Commit 17583c8b authored by Mark Beierl's avatar Mark Beierl
Browse files

Merge branch 'master' into 'master'

Update squid charm

See merge request !105
parents e2197769 9c012576
Pipeline #143 passed with stage
in 1 minute and 34 seconds
# -*- coding: utf-8 -*-
from ._compat import imap
from ._compat import implements_to_string
from ._compat import PY2
from ._compat import text_type
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)
@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
@implements_to_string
class TemplateNotFound(IOError, LookupError, TemplateError):
"""Raised if a template does not exist.
.. versionchanged:: 2.11
If the given name is :class:`Undefined` and no message was
provided, an :exc:`UndefinedError` is raised.
"""
# looks weird, but removes the warning descriptor that just
# bogusly warns us about message being deprecated
message = None
def __init__(self, name, message=None):
IOError.__init__(self, name)
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
class TemplatesNotFound(TemplateNotFound):
"""Like :class:`TemplateNotFound` but raised if multiple templates
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):
if message is None:
from .runtime import Undefined
parts = []
for name in names:
if isinstance(name, Undefined):
parts.append(name._undefined_message)
else:
parts.append(name)
message = u"none of the templates given were found: " + u", ".join(
imap(text_type, parts)
)
TemplateNotFound.__init__(self, names and names[-1] or 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)
self.lineno = lineno
self.name = name
self.filename = filename
self.source = 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):
# for translated errors we only return the message
if self.translated:
return self.message
# otherwise attach some stuff
location = "line %d" % self.lineno
name = self.filename or self.name
if name:
location = 'File "%s", %s' % (name, location)
lines = [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())
return u"\n".join(lines)
def __reduce__(self):
# 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):
"""Like a template syntax error, but covers cases where something in the
template caused an error at compile time that wasn't necessarily caused
by a syntax error. However it's a direct subclass of
:exc:`TemplateSyntaxError` and has the same attributes.
"""
class TemplateRuntimeError(TemplateError):
"""A generic runtime error in the template engine. Under some situations
Jinja may raise this exception.
"""
class UndefinedError(TemplateRuntimeError):
"""Raised if a template tries to operate on :class:`Undefined`."""
class SecurityError(TemplateRuntimeError):
"""Raised if a template tries to do something insecure if the
sandbox is enabled.
"""
class FilterArgumentError(TemplateRuntimeError):
"""This error is raised if a filter was called with inappropriate
arguments
"""
# -*- coding: utf-8 -*-
"""Extension API for adding custom tags and behavior."""
import pprint
import re
from sys import version_info
from markupsafe import Markup
from . import nodes
from ._compat import iteritems
from ._compat import string_types
from ._compat import with_metaclass
from .defaults import BLOCK_END_STRING
from .defaults import BLOCK_START_STRING
from .defaults import COMMENT_END_STRING
from .defaults import COMMENT_START_STRING
from .defaults import KEEP_TRAILING_NEWLINE
from .defaults import LINE_COMMENT_PREFIX
from .defaults import LINE_STATEMENT_PREFIX
from .defaults import LSTRIP_BLOCKS
from .defaults import NEWLINE_SEQUENCE
from .defaults import TRIM_BLOCKS
from .defaults import VARIABLE_END_STRING
from .defaults import VARIABLE_START_STRING
from .environment import Environment
from .exceptions import TemplateAssertionError
from .exceptions import TemplateSyntaxError
from .nodes import ContextReference
from .runtime import concat
from .utils import contextfunction
from .utils import import_string
# the only real useful gettext functions for a Jinja template. Note
# that ugettext must be assigned to gettext as Jinja doesn't support
# non unicode strings.
GETTEXT_FUNCTIONS = ("_", "gettext", "ngettext")
_ws_re = re.compile(r"\s*\n\s*")
class ExtensionRegistry(type):
"""Gives the extension an unique identifier."""
def __new__(mcs, name, bases, d):
rv = type.__new__(mcs, name, bases, d)
rv.identifier = rv.__module__ + "." + rv.__name__
return rv
class Extension(with_metaclass(ExtensionRegistry, object)):
"""Extensions can be used to add extra functionality to the Jinja template
system at the parser level. Custom extensions are bound to an environment
but may not store environment specific data on `self`. The reason for
this is that an extension can be bound to another environment (for
overlays) by creating a copy and reassigning the `environment` attribute.
