async_iterator = async_iterator
self._after = after
self._length = length
@property
def length(self):
if self._length is None:
raise TypeError('Loop length for some iterators cannot be '
'lazily calculated in async mode')
return self._length
def __aiter__(self):
return AsyncLoopContextIterator(self)
class AsyncLoopContextIterator(object):
__slots__ = ('context',)
def __init__(self, context):
self.context = context
def __aiter__(self):
return self
async def __anext__(self):
ctx = self.context
ctx.index0 += 1
if ctx._after is _last_iteration:
raise StopAsyncIteration()
ctx._before = ctx._current
ctx._current = ctx._after
try:
ctx._after = await ctx._async_iterator.__anext__()
except StopAsyncIteration:
ctx._after = _last_iteration
return ctx._current, ctx
async def make_async_loop_context(iterable, undefined, recurse=None, depth0=0):
# Length is more complicated and less efficient in async mode. The
# reason for this is that we cannot know if length will be used
# upfront but because length is a property we cannot lazily execute it
# later. This means that we need to buffer it up and measure :(
#
# We however only do this for actual iterators, not for async
# iterators as blocking here does not seem like the best idea in the
# world.
try:
length = len(iterable)
except (TypeError, AttributeError):
if not hasattr(iterable, '__aiter__'):
iterable = tuple(iterable)
length = len(iterable)
else:
length = None
async_iterator = auto_aiter(iterable)
try:
after = await async_iterator.__anext__()
except StopAsyncIteration:
after = _last_iteration
return AsyncLoopContext(async_iterator, undefined, after, length, recurse,
depth0)

View File

@ -1,63 +1,47 @@
# -*- coding: utf-8 -*-
"""The optional bytecode cache system. This is useful if you have very
complex template situations and the compilation of all those templates
slows down your application too much.
Situations where this is useful are often forking web applications that
are initialized on the first request.
"""
jinja2.bccache
~~~~~~~~~~~~~~
This module implements the bytecode cache system Jinja is optionally
using. This is useful if you have very complex template situations and
the compiliation of all those templates slow down your application too
much.
Situations where this is useful are often forking web applications that
are initialized on the first request.
:copyright: (c) 2017 by the Jinja Team.
:license: BSD.
"""
from os import path, listdir
import os
import sys
import stat
import errno
import marshal
import tempfile
import fnmatch
import marshal
import os
import pickle
import stat
import sys
import tempfile
import typing as t
from hashlib import sha1
from jinja2.utils import open_if_exists
from jinja2._compat import BytesIO, pickle, PY2, text_type
from io import BytesIO
from types import CodeType
if t.TYPE_CHECKING:
import typing_extensions as te
from .environment import Environment
class _MemcachedClient(te.Protocol):
def get(self, key: str) -> bytes:
...
def set(self, key: str, value: bytes, timeout: t.Optional[int] = None) -> None:
...
# marshal works better on 3.x, one hack less required
if not PY2:
marshal_dump = marshal.dump
marshal_load = marshal.load
else:
def marshal_dump(code, f):
if isinstance(f, file):
marshal.dump(code, f)
else:
f.write(marshal.dumps(code))
def marshal_load(f):
if isinstance(f, file):
return marshal.load(f)
return marshal.loads(f.read())
bc_version = 5
# Magic bytes to identify Jinja bytecode cache files. Contains the
# Python major and minor version to avoid loading incompatible bytecode
# if a project upgrades its Python version.
bc_magic = (
b"j2"
+ pickle.dumps(bc_version, 2)
+ pickle.dumps((sys.version_info[0] << 24) | sys.version_info[1], 2)
)
bc_version = 3
# magic version used to only change with new jinja versions. With 2.6
# we change this to also take Python version changes into account. The
# reason for this is that Python tends to segfault if fed earlier bytecode
# versions because someone thought it would be a good idea to reuse opcodes
# or make Python incompatible with earlier versions.
bc_magic = 'j2'.encode('ascii') + \
pickle.dumps(bc_version, 2) + \
pickle.dumps((sys.version_info[0] << 24) | sys.version_info[1])
class Bucket(object):
class Bucket:
"""Buckets are used to store the bytecode for one template. It's created
and initialized by the bytecode cache and passed to the loading functions.
@ -66,17 +50,17 @@ class Bucket(object):
cache subclasses don't have to care about cache invalidation.
"""
def __init__(self, environment, key, checksum):
def __init__(self, environment: "Environment", key: str, checksum: str) -> None:
self.environment = environment
self.key = key
self.checksum = checksum
self.reset()
def reset(self):
def reset(self) -> None:
"""Resets the bucket (unloads the bytecode)."""
self.code = None
self.code: t.Optional[CodeType] = None
def load_bytecode(self, f):
def load_bytecode(self, f: t.BinaryIO) -> None:
"""Loads bytecode from a file or file like object."""
# make sure the magic header is correct
magic = f.read(len(bc_magic))
@ -90,31 +74,31 @@ class Bucket(object):
return
# if marshal_load fails then we need to reload
try:
self.code = marshal_load(f)
self.code = marshal.load(f)
except (EOFError, ValueError, TypeError):
self.reset()
return
def write_bytecode(self, f):
def write_bytecode(self, f: t.IO[bytes]) -> None:
"""Dump the bytecode into the file or file like object passed."""
if self.code is None:
raise TypeError('can\'t write empty bucket')
raise TypeError("can't write empty bucket")
f.write(bc_magic)
pickle.dump(self.checksum, f, 2)
marshal_dump(self.code, f)
marshal.dump(self.code, f)
def bytecode_from_string(self, string):
"""Load bytecode from a string."""
def bytecode_from_string(self, string: bytes) -> None:
"""Load bytecode from bytes."""
self.load_bytecode(BytesIO(string))
def bytecode_to_string(self):
"""Return the bytecode as string."""
def bytecode_to_string(self) -> bytes:
"""Return the bytecode as bytes."""
out = BytesIO()
self.write_bytecode(out)
return out.getvalue()
class BytecodeCache(object):
class BytecodeCache:
"""To implement your own bytecode cache you have to subclass this class
and override :meth:`load_bytecode` and :meth:`dump_bytecode`. Both of
these methods are passed a :class:`~jinja2.bccache.Bucket`.
@ -140,44 +124,51 @@ class BytecodeCache(object):
bucket.write_bytecode(f)
A more advanced version of a filesystem based bytecode cache is part of
Jinja2.
Jinja.
"""
def load_bytecode(self, bucket):
def load_bytecode(self, bucket: Bucket) -> None:
"""Subclasses have to override this method to load bytecode into a
bucket. If they are not able to find code in the cache for the
bucket, it must not do anything.
"""
raise NotImplementedError()
def dump_bytecode(self, bucket):
def dump_bytecode(self, bucket: Bucket) -> None:
"""Subclasses have to override this method to write the bytecode
from a bucket back to the cache. If it unable to do so it must not
fail silently but raise an exception.
"""
raise NotImplementedError()
def clear(self):
"""Clears the cache. This method is not used by Jinja2 but should be
def clear(self) -> None:
"""Clears the cache. This method is not used by Jinja but should be
implemented to allow applications to clear the bytecode cache used
by a particular environment.
"""
def get_cache_key(self, name, filename=None):
def get_cache_key(
self, name: str, filename: t.Optional[t.Union[str]] = None
) -> str:
"""Returns the unique hash key for this template name."""
hash = sha1(name.encode('utf-8'))
hash = sha1(name.encode("utf-8"))
if filename is not None:
filename = '|' + filename
if isinstance(filename, text_type):
filename = filename.encode('utf-8')
hash.update(filename)
hash.update(f"|{filename}".encode())
return hash.hexdigest()
def get_source_checksum(self, source):
def get_source_checksum(self, source: str) -> str:
"""Returns a checksum for the source."""
return sha1(source.encode('utf-8')).hexdigest()
return sha1(source.encode("utf-8")).hexdigest()
def get_bucket(self, environment, name, filename, source):
def get_bucket(
self,
environment: "Environment",
name: str,
filename: t.Optional[str],
source: str,
) -> Bucket:
"""Return a cache bucket for the given template. All arguments are
mandatory but filename may be `None`.
"""
@ -187,7 +178,7 @@ class BytecodeCache(object):
self.load_bytecode(bucket)
return bucket
def set_bucket(self, bucket):
def set_bucket(self, bucket: Bucket) -> None:
"""Put the bucket into the cache."""
self.dump_bytecode(bucket)
@ -210,27 +201,31 @@ class FileSystemBytecodeCache(BytecodeCache):
This bytecode cache supports clearing of the cache using the clear method.
"""
def __init__(self, directory=None, pattern='__jinja2_%s.cache'):
def __init__(
self, directory: t.Optional[str] = None, pattern: str = "__jinja2_%s.cache"
) -> None:
if directory is None:
directory = self._get_default_cache_dir()
self.directory = directory
self.pattern = pattern
def _get_default_cache_dir(self):
def _unsafe_dir():
raise RuntimeError('Cannot determine safe temp directory. You '
'need to explicitly provide one.')
def _get_default_cache_dir(self) -> str:
def _unsafe_dir() -> "te.NoReturn":
raise RuntimeError(
"Cannot determine safe temp directory. You "
"need to explicitly provide one."
)
tmpdir = tempfile.gettempdir()
# On windows the temporary directory is used specific unless
# explicitly forced otherwise. We can just use that.
if os.name == 'nt':
if os.name == "nt":
return tmpdir
if not hasattr(os, 'getuid'):
if not hasattr(os, "getuid"):
_unsafe_dir()
dirname = '_jinja2-cache-%d' % os.getuid()
dirname = f"_jinja2-cache-{os.getuid()}"
actual_dir = os.path.join(tmpdir, dirname)
try:
@ -241,49 +236,92 @@ class FileSystemBytecodeCache(BytecodeCache):
try:
os.chmod(actual_dir, stat.S_IRWXU)
actual_dir_stat = os.lstat(actual_dir)
if actual_dir_stat.st_uid != os.getuid() \
or not stat.S_ISDIR(actual_dir_stat.st_mode) \
or stat.S_IMODE(actual_dir_stat.st_mode) != stat.S_IRWXU:
if (
actual_dir_stat.st_uid != os.getuid()
or not stat.S_ISDIR(actual_dir_stat.st_mode)
or stat.S_IMODE(actual_dir_stat.st_mode) != stat.S_IRWXU
):
_unsafe_dir()
except OSError as e:
if e.errno != errno.EEXIST:
raise
actual_dir_stat = os.lstat(actual_dir)
if actual_dir_stat.st_uid != os.getuid() \
or not stat.S_ISDIR(actual_dir_stat.st_mode) \
or stat.S_IMODE(actual_dir_stat.st_mode) != stat.S_IRWXU:
if (
actual_dir_stat.st_uid != os.getuid()
or not stat.S_ISDIR(actual_dir_stat.st_mode)
or stat.S_IMODE(actual_dir_stat.st_mode) != stat.S_IRWXU
):
_unsafe_dir()
return actual_dir
def _get_cache_filename(self, bucket):
return path.join(self.directory, self.pattern % bucket.key)
def _get_cache_filename(self, bucket: Bucket) -> str:
return os.path.join(self.directory, self.pattern % (bucket.key,))
def load_bytecode(self, bucket):
f = open_if_exists(self._get_cache_filename(bucket), 'rb')
if f is not None:
try:
bucket.load_bytecode(f)
finally:
f.close()
def load_bytecode(self, bucket: Bucket) -> None:
filename = self._get_cache_filename(bucket)
def dump_bytecode(self, bucket):
f = open(self._get_cache_filename(bucket), 'wb')
# Don't test for existence before opening the file, since the
# file could disappear after the test before the open.
try:
bucket.write_bytecode(f)
finally:
f.close()
f = open(filename, "rb")
except (FileNotFoundError, IsADirectoryError, PermissionError):
# PermissionError can occur on Windows when an operation is
# in progress, such as calling clear().
return
def clear(self):
with f:
bucket.load_bytecode(f)
def dump_bytecode(self, bucket: Bucket) -> None:
# Write to a temporary file, then rename to the real name after
# writing. This avoids another process reading the file before
# it is fully written.
name = self._get_cache_filename(bucket)
f = tempfile.NamedTemporaryFile(
mode="wb",
dir=os.path.dirname(name),
prefix=os.path.basename(name),
suffix=".tmp",
delete=False,
)
def remove_silent() -> None:
try:
os.remove(f.name)
except OSError:
# Another process may have called clear(). On Windows,
# another program may be holding the file open.
pass
try:
with f:
bucket.write_bytecode(f)
except BaseException:
remove_silent()
raise
try:
os.replace(f.name, name)
except OSError:
# Another process may have called clear(). On Windows,
# another program may be holding the file open.
remove_silent()
except BaseException:
remove_silent()
raise
def clear(self) -> None:
# imported lazily here because google app-engine doesn't support
# write access on the file system and the function does not exist
# normally.
from os import remove
files = fnmatch.filter(listdir(self.directory), self.pattern % '*')
files = fnmatch.filter(os.listdir(self.directory), self.pattern % ("*",))
for filename in files:
try:
remove(path.join(self.directory, filename))
remove(os.path.join(self.directory, filename))
except OSError:
pass
@ -296,12 +334,11 @@ class MemcachedBytecodeCache(BytecodeCache):
Libraries compatible with this class:
- `werkzeug <http://werkzeug.pocoo.org/>`_.contrib.cache
- `python-memcached <https://www.tummy.com/Community/software/python-memcached/>`_
- `cmemcache <http://gijsbert.org/cmemcache/>`_
- `cachelib <https://github.com/pallets/cachelib>`_
- `python-memcached <https://pypi.org/project/python-memcached/>`_
(Unfortunately the django cache interface is not compatible because it
does not support storing binary data, only unicode. You can however pass
does not support storing binary data, only text. You can however pass
the underlying cache client to the bytecode cache which is available
as `django.core.cache.cache._client`.)
@ -334,29 +371,36 @@ class MemcachedBytecodeCache(BytecodeCache):
`ignore_memcache_errors` parameter.
"""
def __init__(self, client, prefix='jinja2/bytecode/', timeout=None,
ignore_memcache_errors=True):
def __init__(
self,
client: "_MemcachedClient",
prefix: str = "jinja2/bytecode/",
timeout: t.Optional[int] = None,
ignore_memcache_errors: bool = True,
):
self.client = client
self.prefix = prefix
self.timeout = timeout
self.ignore_memcache_errors = ignore_memcache_errors
def load_bytecode(self, bucket):
def load_bytecode(self, bucket: Bucket) -> None:
try:
code = self.client.get(self.prefix + bucket.key)
except Exception:
if not self.ignore_memcache_errors:
raise
code = None
if code is not None:
else:
bucket.bytecode_from_string(code)
def dump_bytecode(self, bucket):
args = (self.prefix + bucket.key, bucket.bytecode_to_string())
if self.timeout is not None:
args += (self.timeout,)
def dump_bytecode(self, bucket: Bucket) -> None:
key = self.prefix + bucket.key
value = bucket.bytecode_to_string()
try:
self.client.set(*args)
if self.timeout is not None:
self.client.set(key, value, self.timeout)
else:
self.client.set(key, value)
except Exception:
if not self.ignore_memcache_errors:
raise

File diff suppressed because it is too large Load Diff

View File

@ -1,17 +1,5 @@
# -*- coding: utf-8 -*-
"""
jinja.constants
~~~~~~~~~~~~~~~
Various constants.
:copyright: (c) 2017 by the Jinja Team.
:license: BSD, see LICENSE for more details.
"""
#: list of lorem ipsum words used by the lipsum() helper function
LOREM_IPSUM_WORDS = u'''\
LOREM_IPSUM_WORDS = """\
a ac accumsan ad adipiscing aenean aliquam aliquet amet ante aptent arcu at
auctor augue bibendum blandit class commodo condimentum congue consectetuer
consequat conubia convallis cras cubilia cum curabitur curae cursus dapibus
@ -29,4 +17,4 @@ ridiculus risus rutrum sagittis sapien scelerisque sed sem semper senectus sit
sociis sociosqu sodales sollicitudin suscipit suspendisse taciti tellus tempor
tempus tincidunt torquent tortor tristique turpis ullamcorper ultrices
ultricies urna ut varius vehicula vel velit venenatis vestibulum vitae vivamus
viverra volutpat vulputate'''
viverra volutpat vulputate"""

View File

@ -1,372 +1,191 @@
# -*- coding: utf-8 -*-
"""
jinja2.debug
~~~~~~~~~~~~
Implements the debug interface for Jinja. This module does some pretty
ugly stuff with the Python traceback system in order to achieve tracebacks
with correct line numbers, locals and contents.
:copyright: (c) 2017 by the Jinja Team.
:license: BSD, see LICENSE for more details.
"""
import sys
import traceback
from types import TracebackType, CodeType
from jinja2.utils import missing, internal_code
from jinja2.exceptions import TemplateSyntaxError
from jinja2._compat import iteritems, reraise, PY2
import typing as t
from types import CodeType
from types import TracebackType
# on pypy we can take advantage of transparent proxies
try:
from __pypy__ import tproxy
except ImportError:
tproxy = None
from .exceptions import TemplateSyntaxError
from .utils import internal_code
from .utils import missing
if t.TYPE_CHECKING:
from .runtime import Context
# how does the raise helper look like?
try:
exec("raise TypeError, 'foo'")
except SyntaxError:
raise_helper = 'raise __jinja_exception__[1]'
except TypeError:
raise_helper = 'raise __jinja_exception__[0], __jinja_exception__[1]'
def rewrite_traceback_stack(source: t.Optional[str] = None) -> BaseException:
"""Rewrite the current exception to replace any tracebacks from
within compiled template code with tracebacks that look like they
came from the template source.
This must be called within an ``except`` block.
class TracebackFrameProxy(object):
"""Proxies a traceback frame."""
def __init__(self, tb):
self.tb = tb
self._tb_next = None
@property
def tb_next(self):
return self._tb_next
def set_next(self, next):
if tb_set_next is not None:
try:
tb_set_next(self.tb, next and next.tb or None)
except Exception:
# this function can fail due to all the hackery it does
# on various python implementations. We just catch errors
# down and ignore them if necessary.
pass
self._tb_next = next
@property
def is_jinja_frame(self):
return '__jinja_template__' in self.tb.tb_frame.f_globals
def __getattr__(self, name):
return getattr(self.tb, name)
def make_frame_proxy(frame):
proxy = TracebackFrameProxy(frame)
if tproxy is None:
return proxy
def operation_handler(operation, *args, **kwargs):
if operation in ('__getattribute__', '__getattr__'):
return getattr(proxy, args[0])
elif operation == '__setattr__':
proxy.__setattr__(*args, **kwargs)
else:
return getattr(proxy, operation)(*args, **kwargs)
return tproxy(TracebackType, operation_handler)
class ProcessedTraceback(object):
"""Holds a Jinja preprocessed traceback for printing or reraising."""
def __init__(self, exc_type, exc_value, frames):
assert frames, 'no frames for this traceback?'
self.exc_type = exc_type
self.exc_value = exc_value
self.frames = frames
# newly concatenate the frames (which are proxies)
prev_tb = None
for tb in self.frames:
if prev_tb is not None:
prev_tb.set_next(tb)
prev_tb = tb
prev_tb.set_next(None)
def render_as_text(self, limit=None):
"""Return a string with the traceback."""
lines = traceback.format_exception(self.exc_type, self.exc_value,
self.frames[0], limit=limit)
return ''.join(lines).rstrip()
def render_as_html(self, full=False):
"""Return a unicode string with the traceback as rendered HTML."""
from jinja2.debugrenderer import render_traceback
return u'%s\n\n<!--\n%s\n-->' % (
render_traceback(self, full=full),
self.render_as_text().decode('utf-8', 'replace')
)
@property
def is_template_syntax_error(self):
"""`True` if this is a template syntax error."""
return isinstance(self.exc_value, TemplateSyntaxError)
@property
def exc_info(self):
"""Exception info tuple with a proxy around the frame objects."""
return self.exc_type, self.exc_value, self.frames[0]
@property
def standard_exc_info(self):
"""Standard python exc_info for re-raising"""
tb = self.frames[0]
# the frame will be an actual traceback (or transparent proxy) if
# we are on pypy or a python implementation with support for tproxy
if type(tb) is not TracebackType:
tb = tb.tb
return self.exc_type, self.exc_value, tb
def make_traceback(exc_info, source_hint=None):
"""Creates a processed traceback object from the exc_info."""
exc_type, exc_value, tb = exc_info
if isinstance(exc_value, TemplateSyntaxError):
exc_info = translate_syntax_error(exc_value, source_hint)
initial_skip = 0
else:
initial_skip = 1
return translate_exception(exc_info, initial_skip)
def translate_syntax_error(error, source=None):
"""Rewrites a syntax error to please traceback systems."""
error.source = source
error.translated = True
exc_info = (error.__class__, error, None)
filename = error.filename
if filename is None:
filename = '<unknown>'
return fake_exc_info(exc_info, filename, error.lineno)
def translate_exception(exc_info, initial_skip=0):
"""If passed an exc_info it will automatically rewrite the exceptions
all the way down to the correct line numbers and frames.
:param source: For ``TemplateSyntaxError``, the original source if
known.
:return: The original exception with the rewritten traceback.
"""
tb = exc_info[2]
frames = []
_, exc_value, tb = sys.exc_info()
exc_value = t.cast(BaseException, exc_value)
tb = t.cast(TracebackType, tb)
# skip some internal frames if wanted
for x in range(initial_skip):
if tb is not None:
tb = tb.tb_next
initial_tb = tb
if isinstance(exc_value, TemplateSyntaxError) and not exc_value.translated:
exc_value.translated = True
exc_value.source = source
# Remove the old traceback, otherwise the frames from the
# compiler still show up.
exc_value.with_traceback(None)
# Outside of runtime, so the frame isn't executing template
# code, but it still needs to point at the template.
tb = fake_traceback(
exc_value, None, exc_value.filename or "<unknown>", exc_value.lineno
)
else:
# Skip the frame for the render function.
tb = tb.tb_next
stack = []
# Build the stack of traceback object, replacing any in template
# code with the source file and line information.
while tb is not None:
# skip frames decorated with @internalcode. These are internal
# calls we can't avoid and that are useless in template debugging
# output.
# Skip frames decorated with @internalcode. These are internal
# calls that aren't useful in template debugging output.
if tb.tb_frame.f_code in internal_code:
tb = tb.tb_next
continue
# save a reference to the next frame if we override the current
# one with a faked one.
next = tb.tb_next
template = tb.tb_frame.f_globals.get("__jinja_template__")
# fake template exceptions
template = tb.tb_frame.f_globals.get('__jinja_template__')
if template is not None:
lineno = template.get_corresponding_lineno(tb.tb_lineno)
tb = fake_exc_info(exc_info[:2] + (tb,), template.filename,
lineno)[2]
fake_tb = fake_traceback(exc_value, tb, template.filename, lineno)
stack.append(fake_tb)
else:
stack.append(tb)
frames.append(make_frame_proxy(tb))
tb = next
tb = tb.tb_next
# if we don't have any exceptions in the frames left, we have to
# reraise it unchanged.
# XXX: can we backup here? when could this happen?
if not frames:
reraise(exc_info[0], exc_info[1], exc_info[2])
tb_next = None
return ProcessedTraceback(exc_info[0], exc_info[1], frames)
# Assign tb_next in reverse to avoid circular references.
for tb in reversed(stack):
tb.tb_next = tb_next
tb_next = tb
return exc_value.with_traceback(tb_next)
def get_jinja_locals(real_locals):
ctx = real_locals.get('context')
if ctx:
locals = ctx.get_all().copy()
def fake_traceback( # type: ignore
exc_value: BaseException, tb: t.Optional[TracebackType], filename: str, lineno: int
) -> TracebackType:
"""Produce a new traceback object that looks like it came from the
template source instead of the compiled code. The filename, line
number, and location name will point to the template, and the local
variables will be the current template context.
:param exc_value: The original exception to be re-raised to create
the new traceback.
:param tb: The original traceback to get the local variables and
code info from.
:param filename: The template filename.
:param lineno: The line number in the template source.
"""
if tb is not None:
# Replace the real locals with the context that would be
# available at that point in the template.
locals = get_template_locals(tb.tb_frame.f_locals)
locals.pop("__jinja_exception__", None)
else:
locals = {}
local_overrides = {}
globals = {
"__name__": filename,
"__file__": filename,
"__jinja_exception__": exc_value,
}
# Raise an exception at the correct line number.
code: CodeType = compile(
"\n" * (lineno - 1) + "raise __jinja_exception__", filename, "exec"
)
for name, value in iteritems(real_locals):
if not name.startswith('l_') or value is missing:
# Build a new code object that points to the template file and
# replaces the location with a block name.
location = "template"
if tb is not None:
function = tb.tb_frame.f_code.co_name
if function == "root":
location = "top-level template code"
elif function.startswith("block_"):
location = f"block {function[6:]!r}"
if sys.version_info >= (3, 8):
code = code.replace(co_name=location)
else:
code = CodeType(
code.co_argcount,
code.co_kwonlyargcount,
code.co_nlocals,
code.co_stacksize,
code.co_flags,
code.co_code,
code.co_consts,
code.co_names,
code.co_varnames,
code.co_filename,
location,
code.co_firstlineno,
code.co_lnotab,
code.co_freevars,
code.co_cellvars,
)
# Execute the new code, which is guaranteed to raise, and return
# the new traceback without this frame.
try:
exec(code, globals, locals)
except BaseException:
return sys.exc_info()[2].tb_next # type: ignore
def get_template_locals(real_locals: t.Mapping[str, t.Any]) -> t.Dict[str, t.Any]:
"""Based on the runtime locals, get the context that would be
available at that point in the template.
"""
# Start with the current template context.
ctx: "t.Optional[Context]" = real_locals.get("context")
if ctx is not None:
data: t.Dict[str, t.Any] = ctx.get_all().copy()
else:
data = {}
# Might be in a derived context that only sets local variables
# rather than pushing a context. Local variables follow the scheme
# l_depth_name. Find the highest-depth local that has a value for
# each name.
local_overrides: t.Dict[str, t.Tuple[int, t.Any]] = {}
for name, value in real_locals.items():
if not name.startswith("l_") or value is missing:
# Not a template variable, or no longer relevant.
continue
try:
_, depth, name = name.split('_', 2)
depth = int(depth)
_, depth_str, name = name.split("_", 2)
depth = int(depth_str)
except ValueError:
continue
cur_depth = local_overrides.get(name, (-1,))[0]
if cur_depth < depth:
local_overrides[name] = (depth, value)
for name, (_, value) in iteritems(local_overrides):
# Modify the context with any derived context.
for name, (_, value) in local_overrides.items():
if value is missing:
locals.pop(name, None)
data.pop(name, None)
else:
locals[name] = value
data[name] = value
return locals
def fake_exc_info(exc_info, filename, lineno):
"""Helper for `translate_exception`."""
exc_type, exc_value, tb = exc_info
# figure the real context out
if tb is not None:
locals = get_jinja_locals(tb.tb_frame.f_locals)
# if there is a local called __jinja_exception__, we get
# rid of it to not break the debug functionality.
locals.pop('__jinja_exception__', None)
else:
locals = {}
# assamble fake globals we need
globals = {
'__name__': filename,
'__file__': filename,
'__jinja_exception__': exc_info[:2],
# we don't want to keep the reference to the template around
# to not cause circular dependencies, but we mark it as Jinja
# frame for the ProcessedTraceback
'__jinja_template__': None
}
# and fake the exception
code = compile('\n' * (lineno - 1) + raise_helper, filename, 'exec')
# if it's possible, change the name of the code. This won't work
# on some python environments such as google appengine
try:
if tb is None:
location = 'template'
else:
function = tb.tb_frame.f_code.co_name
if function == 'root':
location = 'top-level template code'
elif function.startswith('block_'):
location = 'block "%s"' % function[6:]
else:
location = 'template'
if PY2:
code = CodeType(0, code.co_nlocals, code.co_stacksize,
code.co_flags, code.co_code, code.co_consts,
code.co_names, code.co_varnames, filename,
location, code.co_firstlineno,
code.co_lnotab, (), ())
else:
code = CodeType(0, code.co_kwonlyargcount,
code.co_nlocals, code.co_stacksize,
code.co_flags, code.co_code, code.co_consts,
code.co_names, code.co_varnames, filename,
location, code.co_firstlineno,
code.co_lnotab, (), ())
except Exception as e:
pass
# execute the code and catch the new traceback
try:
exec(code, globals, locals)
except:
exc_info = sys.exc_info()
new_tb = exc_info[2].tb_next
# return without this frame
return exc_info[:2] + (new_tb,)
def _init_ugly_crap():
"""This function implements a few ugly things so that we can patch the
traceback objects. The function returned allows resetting `tb_next` on
any python traceback object. Do not attempt to use this on non cpython
interpreters
"""
import ctypes
from types import TracebackType
if PY2:
# figure out size of _Py_ssize_t for Python 2:
if hasattr(ctypes.pythonapi, 'Py_InitModule4_64'):
_Py_ssize_t = ctypes.c_int64
else:
_Py_ssize_t = ctypes.c_int
else:
# platform ssize_t on Python 3
_Py_ssize_t = ctypes.c_ssize_t
# regular python
class _PyObject(ctypes.Structure):
pass
_PyObject._fields_ = [
('ob_refcnt', _Py_ssize_t),
('ob_type', ctypes.POINTER(_PyObject))
]
# python with trace
if hasattr(sys, 'getobjects'):
class _PyObject(ctypes.Structure):
pass
_PyObject._fields_ = [
('_ob_next', ctypes.POINTER(_PyObject)),
('_ob_prev', ctypes.POINTER(_PyObject)),
('ob_refcnt', _Py_ssize_t),
('ob_type', ctypes.POINTER(_PyObject))
]
class _Traceback(_PyObject):
pass
_Traceback._fields_ = [
('tb_next', ctypes.POINTER(_Traceback)),
('tb_frame', ctypes.POINTER(_PyObject)),
('tb_lasti', ctypes.c_int),
('tb_lineno', ctypes.c_int)
]
def tb_set_next(tb, next):
"""Set the tb_next attribute of a traceback object."""
if not (isinstance(tb, TracebackType) and
(next is None or isinstance(next, TracebackType))):
raise TypeError('tb_set_next arguments must be traceback objects')
obj = _Traceback.from_address(id(tb))
if tb.tb_next is not None:
old = _Traceback.from_address(id(tb.tb_next))
old.ob_refcnt -= 1
if next is None:
obj.tb_next = ctypes.POINTER(_Traceback)()
else:
next = _Traceback.from_address(id(next))
next.ob_refcnt += 1
obj.tb_next = ctypes.pointer(next)
return tb_set_next
# try to get a tb_set_next implementation if we don't have transparent
# proxies.
tb_set_next = None
if tproxy is None:
try:
tb_set_next = _init_ugly_crap()
except:
pass
del _init_ugly_crap
return data

View File

@ -1,56 +1,48 @@
# -*- coding: utf-8 -*-
"""
jinja2.defaults
~~~~~~~~~~~~~~~
import typing as t
Jinja default filters and tags.
:copyright: (c) 2017 by the Jinja Team.
:license: BSD, see LICENSE for more details.
"""
from jinja2._compat import range_type
from jinja2.utils import generate_lorem_ipsum, Cycler, Joiner, Namespace
from .filters import FILTERS as DEFAULT_FILTERS # noqa: F401
from .tests import TESTS as DEFAULT_TESTS # noqa: F401
from .utils import Cycler
from .utils import generate_lorem_ipsum
from .utils import Joiner
from .utils import Namespace
if t.TYPE_CHECKING:
import typing_extensions as te
# defaults for the parser / lexer
BLOCK_START_STRING = '{%'
BLOCK_END_STRING = '%}'
VARIABLE_START_STRING = '{{'
VARIABLE_END_STRING = '}}'
COMMENT_START_STRING = '{#'
COMMENT_END_STRING = '#}'
LINE_STATEMENT_PREFIX = None
LINE_COMMENT_PREFIX = None
BLOCK_START_STRING = "{%"
BLOCK_END_STRING = "%}"
VARIABLE_START_STRING = "{{"
VARIABLE_END_STRING = "}}"
COMMENT_START_STRING = "{#"
COMMENT_END_STRING = "#}"
LINE_STATEMENT_PREFIX: t.Optional[str] = None
LINE_COMMENT_PREFIX: t.Optional[str] = None
TRIM_BLOCKS = False
LSTRIP_BLOCKS = False
NEWLINE_SEQUENCE = '\n'
NEWLINE_SEQUENCE: "te.Literal['\\n', '\\r\\n', '\\r']" = "\n"
KEEP_TRAILING_NEWLINE = False
# default filters, tests and namespace
from jinja2.filters import FILTERS as DEFAULT_FILTERS
from jinja2.tests import TESTS as DEFAULT_TESTS
DEFAULT_NAMESPACE = {
'range': range_type,
'dict': dict,
'lipsum': generate_lorem_ipsum,
'cycler': Cycler,
'joiner': Joiner,
'namespace': Namespace
}
DEFAULT_NAMESPACE = {
"range": range,
"dict": dict,
"lipsum": generate_lorem_ipsum,
"cycler": Cycler,
"joiner": Joiner,
"namespace": Namespace,
}
# default policies
DEFAULT_POLICIES = {
'compiler.ascii_str': True,
'urlize.rel': 'noopener',
'urlize.target': None,
'truncate.leeway': 5,
'json.dumps_function': None,
'json.dumps_kwargs': {'sort_keys': True},
'ext.i18n.trimmed': False,
DEFAULT_POLICIES: t.Dict[str, t.Any] = {
"compiler.ascii_str": True,
"urlize.rel": "noopener",
"urlize.target": None,
"urlize.extra_schemes": None,
"truncate.leeway": 5,
"json.dumps_function": None,
"json.dumps_kwargs": {"sort_keys": True},
"ext.i18n.trimmed": False,
}
# export all constants
__all__ = tuple(x for x in locals().keys() if x.isupper())