As extensions are created by the environment they cannot accept any
arguments for configuration. One may want to work around that by using
a factory function, but that is not possible as extensions are identified
by their import name. The correct way to configure the extension is
storing the configuration values on the environment. Because this way the
environment ends up acting as central configuration storage the
attributes may clash which is why extensions have to ensure that the names
they choose for configuration are not too generic. ``prefix`` for example
is a terrible name, ``fragment_cache_prefix`` on the other hand is a good
name as includes the name of the extension (fragment cache).
"""
#: if this extension parses this is the list of tags it's listening to.
tags = set()
#: the priority of that extension. This is especially useful for
#: extensions that preprocess values. A lower value means higher
#: priority.
#:
#: .. versionadded:: 2.4
priority = 100
def __init__(self, environment):
self.environment = environment
def bind(self, environment):
"""Create a copy of this extension bound to another environment."""
rv = object.__new__(self.__class__)
rv.__dict__.update(self.__dict__)
rv.environment = environment
return rv
def preprocess(self, source, name, filename=None):
"""This method is called before the actual lexing and can be used to
preprocess the source. The `filename` is optional. The return value
must be the preprocessed source.
"""
return source
def filter_stream(self, stream):
"""It's passed a :class:`~jinja2.lexer.TokenStream` that can be used
to filter tokens returned. This method has to return an iterable of
:class:`~jinja2.lexer.Token`\\s, but it doesn't have to return a
:class:`~jinja2.lexer.TokenStream`.
"""
return stream
def parse(self, parser):
"""If any of the :attr:`tags` matched this method is called with the
parser as first argument. The token the parser stream is pointing at
is the name token that matched. This method has to return one or a
list of multiple nodes.
"""
raise NotImplementedError()
def attr(self, name, lineno=None):
"""Return an attribute node for the current extension. This is useful
to pass constants on extensions to generated template code.
::
self.attr('_my_attribute', lineno=lineno)
"""
return nodes.ExtensionAttribute(self.identifier, name, lineno=lineno)
def call_method(
self, name, args=None, kwargs=None, dyn_args=None, dyn_kwargs=None, lineno=None
):
"""Call a method of the extension. This is a shortcut for
:meth:`attr` + :class:`jinja2.nodes.Call`.
"""
if args is None:
args = []
if kwargs is None:
kwargs = []
return nodes.Call(
self.attr(name, lineno=lineno),
args,
kwargs,
dyn_args,
dyn_kwargs,
lineno=lineno,
)
@contextfunction
def _gettext_alias(__context, *args, **kwargs):
return __context.call(__context.resolve("gettext"), *args, **kwargs)
def _make_new_gettext(func):
@contextfunction
def gettext(__context, __string, **variables):
rv = __context.call(func, __string)
if __context.eval_ctx.autoescape:
rv = Markup(rv)
# Always treat as a format string, even if there are no
# variables. This makes translation strings more consistent
# and predictable. This requires escaping
return rv % variables
return gettext
def _make_new_ngettext(func):
@contextfunction
def ngettext(__context, __singular, __plural, __num, **variables):
variables.setdefault("num", __num)
rv = __context.call(func, __singular, __plural, __num)
if __context.eval_ctx.autoescape:
rv = Markup(rv)
# Always treat as a format string, see gettext comment above.
return rv % variables
return ngettext
class InternationalizationExtension(Extension):
"""This extension adds gettext support to Jinja."""
tags = {"trans"}
# TODO: the i18n extension is currently reevaluating values in a few
# situations. Take this example:
# {% trans count=something() %}{{ count }} foo{% pluralize
# %}{{ count }} fooss{% endtrans %}
# something is called twice here. One time for the gettext value and
# the other time for the n-parameter of the ngettext function.
def __init__(self, environment):
Extension.__init__(self, environment)
environment.globals["_"] = _gettext_alias
environment.extend(
install_gettext_translations=self._install,
install_null_translations=self._install_null,
install_gettext_callables=self._install_callables,
uninstall_gettext_translations=self._uninstall,
extract_translations=self._extract,
newstyle_gettext=False,
)
def _install(self, translations, newstyle=None):
gettext = getattr(translations, "ugettext", None)
if gettext is None:
gettext = translations.gettext
ngettext = getattr(translations, "ungettext", None)
if ngettext is None:
ngettext = translations.ngettext
self._install_callables(gettext, ngettext, newstyle)
def _install_null(self, newstyle=None):
self._install_callables(
lambda x: x, lambda s, p, n: (n != 1 and (p,) or (s,))[0], newstyle
)
def _install_callables(self, gettext, ngettext, newstyle=None):
if newstyle is not None:
self.environment.newstyle_gettext = newstyle
if self.environment.newstyle_gettext:
gettext = _make_new_gettext(gettext)
ngettext = _make_new_ngettext(ngettext)
self.environment.globals.update(gettext=gettext, ngettext=ngettext)
def _uninstall(self, translations):
for key in "gettext", "ngettext":
self.environment.globals.pop(key, None)
def _extract(self, source, gettext_functions=GETTEXT_FUNCTIONS):
if isinstance(source, string_types):
source = self.environment.parse(source)
return extract_from_ast(source, gettext_functions)
def parse(self, parser):
"""Parse a translatable tag."""
lineno = next(parser.stream).lineno
num_called_num = False
# find all the variables referenced. Additionally a variable can be
# defined in the body of the trans block too, but this is checked at
# a later state.
plural_expr = None
plural_expr_assignment = None
variables = {}
trimmed = None
while parser.stream.current.type != "block_end":
if variables:
parser.stream.expect("comma")
# skip colon for python compatibility
if parser.stream.skip_if("colon"):
break
name = parser.stream.expect("name")
if name.value in variables:
parser.fail(
"translatable variable %r defined twice." % name.value,
name.lineno,
exc=TemplateAssertionError,
)
# expressions
if parser.stream.current.type == "assign":
next(parser.stream)
variables[name.value] = var = parser.parse_expression()
elif trimmed is None and name.value in ("trimmed", "notrimmed"):
trimmed = name.value == "trimmed"
continue
else:
variables[name.value] = var = nodes.Name(name.value, "load")
if plural_expr is None:
if isinstance(var, nodes.Call):
plural_expr = nodes.Name("_trans", "load")
variables[name.value] = plural_expr
plural_expr_assignment = nodes.Assign(
nodes.Name("_trans", "store"), var
)
else:
plural_expr = var
num_called_num = name.value == "num"
parser.stream.expect("block_end")
plural = None
have_plural = False
referenced = set()
# now parse until endtrans or pluralize
singular_names, singular = self._parse_block(parser, True)
if singular_names:
referenced.update(singular_names)
if plural_expr is None:
plural_expr = nodes.Name(singular_names[0], "load")
num_called_num = singular_names[0] == "num"
# if we have a pluralize block, we parse that too
if parser.stream.current.test("name:pluralize"):
have_plural = True
next(parser.stream)
if parser.stream.current.type != "block_end":
name = parser.stream.expect("name")
if name.value not in variables:
parser.fail(
"unknown variable %r for pluralization" % name.value,
name.lineno,
exc=TemplateAssertionError,
)
plural_expr = variables[name.value]
num_called_num = name.value == "num"
parser.stream.expect("block_end")
plural_names, plural = self._parse_block(parser, False)
next(parser.stream)
referenced.update(plural_names)
else:
next(parser.stream)
# register free names as simple name expressions
for var in referenced:
if var not in variables:
variables[var] = nodes.Name(var, "load")
if not have_plural:
plural_expr = None
elif plural_expr is None:
parser.fail("pluralize without variables", lineno)
if trimmed is None:
trimmed = self.environment.policies["ext.i18n.trimmed"]
if trimmed:
singular = self._trim_whitespace(singular)
if plural:
plural = self._trim_whitespace(plural)
node = self._make_node(
singular,
plural,
variables,
plural_expr,
bool(referenced),
num_called_num and have_plural,
)
node.set_lineno(lineno)
if plural_expr_assignment is not None:
return [plural_expr_assignment, node]
else:
return node
def _trim_whitespace(self, string, _ws_re=_ws_re):
return _ws_re.sub(" ", string.strip())
def _parse_block(self, parser, allow_pluralize):
"""Parse until the next block tag with a given name."""
referenced = []
buf = []
while 1:
if parser.stream.current.type == "data":
buf.append(parser.stream.current.value.replace("%", "%%"))
next(parser.stream)
elif parser.stream.current.type == "variable_begin":
next(parser.stream)
name = parser.stream.expect("name").value
referenced.append(name)
buf.append("%%(%s)s" % name)
parser.stream.expect("variable_end")
elif parser.stream.current.type == "block_begin":
next(parser.stream)
if parser.stream.current.test("name:endtrans"):
break
elif parser.stream.current.test("name:pluralize"):
if allow_pluralize:
break
parser.fail(
"a translatable section can have only one pluralize section"
)
parser.fail(
"control structures in translatable sections are not allowed"
)
elif parser.stream.eos:
parser.fail("unclosed translation block")
else:
raise RuntimeError("internal parser error")
return referenced, concat(buf)
def _make_node(
self, singular, plural, variables, plural_expr, vars_referenced, num_called_num
):
"""Generates a useful node from the data provided."""