File diff suppressed because it is too large Load Diff

View File

@ -1,64 +1,53 @@
# -*- coding: utf-8 -*-
"""
jinja2.exceptions
~~~~~~~~~~~~~~~~~
import typing as t
Jinja exceptions.
:copyright: (c) 2017 by the Jinja Team.
:license: BSD, see LICENSE for more details.
"""
from jinja2._compat import imap, text_type, PY2, implements_to_string
if t.TYPE_CHECKING:
from .runtime import Undefined
class TemplateError(Exception):
"""Baseclass for all template errors."""
if PY2:
def __init__(self, message=None):
if message is not None:
message = text_type(message).encode('utf-8')
Exception.__init__(self, message)
def __init__(self, message: t.Optional[str] = None) -> None:
super().__init__(message)
@property
def message(self):
if self.args:
message = self.args[0]
if message is not None:
return message.decode('utf-8', 'replace')
def __unicode__(self):
return self.message or u''
else:
def __init__(self, message=None):
Exception.__init__(self, message)
@property
def message(self):
if self.args:
message = self.args[0]
if message is not None:
return message
@property
def message(self) -> t.Optional[str]:
return self.args[0] if self.args else None
@implements_to_string
class TemplateNotFound(IOError, LookupError, TemplateError):
"""Raised if a template does not exist."""
"""Raised if a template does not exist.
# looks weird, but removes the warning descriptor that just
# bogusly warns us about message being deprecated
message = None
.. versionchanged:: 2.11
If the given name is :class:`Undefined` and no message was
provided, an :exc:`UndefinedError` is raised.
"""
# Silence the Python warning about message being deprecated since
# it's not valid here.
message: t.Optional[str] = None
def __init__(
self,
name: t.Optional[t.Union[str, "Undefined"]],
message: t.Optional[str] = None,
) -> None:
IOError.__init__(self, name)
def __init__(self, name, message=None):
IOError.__init__(self)
if message is None:
from .runtime import Undefined
if isinstance(name, Undefined):
name._fail_with_undefined_error()
message = name
self.message = message
self.name = name
self.templates = [name]
def __str__(self):
return self.message
def __str__(self) -> str:
return str(self.message)
class TemplatesNotFound(TemplateNotFound):
@ -66,54 +55,85 @@ class TemplatesNotFound(TemplateNotFound):
are selected. This is a subclass of :class:`TemplateNotFound`
exception, so just catching the base exception will catch both.
.. versionchanged:: 2.11
If a name in the list of names is :class:`Undefined`, a message
about it being undefined is shown rather than the empty string.
.. versionadded:: 2.2
"""
def __init__(self, names=(), message=None):
def __init__(
self,
names: t.Sequence[t.Union[str, "Undefined"]] = (),
message: t.Optional[str] = None,
) -> None:
if message is None:
message = u'none of the templates given were found: ' + \
u', '.join(imap(text_type, names))
TemplateNotFound.__init__(self, names and names[-1] or None, message)
from .runtime import Undefined
parts = []
for name in names:
if isinstance(name, Undefined):
parts.append(name._undefined_message)
else:
parts.append(name)
parts_str = ", ".join(map(str, parts))
message = f"none of the templates given were found: {parts_str}"
super().__init__(names[-1] if names else None, message)
self.templates = list(names)
@implements_to_string
class TemplateSyntaxError(TemplateError):
"""Raised to tell the user that there is a problem with the template."""
def __init__(self, message, lineno, name=None, filename=None):
TemplateError.__init__(self, message)
def __init__(
self,
message: str,
lineno: int,
name: t.Optional[str] = None,
filename: t.Optional[str] = None,
) -> None:
super().__init__(message)
self.lineno = lineno
self.name = name
self.filename = filename
self.source = None
self.source: t.Optional[str] = None
# this is set to True if the debug.translate_syntax_error
# function translated the syntax error into a new traceback
self.translated = False
def __str__(self):
def __str__(self) -> str:
# for translated errors we only return the message
if self.translated:
return self.message
return t.cast(str, self.message)
# otherwise attach some stuff
location = 'line %d' % self.lineno
location = f"line {self.lineno}"
name = self.filename or self.name
if name:
location = 'File "%s", %s' % (name, location)
lines = [self.message, ' ' + location]
location = f'File "{name}", {location}'
lines = [t.cast(str, self.message), " " + location]
# if the source is set, add the line to the output
if self.source is not None:
try:
line = self.source.splitlines()[self.lineno - 1]
except IndexError:
line = None
if line:
lines.append(' ' + line.strip())
pass
else:
lines.append(" " + line.strip())
return u'\n'.join(lines)
return "\n".join(lines)
def __reduce__(self): # type: ignore
# https://bugs.python.org/issue1692335 Exceptions that take
# multiple required arguments have problems with pickling.
# Without this, raises TypeError: __init__() missing 1 required
# positional argument: 'lineno'
return self.__class__, (self.message, self.lineno, self.name, self.filename)
class TemplateAssertionError(TemplateSyntaxError):

File diff suppressed because it is too large Load Diff

File diff suppressed because it is too large Load Diff

View File

@ -1,14 +1,17 @@
from jinja2.visitor import NodeVisitor
from jinja2._compat import iteritems
import typing as t
from . import nodes
from .visitor import NodeVisitor
VAR_LOAD_PARAMETER = "param"
VAR_LOAD_RESOLVE = "resolve"
VAR_LOAD_ALIAS = "alias"
VAR_LOAD_UNDEFINED = "undefined"
VAR_LOAD_PARAMETER = 'param'
VAR_LOAD_RESOLVE = 'resolve'
VAR_LOAD_ALIAS = 'alias'
VAR_LOAD_UNDEFINED = 'undefined'
def find_symbols(nodes, parent_symbols=None):
def find_symbols(
nodes: t.Iterable[nodes.Node], parent_symbols: t.Optional["Symbols"] = None
) -> "Symbols":
sym = Symbols(parent=parent_symbols)
visitor = FrameSymbolVisitor(sym)
for node in nodes:
@ -16,57 +19,71 @@ def find_symbols(nodes, parent_symbols=None):
return sym
def symbols_for_node(node, parent_symbols=None):
def symbols_for_node(
node: nodes.Node, parent_symbols: t.Optional["Symbols"] = None
) -> "Symbols":
sym = Symbols(parent=parent_symbols)
sym.analyze_node(node)
return sym
class Symbols(object):
def __init__(self, parent=None, level=None):
class Symbols:
def __init__(
self, parent: t.Optional["Symbols"] = None, level: t.Optional[int] = None
) -> None:
if level is None:
if parent is None:
level = 0
else:
level = parent.level + 1
self.level = level
self.parent = parent
self.refs = {}
self.loads = {}
self.stores = set()
def analyze_node(self, node, **kwargs):
self.level: int = level
self.parent = parent
self.refs: t.Dict[str, str] = {}
self.loads: t.Dict[str, t.Any] = {}
self.stores: t.Set[str] = set()
def analyze_node(self, node: nodes.Node, **kwargs: t.Any) -> None:
visitor = RootVisitor(self)
visitor.visit(node, **kwargs)
def _define_ref(self, name, load=None):
ident = 'l_%d_%s' % (self.level, name)
def _define_ref(
self, name: str, load: t.Optional[t.Tuple[str, t.Optional[str]]] = None
) -> str:
ident = f"l_{self.level}_{name}"
self.refs[name] = ident
if load is not None:
self.loads[ident] = load
return ident
def find_load(self, target):
def find_load(self, target: str) -> t.Optional[t.Any]:
if target in self.loads:
return self.loads[target]
if self.parent is not None:
return self.parent.find_load(target)
def find_ref(self, name):
return None
def find_ref(self, name: str) -> t.Optional[str]:
if name in self.refs:
return self.refs[name]
if self.parent is not None:
return self.parent.find_ref(name)
def ref(self, name):
return None
def ref(self, name: str) -> str:
rv = self.find_ref(name)
if rv is None:
raise AssertionError('Tried to resolve a name to a reference that '
'was unknown to the frame (%r)' % name)
raise AssertionError(
"Tried to resolve a name to a reference that was"
f" unknown to the frame ({name!r})"
)
return rv
def copy(self):
def copy(self) -> "Symbols":
rv = object.__new__(self.__class__)
rv.__dict__.update(self.__dict__)
rv.refs = self.refs.copy()
@ -74,7 +91,7 @@ class Symbols(object):
rv.stores = self.stores.copy()
return rv
def store(self, name):
def store(self, name: str) -> None:
self.stores.add(name)
# If we have not see the name referenced yet, we need to figure
@ -92,17 +109,16 @@ class Symbols(object):
# Otherwise we can just set it to undefined.
self._define_ref(name, load=(VAR_LOAD_UNDEFINED, None))
def declare_parameter(self, name):
def declare_parameter(self, name: str) -> str:
self.stores.add(name)
return self._define_ref(name, load=(VAR_LOAD_PARAMETER, None))
def load(self, name):
target = self.find_ref(name)
if target is None:
def load(self, name: str) -> None:
if self.find_ref(name) is None:
self._define_ref(name, load=(VAR_LOAD_RESOLVE, name))
def branch_update(self, branch_symbols):
stores = {}
def branch_update(self, branch_symbols: t.Sequence["Symbols"]) -> None:
stores: t.Dict[str, int] = {}
for branch in branch_symbols:
for target in branch.stores:
if target in self.stores:
@ -114,11 +130,12 @@ class Symbols(object):
self.loads.update(sym.loads)
self.stores.update(sym.stores)
for name, branch_count in iteritems(stores):
for name, branch_count in stores.items():
if branch_count == len(branch_symbols):
continue
target = self.find_ref(name)
assert target is not None, 'should not happen'
target = self.find_ref(name) # type: ignore
assert target is not None, "should not happen"
if self.parent is not None:
outer_target = self.parent.find_ref(name)
@ -127,160 +144,175 @@ class Symbols(object):
continue
self.loads[target] = (VAR_LOAD_RESOLVE, name)
def dump_stores(self):
rv = {}
node = self
def dump_stores(self) -> t.Dict[str, str]:
rv: t.Dict[str, str] = {}
node: t.Optional["Symbols"] = self
while node is not None:
for name in node.stores:
for name in sorted(node.stores):
if name not in rv:
rv[name] = self.find_ref(name)
rv[name] = self.find_ref(name) # type: ignore
node = node.parent
return rv
def dump_param_targets(self):
def dump_param_targets(self) -> t.Set[str]:
rv = set()
node = self
node: t.Optional["Symbols"] = self
while node is not None:
for target, (instr, _) in iteritems(self.loads):
for target, (instr, _) in self.loads.items():
if instr == VAR_LOAD_PARAMETER:
rv.add(target)
node = node.parent
return rv
class RootVisitor(NodeVisitor):
def __init__(self, symbols):
def __init__(self, symbols: "Symbols") -> None:
self.sym_visitor = FrameSymbolVisitor(symbols)
def _simple_visit(self, node, **kwargs):
def _simple_visit(self, node: nodes.Node, **kwargs: t.Any) -> None:
for child in node.iter_child_nodes():
self.sym_visitor.visit(child)
visit_Template = visit_Block = visit_Macro = visit_FilterBlock = \
visit_Scope = visit_If = visit_ScopedEvalContextModifier = \
_simple_visit
visit_Template = _simple_visit
visit_Block = _simple_visit
visit_Macro = _simple_visit
visit_FilterBlock = _simple_visit
visit_Scope = _simple_visit
visit_If = _simple_visit
visit_ScopedEvalContextModifier = _simple_visit
def visit_AssignBlock(self, node, **kwargs):
def visit_AssignBlock(self, node: nodes.AssignBlock, **kwargs: t.Any) -> None:
for child in node.body:
self.sym_visitor.visit(child)
def visit_CallBlock(self, node, **kwargs):
for child in node.iter_child_nodes(exclude=('call',)):
def visit_CallBlock(self, node: nodes.CallBlock, **kwargs: t.Any) -> None:
for child in node.iter_child_nodes(exclude=("call",)):
self.sym_visitor.visit(child)
def visit_OverlayScope(self, node, **kwargs):
def visit_OverlayScope(self, node: nodes.OverlayScope, **kwargs: t.Any) -> None:
for child in node.body:
self.sym_visitor.visit(child)
def visit_For(self, node, for_branch='body', **kwargs):
if for_branch == 'body':
def visit_For(
self, node: nodes.For, for_branch: str = "body", **kwargs: t.Any
) -> None:
if for_branch == "body":
self.sym_visitor.visit(node.target, store_as_param=True)
branch = node.body
elif for_branch == 'else':
elif for_branch == "else":
branch = node.else_
elif for_branch == 'test':
elif for_branch == "test":
self.sym_visitor.visit(node.target, store_as_param=True)
if node.test is not None:
self.sym_visitor.visit(node.test)
return
else:
raise RuntimeError('Unknown for branch')
for item in branch or ():
self.sym_visitor.visit(item)
raise RuntimeError("Unknown for branch")
def visit_With(self, node, **kwargs):
if branch:
for item in branch:
self.sym_visitor.visit(item)
def visit_With(self, node: nodes.With, **kwargs: t.Any) -> None:
for target in node.targets:
self.sym_visitor.visit(target)
for child in node.body:
self.sym_visitor.visit(child)
def generic_visit(self, node, *args, **kwargs):
raise NotImplementedError('Cannot find symbols for %r' %
node.__class__.__name__)
def generic_visit(self, node: nodes.Node, *args: t.Any, **kwargs: t.Any) -> None:
raise NotImplementedError(f"Cannot find symbols for {type(node).__name__!r}")
class FrameSymbolVisitor(NodeVisitor):
"""A visitor for `Frame.inspect`."""
def __init__(self, symbols):
def __init__(self, symbols: "Symbols") -> None:
self.symbols = symbols
def visit_Name(self, node, store_as_param=False, **kwargs):
def visit_Name(
self, node: nodes.Name, store_as_param: bool = False, **kwargs: t.Any
) -> None:
"""All assignments to names go through this function."""
if store_as_param or node.ctx == 'param':
if store_as_param or node.ctx == "param":
self.symbols.declare_parameter(node.name)
elif node.ctx == 'store':
elif node.ctx == "store":
self.symbols.store(node.name)
elif node.ctx == 'load':
elif node.ctx == "load":
self.symbols.load(node.name)
def visit_NSRef(self, node, **kwargs):
def visit_NSRef(self, node: nodes.NSRef, **kwargs: t.Any) -> None:
self.symbols.load(node.name)
def visit_If(self, node, **kwargs):
def visit_If(self, node: nodes.If, **kwargs: t.Any) -> None:
self.visit(node.test, **kwargs)
original_symbols = self.symbols
def inner_visit(nodes):
def inner_visit(nodes: t.Iterable[nodes.Node]) -> "Symbols":
self.symbols = rv = original_symbols.copy()
for subnode in nodes:
self.visit(subnode, **kwargs)
self.symbols = original_symbols
return rv
body_symbols = inner_visit(node.body)
elif_symbols = inner_visit(node.elif_)
else_symbols = inner_visit(node.else_ or ())
self.symbols.branch_update([body_symbols, elif_symbols, else_symbols])
def visit_Macro(self, node, **kwargs):
def visit_Macro(self, node: nodes.Macro, **kwargs: t.Any) -> None:
self.symbols.store(node.name)
def visit_Import(self, node, **kwargs):
def visit_Import(self, node: nodes.Import, **kwargs: t.Any) -> None:
self.generic_visit(node, **kwargs)
self.symbols.store(node.target)
def visit_FromImport(self, node, **kwargs):
def visit_FromImport(self, node: nodes.FromImport, **kwargs: t.Any) -> None:
self.generic_visit(node, **kwargs)
for name in node.names:
if isinstance(name, tuple):
self.symbols.store(name[1])
else:
self.symbols.store(name)
def visit_Assign(self, node, **kwargs):
def visit_Assign(self, node: nodes.Assign, **kwargs: t.Any) -> None:
"""Visit assignments in the correct order."""
self.visit(node.node, **kwargs)
self.visit(node.target, **kwargs)
def visit_For(self, node, **kwargs):
def visit_For(self, node: nodes.For, **kwargs: t.Any) -> None:
"""Visiting stops at for blocks. However the block sequence
is visited as part of the outer scope.
"""
self.visit(node.iter, **kwargs)
def visit_CallBlock(self, node, **kwargs):
def visit_CallBlock(self, node: nodes.CallBlock, **kwargs: t.Any) -> None:
self.visit(node.call, **kwargs)
def visit_FilterBlock(self, node, **kwargs):
def visit_FilterBlock(self, node: nodes.FilterBlock, **kwargs: t.Any) -> None:
self.visit(node.filter, **kwargs)
def visit_With(self, node, **kwargs):
def visit_With(self, node: nodes.With, **kwargs: t.Any) -> None:
for target in node.values:
self.visit(target)
def visit_AssignBlock(self, node, **kwargs):
def visit_AssignBlock(self, node: nodes.AssignBlock, **kwargs: t.Any) -> None:
"""Stop visiting at block assigns."""
self.visit(node.target, **kwargs)
def visit_Scope(self, node, **kwargs):
def visit_Scope(self, node: nodes.Scope, **kwargs: t.Any) -> None:
"""Stop visiting at scopes."""
def visit_Block(self, node, **kwargs):
def visit_Block(self, node: nodes.Block, **kwargs: t.Any) -> None:
"""Stop visiting at blocks."""
def visit_OverlayScope(self, node, **kwargs):
def visit_OverlayScope(self, node: nodes.OverlayScope, **kwargs: t.Any) -> None:
"""Do not visit into overlay scopes."""

File diff suppressed because it is too large Load Diff

View File

@ -1,40 +1,45 @@
# -*- coding: utf-8 -*-
"""
jinja2.loaders
~~~~~~~~~~~~~~
Jinja loader classes.
:copyright: (c) 2017 by the Jinja Team.
:license: BSD, see LICENSE for more details.
"""API and implementations for loading templates from different data
sources.
"""
import importlib.util
import os
import posixpath
import sys
import typing as t
import weakref
from types import ModuleType
from os import path
import zipimport
from collections import abc
from hashlib import sha1
from jinja2.exceptions import TemplateNotFound
from jinja2.utils import open_if_exists, internalcode
from jinja2._compat import string_types, iteritems
from importlib import import_module
from types import ModuleType
from .exceptions import TemplateNotFound
from .utils import internalcode
from .utils import open_if_exists
if t.TYPE_CHECKING:
from .environment import Environment
from .environment import Template
def split_template_path(template):
def split_template_path(template: str) -> t.List[str]:
"""Split a path into segments and perform a sanity check. If it detects
'..' in the path it will raise a `TemplateNotFound` error.
"""
pieces = []
for piece in template.split('/'):
if path.sep in piece \
or (path.altsep and path.altsep in piece) or \
piece == path.pardir:
for piece in template.split("/"):
if (
os.path.sep in piece
or (os.path.altsep and os.path.altsep in piece)
or piece == os.path.pardir
):
raise TemplateNotFound(template)
elif piece and piece != '.':
elif piece and piece != ".":
pieces.append(piece)
return pieces
class BaseLoader(object):
class BaseLoader:
"""Baseclass for all loaders. Subclass this and override `get_source` to
implement a custom loading mechanism. The environment provides a
`get_template` method that calls the loader's `load` method to get the
@ -56,8 +61,8 @@ class BaseLoader(object):
if not exists(path):
raise TemplateNotFound(template)
mtime = getmtime(path)
with file(path) as f:
source = f.read().decode('utf-8')
with open(path) as f:
source = f.read()
return source, path, lambda: mtime == getmtime(path)
"""
@ -67,16 +72,18 @@ class BaseLoader(object):
#: .. versionadded:: 2.4
has_source_access = True
def get_source(self, environment, template):
def get_source(
self, environment: "Environment", template: str
) -> t.Tuple[str, t.Optional[str], t.Optional[t.Callable[[], bool]]]:
"""Get the template source, filename and reload helper for a template.
It's passed the environment and template name and has to return a
tuple in the form ``(source, filename, uptodate)`` or raise a
`TemplateNotFound` error if it can't locate the template.
The source part of the returned tuple must be the source of the
template as unicode string or a ASCII bytestring. The filename should
be the name of the file on the filesystem if it was loaded from there,
otherwise `None`. The filename is used by python for the tracebacks
template as a string. The filename should be the name of the
file on the filesystem if it was loaded from there, otherwise
``None``. The filename is used by Python for the tracebacks
if no loader extension is used.
The last item in the tuple is the `uptodate` function. If auto
@ -86,18 +93,24 @@ class BaseLoader(object):
the template will be reloaded.
"""
if not self.has_source_access:
raise RuntimeError('%s cannot provide access to the source' %
self.__class__.__name__)
raise RuntimeError(
f"{type(self).__name__} cannot provide access to the source"
)
raise TemplateNotFound(template)
def list_templates(self):
def list_templates(self) -> t.List[str]:
"""Iterates over all templates. If the loader does not support that
it should raise a :exc:`TypeError` which is the default behavior.
"""
raise TypeError('this loader cannot iterate over all templates')
raise TypeError("this loader cannot iterate over all templates")
@internalcode
def load(self, environment, name, globals=None):
def load(
self,
environment: "Environment",
name: str,
globals: t.Optional[t.MutableMapping[str, t.Any]] = None,
) -> "Template":
"""Loads a template. This method looks up the template in the cache
or loads one by calling :meth:`get_source`. Subclasses should not
override this method as loaders working on collections of other
@ -131,43 +144,59 @@ class BaseLoader(object):
bucket.code = code
bcc.set_bucket(bucket)
return environment.template_class.from_code(environment, code,
globals, uptodate)
return environment.template_class.from_code(
environment, code, globals, uptodate
)
class FileSystemLoader(BaseLoader):
"""Loads templates from the file system. This loader can find templates
in folders on the file system and is the preferred way to load them.
"""Load templates from a directory in the file system.
The loader takes the path to the templates as string, or if multiple
locations are wanted a list of them which is then looked up in the
given order::
The path can be relative or absolute. Relative paths are relative to
the current working directory.
>>> loader = FileSystemLoader('/path/to/templates')
>>> loader = FileSystemLoader(['/path/to/templates', '/other/path'])
.. code-block:: python
Per default the template encoding is ``'utf-8'`` which can be changed
by setting the `encoding` parameter to something else.
loader = FileSystemLoader("templates")
To follow symbolic links, set the *followlinks* parameter to ``True``::
A list of paths can be given. The directories will be searched in
order, stopping at the first matching template.
>>> loader = FileSystemLoader('/path/to/templates', followlinks=True)
.. code-block:: python
.. versionchanged:: 2.8+
The *followlinks* parameter was added.
loader = FileSystemLoader(["/override/templates", "/default/templates"])
:param searchpath: A path, or list of paths, to the directory that
contains the templates.
:param encoding: Use this encoding to read the text from template
files.
:param followlinks: Follow symbolic links in the path.
.. versionchanged:: 2.8
Added the ``followlinks`` parameter.
"""
def __init__(self, searchpath, encoding='utf-8', followlinks=False):
if isinstance(searchpath, string_types):
def __init__(
self,
searchpath: t.Union[str, os.PathLike, t.Sequence[t.Union[str, os.PathLike]]],
encoding: str = "utf-8",
followlinks: bool = False,
) -> None:
if not isinstance(searchpath, abc.Iterable) or isinstance(searchpath, str):
searchpath = [searchpath]
self.searchpath = list(searchpath)
self.searchpath = [os.fspath(p) for p in searchpath]
self.encoding = encoding
self.followlinks = followlinks
def get_source(self, environment, template):
def get_source(
self, environment: "Environment", template: str
) -> t.Tuple[str, str, t.Callable[[], bool]]:
pieces = split_template_path(template)
for searchpath in self.searchpath:
filename = path.join(searchpath, *pieces)
# Use posixpath even on Windows to avoid "drive:" or UNC
# segments breaking out of the search directory.
filename = posixpath.join(searchpath, *pieces)
f = open_if_exists(filename)
if f is None:
continue
@ -176,26 +205,30 @@ class FileSystemLoader(BaseLoader):
finally:
f.close()
mtime = path.getmtime(filename)
mtime = os.path.getmtime(filename)
def uptodate():
def uptodate() -> bool:
try:
return path.getmtime(filename) == mtime
return os.path.getmtime(filename) == mtime
except OSError:
return False
return contents, filename, uptodate
# Use normpath to convert Windows altsep to sep.
return contents, os.path.normpath(filename), uptodate
raise TemplateNotFound(template)
def list_templates(self):
def list_templates(self) -> t.List[str]:
found = set()
for searchpath in self.searchpath:
walk_dir = os.walk(searchpath, followlinks=self.followlinks)
for dirpath, dirnames, filenames in walk_dir:
for dirpath, _, filenames in walk_dir:
for filename in filenames:
template = os.path.join(dirpath, filename) \
[len(searchpath):].strip(os.path.sep) \
.replace(os.path.sep, '/')
if template[:2] == './':
template = (
os.path.join(dirpath, filename)[len(searchpath) :]
.strip(os.path.sep)
.replace(os.path.sep, "/")
)
if template[:2] == "./":
template = template[2:]
if template not in found:
found.add(template)
@ -203,96 +236,199 @@ class FileSystemLoader(BaseLoader):
class PackageLoader(BaseLoader):
"""Load templates from python eggs or packages. It is constructed with
the name of the python package and the path to the templates in that
package::
"""Load templates from a directory in a Python package.
loader = PackageLoader('mypackage', 'views')
:param package_name: Import name of the package that contains the
template directory.
:param package_path: Directory within the imported package that
contains the templates.
:param encoding: Encoding of template files.
If the package path is not given, ``'templates'`` is assumed.
The following example looks up templates in the ``pages`` directory
within the ``project.ui`` package.
Per default the template encoding is ``'utf-8'`` which can be changed
by setting the `encoding` parameter to something else. Due to the nature
of eggs it's only possible to reload templates if the package was loaded
from the file system and not a zip file.
.. code-block:: python
loader = PackageLoader("project.ui", "pages")
Only packages installed as directories (standard pip behavior) or
zip/egg files (less common) are supported. The Python API for
introspecting data in packages is too limited to support other
installation methods the way this loader requires.
There is limited support for :pep:`420` namespace packages. The
template directory is assumed to only be in one namespace
contributor. Zip files contributing to a namespace are not
supported.
.. versionchanged:: 3.0
No longer uses ``setuptools`` as a dependency.
.. versionchanged:: 3.0
Limited PEP 420 namespace package support.
"""
def __init__(self, package_name, package_path='templates',
encoding='utf-8'):
from pkg_resources import DefaultProvider, ResourceManager, \
get_provider
provider = get_provider(package_name)
self.encoding = encoding
self.manager = ResourceManager()
self.filesystem_bound = isinstance(provider, DefaultProvider)
self.provider = provider
def __init__(
self,
package_name: str,
package_path: "str" = "templates",
encoding: str = "utf-8",
) -> None:
package_path = os.path.normpath(package_path).rstrip(os.path.sep)
# normpath preserves ".", which isn't valid in zip paths.
if package_path == os.path.curdir:
package_path = ""
elif package_path[:2] == os.path.curdir + os.path.sep:
package_path = package_path[2:]
self.package_path = package_path
self.package_name = package_name
self.encoding = encoding
def get_source(self, environment, template):
pieces = split_template_path(template)
p = '/'.join((self.package_path,) + tuple(pieces))
if not self.provider.has_resource(p):
raise TemplateNotFound(template)
# Make sure the package exists. This also makes namespace
# packages work, otherwise get_loader returns None.
import_module(package_name)
spec = importlib.util.find_spec(package_name)
assert spec is not None, "An import spec was not found for the package."
loader = spec.loader
assert loader is not None, "A loader was not found for the package."
self._loader = loader
self._archive = None
template_root = None
filename = uptodate = None
if self.filesystem_bound:
filename = self.provider.get_resource_filename(self.manager, p)
mtime = path.getmtime(filename)
def uptodate():
try:
return path.getmtime(filename) == mtime
except OSError:
return False
if isinstance(loader, zipimport.zipimporter):
self._archive = loader.archive
pkgdir = next(iter(spec.submodule_search_locations)) # type: ignore
template_root = os.path.join(pkgdir, package_path).rstrip(os.path.sep)
else:
roots: t.List[str] = []
source = self.provider.get_resource_string(self.manager, p)
return source.decode(self.encoding), filename, uptodate
# One element for regular packages, multiple for namespace
# packages, or None for single module file.
if spec.submodule_search_locations:
roots.extend(spec.submodule_search_locations)
# A single module file, use the parent directory instead.
elif spec.origin is not None:
roots.append(os.path.dirname(spec.origin))
for root in roots:
root = os.path.join(root, package_path)
if os.path.isdir(root):
template_root = root
break
if template_root is None:
raise ValueError(
f"The {package_name!r} package was not installed in a"
" way that PackageLoader understands."
)
self._template_root = template_root
def get_source(
self, environment: "Environment", template: str
) -> t.Tuple[str, str, t.Optional[t.Callable[[], bool]]]:
# Use posixpath even on Windows to avoid "drive:" or UNC
# segments breaking out of the search directory. Use normpath to
# convert Windows altsep to sep.
p = os.path.normpath(
posixpath.join(self._template_root, *split_template_path(template))
)
up_to_date: t.Optional[t.Callable[[], bool]]
if self._archive is None:
# Package is a directory.
if not os.path.isfile(p):
raise TemplateNotFound(template)
with open(p, "rb") as f:
source = f.read()
mtime = os.path.getmtime(p)
def up_to_date() -> bool:
return os.path.isfile(p) and os.path.getmtime(p) == mtime
else:
# Package is a zip file.
try:
source = self._loader.get_data(p) # type: ignore
except OSError as e:
raise TemplateNotFound(template) from e
# Could use the zip's mtime for all template mtimes, but
# would need to safely reload the module if it's out of
# date, so just report it as always current.
up_to_date = None
return source.decode(self.encoding), p, up_to_date
def list_templates(self) -> t.List[str]:
results: t.List[str] = []
if self._archive is None:
# Package is a directory.
offset = len(self._template_root)
for dirpath, _, filenames in os.walk(self._template_root):
dirpath = dirpath[offset:].lstrip(os.path.sep)
results.extend(
os.path.join(dirpath, name).replace(os.path.sep, "/")
for name in filenames
)
else:
if not hasattr(self._loader, "_files"):
raise TypeError(
"This zip import does not have the required"
" metadata to list templates."
)
# Package is a zip file.
prefix = (
self._template_root[len(self._archive) :].lstrip(os.path.sep)
+ os.path.sep
)
offset = len(prefix)
for name in self._loader._files.keys(): # type: ignore
# Find names under the templates directory that aren't directories.
if name.startswith(prefix) and name[-1] != os.path.sep:
results.append(name[offset:].replace(os.path.sep, "/"))
def list_templates(self):
path = self.package_path
if path[:2] == './':
path = path[2:]
elif path == '.':
path = ''
offset = len(path)
results = []
def _walk(path):
for filename in self.provider.resource_listdir(path):
fullname = path + '/' + filename
if self.provider.resource_isdir(fullname):
_walk(fullname)
else:
results.append(fullname[offset:].lstrip('/'))
_walk(path)
results.sort()
return results
class DictLoader(BaseLoader):
"""Loads a template from a python dict. It's passed a dict of unicode
strings bound to template names. This loader is useful for unittesting:
"""Loads a template from a Python dict mapping template names to
template source. This loader is useful for unittesting:
>>> loader = DictLoader({'index.html': 'source here'})
Because auto reloading is rarely useful this is disabled per default.
"""
def __init__(self, mapping):
def __init__(self, mapping: t.Mapping[str, str]) -> None:
self.mapping = mapping
def get_source(self, environment, template):
def get_source(
self, environment: "Environment", template: str
) -> t.Tuple[str, None, t.Callable[[], bool]]:
if template in self.mapping:
source = self.mapping[template]
return source, None, lambda: source == self.mapping.get(template)
raise TemplateNotFound(template)
def list_templates(self):
def list_templates(self) -> t.List[str]:
return sorted(self.mapping)
class FunctionLoader(BaseLoader):
"""A loader that is passed a function which does the loading. The
function receives the name of the template and has to return either
an unicode string with the template source, a tuple in the form ``(source,
a string with the template source, a tuple in the form ``(source,
filename, uptodatefunc)`` or `None` if the template does not exist.
>>> def load_template(name):
@ -307,15 +443,30 @@ class FunctionLoader(BaseLoader):
return value.
"""
def __init__(self, load_func):
def __init__(
self,
load_func: t.Callable[
[str],
t.Optional[
t.Union[
str, t.Tuple[str, t.Optional[str], t.Optional[t.Callable[[], bool]]]
]
],
],
) -> None:
self.load_func = load_func
def get_source(self, environment, template):
def get_source(
self, environment: "Environment", template: str
) -> t.Tuple[str, t.Optional[str], t.Optional[t.Callable[[], bool]]]:
rv = self.load_func(template)
if rv is None:
raise TemplateNotFound(template)
elif isinstance(rv, string_types):
if isinstance(rv, str):
return rv, None, None
return rv
@ -334,40 +485,49 @@ class PrefixLoader(BaseLoader):
by loading ``'app2/index.html'`` the file from the second.
"""
def __init__(self, mapping, delimiter='/'):
def __init__(
self, mapping: t.Mapping[str, BaseLoader], delimiter: str = "/"
) -> None:
self.mapping = mapping
self.delimiter = delimiter
def get_loader(self, template):
def get_loader(self, template: str) -> t.Tuple[BaseLoader, str]:
try:
prefix, name = template.split(self.delimiter, 1)
loader = self.mapping[prefix]
except (ValueError, KeyError):
raise TemplateNotFound(template)
except (ValueError, KeyError) as e:
raise TemplateNotFound(template) from e
return loader, name
def get_source(self, environment, template):
def get_source(
self, environment: "Environment", template: str
) -> t.Tuple[str, t.Optional[str], t.Optional[t.Callable[[], bool]]]:
loader, name = self.get_loader(template)
try:
return loader.get_source(environment, name)
except TemplateNotFound:
except TemplateNotFound as e:
# re-raise the exception with the correct filename here.
# (the one that includes the prefix)
raise TemplateNotFound(template)
raise TemplateNotFound(template) from e
@internalcode
def load(self, environment, name, globals=None):
def load(
self,
environment: "Environment",
name: str,
globals: t.Optional[t.MutableMapping[str, t.Any]] = None,
) -> "Template":
loader, local_name = self.get_loader(name)
try:
return loader.load(environment, local_name, globals)
except TemplateNotFound:
except TemplateNotFound as e:
# re-raise the exception with the correct filename here.
# (the one that includes the prefix)
raise TemplateNotFound(name)
raise TemplateNotFound(name) from e
def list_templates(self):
def list_templates(self) -> t.List[str]:
result = []
for prefix, loader in iteritems(self.mapping):
for prefix, loader in self.mapping.items():
for template in loader.list_templates():
result.append(prefix + self.delimiter + template)
return result
@ -387,10 +547,12 @@ class ChoiceLoader(BaseLoader):
from a different location.
"""
def __init__(self, loaders):
def __init__(self, loaders: t.Sequence[BaseLoader]) -> None:
self.loaders = loaders
def get_source(self, environment, template):
def get_source(
self, environment: "Environment", template: str
) -> t.Tuple[str, t.Optional[str], t.Optional[t.Callable[[], bool]]]:
for loader in self.loaders:
try:
return loader.get_source(environment, template)
@ -399,7 +561,12 @@ class ChoiceLoader(BaseLoader):
raise TemplateNotFound(template)
@internalcode
def load(self, environment, name, globals=None):
def load(
self,
environment: "Environment",
name: str,
globals: t.Optional[t.MutableMapping[str, t.Any]] = None,
) -> "Template":
for loader in self.loaders:
try:
return loader.load(environment, name, globals)
@ -407,7 +574,7 @@ class ChoiceLoader(BaseLoader):
pass
raise TemplateNotFound(name)
def list_templates(self):
def list_templates(self) -> t.List[str]:
found = set()
for loader in self.loaders:
found.update(loader.list_templates())
@ -433,20 +600,23 @@ class ModuleLoader(BaseLoader):
has_source_access = False
def __init__(self, path):
package_name = '_jinja2_module_templates_%x' % id(self)
def __init__(
self, path: t.Union[str, os.PathLike, t.Sequence[t.Union[str, os.PathLike]]]
) -> None:
package_name = f"_jinja2_module_templates_{id(self):x}"
# create a fake module that looks for the templates in the
# path given.
mod = _TemplateModule(package_name)
if isinstance(path, string_types):
path = [path]
else:
path = list(path)
mod.__path__ = path
sys.modules[package_name] = weakref.proxy(mod,
lambda x: sys.modules.pop(package_name, None))
if not isinstance(path, abc.Iterable) or isinstance(path, str):
path = [path]
mod.__path__ = [os.fspath(p) for p in path]
sys.modules[package_name] = weakref.proxy(
mod, lambda x: sys.modules.pop(package_name, None)
)
# the only strong reference, the sys.modules entry is weak
# so that the garbage collector can remove it once the
@ -455,27 +625,37 @@ class ModuleLoader(BaseLoader):
self.package_name = package_name
@staticmethod
def get_template_key(name):
return 'tmpl_' + sha1(name.encode('utf-8')).hexdigest()
def get_template_key(name: str) -> str:
return "tmpl_" + sha1(name.encode("utf-8")).hexdigest()
@staticmethod
def get_module_filename(name):
return ModuleLoader.get_template_key(name) + '.py'
def get_module_filename(name: str) -> str:
return ModuleLoader.get_template_key(name) + ".py"
@internalcode
def load(self, environment, name, globals=None):
def load(
self,
environment: "Environment",
name: str,
globals: t.Optional[t.MutableMapping[str, t.Any]] = None,
) -> "Template":
key = self.get_template_key(name)
module = '%s.%s' % (self.package_name, key)
module = f"{self.package_name}.{key}"
mod = getattr(self.module, module, None)
if mod is None:
try:
mod = __import__(module, None, None, ['root'])
except ImportError:
raise TemplateNotFound(name)
mod = __import__(module, None, None, ["root"])
except ImportError as e:
raise TemplateNotFound(name) from e
# remove the entry from sys.modules, we only want the attribute
# on the module object we have stored on the loader.
sys.modules.pop(module, None)
if globals is None:
globals = {}
return environment.template_class.from_module_dict(
environment, mod.__dict__, globals)
environment, mod.__dict__, globals
)

View File

@ -1,39 +1,36 @@
# -*- coding: utf-8 -*-
"""Functions that expose information about templates that might be
interesting for introspection.
"""
jinja2.meta
~~~~~~~~~~~
import typing as t
This module implements various functions that exposes information about
templates that might be interesting for various kinds of applications.
from . import nodes
from .compiler import CodeGenerator
from .compiler import Frame
:copyright: (c) 2017 by the Jinja Team, see AUTHORS for more details.
:license: BSD, see LICENSE for more details.
"""
from jinja2 import nodes
from jinja2.compiler import CodeGenerator
from jinja2._compat import string_types, iteritems
if t.TYPE_CHECKING:
from .environment import Environment
class TrackingCodeGenerator(CodeGenerator):
"""We abuse the code generator for introspection."""
def __init__(self, environment):
CodeGenerator.__init__(self, environment, '<introspection>',
'<introspection>')
self.undeclared_identifiers = set()
def __init__(self, environment: "Environment") -> None:
super().__init__(environment, "<introspection>", "<introspection>")
self.undeclared_identifiers: t.Set[str] = set()
def write(self, x):
def write(self, x: str) -> None:
"""Don't write."""
def enter_frame(self, frame):
def enter_frame(self, frame: Frame) -> None:
"""Remember all undeclared identifiers."""
CodeGenerator.enter_frame(self, frame)
for _, (action, param) in iteritems(frame.symbols.loads):
if action == 'resolve':
super().enter_frame(frame)
for _, (action, param) in frame.symbols.loads.items():
if action == "resolve" and param not in self.environment.globals:
self.undeclared_identifiers.add(param)
def find_undeclared_variables(ast):
def find_undeclared_variables(ast: nodes.Template) -> t.Set[str]:
"""Returns a set of all variables in the AST that will be looked up from
the context at runtime. Because at compile time it's not known which
variables will be used depending on the path the execution takes at
@ -42,7 +39,7 @@ def find_undeclared_variables(ast):
>>> from jinja2 import Environment, meta
>>> env = Environment()
>>> ast = env.parse('{% set foo = 42 %}{{ bar + foo }}')
>>> meta.find_undeclared_variables(ast) == set(['bar'])
>>> meta.find_undeclared_variables(ast) == {'bar'}
True
.. admonition:: Implementation
@ -52,12 +49,16 @@ def find_undeclared_variables(ast):
:exc:`TemplateAssertionError` during compilation and as a matter of
fact this function can currently raise that exception as well.
"""
codegen = TrackingCodeGenerator(ast.environment)
codegen = TrackingCodeGenerator(ast.environment) # type: ignore
codegen.visit(ast)
return codegen.undeclared_identifiers
def find_referenced_templates(ast):
_ref_types = (nodes.Extends, nodes.FromImport, nodes.Import, nodes.Include)
_RefType = t.Union[nodes.Extends, nodes.FromImport, nodes.Import, nodes.Include]
def find_referenced_templates(ast: nodes.Template) -> t.Iterator[t.Optional[str]]:
"""Finds all the referenced templates from the AST. This will return an
iterator over all the hardcoded template extensions, inclusions and
imports. If dynamic inheritance or inclusion is used, `None` will be
@ -72,16 +73,19 @@ def find_referenced_templates(ast):
This function is useful for dependency tracking. For example if you want
to rebuild parts of the website after a layout template has changed.
"""
for node in ast.find_all((nodes.Extends, nodes.FromImport, nodes.Import,
nodes.Include)):
if not isinstance(node.template, nodes.Const):
template_name: t.Any
for node in ast.find_all(_ref_types):
template: nodes.Expr = node.template # type: ignore
if not isinstance(template, nodes.Const):
# a tuple with some non consts in there
if isinstance(node.template, (nodes.Tuple, nodes.List)):
for template_name in node.template.items:
if isinstance(template, (nodes.Tuple, nodes.List)):
for template_name in template.items:
# something const, only yield the strings and ignore
# non-string consts that really just make no sense
if isinstance(template_name, nodes.Const):
if isinstance(template_name.value, string_types):
if isinstance(template_name.value, str):
yield template_name.value
# something dynamic in there
else:
@ -91,15 +95,16 @@ def find_referenced_templates(ast):
yield None
continue
# constant is a basestring, direct template name
if isinstance(node.template.value, string_types):
yield node.template.value
if isinstance(template.value, str):
yield template.value
# a tuple or list (latter *should* not happen) made of consts,
# yield the consts that are strings. We could warn here for
# non string values
elif isinstance(node, nodes.Include) and \
isinstance(node.template.value, (tuple, list)):
for template_name in node.template.value:
if isinstance(template_name, string_types):
elif isinstance(node, nodes.Include) and isinstance(
template.value, (tuple, list)
):
for template_name in template.value:
if isinstance(template_name, str):
yield template_name
# something else we don't care about, we could warn here
else:

View File

@ -1,220 +1,130 @@
import sys
import typing as t
from ast import literal_eval
from itertools import islice, chain
from jinja2 import nodes
from jinja2._compat import text_type
from jinja2.compiler import CodeGenerator, has_safe_repr
from jinja2.environment import Environment, Template
from jinja2.utils import concat, escape
from ast import parse
from itertools import chain
from itertools import islice
from types import GeneratorType
from . import nodes
from .compiler import CodeGenerator
from .compiler import Frame
from .compiler import has_safe_repr
from .environment import Environment
from .environment import Template
def native_concat(nodes):
"""Return a native Python type from the list of compiled nodes. If the
result is a single node, its value is returned. Otherwise, the nodes are
concatenated as strings. If the result can be parsed with
:func:`ast.literal_eval`, the parsed value is returned. Otherwise, the
string is returned.
def native_concat(values: t.Iterable[t.Any]) -> t.Optional[t.Any]:
"""Return a native Python type from the list of compiled nodes. If
the result is a single node, its value is returned. Otherwise, the
nodes are concatenated as strings. If the result can be parsed with
:func:`ast.literal_eval`, the parsed value is returned. Otherwise,
the string is returned.
:param values: Iterable of outputs to concatenate.
"""
head = list(islice(nodes, 2))
head = list(islice(values, 2))
if not head:
return None
if len(head) == 1:
out = head[0]
raw = head[0]
if not isinstance(raw, str):
return raw
else:
out = u''.join([text_type(v) for v in chain(head, nodes)])
if isinstance(values, GeneratorType):
values = chain(head, values)
raw = "".join([str(v) for v in values])
try:
return literal_eval(out)
return literal_eval(
# In Python 3.10+ ast.literal_eval removes leading spaces/tabs
# from the given string. For backwards compatibility we need to
# parse the string ourselves without removing leading spaces/tabs.
parse(raw, mode="eval")
)
except (ValueError, SyntaxError, MemoryError):
return out
return raw
class NativeCodeGenerator(CodeGenerator):
"""A code generator which avoids injecting ``to_string()`` calls around the
internal code Jinja uses to render templates.
"""A code generator which renders Python types by not adding
``str()`` around output nodes.
"""
def visit_Output(self, node, frame):
"""Same as :meth:`CodeGenerator.visit_Output`, but do not call
``to_string`` on output nodes in generated code.
"""
if self.has_known_extends and frame.require_output_check:
return
@staticmethod
def _default_finalize(value: t.Any) -> t.Any:
return value
finalize = self.environment.finalize
finalize_context = getattr(finalize, 'contextfunction', False)
finalize_eval = getattr(finalize, 'evalcontextfunction', False)
finalize_env = getattr(finalize, 'environmentfunction', False)
def _output_const_repr(self, group: t.Iterable[t.Any]) -> str:
return repr("".join([str(v) for v in group]))
if finalize is not None:
if finalize_context or finalize_eval:
const_finalize = None
elif finalize_env:
def const_finalize(x):
return finalize(self.environment, x)
else:
const_finalize = finalize
else:
def const_finalize(x):
return x
def _output_child_to_const(
self, node: nodes.Expr, frame: Frame, finalize: CodeGenerator._FinalizeInfo
) -> t.Any:
const = node.as_const(frame.eval_ctx)
# If we are inside a frame that requires output checking, we do so.
outdent_later = False
if not has_safe_repr(const):
raise nodes.Impossible()
if frame.require_output_check:
self.writeline('if parent_template is None:')
self.indent()
outdent_later = True
if isinstance(node, nodes.TemplateData):
return const
# Try to evaluate as many chunks as possible into a static string at
# compile time.
body = []
return finalize.const(const) # type: ignore
for child in node.nodes:
try:
if const_finalize is None:
raise nodes.Impossible()
def _output_child_pre(
self, node: nodes.Expr, frame: Frame, finalize: CodeGenerator._FinalizeInfo
) -> None:
if finalize.src is not None:
self.write(finalize.src)
const = child.as_const(frame.eval_ctx)
if not has_safe_repr(const):
raise nodes.Impossible()
except nodes.Impossible:
body.append(child)
continue
# the frame can't be volatile here, because otherwise the as_const
# function would raise an Impossible exception at that point
try:
if frame.eval_ctx.autoescape:
if hasattr(const, '__html__'):
const = const.__html__()
else:
const = escape(const)
const = const_finalize(const)
except Exception:
# if something goes wrong here we evaluate the node at runtime
# for easier debugging
body.append(child)
continue
if body and isinstance(body[-1], list):
body[-1].append(const)
else:
body.append([const])
# if we have less than 3 nodes or a buffer we yield or extend/append
if len(body) < 3 or frame.buffer is not None:
if frame.buffer is not None:
# for one item we append, for more we extend
if len(body) == 1:
self.writeline('%s.append(' % frame.buffer)
else:
self.writeline('%s.extend((' % frame.buffer)
self.indent()
for item in body:
if isinstance(item, list):
val = repr(native_concat(item))
if frame.buffer is None:
self.writeline('yield ' + val)
else:
self.writeline(val + ',')
else:
if frame.buffer is None:
self.writeline('yield ', item)
else:
self.newline(item)
close = 0
if finalize is not None:
self.write('environment.finalize(')
if finalize_context:
self.write('context, ')
close += 1
self.visit(item, frame)
if close > 0:
self.write(')' * close)
if frame.buffer is not None:
self.write(',')
if frame.buffer is not None:
# close the open parentheses
self.outdent()
self.writeline(len(body) == 1 and ')' or '))')
# otherwise we create a format string as this is faster in that case
else:
format = []
arguments = []
for item in body:
if isinstance(item, list):
format.append(native_concat(item).replace('%', '%%'))
else:
format.append('%s')
arguments.append(item)
self.writeline('yield ')
self.write(repr(concat(format)) + ' % (')
self.indent()
for argument in arguments:
self.newline(argument)
close = 0
if finalize is not None:
self.write('environment.finalize(')
if finalize_context:
self.write('context, ')
elif finalize_eval:
self.write('context.eval_ctx, ')
elif finalize_env:
self.write('environment, ')
close += 1
self.visit(argument, frame)
self.write(')' * close + ', ')
self.outdent()
self.writeline(')')
if outdent_later:
self.outdent()
class NativeTemplate(Template):
def render(self, *args, **kwargs):
"""Render the template to produce a native Python type. If the result
is a single node, its value is returned. Otherwise, the nodes are
concatenated as strings. If the result can be parsed with
:func:`ast.literal_eval`, the parsed value is returned. Otherwise, the
string is returned.
"""
vars = dict(*args, **kwargs)
try:
return native_concat(self.root_render_func(self.new_context(vars)))
except Exception:
exc_info = sys.exc_info()
return self.environment.handle_exception(exc_info, True)
def _output_child_post(
self, node: nodes.Expr, frame: Frame, finalize: CodeGenerator._FinalizeInfo
) -> None:
if finalize.src is not None:
self.write(")")
class NativeEnvironment(Environment):
"""An environment that renders templates to native Python types."""
code_generator_class = NativeCodeGenerator
template_class = NativeTemplate
concat = staticmethod(native_concat) # type: ignore
class NativeTemplate(Template):
environment_class = NativeEnvironment
def render(self, *args: t.Any, **kwargs: t.Any) -> t.Any:
"""Render the template to produce a native Python type. If the
result is a single node, its value is returned. Otherwise, the
nodes are concatenated as strings. If the result can be parsed
with :func:`ast.literal_eval`, the parsed value is returned.
Otherwise, the string is returned.
"""
ctx = self.new_context(dict(*args, **kwargs))
try:
return self.environment_class.concat( # type: ignore
self.root_render_func(ctx) # type: ignore
)
except Exception:
return self.environment.handle_exception()
async def render_async(self, *args: t.Any, **kwargs: t.Any) -> t.Any:
if not self.environment.is_async:
raise RuntimeError(
"The environment was not created with async mode enabled."
)
ctx = self.new_context(dict(*args, **kwargs))
try:
return self.environment_class.concat( # type: ignore
[n async for n in self.root_render_func(ctx)] # type: ignore
)
except Exception:
return self.environment.handle_exception()
NativeEnvironment.template_class = NativeTemplate

File diff suppressed because it is too large Load Diff

View File

@ -1,49 +1,47 @@
# -*- coding: utf-8 -*-
"""The optimizer tries to constant fold expressions and modify the AST
in place so that it should be faster to evaluate.
Because the AST does not contain all the scoping information and the
compiler has to find that out, we cannot do all the optimizations we
want. For example, loop unrolling doesn't work because unrolled loops
would have a different scope. The solution would be a second syntax tree
that stored the scoping rules.
"""
jinja2.optimizer
~~~~~~~~~~~~~~~~
import typing as t
The jinja optimizer is currently trying to constant fold a few expressions
and modify the AST in place so that it should be easier to evaluate it.
from . import nodes
from .visitor import NodeTransformer
Because the AST does not contain all the scoping information and the
compiler has to find that out, we cannot do all the optimizations we
want. For example loop unrolling doesn't work because unrolled loops would
have a different scoping.
The solution would be a second syntax tree that has the scoping rules stored.
:copyright: (c) 2017 by the Jinja Team.
:license: BSD.
"""
from jinja2 import nodes
from jinja2.visitor import NodeTransformer
if t.TYPE_CHECKING:
from .environment import Environment
def optimize(node, environment):
def optimize(node: nodes.Node, environment: "Environment") -> nodes.Node:
"""The context hint can be used to perform an static optimization
based on the context given."""
optimizer = Optimizer(environment)
return optimizer.visit(node)
return t.cast(nodes.Node, optimizer.visit(node))
class Optimizer(NodeTransformer):
def __init__(self, environment):
def __init__(self, environment: "t.Optional[Environment]") -> None:
self.environment = environment
def fold(self, node, eval_ctx=None):
"""Do constant folding."""
node = self.generic_visit(node)
try:
return nodes.Const.from_untrusted(node.as_const(eval_ctx),
lineno=node.lineno,
environment=self.environment)
except nodes.Impossible:
return node
def generic_visit(
self, node: nodes.Node, *args: t.Any, **kwargs: t.Any
) -> nodes.Node:
node = super().generic_visit(node, *args, **kwargs)
visit_Add = visit_Sub = visit_Mul = visit_Div = visit_FloorDiv = \
visit_Pow = visit_Mod = visit_And = visit_Or = visit_Pos = visit_Neg = \
visit_Not = visit_Compare = visit_Getitem = visit_Getattr = visit_Call = \
visit_Filter = visit_Test = visit_CondExpr = fold
del fold
# Do constant folding. Some other nodes besides Expr have
# as_const, but folding them causes errors later on.
if isinstance(node, nodes.Expr):
try:
return nodes.Const.from_untrusted(
node.as_const(args[0] if args else None),
lineno=node.lineno,
environment=self.environment,
)
except nodes.Impossible:
pass
return node