# no variables referenced? no need to escape for old style
# gettext invocations only if there are vars.
if not vars_referenced and not self.environment.newstyle_gettext:
singular = singular.replace("%%", "%")
if plural:
plural = plural.replace("%%", "%")
# singular only:
if plural_expr is None:
gettext = nodes.Name("gettext", "load")
node = nodes.Call(gettext, [nodes.Const(singular)], [], None, None)
# singular and plural
else:
ngettext = nodes.Name("ngettext", "load")
node = nodes.Call(
ngettext,
[nodes.Const(singular), nodes.Const(plural), plural_expr],
[],
None,
None,
)
# in case newstyle gettext is used, the method is powerful
# enough to handle the variable expansion and autoescape
# handling itself
if self.environment.newstyle_gettext:
for key, value in iteritems(variables):
# the function adds that later anyways in case num was
# called num, so just skip it.
if num_called_num and key == "num":
continue
node.kwargs.append(nodes.Keyword(key, value))
# otherwise do that here
else:
# mark the return value as safe if we are in an
# environment with autoescaping turned on
node = nodes.MarkSafeIfAutoescape(node)
if variables:
node = nodes.Mod(
node,
nodes.Dict(
[
nodes.Pair(nodes.Const(key), value)
for key, value in variables.items()
]
),
)
return nodes.Output([node])
class ExprStmtExtension(Extension):
"""Adds a `do` tag to Jinja that works like the print statement just
that it doesn't print the return value.
"""
tags = set(["do"])
def parse(self, parser):
node = nodes.ExprStmt(lineno=next(parser.stream).lineno)
node.node = parser.parse_tuple()
return node
class LoopControlExtension(Extension):
"""Adds break and continue to the template engine."""
tags = set(["break", "continue"])
def parse(self, parser):
token = next(parser.stream)
if token.value == "break":
return nodes.Break(lineno=token.lineno)
return nodes.Continue(lineno=token.lineno)
class WithExtension(Extension):
pass
class AutoEscapeExtension(Extension):
pass
class DebugExtension(Extension):
"""A ``{% debug %}`` tag that dumps the available variables,
filters, and tests.
.. code-block:: html+jinja
<pre>{% debug %}</pre>
.. code-block:: text
{'context': {'cycler': <class 'jinja2.utils.Cycler'>,
...,
'namespace': <class 'jinja2.utils.Namespace'>},
'filters': ['abs', 'attr', 'batch', 'capitalize', 'center', 'count', 'd',
..., 'urlencode', 'urlize', 'wordcount', 'wordwrap', 'xmlattr'],
'tests': ['!=', '<', '<=', '==', '>', '>=', 'callable', 'defined',
..., 'odd', 'sameas', 'sequence', 'string', 'undefined', 'upper']}
.. versionadded:: 2.11.0
"""
tags = {"debug"}
def parse(self, parser):
lineno = parser.stream.expect("name:debug").lineno
context = ContextReference()
result = self.call_method("_render", [context], lineno=lineno)
return nodes.Output([result], lineno=lineno)
def _render(self, context):
result = {
"context": context.get_all(),
"filters": sorted(self.environment.filters.keys()),
"tests": sorted(self.environment.tests.keys()),
}
# Set the depth since the intent is to show the top few names.
if version_info[:2] >= (3, 4):
return pprint.pformat(result, depth=3, compact=True)
else:
return pprint.pformat(result, depth=3)
def extract_from_ast(node, gettext_functions=GETTEXT_FUNCTIONS, babel_style=True):
"""Extract localizable strings from the given template node. Per
default this function returns matches in babel style that means non string
parameters as well as keyword arguments are returned as `None`. This
allows Babel to figure out what you really meant if you are using
gettext functions that allow keyword arguments for placeholder expansion.
If you don't want that behavior set the `babel_style` parameter to `False`
which causes only strings to be returned and parameters are always stored
in tuples. As a consequence invalid gettext calls (calls without a single
string parameter or string parameters after non-string parameters) are
skipped.
This example explains the behavior:
>>> from jinja2 import Environment
>>> env = Environment()
>>> node = env.parse('{{ (_("foo"), _(), ngettext("foo", "bar", 42)) }}')
>>> list(extract_from_ast(node))
[(1, '_', 'foo'), (1, '_', ()), (1, 'ngettext', ('foo', 'bar', None))]
>>> list(extract_from_ast(node, babel_style=False))
[(1, '_', ('foo',)), (1, 'ngettext', ('foo', 'bar'))]
For every string found this function yields a ``(lineno, function,
message)`` tuple, where:
* ``lineno`` is the number of the line on which the string was found,
* ``function`` is the name of the ``gettext`` function used (if the
string was extracted from embedded Python code), and
* ``message`` is the string itself (a ``unicode`` object, or a tuple
of ``unicode`` objects for functions with multiple string arguments).
This extraction function operates on the AST and is because of that unable
to extract any comments. For comment support you have to use the babel
extraction interface or extract comments yourself.
"""
for node in node.find_all(nodes.Call):
if (
not isinstance(node.node, nodes.Name)
or node.node.name not in gettext_functions
):
continue
strings = []
for arg in node.args:
if isinstance(arg, nodes.Const) and isinstance(arg.value, string_types):
strings.append(arg.value)
else:
strings.append(None)
for _ in node.kwargs:
strings.append(None)
if node.dyn_args is not None:
strings.append(None)
if node.dyn_kwargs is not None:
strings.append(None)
if not babel_style:
strings = tuple(x for x in strings if x is not None)
if not strings:
continue
else:
if len(strings) == 1:
strings = strings[0]
else:
strings = tuple(strings)
yield node.lineno, node.node.name, strings
class _CommentFinder(object):
"""Helper class to find comments in a token stream. Can only
find comments for gettext calls forwards. Once the comment
from line 4 is found, a comment for line 1 will not return a
usable value.
"""
def __init__(self, tokens, comment_tags):
self.tokens = tokens
self.comment_tags = comment_tags
self.offset = 0
self.last_lineno = 0
def find_backwards(self, offset):
try:
for _, token_type, token_value in reversed(
self.tokens[self.offset : offset]
):
if token_type in ("comment", "linecomment"):
try:
prefix, comment = token_value.split(None, 1)
except ValueError:
continue
if prefix in self.comment_tags:
return [comment.rstrip()]
return []
finally:
self.offset = offset
def find_comments(self, lineno):
if not self.comment_tags or self.last_lineno > lineno:
return []
for idx, (token_lineno, _, _) in enumerate(self.tokens[self.offset :]):
if token_lineno > lineno:
return self.find_backwards(self.offset + idx)
return self.find_backwards(len(self.tokens))
def babel_extract(fileobj, keywords, comment_tags, options):
"""Babel extraction method for Jinja templates.
.. versionchanged:: 2.3
Basic support for translation comments was added. If `comment_tags`
is now set to a list of keywords for extraction, the extractor will
try to find the best preceding comment that begins with one of the
keywords. For best results, make sure to not have more than one
gettext call in one line of code and the matching comment in the
same line or the line before.
.. versionchanged:: 2.5.1
The `newstyle_gettext` flag can be set to `True` to enable newstyle
gettext calls.
.. versionchanged:: 2.7
A `silent` option can now be provided. If set to `False` template
syntax errors are propagated instead of being ignored.
:param fileobj: the file-like object the messages should be extracted from
:param keywords: a list of keywords (i.e. function names) that should be
recognized as translation functions
:param comment_tags: a list of translator tags to search for and include
in the results.
:param options: a dictionary of additional options (optional)
:return: an iterator over ``(lineno, funcname, message, comments)`` tuples.