File diff suppressed because it is too large Load Diff

View File

File diff suppressed because it is too large Load Diff

View File

@ -1,174 +1,128 @@
# -*- coding: utf-8 -*-
"""A sandbox layer that ensures unsafe operations cannot be performed.
Useful when the template itself comes from an untrusted source.
"""
jinja2.sandbox
~~~~~~~~~~~~~~
Adds a sandbox layer to Jinja as it was the default behavior in the old
Jinja 1 releases. This sandbox is slightly different from Jinja 1 as the
default behavior is easier to use.
The behavior can be changed by subclassing the environment.
:copyright: (c) 2017 by the Jinja Team.
:license: BSD.
"""
import types
import operator
from collections import Mapping
from jinja2.environment import Environment
from jinja2.exceptions import SecurityError
from jinja2._compat import string_types, PY2
from jinja2.utils import Markup
from markupsafe import EscapeFormatter
import types
import typing as t
from _string import formatter_field_name_split # type: ignore
from collections import abc
from collections import deque
from string import Formatter
from markupsafe import EscapeFormatter
from markupsafe import Markup
from .environment import Environment
from .exceptions import SecurityError
from .runtime import Context
from .runtime import Undefined
F = t.TypeVar("F", bound=t.Callable[..., t.Any])
#: maximum number of items a range may produce
MAX_RANGE = 100000
#: attributes of function objects that are considered unsafe.
if PY2:
UNSAFE_FUNCTION_ATTRIBUTES = set(['func_closure', 'func_code', 'func_dict',
'func_defaults', 'func_globals'])
else:
# On versions > python 2 the special attributes on functions are gone,
# but they remain on methods and generators for whatever reason.
UNSAFE_FUNCTION_ATTRIBUTES = set()
#: Unsafe function attributes.
UNSAFE_FUNCTION_ATTRIBUTES: t.Set[str] = set()
#: Unsafe method attributes. Function attributes are unsafe for methods too.
UNSAFE_METHOD_ATTRIBUTES: t.Set[str] = set()
#: unsafe method attributes. function attributes are unsafe for methods too
UNSAFE_METHOD_ATTRIBUTES = set(['im_class', 'im_func', 'im_self'])
#: unsafe generator attirbutes.
UNSAFE_GENERATOR_ATTRIBUTES = set(['gi_frame', 'gi_code'])
#: unsafe generator attributes.
UNSAFE_GENERATOR_ATTRIBUTES = {"gi_frame", "gi_code"}
#: unsafe attributes on coroutines
UNSAFE_COROUTINE_ATTRIBUTES = set(['cr_frame', 'cr_code'])
UNSAFE_COROUTINE_ATTRIBUTES = {"cr_frame", "cr_code"}
#: unsafe attributes on async generators
UNSAFE_ASYNC_GENERATOR_ATTRIBUTES = set(['ag_code', 'ag_frame'])
UNSAFE_ASYNC_GENERATOR_ATTRIBUTES = {"ag_code", "ag_frame"}
import warnings
# make sure we don't warn in python 2.6 about stuff we don't care about
warnings.filterwarnings('ignore', 'the sets module', DeprecationWarning,
module='jinja2.sandbox')
from collections import deque
_mutable_set_types = (set,)
_mutable_mapping_types = (dict,)
_mutable_sequence_types = (list,)
# on python 2.x we can register the user collection types
try:
from UserDict import UserDict, DictMixin
from UserList import UserList
_mutable_mapping_types += (UserDict, DictMixin)
_mutable_set_types += (UserList,)
except ImportError:
pass
# if sets is still available, register the mutable set from there as well
try:
from sets import Set
_mutable_set_types += (Set,)
except ImportError:
pass
#: register Python 2.6 abstract base classes
from collections import MutableSet, MutableMapping, MutableSequence
_mutable_set_types += (MutableSet,)
_mutable_mapping_types += (MutableMapping,)
_mutable_sequence_types += (MutableSequence,)
_mutable_spec = (
(_mutable_set_types, frozenset([
'add', 'clear', 'difference_update', 'discard', 'pop', 'remove',
'symmetric_difference_update', 'update'
])),
(_mutable_mapping_types, frozenset([
'clear', 'pop', 'popitem', 'setdefault', 'update'
])),
(_mutable_sequence_types, frozenset([
'append', 'reverse', 'insert', 'sort', 'extend', 'remove'
])),
(deque, frozenset([
'append', 'appendleft', 'clear', 'extend', 'extendleft', 'pop',
'popleft', 'remove', 'rotate'
]))
_mutable_spec: t.Tuple[t.Tuple[t.Type, t.FrozenSet[str]], ...] = (
(
abc.MutableSet,
frozenset(
[
"add",
"clear",
"difference_update",
"discard",
"pop",
"remove",
"symmetric_difference_update",
"update",
]
),
),
(
abc.MutableMapping,
frozenset(["clear", "pop", "popitem", "setdefault", "update"]),
),
(
abc.MutableSequence,
frozenset(["append", "reverse", "insert", "sort", "extend", "remove"]),
),
(
deque,
frozenset(
[
"append",
"appendleft",
"clear",
"extend",
"extendleft",
"pop",
"popleft",
"remove",
"rotate",
]
),
),
)
class _MagicFormatMapping(Mapping):
"""This class implements a dummy wrapper to fix a bug in the Python
standard library for string formatting.
See https://bugs.python.org/issue13598 for information about why
this is necessary.
"""
def __init__(self, args, kwargs):
self._args = args
self._kwargs = kwargs
self._last_index = 0
def __getitem__(self, key):
if key == '':
idx = self._last_index
self._last_index += 1
try:
return self._args[idx]
except LookupError:
pass
key = str(idx)
return self._kwargs[key]
def __iter__(self):
return iter(self._kwargs)
def __len__(self):
return len(self._kwargs)
def inspect_format_method(callable):
if not isinstance(callable, (types.MethodType,
types.BuiltinMethodType)) or \
callable.__name__ != 'format':
def inspect_format_method(callable: t.Callable) -> t.Optional[str]:
if not isinstance(
callable, (types.MethodType, types.BuiltinMethodType)
) or callable.__name__ not in ("format", "format_map"):
return None
obj = callable.__self__
if isinstance(obj, string_types):
if isinstance(obj, str):
return obj
return None
def safe_range(*args):
def safe_range(*args: int) -> range:
"""A range that can't generate ranges with a length of more than
MAX_RANGE items.
"""
rng = range(*args)
if len(rng) > MAX_RANGE:
raise OverflowError('range too big, maximum size for range is %d' %
MAX_RANGE)
raise OverflowError(
"Range too big. The sandbox blocks ranges larger than"
f" MAX_RANGE ({MAX_RANGE})."
)
return rng
def unsafe(f):
def unsafe(f: F) -> F:
"""Marks a function or method as unsafe.
::
.. code-block: python
@unsafe
def delete(self):
pass
"""
f.unsafe_callable = True
f.unsafe_callable = True # type: ignore
return f
def is_internal_attribute(obj, attr):
def is_internal_attribute(obj: t.Any, attr: str) -> bool:
"""Test if the attribute given is an internal python attribute. For
example this function returns `True` for the `func_code` attribute of
python objects. This is useful if the environment method
@ -184,32 +138,31 @@ def is_internal_attribute(obj, attr):
if attr in UNSAFE_FUNCTION_ATTRIBUTES:
return True
elif isinstance(obj, types.MethodType):
if attr in UNSAFE_FUNCTION_ATTRIBUTES or \
attr in UNSAFE_METHOD_ATTRIBUTES:
if attr in UNSAFE_FUNCTION_ATTRIBUTES or attr in UNSAFE_METHOD_ATTRIBUTES:
return True
elif isinstance(obj, type):
if attr == 'mro':
if attr == "mro":
return True
elif isinstance(obj, (types.CodeType, types.TracebackType, types.FrameType)):
return True
elif isinstance(obj, types.GeneratorType):
if attr in UNSAFE_GENERATOR_ATTRIBUTES:
return True
elif hasattr(types, 'CoroutineType') and isinstance(obj, types.CoroutineType):
elif hasattr(types, "CoroutineType") and isinstance(obj, types.CoroutineType):
if attr in UNSAFE_COROUTINE_ATTRIBUTES:
return True
elif hasattr(types, 'AsyncGeneratorType') and isinstance(obj, types.AsyncGeneratorType):
elif hasattr(types, "AsyncGeneratorType") and isinstance(
obj, types.AsyncGeneratorType
):
if attr in UNSAFE_ASYNC_GENERATOR_ATTRIBUTES:
return True
return attr.startswith('__')
return attr.startswith("__")
def modifies_known_mutable(obj, attr):
def modifies_known_mutable(obj: t.Any, attr: str) -> bool:
"""This function checks if an attribute on a builtin mutable object
(list, dict, set or deque) would modify it if called. It also supports
the "user"-versions of the objects (`sets.Set`, `UserDict.*` etc.) and
with Python 2.6 onwards the abstract base classes `MutableSet`,
`MutableMapping`, and `MutableSequence`.
(list, dict, set or deque) or the corresponding ABCs would modify it
if called.
>>> modifies_known_mutable({}, "clear")
True
@ -220,8 +173,7 @@ def modifies_known_mutable(obj, attr):
>>> modifies_known_mutable([], "index")
False
If called with an unsupported object (such as unicode) `False` is
returned.
If called with an unsupported object, ``False`` is returned.
>>> modifies_known_mutable("foo", "upper")
False
@ -242,27 +194,28 @@ class SandboxedEnvironment(Environment):
raised. However also other exceptions may occur during the rendering so
the caller has to ensure that all exceptions are caught.
"""
sandboxed = True
#: default callback table for the binary operators. A copy of this is
#: available on each instance of a sandboxed environment as
#: :attr:`binop_table`
default_binop_table = {
'+': operator.add,
'-': operator.sub,
'*': operator.mul,
'/': operator.truediv,
'//': operator.floordiv,
'**': operator.pow,
'%': operator.mod
default_binop_table: t.Dict[str, t.Callable[[t.Any, t.Any], t.Any]] = {
"+": operator.add,
"-": operator.sub,
"*": operator.mul,
"/": operator.truediv,
"//": operator.floordiv,
"**": operator.pow,
"%": operator.mod,
}
#: default callback table for the unary operators. A copy of this is
#: available on each instance of a sandboxed environment as
#: :attr:`unop_table`
default_unop_table = {
'+': operator.pos,
'-': operator.neg
default_unop_table: t.Dict[str, t.Callable[[t.Any], t.Any]] = {
"+": operator.pos,
"-": operator.neg,
}
#: a set of binary operators that should be intercepted. Each operator
@ -279,7 +232,7 @@ class SandboxedEnvironment(Environment):
#: interested in.
#:
#: .. versionadded:: 2.6
intercepted_binops = frozenset()
intercepted_binops: t.FrozenSet[str] = frozenset()
#: a set of unary operators that should be intercepted. Each operator
#: that is added to this set (empty by default) is delegated to the
@ -294,51 +247,37 @@ class SandboxedEnvironment(Environment):
#: interested in.
#:
#: .. versionadded:: 2.6
intercepted_unops = frozenset()
intercepted_unops: t.FrozenSet[str] = frozenset()
def intercept_unop(self, operator):
"""Called during template compilation with the name of a unary
operator to check if it should be intercepted at runtime. If this
method returns `True`, :meth:`call_unop` is excuted for this unary
operator. The default implementation of :meth:`call_unop` will use
the :attr:`unop_table` dictionary to perform the operator with the
same logic as the builtin one.
The following unary operators are interceptable: ``+`` and ``-``
Intercepted calls are always slower than the native operator call,
so make sure only to intercept the ones you are interested in.
.. versionadded:: 2.6
"""
return False
def __init__(self, *args, **kwargs):
Environment.__init__(self, *args, **kwargs)
self.globals['range'] = safe_range
def __init__(self, *args: t.Any, **kwargs: t.Any) -> None:
super().__init__(*args, **kwargs)
self.globals["range"] = safe_range
self.binop_table = self.default_binop_table.copy()
self.unop_table = self.default_unop_table.copy()
def is_safe_attribute(self, obj, attr, value):
def is_safe_attribute(self, obj: t.Any, attr: str, value: t.Any) -> bool:
"""The sandboxed environment will call this method to check if the
attribute of an object is safe to access. Per default all attributes
starting with an underscore are considered private as well as the
special attributes of internal python objects as returned by the
:func:`is_internal_attribute` function.
"""
return not (attr.startswith('_') or is_internal_attribute(obj, attr))
return not (attr.startswith("_") or is_internal_attribute(obj, attr))
def is_safe_callable(self, obj):
"""Check if an object is safely callable. Per default a function is
considered safe unless the `unsafe_callable` attribute exists and is
True. Override this method to alter the behavior, but this won't
affect the `unsafe` decorator from this module.
def is_safe_callable(self, obj: t.Any) -> bool:
"""Check if an object is safely callable. By default callables
are considered safe unless decorated with :func:`unsafe`.
This also recognizes the Django convention of setting
``func.alters_data = True``.
"""
return not (getattr(obj, 'unsafe_callable', False) or
getattr(obj, 'alters_data', False))
return not (
getattr(obj, "unsafe_callable", False) or getattr(obj, "alters_data", False)
)
def call_binop(self, context, operator, left, right):
def call_binop(
self, context: Context, operator: str, left: t.Any, right: t.Any
) -> t.Any:
"""For intercepted binary operator calls (:meth:`intercepted_binops`)
this function is executed instead of the builtin operator. This can
be used to fine tune the behavior of certain operators.
@ -347,7 +286,7 @@ class SandboxedEnvironment(Environment):
"""
return self.binop_table[operator](left, right)
def call_unop(self, context, operator, arg):
def call_unop(self, context: Context, operator: str, arg: t.Any) -> t.Any:
"""For intercepted unary operator calls (:meth:`intercepted_unops`)
this function is executed instead of the builtin operator. This can
be used to fine tune the behavior of certain operators.
@ -356,12 +295,14 @@ class SandboxedEnvironment(Environment):
"""
return self.unop_table[operator](arg)
def getitem(self, obj, argument):
def getitem(
self, obj: t.Any, argument: t.Union[str, t.Any]
) -> t.Union[t.Any, Undefined]:
"""Subscribe an object from sandboxed code."""
try:
return obj[argument]
except (TypeError, LookupError):
if isinstance(argument, string_types):
if isinstance(argument, str):
try:
attr = str(argument)
except Exception:
@ -377,7 +318,7 @@ class SandboxedEnvironment(Environment):
return self.unsafe_undefined(obj, argument)
return self.undefined(obj=obj, name=argument)
def getattr(self, obj, attribute):
def getattr(self, obj: t.Any, attribute: str) -> t.Union[t.Any, Undefined]:
"""Subscribe an object from sandboxed code and prefer the
attribute. The attribute passed *must* be a bytestring.
"""
@ -394,36 +335,61 @@ class SandboxedEnvironment(Environment):
return self.unsafe_undefined(obj, attribute)
return self.undefined(obj=obj, name=attribute)
def unsafe_undefined(self, obj, attribute):
def unsafe_undefined(self, obj: t.Any, attribute: str) -> Undefined:
"""Return an undefined object for unsafe attributes."""
return self.undefined('access to attribute %r of %r '
'object is unsafe.' % (
attribute,
obj.__class__.__name__
), name=attribute, obj=obj, exc=SecurityError)
return self.undefined(
f"access to attribute {attribute!r} of"
f" {type(obj).__name__!r} object is unsafe.",
name=attribute,
obj=obj,
exc=SecurityError,
)
def format_string(self, s, args, kwargs):
def format_string(
self,
s: str,
args: t.Tuple[t.Any, ...],
kwargs: t.Dict[str, t.Any],
format_func: t.Optional[t.Callable] = None,
) -> str:
"""If a format call is detected, then this is routed through this
method so that our safety sandbox can be used for it.
"""
formatter: SandboxedFormatter
if isinstance(s, Markup):
formatter = SandboxedEscapeFormatter(self, s.escape)
formatter = SandboxedEscapeFormatter(self, escape=s.escape)
else:
formatter = SandboxedFormatter(self)
kwargs = _MagicFormatMapping(args, kwargs)
if format_func is not None and format_func.__name__ == "format_map":
if len(args) != 1 or kwargs:
raise TypeError(
"format_map() takes exactly one argument"
f" {len(args) + (kwargs is not None)} given"
)
kwargs = args[0]
args = ()
rv = formatter.vformat(s, args, kwargs)
return type(s)(rv)
def call(__self, __context, __obj, *args, **kwargs):
def call(
__self, # noqa: B902
__context: Context,
__obj: t.Any,
*args: t.Any,
**kwargs: t.Any,
) -> t.Any:
"""Call an object from sandboxed code."""
fmt = inspect_format_method(__obj)
if fmt is not None:
return __self.format_string(fmt, args, kwargs)
return __self.format_string(fmt, args, kwargs, __obj)
# the double prefixes are to avoid double keyword argument
# errors when proxying the call.
if not __self.is_safe_callable(__obj):
raise SecurityError('%r is not safely callable' % (__obj,))
raise SecurityError(f"{__obj!r} is not safely callable")
return __context.call(__obj, *args, **kwargs)
@ -433,26 +399,21 @@ class ImmutableSandboxedEnvironment(SandboxedEnvironment):
`dict` by using the :func:`modifies_known_mutable` function.
"""
def is_safe_attribute(self, obj, attr, value):
if not SandboxedEnvironment.is_safe_attribute(self, obj, attr, value):
def is_safe_attribute(self, obj: t.Any, attr: str, value: t.Any) -> bool:
if not super().is_safe_attribute(obj, attr, value):
return False
return not modifies_known_mutable(obj, attr)
# This really is not a public API apparenlty.
try:
from _string import formatter_field_name_split
except ImportError:
def formatter_field_name_split(field_name):
return field_name._formatter_field_name_split()
class SandboxedFormatterMixin(object):
def __init__(self, env):
class SandboxedFormatter(Formatter):
def __init__(self, env: Environment, **kwargs: t.Any) -> None:
self._env = env
super().__init__(**kwargs)
def get_field(self, field_name, args, kwargs):
def get_field(
self, field_name: str, args: t.Sequence[t.Any], kwargs: t.Mapping[str, t.Any]
) -> t.Tuple[t.Any, str]:
first, rest = formatter_field_name_split(field_name)
obj = self.get_value(first, args, kwargs)
for is_attr, i in rest:
@ -462,14 +423,6 @@ class SandboxedFormatterMixin(object):
obj = self._env.getitem(obj, i)
return obj, first
class SandboxedFormatter(SandboxedFormatterMixin, Formatter):
def __init__(self, env):
SandboxedFormatterMixin.__init__(self, env)
Formatter.__init__(self)
class SandboxedEscapeFormatter(SandboxedFormatterMixin, EscapeFormatter):
def __init__(self, env, escape):
SandboxedFormatterMixin.__init__(self, env)
EscapeFormatter.__init__(self, escape)
class SandboxedEscapeFormatter(SandboxedFormatter, EscapeFormatter):
pass

View File

@ -1,43 +1,32 @@
# -*- coding: utf-8 -*-
"""
jinja2.tests
~~~~~~~~~~~~
Jinja test functions. Used with the "is" operator.
:copyright: (c) 2017 by the Jinja Team.
:license: BSD, see LICENSE for more details.
"""
"""Built-in template tests used with the ``is`` operator."""
import operator
import re
from collections import Mapping
from jinja2.runtime import Undefined
from jinja2._compat import text_type, string_types, integer_types
import decimal
import typing as t
from collections import abc
from numbers import Number
number_re = re.compile(r'^-?\d+(\.\d+)?$')
regex_type = type(number_re)
from .runtime import Undefined
from .utils import pass_environment
if t.TYPE_CHECKING:
from .environment import Environment
test_callable = callable
def test_odd(value):
def test_odd(value: int) -> bool:
"""Return true if the variable is odd."""
return value % 2 == 1
def test_even(value):
def test_even(value: int) -> bool:
"""Return true if the variable is even."""
return value % 2 == 0
def test_divisibleby(value, num):
def test_divisibleby(value: int, num: int) -> bool:
"""Check if a variable is divisible by a number."""
return value % num == 0
def test_defined(value):
def test_defined(value: t.Any) -> bool:
"""Return true if the variable is defined:
.. sourcecode:: jinja
@ -54,57 +43,140 @@ def test_defined(value):
return not isinstance(value, Undefined)
def test_undefined(value):
def test_undefined(value: t.Any) -> bool:
"""Like :func:`defined` but the other way round."""
return isinstance(value, Undefined)
def test_none(value):
@pass_environment
def test_filter(env: "Environment", value: str) -> bool:
"""Check if a filter exists by name. Useful if a filter may be
optionally available.
.. code-block:: jinja
{% if 'markdown' is filter %}
{{ value | markdown }}
{% else %}
{{ value }}
{% endif %}
.. versionadded:: 3.0
"""
return value in env.filters
@pass_environment
def test_test(env: "Environment", value: str) -> bool:
"""Check if a test exists by name. Useful if a test may be
optionally available.
.. code-block:: jinja
{% if 'loud' is test %}
{% if value is loud %}
{{ value|upper }}
{% else %}
{{ value|lower }}
{% endif %}
{% else %}
{{ value }}
{% endif %}
.. versionadded:: 3.0
"""
return value in env.tests
def test_none(value: t.Any) -> bool:
"""Return true if the variable is none."""
return value is None
def test_lower(value):
def test_boolean(value: t.Any) -> bool:
"""Return true if the object is a boolean value.
.. versionadded:: 2.11
"""
return value is True or value is False
def test_false(value: t.Any) -> bool:
"""Return true if the object is False.
.. versionadded:: 2.11
"""
return value is False
def test_true(value: t.Any) -> bool:
"""Return true if the object is True.
.. versionadded:: 2.11
"""
return value is True
# NOTE: The existing 'number' test matches booleans and floats
def test_integer(value: t.Any) -> bool:
"""Return true if the object is an integer.
.. versionadded:: 2.11
"""
return isinstance(value, int) and value is not True and value is not False
# NOTE: The existing 'number' test matches booleans and integers
def test_float(value: t.Any) -> bool:
"""Return true if the object is a float.
.. versionadded:: 2.11
"""
return isinstance(value, float)
def test_lower(value: str) -> bool:
"""Return true if the variable is lowercased."""
return text_type(value).islower()
return str(value).islower()
def test_upper(value):
def test_upper(value: str) -> bool:
"""Return true if the variable is uppercased."""
return text_type(value).isupper()
return str(value).isupper()
def test_string(value):
def test_string(value: t.Any) -> bool:
"""Return true if the object is a string."""
return isinstance(value, string_types)
return isinstance(value, str)
def test_mapping(value):
def test_mapping(value: t.Any) -> bool:
"""Return true if the object is a mapping (dict etc.).
.. versionadded:: 2.6
"""
return isinstance(value, Mapping)
return isinstance(value, abc.Mapping)
def test_number(value):
def test_number(value: t.Any) -> bool:
"""Return true if the variable is a number."""
return isinstance(value, integer_types + (float, complex, decimal.Decimal))
return isinstance(value, Number)
def test_sequence(value):
def test_sequence(value: t.Any) -> bool:
"""Return true if the variable is a sequence. Sequences are variables
that are iterable.
"""
try:
len(value)
value.__getitem__
except:
except Exception:
return False
return True
def test_sameas(value, other):
def test_sameas(value: t.Any, other: t.Any) -> bool:
"""Check if an object points to the same memory address than another
object:
@ -117,21 +189,22 @@ def test_sameas(value, other):
return value is other
def test_iterable(value):
def test_iterable(value: t.Any) -> bool:
"""Check if it's possible to iterate over an object."""
try:
iter(value)
except TypeError:
return False
return True
def test_escaped(value):
def test_escaped(value: t.Any) -> bool:
"""Check if the value is escaped."""
return hasattr(value, '__html__')
return hasattr(value, "__html__")
def test_in(value, seq):
def test_in(value: t.Any, seq: t.Container) -> bool:
"""Check if value is in seq.
.. versionadded:: 2.10
@ -140,36 +213,43 @@ def test_in(value, seq):
TESTS = {
'odd': test_odd,
'even': test_even,
'divisibleby': test_divisibleby,
'defined': test_defined,
'undefined': test_undefined,
'none': test_none,
'lower': test_lower,
'upper': test_upper,
'string': test_string,
'mapping': test_mapping,
'number': test_number,
'sequence': test_sequence,
'iterable': test_iterable,
'callable': test_callable,
'sameas': test_sameas,
'escaped': test_escaped,
'in': test_in,
'==': operator.eq,
'eq': operator.eq,
'equalto': operator.eq,
'!=': operator.ne,
'ne': operator.ne,
'>': operator.gt,
'gt': operator.gt,
'greaterthan': operator.gt,
'ge': operator.ge,
'>=': operator.ge,
'<': operator.lt,
'lt': operator.lt,
'lessthan': operator.lt,
'<=': operator.le,
'le': operator.le,
"odd": test_odd,
"even": test_even,
"divisibleby": test_divisibleby,
"defined": test_defined,
"undefined": test_undefined,
"filter": test_filter,
"test": test_test,
"none": test_none,
"boolean": test_boolean,
"false": test_false,
"true": test_true,
"integer": test_integer,
"float": test_float,
"lower": test_lower,
"upper": test_upper,
"string": test_string,
"mapping": test_mapping,
"number": test_number,
"sequence": test_sequence,
"iterable": test_iterable,
"callable": callable,
"sameas": test_sameas,
"escaped": test_escaped,
"in": test_in,
"==": operator.eq,
"eq": operator.eq,
"equalto": operator.eq,
"!=": operator.ne,
"ne": operator.ne,
">": operator.gt,
"gt": operator.gt,
"greaterthan": operator.gt,
"ge": operator.ge,
">=": operator.ge,
"<": operator.lt,
"lt": operator.lt,
"lessthan": operator.lt,
"<=": operator.le,
"le": operator.le,
}

File diff suppressed because it is too large Load Diff

View File

@ -1,17 +1,19 @@
# -*- coding: utf-8 -*-
"""API for traversing the AST nodes. Implemented by the compiler and
meta introspection.
"""
jinja2.visitor
~~~~~~~~~~~~~~
import typing as t
This module implements a visitor for the nodes.
from .nodes import Node
:copyright: (c) 2017 by the Jinja Team.
:license: BSD.
"""
from jinja2.nodes import Node
if t.TYPE_CHECKING:
import typing_extensions as te
class VisitCallable(te.Protocol):
def __call__(self, node: Node, *args: t.Any, **kwargs: t.Any) -> t.Any:
...
class NodeVisitor(object):
class NodeVisitor:
"""Walks the abstract syntax tree and call visitor functions for every
node found. The visitor functions may return values which will be
forwarded by the `visit` method.
@ -23,25 +25,26 @@ class NodeVisitor(object):
(return value `None`) the `generic_visit` visitor is used instead.
"""
def get_visitor(self, node):
def get_visitor(self, node: Node) -> "t.Optional[VisitCallable]":
"""Return the visitor function for this node or `None` if no visitor
exists for this node. In that case the generic visit function is
used instead.
"""
method = 'visit_' + node.__class__.__name__
return getattr(self, method, None)
return getattr(self, f"visit_{type(node).__name__}", None)
def visit(self, node, *args, **kwargs):
def visit(self, node: Node, *args: t.Any, **kwargs: t.Any) -> t.Any:
"""Visit a node."""
f = self.get_visitor(node)
if f is not None:
return f(node, *args, **kwargs)
return self.generic_visit(node, *args, **kwargs)
def generic_visit(self, node, *args, **kwargs):
def generic_visit(self, node: Node, *args: t.Any, **kwargs: t.Any) -> t.Any:
"""Called if no explicit visitor function exists for a node."""
for node in node.iter_child_nodes():
self.visit(node, *args, **kwargs)
for child_node in node.iter_child_nodes():
self.visit(child_node, *args, **kwargs)
class NodeTransformer(NodeVisitor):
@ -55,7 +58,7 @@ class NodeTransformer(NodeVisitor):
replacement takes place.
"""
def generic_visit(self, node, *args, **kwargs):
def generic_visit(self, node: Node, *args: t.Any, **kwargs: t.Any) -> Node:
for field, old_value in node.iter_fields():
if isinstance(old_value, list):
new_values = []
@ -77,11 +80,13 @@ class NodeTransformer(NodeVisitor):
setattr(node, field, new_node)
return node
def visit_list(self, node, *args, **kwargs):
def visit_list(self, node: Node, *args: t.Any, **kwargs: t.Any) -> t.List[Node]:
"""As transformers may return lists in some places this method
can be used to enforce a list as return value.
"""
rv = self.visit(node, *args, **kwargs)
if not isinstance(rv, list):
rv = [rv]
return [rv]
return rv

View File

@ -0,0 +1,28 @@
Copyright 2010 Pallets
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
1. Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
2. Redistributions in binary form must reproduce the above copyright
notice, this list of conditions and the following disclaimer in the
documentation and/or other materials provided with the distribution.
3. Neither the name of the copyright holder nor the names of its
contributors may be used to endorse or promote products derived from
this software without specific prior written permission.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED
TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

View File

@ -1,305 +1,295 @@
# -*- coding: utf-8 -*-
"""
markupsafe
~~~~~~~~~~
Implements a Markup string.
:copyright: (c) 2010 by Armin Ronacher.
:license: BSD, see LICENSE for more details.
"""
import functools
import re
import string
from collections import Mapping
from markupsafe._compat import text_type, string_types, int_types, \
unichr, iteritems, PY2
import typing as t
__version__ = "1.0"
if t.TYPE_CHECKING:
import typing_extensions as te
__all__ = ['Markup', 'soft_unicode', 'escape', 'escape_silent']
class HasHTML(te.Protocol):
def __html__(self) -> str:
pass
_striptags_re = re.compile(r'(<!--.*?-->|<[^>]*>)')
_entity_re = re.compile(r'&([^& ;]+);')
__version__ = "2.1.1"
_strip_comments_re = re.compile(r"<!--.*?-->")
_strip_tags_re = re.compile(r"<.*?>")
class Markup(text_type):
r"""Marks a string as being safe for inclusion in HTML/XML output without
needing to be escaped. This implements the `__html__` interface a couple
of frameworks and web applications use. :class:`Markup` is a direct
subclass of `unicode` and provides all the methods of `unicode` just that
it escapes arguments passed and always returns `Markup`.
def _simple_escaping_wrapper(name: str) -> t.Callable[..., "Markup"]:
orig = getattr(str, name)
The `escape` function returns markup objects so that double escaping can't
happen.
@functools.wraps(orig)
def wrapped(self: "Markup", *args: t.Any, **kwargs: t.Any) -> "Markup":
args = _escape_argspec(list(args), enumerate(args), self.escape) # type: ignore
_escape_argspec(kwargs, kwargs.items(), self.escape)
return self.__class__(orig(self, *args, **kwargs))
The constructor of the :class:`Markup` class can be used for three
different things: When passed an unicode object it's assumed to be safe,
when passed an object with an HTML representation (has an `__html__`
method) that representation is used, otherwise the object passed is
converted into a unicode string and then assumed to be safe:
return wrapped
>>> Markup("Hello <em>World</em>!")
Markup(u'Hello <em>World</em>!')
>>> class Foo(object):
... def __html__(self):
... return '<a href="#">foo</a>'
class Markup(str):
"""A string that is ready to be safely inserted into an HTML or XML
document, either because it was escaped or because it was marked
safe.
Passing an object to the constructor converts it to text and wraps
it to mark it safe without escaping. To escape the text, use the
:meth:`escape` class method instead.
>>> Markup("Hello, <em>World</em>!")
Markup('Hello, <em>World</em>!')
>>> Markup(42)
Markup('42')
>>> Markup.escape("Hello, <em>World</em>!")
Markup('Hello &lt;em&gt;World&lt;/em&gt;!')
This implements the ``__html__()`` interface that some frameworks
use. Passing an object that implements ``__html__()`` will wrap the
output of that method, marking it safe.
>>> class Foo:
... def __html__(self):
... return '<a href="/foo">foo</a>'
...
>>> Markup(Foo())
Markup(u'<a href="#">foo</a>')
Markup('<a href="/foo">foo</a>')
If you want object passed being always treated as unsafe you can use the
:meth:`escape` classmethod to create a :class:`Markup` object:
This is a subclass of :class:`str`. It has the same methods, but
escapes their arguments and returns a ``Markup`` instance.
>>> Markup.escape("Hello <em>World</em>!")
Markup(u'Hello &lt;em&gt;World&lt;/em&gt;!')
Operations on a markup string are markup aware which means that all
arguments are passed through the :func:`escape` function:
>>> em = Markup("<em>%s</em>")
>>> em % "foo & bar"
Markup(u'<em>foo &amp; bar</em>')
>>> strong = Markup("<strong>%(text)s</strong>")
>>> strong % {'text': '<blink>hacker here</blink>'}
Markup(u'<strong>&lt;blink&gt;hacker here&lt;/blink&gt;</strong>')
>>> Markup("<em>%s</em>") % ("foo & bar",)
Markup('<em>foo &amp; bar</em>')
>>> Markup("<em>Hello</em> ") + "<foo>"
Markup(u'<em>Hello</em> &lt;foo&gt;')
Markup('<em>Hello</em> &lt;foo&gt;')
"""
__slots__ = ()
def __new__(cls, base=u'', encoding=None, errors='strict'):
if hasattr(base, '__html__'):
def __new__(
cls, base: t.Any = "", encoding: t.Optional[str] = None, errors: str = "strict"
) -> "Markup":
if hasattr(base, "__html__"):
base = base.__html__()
if encoding is None:
return text_type.__new__(cls, base)
return text_type.__new__(cls, base, encoding, errors)
def __html__(self):
if encoding is None:
return super().__new__(cls, base)
return super().__new__(cls, base, encoding, errors)
def __html__(self) -> "Markup":
return self
def __add__(self, other):
if isinstance(other, string_types) or hasattr(other, '__html__'):
return self.__class__(super(Markup, self).__add__(self.escape(other)))
def __add__(self, other: t.Union[str, "HasHTML"]) -> "Markup":
if isinstance(other, str) or hasattr(other, "__html__"):
return self.__class__(super().__add__(self.escape(other)))
return NotImplemented
def __radd__(self, other):
if hasattr(other, '__html__') or isinstance(other, string_types):
def __radd__(self, other: t.Union[str, "HasHTML"]) -> "Markup":
if isinstance(other, str) or hasattr(other, "__html__"):
return self.escape(other).__add__(self)
return NotImplemented
def __mul__(self, num):
if isinstance(num, int_types):
return self.__class__(text_type.__mul__(self, num))
def __mul__(self, num: "te.SupportsIndex") -> "Markup":
if isinstance(num, int):
return self.__class__(super().__mul__(num))
return NotImplemented
__rmul__ = __mul__
def __mod__(self, arg):
def __mod__(self, arg: t.Any) -> "Markup":
if isinstance(arg, tuple):
# a tuple of arguments, each wrapped
arg = tuple(_MarkupEscapeHelper(x, self.escape) for x in arg)
else:
elif hasattr(type(arg), "__getitem__") and not isinstance(arg, str):
# a mapping of arguments, wrapped
arg = _MarkupEscapeHelper(arg, self.escape)
return self.__class__(text_type.__mod__(self, arg))
else:
# a single argument, wrapped with the helper and a tuple
arg = (_MarkupEscapeHelper(arg, self.escape),)
def __repr__(self):
return '%s(%s)' % (
self.__class__.__name__,
text_type.__repr__(self)
)
return self.__class__(super().__mod__(arg))
def join(self, seq):
return self.__class__(text_type.join(self, map(self.escape, seq)))
join.__doc__ = text_type.join.__doc__
def __repr__(self) -> str:
return f"{self.__class__.__name__}({super().__repr__()})"
def split(self, *args, **kwargs):
return list(map(self.__class__, text_type.split(self, *args, **kwargs)))
split.__doc__ = text_type.split.__doc__
def join(self, seq: t.Iterable[t.Union[str, "HasHTML"]]) -> "Markup":
return self.__class__(super().join(map(self.escape, seq)))
def rsplit(self, *args, **kwargs):
return list(map(self.__class__, text_type.rsplit(self, *args, **kwargs)))
rsplit.__doc__ = text_type.rsplit.__doc__
join.__doc__ = str.join.__doc__
def splitlines(self, *args, **kwargs):
return list(map(self.__class__, text_type.splitlines(
self, *args, **kwargs)))
splitlines.__doc__ = text_type.splitlines.__doc__
def split( # type: ignore
self, sep: t.Optional[str] = None, maxsplit: int = -1
) -> t.List["Markup"]:
return [self.__class__(v) for v in super().split(sep, maxsplit)]
def unescape(self):
r"""Unescape markup again into an text_type string. This also resolves
known HTML4 and XHTML entities:
split.__doc__ = str.split.__doc__
def rsplit( # type: ignore
self, sep: t.Optional[str] = None, maxsplit: int = -1
) -> t.List["Markup"]:
return [self.__class__(v) for v in super().rsplit(sep, maxsplit)]
rsplit.__doc__ = str.rsplit.__doc__
def splitlines(self, keepends: bool = False) -> t.List["Markup"]: # type: ignore
return [self.__class__(v) for v in super().splitlines(keepends)]
splitlines.__doc__ = str.splitlines.__doc__
def unescape(self) -> str:
"""Convert escaped markup back into a text string. This replaces
HTML entities with the characters they represent.
>>> Markup("Main &raquo; <em>About</em>").unescape()
u'Main \xbb <em>About</em>'
'Main » <em>About</em>'
"""
from markupsafe._constants import HTML_ENTITIES
def handle_match(m):
name = m.group(1)
if name in HTML_ENTITIES:
return unichr(HTML_ENTITIES[name])
try:
if name[:2] in ('#x', '#X'):
return unichr(int(name[2:], 16))
elif name.startswith('#'):
return unichr(int(name[1:]))
except ValueError:
pass
# Don't modify unexpected input.
return m.group()
return _entity_re.sub(handle_match, text_type(self))
from html import unescape
def striptags(self):
r"""Unescape markup into an text_type string and strip all tags. This
also resolves known HTML4 and XHTML entities. Whitespace is
normalized to one:
return unescape(str(self))
>>> Markup("Main &raquo; <em>About</em>").striptags()
u'Main \xbb About'
def striptags(self) -> str:
""":meth:`unescape` the markup, remove tags, and normalize
whitespace to single spaces.
>>> Markup("Main &raquo;\t<em>About</em>").striptags()
'Main » About'
"""
stripped = u' '.join(_striptags_re.sub('', self).split())
return Markup(stripped).unescape()
# Use two regexes to avoid ambiguous matches.
value = _strip_comments_re.sub("", self)
value = _strip_tags_re.sub("", value)
value = " ".join(value.split())
return Markup(value).unescape()
@classmethod
def escape(cls, s):
"""Escape the string. Works like :func:`escape` with the difference
that for subclasses of :class:`Markup` this function would return the
correct subclass.
def escape(cls, s: t.Any) -> "Markup":
"""Escape a string. Calls :func:`escape` and ensures that for
subclasses the correct type is returned.
"""
rv = escape(s)
if rv.__class__ is not cls:
return cls(rv)
return rv
def make_simple_escaping_wrapper(name):
orig = getattr(text_type, name)
def func(self, *args, **kwargs):
args = _escape_argspec(list(args), enumerate(args), self.escape)
_escape_argspec(kwargs, iteritems(kwargs), self.escape)
return self.__class__(orig(self, *args, **kwargs))
func.__name__ = orig.__name__
func.__doc__ = orig.__doc__
return func
for method in (
"__getitem__",
"capitalize",
"title",
"lower",
"upper",
"replace",
"ljust",
"rjust",
"lstrip",
"rstrip",
"center",
"strip",
"translate",
"expandtabs",
"swapcase",
"zfill",
):
locals()[method] = _simple_escaping_wrapper(method)
for method in '__getitem__', 'capitalize', \
'title', 'lower', 'upper', 'replace', 'ljust', \
'rjust', 'lstrip', 'rstrip', 'center', 'strip', \
'translate', 'expandtabs', 'swapcase', 'zfill':
locals()[method] = make_simple_escaping_wrapper(method)
del method
# new in python 2.5
if hasattr(text_type, 'partition'):
def partition(self, sep):
return tuple(map(self.__class__,
text_type.partition(self, self.escape(sep))))
def rpartition(self, sep):
return tuple(map(self.__class__,
text_type.rpartition(self, self.escape(sep))))
def partition(self, sep: str) -> t.Tuple["Markup", "Markup", "Markup"]:
l, s, r = super().partition(self.escape(sep))
cls = self.__class__
return cls(l), cls(s), cls(r)
# new in python 2.6
if hasattr(text_type, 'format'):
def format(*args, **kwargs):
self, args = args[0], args[1:]
formatter = EscapeFormatter(self.escape)
kwargs = _MagicFormatMapping(args, kwargs)
return self.__class__(formatter.vformat(self, args, kwargs))
def rpartition(self, sep: str) -> t.Tuple["Markup", "Markup", "Markup"]:
l, s, r = super().rpartition(self.escape(sep))
cls = self.__class__
return cls(l), cls(s), cls(r)
def __html_format__(self, format_spec):
def format(self, *args: t.Any, **kwargs: t.Any) -> "Markup":
formatter = EscapeFormatter(self.escape)
return self.__class__(formatter.vformat(self, args, kwargs))
def __html_format__(self, format_spec: str) -> "Markup":
if format_spec:
raise ValueError("Unsupported format specification for Markup.")
return self
class EscapeFormatter(string.Formatter):
__slots__ = ("escape",)
def __init__(self, escape: t.Callable[[t.Any], Markup]) -> None:
self.escape = escape
super().__init__()
def format_field(self, value: t.Any, format_spec: str) -> str:
if hasattr(value, "__html_format__"):
rv = value.__html_format__(format_spec)
elif hasattr(value, "__html__"):
if format_spec:
raise ValueError('Unsupported format specification '
'for Markup.')
return self
# not in python 3
if hasattr(text_type, '__getslice__'):
__getslice__ = make_simple_escaping_wrapper('__getslice__')
del method, make_simple_escaping_wrapper
raise ValueError(
f"Format specifier {format_spec} given, but {type(value)} does not"
" define __html_format__. A class that defines __html__ must define"
" __html_format__ to work with format specifiers."
)
rv = value.__html__()
else:
# We need to make sure the format spec is str here as
# otherwise the wrong callback methods are invoked.
rv = string.Formatter.format_field(self, value, str(format_spec))
return str(self.escape(rv))
class _MagicFormatMapping(Mapping):
"""This class implements a dummy wrapper to fix a bug in the Python
standard library for string formatting.
See http://bugs.python.org/issue13598 for information about why
this is necessary.
"""
def __init__(self, args, kwargs):
self._args = args
self._kwargs = kwargs
self._last_index = 0
def __getitem__(self, key):
if key == '':
idx = self._last_index
self._last_index += 1
try:
return self._args[idx]
except LookupError:
pass
key = str(idx)
return self._kwargs[key]
def __iter__(self):
return iter(self._kwargs)
def __len__(self):
return len(self._kwargs)
_ListOrDict = t.TypeVar("_ListOrDict", list, dict)
if hasattr(text_type, 'format'):
class EscapeFormatter(string.Formatter):
def __init__(self, escape):
self.escape = escape
def format_field(self, value, format_spec):
if hasattr(value, '__html_format__'):
rv = value.__html_format__(format_spec)
elif hasattr(value, '__html__'):
if format_spec:
raise ValueError('No format specification allowed '
'when formatting an object with '
'its __html__ method.')
rv = value.__html__()
else:
# We need to make sure the format spec is unicode here as
# otherwise the wrong callback methods are invoked. For
# instance a byte string there would invoke __str__ and
# not __unicode__.
rv = string.Formatter.format_field(
self, value, text_type(format_spec))
return text_type(self.escape(rv))
def _escape_argspec(obj, iterable, escape):
def _escape_argspec(
obj: _ListOrDict, iterable: t.Iterable[t.Any], escape: t.Callable[[t.Any], Markup]
) -> _ListOrDict:
"""Helper for various string-wrapped functions."""
for key, value in iterable:
if hasattr(value, '__html__') or isinstance(value, string_types):
if isinstance(value, str) or hasattr(value, "__html__"):
obj[key] = escape(value)
return obj
class _MarkupEscapeHelper(object):
"""Helper for Markup.__mod__"""
class _MarkupEscapeHelper:
"""Helper for :meth:`Markup.__mod__`."""
def __init__(self, obj, escape):
__slots__ = ("obj", "escape")
def __init__(self, obj: t.Any, escape: t.Callable[[t.Any], Markup]) -> None:
self.obj = obj
self.escape = escape
__getitem__ = lambda s, x: _MarkupEscapeHelper(s.obj[x], s.escape)
__unicode__ = __str__ = lambda s: text_type(s.escape(s.obj))
__repr__ = lambda s: str(s.escape(repr(s.obj)))
__int__ = lambda s: int(s.obj)
__float__ = lambda s: float(s.obj)
def __getitem__(self, item: t.Any) -> "_MarkupEscapeHelper":
return _MarkupEscapeHelper(self.obj[item], self.escape)
def __str__(self) -> str:
return str(self.escape(self.obj))
def __repr__(self) -> str:
return str(self.escape(repr(self.obj)))
def __int__(self) -> int:
return int(self.obj)
def __float__(self) -> float:
return float(self.obj)
# we have to import it down here as the speedups and native
# modules imports the markup type which is define above.
# circular import
try:
from markupsafe._speedups import escape, escape_silent, soft_unicode
from ._speedups import escape as escape
from ._speedups import escape_silent as escape_silent
from ._speedups import soft_str as soft_str
except ImportError:
from markupsafe._native import escape, escape_silent, soft_unicode
if not PY2:
soft_str = soft_unicode
__all__.append('soft_str')
from ._native import escape as escape
from ._native import escape_silent as escape_silent # noqa: F401
from ._native import soft_str as soft_str # noqa: F401

View File

@ -1,26 +0,0 @@
# -*- coding: utf-8 -*-
"""
markupsafe._compat
~~~~~~~~~~~~~~~~~~
Compatibility module for different Python versions.
:copyright: (c) 2013 by Armin Ronacher.
:license: BSD, see LICENSE for more details.
"""
import sys
PY2 = sys.version_info[0] == 2
if not PY2:
text_type = str
string_types = (str,)
unichr = chr
int_types = (int,)
iteritems = lambda x: iter(x.items())
else:
text_type = unicode
string_types = (str, unicode)
unichr = unichr
int_types = (int, long)
iteritems = lambda x: x.iteritems()

View File

@ -1,267 +0,0 @@
# -*- coding: utf-8 -*-
"""
markupsafe._constants
~~~~~~~~~~~~~~~~~~~~~
Highlevel implementation of the Markup string.
:copyright: (c) 2010 by Armin Ronacher.
:license: BSD, see LICENSE for more details.
"""
HTML_ENTITIES = {
'AElig': 198,
'Aacute': 193,
'Acirc': 194,
'Agrave': 192,
'Alpha': 913,
'Aring': 197,
'Atilde': 195,
'Auml': 196,
'Beta': 914,
'Ccedil': 199,
'Chi': 935,
'Dagger': 8225,
'Delta': 916,
'ETH': 208,
'Eacute': 201,
'Ecirc': 202,
'Egrave': 200,
'Epsilon': 917,
'Eta': 919,
'Euml': 203,
'Gamma': 915,
'Iacute': 205,
'Icirc': 206,
'Igrave': 204,
'Iota': 921,
'Iuml': 207,
'Kappa': 922,
'Lambda': 923,
'Mu': 924,
'Ntilde': 209,
'Nu': 925,
'OElig': 338,
'Oacute': 211,
'Ocirc': 212,
'Ograve': 210,
'Omega': 937,
'Omicron': 927,
'Oslash': 216,
'Otilde': 213,
'Ouml': 214,
'Phi': 934,
'Pi': 928,
'Prime': 8243,
'Psi': 936,
'Rho': 929,
'Scaron': 352,
'Sigma': 931,
'THORN': 222,
'Tau': 932,
'Theta': 920,
'Uacute': 218,
'Ucirc': 219,
'Ugrave': 217,
'Upsilon': 933,
'Uuml': 220,
'Xi': 926,
'Yacute': 221,
'Yuml': 376,
'Zeta': 918,
'aacute': 225,
'acirc': 226,
'acute': 180,
'aelig': 230,
'agrave': 224,
'alefsym': 8501,
'alpha': 945,
'amp': 38,
'and': 8743,
'ang': 8736,
'apos': 39,
'aring': 229,
'asymp': 8776,
'atilde': 227,
'auml': 228,
'bdquo': 8222,
'beta': 946,
'brvbar': 166,
'bull': 8226,
'cap': 8745,
'ccedil': 231,
'cedil': 184,
'cent': 162,
'chi': 967,
'circ': 710,
'clubs': 9827,
'cong': 8773,
'copy': 169,
'crarr': 8629,
'cup': 8746,
'curren': 164,
'dArr': 8659,
'dagger': 8224,
'darr': 8595,
'deg': 176,
'delta': 948,
'diams': 9830,
'divide': 247,
'eacute': 233,
'ecirc': 234,
'egrave': 232,
'empty': 8709,
'emsp': 8195,
'ensp': 8194,
'epsilon': 949,
'equiv': 8801,
'eta': 951,
'eth': 240,
'euml': 235,
'euro': 8364,
'exist': 8707,
'fnof': 402,
'forall': 8704,
'frac12': 189,
'frac14': 188,
'frac34': 190,
'frasl': 8260,
'gamma': 947,
'ge': 8805,
'gt': 62,
'hArr': 8660,
'harr': 8596,
'hearts': 9829,
'hellip': 8230,
'iacute': 237,
'icirc': 238,
'iexcl': 161,
'igrave': 236,
'image': 8465,
'infin': 8734,
'int': 8747,
'iota': 953,
'iquest': 191,
'isin': 8712,
'iuml': 239,
'kappa': 954,
'lArr': 8656,
'lambda': 955,
'lang': 9001,
'laquo': 171,
'larr': 8592,
'lceil': 8968,
'ldquo': 8220,
'le': 8804,
'lfloor': 8970,
'lowast': 8727,
'loz': 9674,
'lrm': 8206,
'lsaquo': 8249,
'lsquo': 8216,
'lt': 60,
'macr': 175,
'mdash': 8212,
'micro': 181,
'middot': 183,
'minus': 8722,
'mu': 956,
'nabla': 8711,
'nbsp': 160,
'ndash': 8211,
'ne': 8800,
'ni': 8715,
'not': 172,
'notin': 8713,
'nsub': 8836,
'ntilde': 241,
'nu': 957,
'oacute': 243,
'ocirc': 244,
'oelig': 339,
'ograve': 242,
'oline': 8254,
'omega': 969,
'omicron': 959,
'oplus': 8853,
'or': 8744,
'ordf': 170,
'ordm': 186,
'oslash': 248,
'otilde': 245,
'otimes': 8855,
'ouml': 246,
'para': 182,
'part': 8706,
'permil': 8240,
'perp': 8869,
'phi': 966,
'pi': 960,
'piv': 982,
'plusmn': 177,
'pound': 163,
'prime': 8242,
'prod': 8719,
'prop': 8733,
'psi': 968,
'quot': 34,
'rArr': 8658,
'radic': 8730,
'rang': 9002,
'raquo': 187,
'rarr': 8594,
'rceil': 8969,
'rdquo': 8221,
'real': 8476,
'reg': 174,
'rfloor': 8971,
'rho': 961,
'rlm': 8207,
'rsaquo': 8250,
'rsquo': 8217,
'sbquo': 8218,
'scaron': 353,
'sdot': 8901,
'sect': 167,
'shy': 173,
'sigma': 963,
'sigmaf': 962,
'sim': 8764,
'spades': 9824,
'sub': 8834,
'sube': 8838,
'sum': 8721,
'sup': 8835,
'sup1': 185,
'sup2': 178,
'sup3': 179,
'supe': 8839,
'szlig': 223,
'tau': 964,
'there4': 8756,
'theta': 952,
'thetasym': 977,
'thinsp': 8201,
'thorn': 254,
'tilde': 732,
'times': 215,
'trade': 8482,
'uArr': 8657,
'uacute': 250,
'uarr': 8593,
'ucirc': 251,
'ugrave': 249,
'uml': 168,
'upsih': 978,
'upsilon': 965,
'uuml': 252,
'weierp': 8472,
'xi': 958,
'yacute': 253,
'yen': 165,
'yuml': 255,
'zeta': 950,
'zwj': 8205,
'zwnj': 8204
}