(comments will be empty currently)
"""
extensions = set()
for extension in options.get("extensions", "").split(","):
extension = extension.strip()
if not extension:
continue
extensions.add(import_string(extension))
if InternationalizationExtension not in extensions:
extensions.add(InternationalizationExtension)
def getbool(options, key, default=False):
return options.get(key, str(default)).lower() in ("1", "on", "yes", "true")
silent = getbool(options, "silent", True)
environment = Environment(
options.get("block_start_string", BLOCK_START_STRING),
options.get("block_end_string", BLOCK_END_STRING),
options.get("variable_start_string", VARIABLE_START_STRING),
options.get("variable_end_string", VARIABLE_END_STRING),
options.get("comment_start_string", COMMENT_START_STRING),
options.get("comment_end_string", COMMENT_END_STRING),
options.get("line_statement_prefix") or LINE_STATEMENT_PREFIX,
options.get("line_comment_prefix") or LINE_COMMENT_PREFIX,
getbool(options, "trim_blocks", TRIM_BLOCKS),
getbool(options, "lstrip_blocks", LSTRIP_BLOCKS),
NEWLINE_SEQUENCE,
getbool(options, "keep_trailing_newline", KEEP_TRAILING_NEWLINE),
frozenset(extensions),
cache_size=0,
auto_reload=False,
)
if getbool(options, "trimmed"):
environment.policies["ext.i18n.trimmed"] = True
if getbool(options, "newstyle_gettext"):
environment.newstyle_gettext = True
source = fileobj.read().decode(options.get("encoding", "utf-8"))
try:
node = environment.parse(source)
tokens = list(environment.lex(environment.preprocess(source)))
except TemplateSyntaxError:
if not silent:
raise
# skip templates with syntax errors
return
finder = _CommentFinder(tokens, comment_tags)
for lineno, func, message in extract_from_ast(node, keywords):
yield lineno, func, message, finder.find_comments(lineno)
#: nicer import names
i18n = InternationalizationExtension
do = ExprStmtExtension
loopcontrols = LoopControlExtension
with_ = WithExtension
autoescape = AutoEscapeExtension
debug = DebugExtension
# -*- coding: utf-8 -*-
"""Built-in template filters used with the ``|`` operator."""
import math
import random
import re
import warnings
from collections import namedtuple
from itertools import chain
from itertools import groupby
from markupsafe import escape
from markupsafe import Markup
from markupsafe import soft_unicode
from ._compat import abc
from ._compat import imap
from ._compat import iteritems
from ._compat import string_types
from ._compat import text_type
from .exceptions import FilterArgumentError
from .runtime import Undefined
from .utils import htmlsafe_json_dumps
from .utils import pformat
from .utils import unicode_urlencode
from .utils import urlize
_word_re = re.compile(r"\w+", re.UNICODE)
_word_beginning_split_re = re.compile(r"([-\s\(\{\[\<]+)", re.UNICODE)
def contextfilter(f):
"""Decorator for marking context dependent filters. The current
:class:`Context` will be passed as first argument.
"""
f.contextfilter = True
return f
def evalcontextfilter(f):
"""Decorator for marking eval-context dependent filters. An eval
context object is passed as first argument. For more information
about the eval context, see :ref:`eval-context`.
.. versionadded:: 2.4
"""
f.evalcontextfilter = True
return f
def environmentfilter(f):
"""Decorator for marking environment dependent filters. The current
:class:`Environment` is passed to the filter as first argument.
"""
f.environmentfilter = True
return f
def ignore_case(value):
"""For use as a postprocessor for :func:`make_attrgetter`. Converts strings
to lowercase and returns other types as-is."""
return value.lower() if isinstance(value, string_types) else value
def make_attrgetter(environment, attribute, postprocess=None, default=None):
"""Returns a callable that looks up the given attribute from a
passed object with the rules of the environment. Dots are allowed
to access attributes of attributes. Integer parts in paths are
looked up as integers.
"""
attribute = _prepare_attribute_parts(attribute)
def attrgetter(item):
for part in attribute:
item = environment.getitem(item, part)
if default and isinstance(item, Undefined):
item = default
if postprocess is not None:
item = postprocess(item)
return item
return attrgetter
def make_multi_attrgetter(environment, attribute, postprocess=None):
"""Returns a callable that looks up the given comma separated
attributes from a passed object with the rules of the environment.
Dots are allowed to access attributes of each attribute. Integer
parts in paths are looked up as integers.
The value returned by the returned callable is a list of extracted
attribute values.
Examples of attribute: "attr1,attr2", "attr1.inner1.0,attr2.inner2.0", etc.
"""
attribute_parts = (
attribute.split(",") if isinstance(attribute, string_types) else [attribute]
)
attribute = [
_prepare_attribute_parts(attribute_part) for attribute_part in attribute_parts
]
def attrgetter(item):
items = [None] * len(attribute)
for i, attribute_part in enumerate(attribute):
item_i = item
for part in attribute_part:
item_i = environment.getitem(item_i, part)
if postprocess is not None:
item_i = postprocess(item_i)
items[i] = item_i
return items
return attrgetter
def _prepare_attribute_parts(attr):
if attr is None:
return []
elif isinstance(attr, string_types):
return [int(x) if x.isdigit() else x for x in attr.split(".")]
else:
return [attr]
def do_forceescape(value):
"""Enforce HTML escaping. This will probably double escape variables."""
if hasattr(value, "__html__"):
value = value.__html__()
return escape(text_type(value))
def do_urlencode(value):
"""Quote data for use in a URL path or query using UTF-8.