View File

@ -1,46 +1,63 @@
# -*- coding: utf-8 -*-
"""
markupsafe._native
~~~~~~~~~~~~~~~~~~
import typing as t
Native Python implementation the C module is not compiled.
:copyright: (c) 2010 by Armin Ronacher.
:license: BSD, see LICENSE for more details.
"""
from markupsafe import Markup
from markupsafe._compat import text_type
from . import Markup
def escape(s):
"""Convert the characters &, <, >, ' and " in string s to HTML-safe
sequences. Use this if you need to display text that might contain
such characters in HTML. Marks return value as markup string.
def escape(s: t.Any) -> Markup:
"""Replace the characters ``&``, ``<``, ``>``, ``'``, and ``"`` in
the string with HTML-safe sequences. Use this if you need to display
text that might contain such characters in HTML.
If the object has an ``__html__`` method, it is called and the
return value is assumed to already be safe for HTML.
:param s: An object to be converted to a string and escaped.
:return: A :class:`Markup` string with the escaped text.
"""
if hasattr(s, '__html__'):
return s.__html__()
return Markup(text_type(s)
.replace('&', '&amp;')
.replace('>', '&gt;')
.replace('<', '&lt;')
.replace("'", '&#39;')
.replace('"', '&#34;')
if hasattr(s, "__html__"):
return Markup(s.__html__())
return Markup(
str(s)
.replace("&", "&amp;")
.replace(">", "&gt;")
.replace("<", "&lt;")
.replace("'", "&#39;")
.replace('"', "&#34;")
)
def escape_silent(s):
"""Like :func:`escape` but converts `None` into an empty
markup string.
def escape_silent(s: t.Optional[t.Any]) -> Markup:
"""Like :func:`escape` but treats ``None`` as the empty string.
Useful with optional values, as otherwise you get the string
``'None'`` when the value is ``None``.
>>> escape(None)
Markup('None')
>>> escape_silent(None)
Markup('')
"""
if s is None:
return Markup()
return escape(s)
def soft_unicode(s):
"""Make a string unicode if it isn't already. That way a markup
string is not converted back to unicode.
def soft_str(s: t.Any) -> str:
"""Convert an object to a string if it isn't already. This preserves
a :class:`Markup` string rather than converting it back to a basic
string, so it will still be marked as safe and won't be escaped
again.
>>> value = escape("<User 1>")
>>> value
Markup('&lt;User 1&gt;')
>>> escape(str(value))
Markup('&amp;lt;User 1&amp;gt;')
>>> escape(soft_str(value))
Markup('&lt;User 1&gt;')
"""
if not isinstance(s, text_type):
s = text_type(s)
if not isinstance(s, str):
return str(s)
return s

View File

@ -1,46 +1,11 @@
/**
* markupsafe._speedups
* ~~~~~~~~~~~~~~~~~~~~
*
* This module implements functions for automatic escaping in C for better
* performance.
*
* :copyright: (c) 2010 by Armin Ronacher.
* :license: BSD.
*/
#include <Python.h>
#define ESCAPED_CHARS_TABLE_SIZE 63
#define UNICHR(x) (PyUnicode_AS_UNICODE((PyUnicodeObject*)PyUnicode_DecodeASCII(x, strlen(x), NULL)));
#if PY_VERSION_HEX < 0x02050000 && !defined(PY_SSIZE_T_MIN)
typedef int Py_ssize_t;
#define PY_SSIZE_T_MAX INT_MAX
#define PY_SSIZE_T_MIN INT_MIN
#endif
static PyObject* markup;
static Py_ssize_t escaped_chars_delta_len[ESCAPED_CHARS_TABLE_SIZE];
static Py_UNICODE *escaped_chars_repl[ESCAPED_CHARS_TABLE_SIZE];
static int
init_constants(void)
{
PyObject *module;
/* mapping of characters to replace */
escaped_chars_repl['"'] = UNICHR("&#34;");
escaped_chars_repl['\''] = UNICHR("&#39;");
escaped_chars_repl['&'] = UNICHR("&amp;");
escaped_chars_repl['<'] = UNICHR("&lt;");
escaped_chars_repl['>'] = UNICHR("&gt;");
/* lengths of those characters when replaced - 1 */
memset(escaped_chars_delta_len, 0, sizeof (escaped_chars_delta_len));
escaped_chars_delta_len['"'] = escaped_chars_delta_len['\''] = \
escaped_chars_delta_len['&'] = 4;
escaped_chars_delta_len['<'] = escaped_chars_delta_len['>'] = 3;
/* import markup type so that we can mark the return value */
module = PyImport_ImportModule("markupsafe");
@ -52,98 +17,209 @@ init_constants(void)
return 1;
}
static PyObject*
escape_unicode(PyUnicodeObject *in)
{
PyUnicodeObject *out;
Py_UNICODE *inp = PyUnicode_AS_UNICODE(in);
const Py_UNICODE *inp_end = PyUnicode_AS_UNICODE(in) + PyUnicode_GET_SIZE(in);
Py_UNICODE *next_escp;
Py_UNICODE *outp;
Py_ssize_t delta=0, erepl=0, delta_len=0;
/* First we need to figure out how long the escaped string will be */
while (*(inp) || inp < inp_end) {
if (*inp < ESCAPED_CHARS_TABLE_SIZE) {
delta += escaped_chars_delta_len[*inp];
erepl += !!escaped_chars_delta_len[*inp];
}
++inp;
#define GET_DELTA(inp, inp_end, delta) \
while (inp < inp_end) { \
switch (*inp++) { \
case '"': \
case '\'': \
case '&': \
delta += 4; \
break; \
case '<': \
case '>': \
delta += 3; \
break; \
} \
}
/* Do we need to escape anything at all? */
if (!erepl) {
#define DO_ESCAPE(inp, inp_end, outp) \
{ \
Py_ssize_t ncopy = 0; \
while (inp < inp_end) { \
switch (*inp) { \
case '"': \
memcpy(outp, inp-ncopy, sizeof(*outp)*ncopy); \
outp += ncopy; ncopy = 0; \
*outp++ = '&'; \
*outp++ = '#'; \
*outp++ = '3'; \
*outp++ = '4'; \
*outp++ = ';'; \
break; \
case '\'': \
memcpy(outp, inp-ncopy, sizeof(*outp)*ncopy); \
outp += ncopy; ncopy = 0; \
*outp++ = '&'; \
*outp++ = '#'; \
*outp++ = '3'; \
*outp++ = '9'; \
*outp++ = ';'; \
break; \
case '&': \
memcpy(outp, inp-ncopy, sizeof(*outp)*ncopy); \
outp += ncopy; ncopy = 0; \
*outp++ = '&'; \
*outp++ = 'a'; \
*outp++ = 'm'; \
*outp++ = 'p'; \
*outp++ = ';'; \
break; \
case '<': \
memcpy(outp, inp-ncopy, sizeof(*outp)*ncopy); \
outp += ncopy; ncopy = 0; \
*outp++ = '&'; \
*outp++ = 'l'; \
*outp++ = 't'; \
*outp++ = ';'; \
break; \
case '>': \
memcpy(outp, inp-ncopy, sizeof(*outp)*ncopy); \
outp += ncopy; ncopy = 0; \
*outp++ = '&'; \
*outp++ = 'g'; \
*outp++ = 't'; \
*outp++ = ';'; \
break; \
default: \
ncopy++; \
} \
inp++; \
} \
memcpy(outp, inp-ncopy, sizeof(*outp)*ncopy); \
}
static PyObject*
escape_unicode_kind1(PyUnicodeObject *in)
{
Py_UCS1 *inp = PyUnicode_1BYTE_DATA(in);
Py_UCS1 *inp_end = inp + PyUnicode_GET_LENGTH(in);
Py_UCS1 *outp;
PyObject *out;
Py_ssize_t delta = 0;
GET_DELTA(inp, inp_end, delta);
if (!delta) {
Py_INCREF(in);
return (PyObject*)in;
}
out = (PyUnicodeObject*)PyUnicode_FromUnicode(NULL, PyUnicode_GET_SIZE(in) + delta);
out = PyUnicode_New(PyUnicode_GET_LENGTH(in) + delta,
PyUnicode_IS_ASCII(in) ? 127 : 255);
if (!out)
return NULL;
outp = PyUnicode_AS_UNICODE(out);
inp = PyUnicode_AS_UNICODE(in);
while (erepl-- > 0) {
/* look for the next substitution */
next_escp = inp;
while (next_escp < inp_end) {
if (*next_escp < ESCAPED_CHARS_TABLE_SIZE &&
(delta_len = escaped_chars_delta_len[*next_escp])) {
++delta_len;
break;
}
++next_escp;
}
inp = PyUnicode_1BYTE_DATA(in);
outp = PyUnicode_1BYTE_DATA(out);
DO_ESCAPE(inp, inp_end, outp);
return out;
}
if (next_escp > inp) {
/* copy unescaped chars between inp and next_escp */
Py_UNICODE_COPY(outp, inp, next_escp-inp);
outp += next_escp - inp;
}
static PyObject*
escape_unicode_kind2(PyUnicodeObject *in)
{
Py_UCS2 *inp = PyUnicode_2BYTE_DATA(in);
Py_UCS2 *inp_end = inp + PyUnicode_GET_LENGTH(in);
Py_UCS2 *outp;
PyObject *out;
Py_ssize_t delta = 0;
/* escape 'next_escp' */
Py_UNICODE_COPY(outp, escaped_chars_repl[*next_escp], delta_len);
outp += delta_len;
inp = next_escp + 1;
GET_DELTA(inp, inp_end, delta);
if (!delta) {
Py_INCREF(in);
return (PyObject*)in;
}
if (inp < inp_end)
Py_UNICODE_COPY(outp, inp, PyUnicode_GET_SIZE(in) - (inp - PyUnicode_AS_UNICODE(in)));
return (PyObject*)out;
out = PyUnicode_New(PyUnicode_GET_LENGTH(in) + delta, 65535);
if (!out)
return NULL;
inp = PyUnicode_2BYTE_DATA(in);
outp = PyUnicode_2BYTE_DATA(out);
DO_ESCAPE(inp, inp_end, outp);
return out;
}
static PyObject*
escape_unicode_kind4(PyUnicodeObject *in)
{
Py_UCS4 *inp = PyUnicode_4BYTE_DATA(in);
Py_UCS4 *inp_end = inp + PyUnicode_GET_LENGTH(in);
Py_UCS4 *outp;
PyObject *out;
Py_ssize_t delta = 0;
GET_DELTA(inp, inp_end, delta);
if (!delta) {
Py_INCREF(in);
return (PyObject*)in;
}
out = PyUnicode_New(PyUnicode_GET_LENGTH(in) + delta, 1114111);
if (!out)
return NULL;
inp = PyUnicode_4BYTE_DATA(in);
outp = PyUnicode_4BYTE_DATA(out);
DO_ESCAPE(inp, inp_end, outp);
return out;
}
static PyObject*
escape_unicode(PyUnicodeObject *in)
{
if (PyUnicode_READY(in))
return NULL;
switch (PyUnicode_KIND(in)) {
case PyUnicode_1BYTE_KIND:
return escape_unicode_kind1(in);
case PyUnicode_2BYTE_KIND:
return escape_unicode_kind2(in);
case PyUnicode_4BYTE_KIND:
return escape_unicode_kind4(in);
}
assert(0); /* shouldn't happen */
return NULL;
}
static PyObject*
escape(PyObject *self, PyObject *text)
{
static PyObject *id_html;
PyObject *s = NULL, *rv = NULL, *html;
if (id_html == NULL) {
id_html = PyUnicode_InternFromString("__html__");
if (id_html == NULL) {
return NULL;
}
}
/* we don't have to escape integers, bools or floats */
if (PyLong_CheckExact(text) ||
#if PY_MAJOR_VERSION < 3
PyInt_CheckExact(text) ||
#endif
PyFloat_CheckExact(text) || PyBool_Check(text) ||
text == Py_None)
PyFloat_CheckExact(text) || PyBool_Check(text) ||
text == Py_None)
return PyObject_CallFunctionObjArgs(markup, text, NULL);
/* if the object has an __html__ method that performs the escaping */
html = PyObject_GetAttrString(text, "__html__");
html = PyObject_GetAttr(text ,id_html);
if (html) {
rv = PyObject_CallObject(html, NULL);
s = PyObject_CallObject(html, NULL);
Py_DECREF(html);
if (s == NULL) {
return NULL;
}
/* Convert to Markup object */
rv = PyObject_CallFunctionObjArgs(markup, (PyObject*)s, NULL);
Py_DECREF(s);
return rv;
}
/* otherwise make the object unicode if it isn't, then escape */
PyErr_Clear();
if (!PyUnicode_Check(text)) {
#if PY_MAJOR_VERSION < 3
PyObject *unicode = PyObject_Unicode(text);
#else
PyObject *unicode = PyObject_Str(text);
#endif
if (!unicode)
return NULL;
s = escape_unicode((PyUnicodeObject*)unicode);
@ -169,54 +245,61 @@ escape_silent(PyObject *self, PyObject *text)
static PyObject*
soft_unicode(PyObject *self, PyObject *s)
soft_str(PyObject *self, PyObject *s)
{
if (!PyUnicode_Check(s))
#if PY_MAJOR_VERSION < 3
return PyObject_Unicode(s);
#else
return PyObject_Str(s);
#endif
Py_INCREF(s);
return s;
}
static PyMethodDef module_methods[] = {
{"escape", (PyCFunction)escape, METH_O,
"escape(s) -> markup\n\n"
"Convert the characters &, <, >, ', and \" in string s to HTML-safe\n"
"sequences. Use this if you need to display text that might contain\n"
"such characters in HTML. Marks return value as markup string."},
{"escape_silent", (PyCFunction)escape_silent, METH_O,
"escape_silent(s) -> markup\n\n"
"Like escape but converts None to an empty string."},
{"soft_unicode", (PyCFunction)soft_unicode, METH_O,
"soft_unicode(object) -> string\n\n"
"Make a string unicode if it isn't already. That way a markup\n"
"string is not converted back to unicode."},
{NULL, NULL, 0, NULL} /* Sentinel */
{
"escape",
(PyCFunction)escape,
METH_O,
"Replace the characters ``&``, ``<``, ``>``, ``'``, and ``\"`` in"
" the string with HTML-safe sequences. Use this if you need to display"
" text that might contain such characters in HTML.\n\n"
"If the object has an ``__html__`` method, it is called and the"
" return value is assumed to already be safe for HTML.\n\n"
":param s: An object to be converted to a string and escaped.\n"
":return: A :class:`Markup` string with the escaped text.\n"
},
{
"escape_silent",
(PyCFunction)escape_silent,
METH_O,
"Like :func:`escape` but treats ``None`` as the empty string."
" Useful with optional values, as otherwise you get the string"
" ``'None'`` when the value is ``None``.\n\n"
">>> escape(None)\n"
"Markup('None')\n"
">>> escape_silent(None)\n"
"Markup('')\n"
},
{
"soft_str",
(PyCFunction)soft_str,
METH_O,
"Convert an object to a string if it isn't already. This preserves"
" a :class:`Markup` string rather than converting it back to a basic"
" string, so it will still be marked as safe and won't be escaped"
" again.\n\n"
">>> value = escape(\"<User 1>\")\n"
">>> value\n"
"Markup('&lt;User 1&gt;')\n"
">>> escape(str(value))\n"
"Markup('&amp;lt;User 1&amp;gt;')\n"
">>> escape(soft_str(value))\n"
"Markup('&lt;User 1&gt;')\n"
},
{NULL, NULL, 0, NULL} /* Sentinel */
};
#if PY_MAJOR_VERSION < 3
#ifndef PyMODINIT_FUNC /* declarations for DLL import/export */
#define PyMODINIT_FUNC void
#endif
PyMODINIT_FUNC
init_speedups(void)
{
if (!init_constants())
return;
Py_InitModule3("markupsafe._speedups", module_methods, "");
}
#else /* Python 3.x module initialization */
static struct PyModuleDef module_definition = {
PyModuleDef_HEAD_INIT,
PyModuleDef_HEAD_INIT,
"markupsafe._speedups",
NULL,
-1,
@ -235,5 +318,3 @@ PyInit__speedups(void)
return PyModule_Create(&module_definition);
}
#endif

View File

@ -0,0 +1,9 @@
from typing import Any
from typing import Optional
from . import Markup
def escape(s: Any) -> Markup: ...
def escape_silent(s: Optional[Any]) -> Markup: ...
def soft_str(s: Any) -> str: ...
def soft_unicode(s: Any) -> str: ...

View File

View File

@ -0,0 +1,20 @@
Copyright (c) 2017-2021 Ingy döt Net
Copyright (c) 2006-2016 Kirill Simonov
Permission is hereby granted, free of charge, to any person obtaining a copy of
this software and associated documentation files (the "Software"), to deal in
the Software without restriction, including without limitation the rights to
use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies
of the Software, and to permit persons to whom the Software is furnished to do
so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.

View File

@ -8,7 +8,7 @@ from .nodes import *
from .loader import *
from .dumper import *
__version__ = '3.12'
__version__ = '6.0'
try:
from .cyaml import *
__with_libyaml__ = True
@ -17,6 +17,15 @@ except ImportError:
import io
#------------------------------------------------------------------------------
# XXX "Warnings control" is now deprecated. Leaving in the API function to not
# break code that uses it.
#------------------------------------------------------------------------------
def warnings(settings=None):
if settings is None:
return {}
#------------------------------------------------------------------------------
def scan(stream, Loader=Loader):
"""
Scan a YAML stream and produce scanning tokens.
@ -62,7 +71,7 @@ def compose_all(stream, Loader=Loader):
finally:
loader.dispose()
def load(stream, Loader=Loader):
def load(stream, Loader):
"""
Parse the first YAML document in a stream
and produce the corresponding Python object.
@ -73,7 +82,7 @@ def load(stream, Loader=Loader):
finally:
loader.dispose()
def load_all(stream, Loader=Loader):
def load_all(stream, Loader):
"""
Parse all YAML documents in a stream
and produce corresponding Python objects.
@ -85,11 +94,33 @@ def load_all(stream, Loader=Loader):
finally:
loader.dispose()
def full_load(stream):
"""
Parse the first YAML document in a stream
and produce the corresponding Python object.
Resolve all tags except those known to be
unsafe on untrusted input.
"""
return load(stream, FullLoader)
def full_load_all(stream):
"""
Parse all YAML documents in a stream
and produce corresponding Python objects.
Resolve all tags except those known to be
unsafe on untrusted input.
"""
return load_all(stream, FullLoader)
def safe_load(stream):
"""
Parse the first YAML document in a stream
and produce the corresponding Python object.
Resolve only basic YAML tags.
Resolve only basic YAML tags. This is known
to be safe for untrusted input.
"""
return load(stream, SafeLoader)
@ -97,10 +128,32 @@ def safe_load_all(stream):
"""
Parse all YAML documents in a stream
and produce corresponding Python objects.
Resolve only basic YAML tags.
Resolve only basic YAML tags. This is known
to be safe for untrusted input.
"""
return load_all(stream, SafeLoader)
def unsafe_load(stream):
"""
Parse the first YAML document in a stream
and produce the corresponding Python object.
Resolve all tags, even those known to be
unsafe on untrusted input.
"""
return load(stream, UnsafeLoader)
def unsafe_load_all(stream):
"""
Parse all YAML documents in a stream
and produce corresponding Python objects.
Resolve all tags, even those known to be
unsafe on untrusted input.
"""
return load_all(stream, UnsafeLoader)
def emit(events, stream=None, Dumper=Dumper,
canonical=None, indent=None, width=None,
allow_unicode=None, line_break=None):
@ -160,11 +213,11 @@ def serialize(node, stream=None, Dumper=Dumper, **kwds):
return serialize_all([node], stream, Dumper=Dumper, **kwds)
def dump_all(documents, stream=None, Dumper=Dumper,
default_style=None, default_flow_style=None,
default_style=None, default_flow_style=False,
canonical=None, indent=None, width=None,
allow_unicode=None, line_break=None,
encoding=None, explicit_start=None, explicit_end=None,
version=None, tags=None):
version=None, tags=None, sort_keys=True):
"""
Serialize a sequence of Python objects into a YAML stream.
If stream is None, return the produced string instead.
@ -181,7 +234,7 @@ def dump_all(documents, stream=None, Dumper=Dumper,
canonical=canonical, indent=indent, width=width,
allow_unicode=allow_unicode, line_break=line_break,
encoding=encoding, version=version, tags=tags,
explicit_start=explicit_start, explicit_end=explicit_end)
explicit_start=explicit_start, explicit_end=explicit_end, sort_keys=sort_keys)
try:
dumper.open()
for data in documents:
@ -216,42 +269,62 @@ def safe_dump(data, stream=None, **kwds):
return dump_all([data], stream, Dumper=SafeDumper, **kwds)
def add_implicit_resolver(tag, regexp, first=None,
Loader=Loader, Dumper=Dumper):
Loader=None, Dumper=Dumper):
"""
Add an implicit scalar detector.
If an implicit scalar value matches the given regexp,
the corresponding tag is assigned to the scalar.
first is a sequence of possible initial characters or None.
"""
Loader.add_implicit_resolver(tag, regexp, first)
if Loader is None:
loader.Loader.add_implicit_resolver(tag, regexp, first)
loader.FullLoader.add_implicit_resolver(tag, regexp, first)
loader.UnsafeLoader.add_implicit_resolver(tag, regexp, first)
else:
Loader.add_implicit_resolver(tag, regexp, first)
Dumper.add_implicit_resolver(tag, regexp, first)
def add_path_resolver(tag, path, kind=None, Loader=Loader, Dumper=Dumper):
def add_path_resolver(tag, path, kind=None, Loader=None, Dumper=Dumper):
"""
Add a path based resolver for the given tag.
A path is a list of keys that forms a path
to a node in the representation tree.
Keys can be string values, integers, or None.
"""
Loader.add_path_resolver(tag, path, kind)
if Loader is None:
loader.Loader.add_path_resolver(tag, path, kind)
loader.FullLoader.add_path_resolver(tag, path, kind)
loader.UnsafeLoader.add_path_resolver(tag, path, kind)
else:
Loader.add_path_resolver(tag, path, kind)
Dumper.add_path_resolver(tag, path, kind)
def add_constructor(tag, constructor, Loader=Loader):
def add_constructor(tag, constructor, Loader=None):
"""
Add a constructor for the given tag.
Constructor is a function that accepts a Loader instance
and a node object and produces the corresponding Python object.
"""
Loader.add_constructor(tag, constructor)
if Loader is None:
loader.Loader.add_constructor(tag, constructor)
loader.FullLoader.add_constructor(tag, constructor)
loader.UnsafeLoader.add_constructor(tag, constructor)
else:
Loader.add_constructor(tag, constructor)
def add_multi_constructor(tag_prefix, multi_constructor, Loader=Loader):
def add_multi_constructor(tag_prefix, multi_constructor, Loader=None):
"""
Add a multi-constructor for the given tag prefix.
Multi-constructor is called for a node if its tag starts with tag_prefix.
Multi-constructor accepts a Loader instance, a tag suffix,
and a node object and produces the corresponding Python object.
"""
Loader.add_multi_constructor(tag_prefix, multi_constructor)
if Loader is None:
loader.Loader.add_multi_constructor(tag_prefix, multi_constructor)
loader.FullLoader.add_multi_constructor(tag_prefix, multi_constructor)
loader.UnsafeLoader.add_multi_constructor(tag_prefix, multi_constructor)
else:
Loader.add_multi_constructor(tag_prefix, multi_constructor)
def add_representer(data_type, representer, Dumper=Dumper):
"""
@ -278,7 +351,12 @@ class YAMLObjectMetaclass(type):
def __init__(cls, name, bases, kwds):
super(YAMLObjectMetaclass, cls).__init__(name, bases, kwds)
if 'yaml_tag' in kwds and kwds['yaml_tag'] is not None:
cls.yaml_loader.add_constructor(cls.yaml_tag, cls.from_yaml)
if isinstance(cls.yaml_loader, list):
for loader in cls.yaml_loader:
loader.add_constructor(cls.yaml_tag, cls.from_yaml)
else:
cls.yaml_loader.add_constructor(cls.yaml_tag, cls.from_yaml)
cls.yaml_dumper.add_representer(cls, cls.to_yaml)
class YAMLObject(metaclass=YAMLObjectMetaclass):
@ -289,7 +367,7 @@ class YAMLObject(metaclass=YAMLObjectMetaclass):
__slots__ = () # no direct instantiation, so allow immutable subclasses
yaml_loader = Loader
yaml_loader = [Loader, FullLoader, UnsafeLoader]
yaml_dumper = Dumper
yaml_tag = None

View File

@ -72,9 +72,9 @@ class Composer:
anchor = event.anchor
if anchor is not None:
if anchor in self.anchors:
raise ComposerError("found duplicate anchor %r; first occurence"
raise ComposerError("found duplicate anchor %r; first occurrence"
% anchor, self.anchors[anchor].start_mark,
"second occurence", event.start_mark)
"second occurrence", event.start_mark)
self.descend_resolver(parent, index)
if self.check_event(ScalarEvent):
node = self.compose_scalar_node(anchor)

View File

@ -1,11 +1,17 @@
__all__ = ['BaseConstructor', 'SafeConstructor', 'Constructor',
'ConstructorError']
__all__ = [
'BaseConstructor',
'SafeConstructor',
'FullConstructor',
'UnsafeConstructor',
'Constructor',
'ConstructorError'
]
from .error import *
from .nodes import *
import collections, datetime, base64, binascii, re, sys, types
import collections.abc, datetime, base64, binascii, re, sys, types
class ConstructorError(MarkedYAMLError):
pass
@ -25,6 +31,14 @@ class BaseConstructor:
# If there are more documents available?
return self.check_node()
def check_state_key(self, key):
"""Block special attributes/methods from being set in a newly created
object, to prevent user-controlled methods from being called during
deserialization"""
if self.get_state_keys_blacklist_regexp().match(key):
raise ConstructorError(None, None,
"blacklisted key '%s' in instance state found" % (key,), None)
def get_data(self):
# Construct and return the next document.
if self.check_node():
@ -66,7 +80,7 @@ class BaseConstructor:
constructor = self.yaml_constructors[node.tag]
else:
for tag_prefix in self.yaml_multi_constructors:
if node.tag.startswith(tag_prefix):
if tag_prefix is not None and node.tag.startswith(tag_prefix):
tag_suffix = node.tag[len(tag_prefix):]
constructor = self.yaml_multi_constructors[tag_prefix]
break
@ -123,7 +137,7 @@ class BaseConstructor:
mapping = {}
for key_node, value_node in node.value:
key = self.construct_object(key_node, deep=deep)
if not isinstance(key, collections.Hashable):
if not isinstance(key, collections.abc.Hashable):
raise ConstructorError("while constructing a mapping", node.start_mark,
"found unhashable key", key_node.start_mark)
value = self.construct_object(value_node, deep=deep)
@ -318,22 +332,23 @@ class SafeConstructor(BaseConstructor):
minute = int(values['minute'])
second = int(values['second'])
fraction = 0
tzinfo = None
if values['fraction']:
fraction = values['fraction'][:6]
while len(fraction) < 6:
fraction += '0'
fraction = int(fraction)
delta = None
if values['tz_sign']:
tz_hour = int(values['tz_hour'])
tz_minute = int(values['tz_minute'] or 0)
delta = datetime.timedelta(hours=tz_hour, minutes=tz_minute)
if values['tz_sign'] == '-':
delta = -delta
data = datetime.datetime(year, month, day, hour, minute, second, fraction)
if delta:
data -= delta
return data
tzinfo = datetime.timezone(delta)
elif values['tz']:
tzinfo = datetime.timezone.utc
return datetime.datetime(year, month, day, hour, minute, second, fraction,
tzinfo=tzinfo)
def construct_yaml_omap(self, node):
# Note: we do not check for duplicate keys, because it's too
@ -464,7 +479,17 @@ SafeConstructor.add_constructor(
SafeConstructor.add_constructor(None,
SafeConstructor.construct_undefined)
class Constructor(SafeConstructor):
class FullConstructor(SafeConstructor):
# 'extend' is blacklisted because it is used by
# construct_python_object_apply to add `listitems` to a newly generate
# python instance
def get_state_keys_blacklist(self):
return ['^extend$', '^__.*__$']
def get_state_keys_blacklist_regexp(self):
if not hasattr(self, 'state_keys_blacklist_regexp'):
self.state_keys_blacklist_regexp = re.compile('(' + '|'.join(self.get_state_keys_blacklist()) + ')')
return self.state_keys_blacklist_regexp
def construct_python_str(self, node):
return self.construct_scalar(node)
@ -497,18 +522,22 @@ class Constructor(SafeConstructor):
def construct_python_tuple(self, node):
return tuple(self.construct_sequence(node))
def find_python_module(self, name, mark):
def find_python_module(self, name, mark, unsafe=False):
if not name:
raise ConstructorError("while constructing a Python module", mark,
"expected non-empty name appended to the tag", mark)
try:
__import__(name)
except ImportError as exc:
if unsafe:
try:
__import__(name)
except ImportError as exc:
raise ConstructorError("while constructing a Python module", mark,
"cannot find module %r (%s)" % (name, exc), mark)
if name not in sys.modules:
raise ConstructorError("while constructing a Python module", mark,
"cannot find module %r (%s)" % (name, exc), mark)
"module %r is not imported" % name, mark)
return sys.modules[name]
def find_python_name(self, name, mark):
def find_python_name(self, name, mark, unsafe=False):
if not name:
raise ConstructorError("while constructing a Python object", mark,
"expected non-empty name appended to the tag", mark)
@ -517,11 +546,15 @@ class Constructor(SafeConstructor):
else:
module_name = 'builtins'
object_name = name
try:
__import__(module_name)
except ImportError as exc:
if unsafe:
try:
__import__(module_name)
except ImportError as exc:
raise ConstructorError("while constructing a Python object", mark,
"cannot find module %r (%s)" % (module_name, exc), mark)
if module_name not in sys.modules:
raise ConstructorError("while constructing a Python object", mark,
"cannot find module %r (%s)" % (module_name, exc), mark)
"module %r is not imported" % module_name, mark)
module = sys.modules[module_name]
if not hasattr(module, object_name):
raise ConstructorError("while constructing a Python object", mark,
@ -544,18 +577,22 @@ class Constructor(SafeConstructor):
return self.find_python_module(suffix, node.start_mark)
def make_python_instance(self, suffix, node,
args=None, kwds=None, newobj=False):
args=None, kwds=None, newobj=False, unsafe=False):
if not args:
args = []
if not kwds:
kwds = {}
cls = self.find_python_name(suffix, node.start_mark)
if not (unsafe or isinstance(cls, type)):
raise ConstructorError("while constructing a Python instance", node.start_mark,
"expected a class, but found %r" % type(cls),
node.start_mark)
if newobj and isinstance(cls, type):
return cls.__new__(cls, *args, **kwds)
else:
return cls(*args, **kwds)
def set_python_instance_state(self, instance, state):
def set_python_instance_state(self, instance, state, unsafe=False):
if hasattr(instance, '__setstate__'):
instance.__setstate__(state)
else:
@ -563,11 +600,16 @@ class Constructor(SafeConstructor):
if isinstance(state, tuple) and len(state) == 2:
state, slotstate = state
if hasattr(instance, '__dict__'):
if not unsafe and state:
for key in state.keys():
self.check_state_key(key)
instance.__dict__.update(state)
elif state:
slotstate.update(state)
for key, value in slotstate.items():
setattr(object, key, value)
if not unsafe:
self.check_state_key(key)
setattr(instance, key, value)
def construct_python_object(self, suffix, node):
# Format:
@ -616,71 +658,91 @@ class Constructor(SafeConstructor):
def construct_python_object_new(self, suffix, node):
return self.construct_python_object_apply(suffix, node, newobj=True)
Constructor.add_constructor(
FullConstructor.add_constructor(
'tag:yaml.org,2002:python/none',
Constructor.construct_yaml_null)
FullConstructor.construct_yaml_null)
Constructor.add_constructor(
FullConstructor.add_constructor(
'tag:yaml.org,2002:python/bool',
Constructor.construct_yaml_bool)
FullConstructor.construct_yaml_bool)
Constructor.add_constructor(
FullConstructor.add_constructor(
'tag:yaml.org,2002:python/str',
Constructor.construct_python_str)
FullConstructor.construct_python_str)
Constructor.add_constructor(
FullConstructor.add_constructor(
'tag:yaml.org,2002:python/unicode',
Constructor.construct_python_unicode)
FullConstructor.construct_python_unicode)
Constructor.add_constructor(
FullConstructor.add_constructor(
'tag:yaml.org,2002:python/bytes',
Constructor.construct_python_bytes)
FullConstructor.construct_python_bytes)
Constructor.add_constructor(
FullConstructor.add_constructor(
'tag:yaml.org,2002:python/int',
Constructor.construct_yaml_int)
FullConstructor.construct_yaml_int)
Constructor.add_constructor(
FullConstructor.add_constructor(
'tag:yaml.org,2002:python/long',
Constructor.construct_python_long)
FullConstructor.construct_python_long)
Constructor.add_constructor(
FullConstructor.add_constructor(
'tag:yaml.org,2002:python/float',
Constructor.construct_yaml_float)
FullConstructor.construct_yaml_float)
Constructor.add_constructor(
FullConstructor.add_constructor(
'tag:yaml.org,2002:python/complex',
Constructor.construct_python_complex)
FullConstructor.construct_python_complex)
Constructor.add_constructor(
FullConstructor.add_constructor(
'tag:yaml.org,2002:python/list',
Constructor.construct_yaml_seq)
FullConstructor.construct_yaml_seq)
Constructor.add_constructor(
FullConstructor.add_constructor(
'tag:yaml.org,2002:python/tuple',
Constructor.construct_python_tuple)
FullConstructor.construct_python_tuple)
Constructor.add_constructor(
FullConstructor.add_constructor(
'tag:yaml.org,2002:python/dict',
Constructor.construct_yaml_map)
FullConstructor.construct_yaml_map)
Constructor.add_multi_constructor(
FullConstructor.add_multi_constructor(
'tag:yaml.org,2002:python/name:',
Constructor.construct_python_name)
FullConstructor.construct_python_name)
Constructor.add_multi_constructor(
class UnsafeConstructor(FullConstructor):
def find_python_module(self, name, mark):
return super(UnsafeConstructor, self).find_python_module(name, mark, unsafe=True)
def find_python_name(self, name, mark):
return super(UnsafeConstructor, self).find_python_name(name, mark, unsafe=True)
def make_python_instance(self, suffix, node, args=None, kwds=None, newobj=False):
return super(UnsafeConstructor, self).make_python_instance(
suffix, node, args, kwds, newobj, unsafe=True)
def set_python_instance_state(self, instance, state):
return super(UnsafeConstructor, self).set_python_instance_state(
instance, state, unsafe=True)
UnsafeConstructor.add_multi_constructor(
'tag:yaml.org,2002:python/module:',
Constructor.construct_python_module)
UnsafeConstructor.construct_python_module)
Constructor.add_multi_constructor(
UnsafeConstructor.add_multi_constructor(
'tag:yaml.org,2002:python/object:',
Constructor.construct_python_object)
UnsafeConstructor.construct_python_object)
Constructor.add_multi_constructor(
'tag:yaml.org,2002:python/object/apply:',
Constructor.construct_python_object_apply)
Constructor.add_multi_constructor(
UnsafeConstructor.add_multi_constructor(
'tag:yaml.org,2002:python/object/new:',
Constructor.construct_python_object_new)
UnsafeConstructor.construct_python_object_new)
UnsafeConstructor.add_multi_constructor(
'tag:yaml.org,2002:python/object/apply:',
UnsafeConstructor.construct_python_object_apply)
# Constructor is same as UnsafeConstructor. Need to leave this in place in case
# people have extended it directly.
class Constructor(UnsafeConstructor):
pass

View File

@ -1,8 +1,10 @@
__all__ = ['CBaseLoader', 'CSafeLoader', 'CLoader',
'CBaseDumper', 'CSafeDumper', 'CDumper']
__all__ = [
'CBaseLoader', 'CSafeLoader', 'CFullLoader', 'CUnsafeLoader', 'CLoader',
'CBaseDumper', 'CSafeDumper', 'CDumper'
]
from _yaml import CParser, CEmitter
from yaml._yaml import CParser, CEmitter
from .constructor import *
@ -25,6 +27,20 @@ class CSafeLoader(CParser, SafeConstructor, Resolver):
SafeConstructor.__init__(self)
Resolver.__init__(self)
class CFullLoader(CParser, FullConstructor, Resolver):
def __init__(self, stream):
CParser.__init__(self, stream)
FullConstructor.__init__(self)
Resolver.__init__(self)
class CUnsafeLoader(CParser, UnsafeConstructor, Resolver):
def __init__(self, stream):
CParser.__init__(self, stream)
UnsafeConstructor.__init__(self)
Resolver.__init__(self)
class CLoader(CParser, Constructor, Resolver):
def __init__(self, stream):
@ -35,51 +51,51 @@ class CLoader(CParser, Constructor, Resolver):
class CBaseDumper(CEmitter, BaseRepresenter, BaseResolver):
def __init__(self, stream,
default_style=None, default_flow_style=None,
default_style=None, default_flow_style=False,
canonical=None, indent=None, width=None,
allow_unicode=None, line_break=None,
encoding=None, explicit_start=None, explicit_end=None,
version=None, tags=None):
version=None, tags=None, sort_keys=True):
CEmitter.__init__(self, stream, canonical=canonical,
indent=indent, width=width, encoding=encoding,
allow_unicode=allow_unicode, line_break=line_break,
explicit_start=explicit_start, explicit_end=explicit_end,
version=version, tags=tags)
Representer.__init__(self, default_style=default_style,
default_flow_style=default_flow_style)
default_flow_style=default_flow_style, sort_keys=sort_keys)
Resolver.__init__(self)
class CSafeDumper(CEmitter, SafeRepresenter, Resolver):
def __init__(self, stream,
default_style=None, default_flow_style=None,
default_style=None, default_flow_style=False,
canonical=None, indent=None, width=None,
allow_unicode=None, line_break=None,
encoding=None, explicit_start=None, explicit_end=None,
version=None, tags=None):
version=None, tags=None, sort_keys=True):
CEmitter.__init__(self, stream, canonical=canonical,
indent=indent, width=width, encoding=encoding,
allow_unicode=allow_unicode, line_break=line_break,
explicit_start=explicit_start, explicit_end=explicit_end,
version=version, tags=tags)
SafeRepresenter.__init__(self, default_style=default_style,
default_flow_style=default_flow_style)
default_flow_style=default_flow_style, sort_keys=sort_keys)
Resolver.__init__(self)
class CDumper(CEmitter, Serializer, Representer, Resolver):
def __init__(self, stream,
default_style=None, default_flow_style=None,
default_style=None, default_flow_style=False,
canonical=None, indent=None, width=None,
allow_unicode=None, line_break=None,
encoding=None, explicit_start=None, explicit_end=None,
version=None, tags=None):
version=None, tags=None, sort_keys=True):
CEmitter.__init__(self, stream, canonical=canonical,
indent=indent, width=width, encoding=encoding,
allow_unicode=allow_unicode, line_break=line_break,
explicit_start=explicit_start, explicit_end=explicit_end,
version=version, tags=tags)
Representer.__init__(self, default_style=default_style,
default_flow_style=default_flow_style)
default_flow_style=default_flow_style, sort_keys=sort_keys)
Resolver.__init__(self)

View File

@ -9,11 +9,11 @@ from .resolver import *
class BaseDumper(Emitter, Serializer, BaseRepresenter, BaseResolver):
def __init__(self, stream,
default_style=None, default_flow_style=None,
default_style=None, default_flow_style=False,
canonical=None, indent=None, width=None,
allow_unicode=None, line_break=None,
encoding=None, explicit_start=None, explicit_end=None,
version=None, tags=None):
version=None, tags=None, sort_keys=True):
Emitter.__init__(self, stream, canonical=canonical,
indent=indent, width=width,
allow_unicode=allow_unicode, line_break=line_break)
@ -21,17 +21,17 @@ class BaseDumper(Emitter, Serializer, BaseRepresenter, BaseResolver):
explicit_start=explicit_start, explicit_end=explicit_end,
version=version, tags=tags)
Representer.__init__(self, default_style=default_style,
default_flow_style=default_flow_style)
default_flow_style=default_flow_style, sort_keys=sort_keys)
Resolver.__init__(self)
class SafeDumper(Emitter, Serializer, SafeRepresenter, Resolver):
def __init__(self, stream,
default_style=None, default_flow_style=None,
default_style=None, default_flow_style=False,
canonical=None, indent=None, width=None,
allow_unicode=None, line_break=None,
encoding=None, explicit_start=None, explicit_end=None,
version=None, tags=None):
version=None, tags=None, sort_keys=True):
Emitter.__init__(self, stream, canonical=canonical,
indent=indent, width=width,
allow_unicode=allow_unicode, line_break=line_break)
@ -39,17 +39,17 @@ class SafeDumper(Emitter, Serializer, SafeRepresenter, Resolver):
explicit_start=explicit_start, explicit_end=explicit_end,
version=version, tags=tags)
SafeRepresenter.__init__(self, default_style=default_style,
default_flow_style=default_flow_style)
default_flow_style=default_flow_style, sort_keys=sort_keys)
Resolver.__init__(self)
class Dumper(Emitter, Serializer, Representer, Resolver):
def __init__(self, stream,
default_style=None, default_flow_style=None,
default_style=None, default_flow_style=False,
canonical=None, indent=None, width=None,
allow_unicode=None, line_break=None,
encoding=None, explicit_start=None, explicit_end=None,
version=None, tags=None):
version=None, tags=None, sort_keys=True):
Emitter.__init__(self, stream, canonical=canonical,
indent=indent, width=width,
allow_unicode=allow_unicode, line_break=line_break)
@ -57,6 +57,6 @@ class Dumper(Emitter, Serializer, Representer, Resolver):
explicit_start=explicit_start, explicit_end=explicit_end,
version=version, tags=tags)
Representer.__init__(self, default_style=default_style,
default_flow_style=default_flow_style)
default_flow_style=default_flow_style, sort_keys=sort_keys)
Resolver.__init__(self)

View File

@ -41,7 +41,7 @@ class Emitter:
# The stream should have the methods `write` and possibly `flush`.
self.stream = stream
# Encoding can be overriden by STREAM-START.
# Encoding can be overridden by STREAM-START.
self.encoding = None
# Emitter is a state machine with a stack of states to handle nested
@ -604,7 +604,7 @@ class Emitter:
start = end = end+1
data = ch.encode('utf-8')
for ch in data:
chunks.append('%%%02X' % ord(ch))
chunks.append('%%%02X' % ch)
if start < end:
chunks.append(suffix[start:end])
suffix_text = ''.join(chunks)
@ -652,7 +652,7 @@ class Emitter:
flow_indicators = True
# First character or preceded by a whitespace.
preceeded_by_whitespace = True
preceded_by_whitespace = True
# Last character or followed by a whitespace.
followed_by_whitespace = (len(scalar) == 1 or
@ -671,7 +671,7 @@ class Emitter:
# Check for indicators.
if index == 0:
# Leading indicators are special characters.
if ch in '#,[]{}&*!|>\'\"%@`':
if ch in '#,[]{}&*!|>\'\"%@`':
flow_indicators = True
block_indicators = True
if ch in '?:':
@ -689,7 +689,7 @@ class Emitter:
flow_indicators = True
if followed_by_whitespace:
block_indicators = True
if ch == '#' and preceeded_by_whitespace:
if ch == '#' and preceded_by_whitespace:
flow_indicators = True
block_indicators = True
@ -698,7 +698,8 @@ class Emitter:
line_breaks = True
if not (ch == '\n' or '\x20' <= ch <= '\x7E'):
if (ch == '\x85' or '\xA0' <= ch <= '\uD7FF'
or '\uE000' <= ch <= '\uFFFD') and ch != '\uFEFF':
or '\uE000' <= ch <= '\uFFFD'
or '\U00010000' <= ch < '\U0010ffff') and ch != '\uFEFF':
unicode_characters = True
if not self.allow_unicode:
special_characters = True
@ -730,7 +731,7 @@ class Emitter:
# Prepare for the next character.
index += 1
preceeded_by_whitespace = (ch in '\0 \t\r\n\x85\u2028\u2029')
preceded_by_whitespace = (ch in '\0 \t\r\n\x85\u2028\u2029')
followed_by_whitespace = (index+1 >= len(scalar) or
scalar[index+1] in '\0 \t\r\n\x85\u2028\u2029')
@ -1134,4 +1135,3 @@ class Emitter:
spaces = (ch == ' ')
breaks = (ch in '\n\x85\u2028\u2029')
end += 1

View File

@ -1,5 +1,5 @@
__all__ = ['BaseLoader', 'SafeLoader', 'Loader']
__all__ = ['BaseLoader', 'FullLoader', 'SafeLoader', 'Loader', 'UnsafeLoader']
from .reader import *
from .scanner import *
@ -18,6 +18,16 @@ class BaseLoader(Reader, Scanner, Parser, Composer, BaseConstructor, BaseResolve
BaseConstructor.__init__(self)
BaseResolver.__init__(self)
class FullLoader(Reader, Scanner, Parser, Composer, FullConstructor, Resolver):
def __init__(self, stream):
Reader.__init__(self, stream)
Scanner.__init__(self)
Parser.__init__(self)
Composer.__init__(self)
FullConstructor.__init__(self)
Resolver.__init__(self)
class SafeLoader(Reader, Scanner, Parser, Composer, SafeConstructor, Resolver):
def __init__(self, stream):
@ -38,3 +48,16 @@ class Loader(Reader, Scanner, Parser, Composer, Constructor, Resolver):
Constructor.__init__(self)
Resolver.__init__(self)
# UnsafeLoader is the same as Loader (which is and was always unsafe on
# untrusted input). Use of either Loader or UnsafeLoader should be rare, since
# FullLoad should be able to load almost all YAML safely. Loader is left intact
# to ensure backwards compatibility.
class UnsafeLoader(Reader, Scanner, Parser, Composer, Constructor, Resolver):
def __init__(self, stream):
Reader.__init__(self, stream)
Scanner.__init__(self)
Parser.__init__(self)
Composer.__init__(self)
Constructor.__init__(self)
Resolver.__init__(self)

View File

@ -134,7 +134,7 @@ class Reader(object):
self.encoding = 'utf-8'
self.update(1)
NON_PRINTABLE = re.compile('[^\x09\x0A\x0D\x20-\x7E\x85\xA0-\uD7FF\uE000-\uFFFD]')
NON_PRINTABLE = re.compile('[^\x09\x0A\x0D\x20-\x7E\x85\xA0-\uD7FF\uE000-\uFFFD\U00010000-\U0010ffff]')
def check_printable(self, data):
match = self.NON_PRINTABLE.search(data)
if match:
@ -183,10 +183,3 @@ class Reader(object):
self.stream_pointer += len(data)
if not data:
self.eof = True
#try:
# import psyco
# psyco.bind(Reader)
#except ImportError:
# pass

View File

@ -5,7 +5,7 @@ __all__ = ['BaseRepresenter', 'SafeRepresenter', 'Representer',
from .error import *
from .nodes import *
import datetime, sys, copyreg, types, base64, collections
import datetime, copyreg, types, base64, collections
class RepresenterError(YAMLError):
pass
@ -15,8 +15,9 @@ class BaseRepresenter:
yaml_representers = {}
yaml_multi_representers = {}
def __init__(self, default_style=None, default_flow_style=None):
def __init__(self, default_style=None, default_flow_style=False, sort_keys=True):
self.default_style = default_style
self.sort_keys = sort_keys
self.default_flow_style = default_flow_style
self.represented_objects = {}
self.object_keeper = []
@ -107,10 +108,11 @@ class BaseRepresenter:
best_style = True
if hasattr(mapping, 'items'):
mapping = list(mapping.items())
try:
mapping = sorted(mapping)
except TypeError:
pass
if self.sort_keys:
try:
mapping = sorted(mapping)
except TypeError:
pass
for item_key, item_value in mapping:
node_key = self.represent_data(item_key)
node_value = self.represent_data(item_value)
@ -226,7 +228,7 @@ class SafeRepresenter(BaseRepresenter):
return self.represent_mapping(tag, state, flow_style=flow_style)
def represent_undefined(self, data):
raise RepresenterError("cannot represent an object: %s" % data)
raise RepresenterError("cannot represent an object", data)
SafeRepresenter.add_representer(type(None),
SafeRepresenter.represent_none)
@ -316,7 +318,7 @@ class Representer(SafeRepresenter):
elif hasattr(data, '__reduce__'):
reduce = data.__reduce__()
else:
raise RepresenterError("cannot represent object: %r" % data)
raise RepresenterError("cannot represent an object", data)
reduce = (list(reduce)+[None]*5)[:5]
function, args, state, listitems, dictitems = reduce
args = list(args)
@ -367,7 +369,7 @@ Representer.add_representer(complex,
Representer.add_representer(tuple,
Representer.represent_tuple)
Representer.add_representer(type,
Representer.add_multi_representer(type,
Representer.represent_name)
Representer.add_representer(collections.OrderedDict,

View File

@ -146,8 +146,8 @@ class BaseResolver:
resolvers = self.yaml_implicit_resolvers.get('', [])
else:
resolvers = self.yaml_implicit_resolvers.get(value[0], [])
resolvers += self.yaml_implicit_resolvers.get(None, [])
for tag, regexp in resolvers:
wildcard_resolvers = self.yaml_implicit_resolvers.get(None, [])
for tag, regexp in resolvers + wildcard_resolvers:
if regexp.match(value):
return tag
implicit = implicit[1]
@ -177,7 +177,7 @@ Resolver.add_implicit_resolver(
Resolver.add_implicit_resolver(
'tag:yaml.org,2002:float',
re.compile(r'''^(?:[-+]?(?:[0-9][0-9_]*)\.[0-9_]*(?:[eE][-+][0-9]+)?
|\.[0-9_]+(?:[eE][-+][0-9]+)?
|\.[0-9][0-9_]*(?:[eE][-+][0-9]+)?
|[-+]?[0-9][0-9_]*(?::[0-5]?[0-9])+\.[0-9_]*
|[-+]?\.(?:inf|Inf|INF)
|\.(?:nan|NaN|NAN))$''', re.X),

View File

@ -124,10 +124,13 @@ class Scanner:
def peek_token(self):
# Return the next token, but do not delete if from the queue.
# Return None if no more tokens.
while self.need_more_tokens():
self.fetch_more_tokens()
if self.tokens:
return self.tokens[0]
else:
return None
def get_token(self):
# Return the next token.
@ -329,7 +332,7 @@ class Scanner:
## }
#if self.flow_level and self.indent > column:
# raise ScannerError(None, None,
# "invalid intendation or unclosed '[' or '{'",
# "invalid indentation or unclosed '[' or '{'",
# self.get_mark())
# In the flow context, indentation is ignored. We make the scanner less
@ -367,7 +370,7 @@ class Scanner:
def fetch_stream_end(self):
# Set the current intendation to -1.
# Set the current indentation to -1.
self.unwind_indent(-1)
# Reset simple keys.
@ -386,7 +389,7 @@ class Scanner:
def fetch_directive(self):
# Set the current intendation to -1.
# Set the current indentation to -1.
self.unwind_indent(-1)
# Reset simple keys.
@ -404,7 +407,7 @@ class Scanner:
def fetch_document_indicator(self, TokenClass):
# Set the current intendation to -1.
# Set the current indentation to -1.
self.unwind_indent(-1)
# Reset simple keys. Note that there could not be a block collection
@ -516,7 +519,7 @@ class Scanner:
# Block context needs additional checks.
if not self.flow_level:
# Are we allowed to start a key (not nessesary a simple)?
# Are we allowed to start a key (not necessary a simple)?
if not self.allow_simple_key:
raise ScannerError(None, None,
"mapping keys are not allowed here",
@ -564,7 +567,7 @@ class Scanner:
else:
# Block context needs additional checks.
# (Do we really need them? They will be catched by the parser
# (Do we really need them? They will be caught by the parser
# anyway.)
if not self.flow_level:
@ -897,7 +900,7 @@ class Scanner:
# The specification does not restrict characters for anchors and
# aliases. This may lead to problems, for instance, the document:
# [ *alias, value ]
# can be interpteted in two ways, as
# can be interpreted in two ways, as
# [ "value" ]
# and
# [ *alias , "value" ]
@ -1166,6 +1169,7 @@ class Scanner:
' ': '\x20',
'\"': '\"',
'\\': '\\',
'/': '/',
'N': '\x85',
'_': '\xA0',
'L': '\u2028',
@ -1207,7 +1211,7 @@ class Scanner:
for k in range(length):
if self.peek(k) not in '0123456789ABCDEFabcdef':
raise ScannerError("while scanning a double-quoted scalar", start_mark,
"expected escape sequence of %d hexdecimal numbers, but found %r" %
"expected escape sequence of %d hexadecimal numbers, but found %r" %
(length, self.peek(k)), self.get_mark())
code = int(self.prefix(length), 16)
chunks.append(chr(code))
@ -1266,7 +1270,7 @@ class Scanner:
def scan_plain(self):
# See the specification for details.
# We add an additional restriction for the flow context:
# plain scalars in the flow context cannot contain ',', ':' and '?'.
# plain scalars in the flow context cannot contain ',' or '?'.
# We also keep track of the `allow_simple_key` flag here.
# Indentation rules are loosed for the flow context.
chunks = []
@ -1285,18 +1289,12 @@ class Scanner:
while True:
ch = self.peek(length)
if ch in '\0 \t\r\n\x85\u2028\u2029' \
or (not self.flow_level and ch == ':' and
self.peek(length+1) in '\0 \t\r\n\x85\u2028\u2029') \
or (self.flow_level and ch in ',:?[]{}'):
or (ch == ':' and
self.peek(length+1) in '\0 \t\r\n\x85\u2028\u2029'
+ (u',[]{}' if self.flow_level else u''))\
or (self.flow_level and ch in ',?[]{}'):
break
length += 1
# It's not clear what we should do with ':' in the flow context.
if (self.flow_level and ch == ':'
and self.peek(length+1) not in '\0 \t\r\n\x85\u2028\u2029,[]{}'):
self.forward(length)
raise ScannerError("while scanning a plain scalar", start_mark,
"found unexpected ':'", self.get_mark(),
"Please check http://pyyaml.org/wiki/YAMLColonInFlowContext for details.")
if length == 0:
break
self.allow_simple_key = False
@ -1405,7 +1403,7 @@ class Scanner:
for k in range(2):
if self.peek(k) not in '0123456789ABCDEFabcdef':
raise ScannerError("while scanning a %s" % name, start_mark,
"expected URI escape sequence of 2 hexdecimal numbers, but found %r"
"expected URI escape sequence of 2 hexadecimal numbers, but found %r"
% self.peek(k), self.get_mark())
codes.append(int(self.prefix(2), 16))
self.forward(2)
@ -1435,10 +1433,3 @@ class Scanner:
self.forward()
return ch
return ''
#try:
# import psyco
# psyco.bind(Scanner)
#except ImportError:
# pass