Basic wrapper around :func:`urllib.parse.quote` when given a
string, or :func:`urllib.parse.urlencode` for a dict or iterable.
:param value: Data to quote. A string will be quoted directly. A
dict or iterable of ``(key, value)`` pairs will be joined as a
query string.
When given a string, "/" is not quoted. HTTP servers treat "/" and
"%2F" equivalently in paths. If you need quoted slashes, use the
``|replace("/", "%2F")`` filter.
.. versionadded:: 2.7
"""
if isinstance(value, string_types) or not isinstance(value, abc.Iterable):
return unicode_urlencode(value)
if isinstance(value, dict):
items = iteritems(value)
else:
items = iter(value)
return u"&".join(
"%s=%s" % (unicode_urlencode(k, for_qs=True), unicode_urlencode(v, for_qs=True))
for k, v in items
)
@evalcontextfilter
def do_replace(eval_ctx, s, old, new, count=None):
"""Return a copy of the value with all occurrences of a substring
replaced with a new one. The first argument is the substring
that should be replaced, the second is the replacement string.
If the optional third argument ``count`` is given, only the first
``count`` occurrences are replaced:
.. sourcecode:: jinja
{{ "Hello World"|replace("Hello", "Goodbye") }}
-> Goodbye World
{{ "aaaaargh"|replace("a", "d'oh, ", 2) }}
-> d'oh, d'oh, aaargh
"""
if count is None:
count = -1
if not eval_ctx.autoescape:
return text_type(s).replace(text_type(old), text_type(new), count)
if (
hasattr(old, "__html__")
or hasattr(new, "__html__")
and not hasattr(s, "__html__")
):
s = escape(s)
else:
s = soft_unicode(s)
return s.replace(soft_unicode(old), soft_unicode(new), count)
def do_upper(s):
"""Convert a value to uppercase."""
return soft_unicode(s).upper()
def do_lower(s):
"""Convert a value to lowercase."""
return soft_unicode(s).lower()
@evalcontextfilter
def do_xmlattr(_eval_ctx, d, autospace=True):
"""Create an SGML/XML attribute string based on the items in a dict.
All values that are neither `none` nor `undefined` are automatically
escaped:
.. sourcecode:: html+jinja
<ul{{ {'class': 'my_list', 'missing': none,
'id': 'list-%d'|format(variable)}|xmlattr }}>
...
</ul>
Results in something like this:
.. sourcecode:: html
<ul class="my_list" id="list-42">
...
</ul>
As you can see it automatically prepends a space in front of the item
if the filter returned something unless the second parameter is false.
"""
rv = u" ".join(
u'%s="%s"' % (escape(key), escape(value))
for key, value in iteritems(d)
if value is not None and not isinstance(value, Undefined)
)
if autospace and rv:
rv = u" " + rv
if _eval_ctx.autoescape:
rv = Markup(rv)
return rv
def do_capitalize(s):
"""Capitalize a value. The first character will be uppercase, all others
lowercase.
"""
return soft_unicode(s).capitalize()
def do_title(s):
"""Return a titlecased version of the value. I.e. words will start with
uppercase letters, all remaining characters are lowercase.
"""
return "".join(
[
item[0].upper() + item[1:].lower()
for item in _word_beginning_split_re.split(soft_unicode(s))
if item
]
)
def do_dictsort(value, case_sensitive=False, by="key", reverse=False):
"""Sort a dict and yield (key, value) pairs. Because python dicts are
unsorted you may want to use this function to order them by either
key or value:
.. sourcecode:: jinja
{% for item in mydict|dictsort %}
sort the dict by key, case insensitive
{% for item in mydict|dictsort(reverse=true) %}
sort the dict by key, case insensitive, reverse order
{% for item in mydict|dictsort(true) %}
sort the dict by key, case sensitive
{% for item in mydict|dictsort(false, 'value') %}
sort the dict by value, case insensitive
"""
if by == "key":
pos = 0
elif by == "value":
pos = 1
else:
raise FilterArgumentError('You can only sort by either "key" or "value"')
def sort_func(item):
value = item[pos]
if not case_sensitive:
value = ignore_case(value)
return value
return sorted(value.items(), key=sort_func, reverse=reverse)
@environmentfilter
def do_sort(environment, value, reverse=False, case_sensitive=False, attribute=None):
"""Sort an iterable using Python's :func:`sorted`.
.. sourcecode:: jinja
{% for city in cities|sort %}
...
{% endfor %}
:param reverse: Sort descending instead of ascending.
:param case_sensitive: When sorting strings, sort upper and lower
case separately.
:param attribute: When sorting objects or dicts, an attribute or
key to sort by. Can use dot notation like ``"address.city"``.
Can be a list of attributes like ``"age,name"``.
The sort is stable, it does not change the relative order of
elements that compare equal. This makes it is possible to chain
sorts on different attributes and ordering.
.. sourcecode:: jinja
{% for user in users|sort(attribute="name")
|sort(reverse=true, attribute="age") %}
...
{% endfor %}
As a shortcut to chaining when the direction is the same for all
attributes, pass a comma separate list of attributes.
.. sourcecode:: jinja
{% for user users|sort(attribute="age,name") %}
...
{% endfor %}
.. versionchanged:: 2.11.0
The ``attribute`` parameter can be a comma separated list of
attributes, e.g. ``"age,name"``.
.. versionchanged:: 2.6
The ``attribute`` parameter was added.
"""
key_func = make_multi_attrgetter(
environment, attribute, postprocess=ignore_case if not case_sensitive else None
)
return sorted(value, key=key_func, reverse=reverse)
@environmentfilter
def do_unique(environment, value, case_sensitive=False, attribute=None):
"""Returns a list of unique items from the given iterable.
.. sourcecode:: jinja
{{ ['foo', 'bar', 'foobar', 'FooBar']|unique|list }}
-> ['foo', 'bar', 'foobar']
The unique items are yielded in the same order as their first occurrence in
the iterable passed to the filter.
:param case_sensitive: Treat upper and lower case strings as distinct.
:param attribute: Filter objects with unique values for this attribute.
"""
getter = make_attrgetter(
environment, attribute, postprocess=ignore_case if not case_sensitive else None
)
seen = set()
for item in value:
key = getter(item)
if key not in seen:
seen.add(key)
yield item
def _min_or_max(environment, value, func, case_sensitive, attribute):
it = iter(value)
try:
first = next(it)
except StopIteration:
return environment.undefined("No aggregated item, sequence was empty.")
key_func = make_attrgetter(
environment, attribute, postprocess=ignore_case if not case_sensitive else None
)
return func(chain([first], it), key=key_func)
@environmentfilter
def do_min(environment, value, case_sensitive=False, attribute=None):
"""Return the smallest item from the sequence.
.. sourcecode:: jinja
{{ [1, 2, 3]|min }}
-> 1
:param case_sensitive: Treat upper and lower case strings as distinct.
:param attribute: Get the object with the min value of this attribute.
"""
return _min_or_max(environment, value, min, case_sensitive, attribute)
@environmentfilter
def do_max(environment, value, case_sensitive=False, attribute=None):
"""Return the largest item from the sequence.
.. sourcecode:: jinja
{{ [1, 2, 3]|max }}
-> 3
:param case_sensitive: Treat upper and lower case strings as distinct.
:param attribute: Get the object with the max value of this attribute.
"""
return _min_or_max(environment, value, max, case_sensitive, attribute)
def do_default(value, default_value=u"", boolean=False):
"""If the value is undefined it will return the passed default value,
otherwise the value of the variable:
.. sourcecode:: jinja
{{ my_variable|default('my_variable is not defined') }}
This will output the value of ``my_variable`` if the variable was
defined, otherwise ``'my_variable is not defined'``. If you want
to use default with variables that evaluate to false you have to
set the second parameter to `true`:
.. sourcecode:: jinja
{{ ''|default('the string was empty', true) }}
.. versionchanged:: 2.11
It's now possible to configure the :class:`~jinja2.Environment` with
:class:`~jinja2.ChainableUndefined` to make the `default` filter work
on nested elements and attributes that may contain undefined values
in the chain without getting an :exc:`~jinja2.UndefinedError`.
"""
if isinstance(value, Undefined) or (boolean and not value):
return default_value
return value
@evalcontextfilter
def do_join(eval_ctx, value, d=u"", attribute=None):
"""Return a string which is the concatenation of the strings in the
sequence. The separator between elements is an empty string per
default, you can define it with the optional parameter:
.. sourcecode:: jinja
{{ [1, 2, 3]|join('|') }}
-> 1|2|3
{{ [1, 2, 3]|join }}
-> 123
It is also possible to join certain attributes of an object:
.. sourcecode:: jinja
{{ users|join(', ', attribute='username') }}
.. versionadded:: 2.6
The `attribute` parameter was added.
"""
if attribute is not None:
value = imap(make_attrgetter(eval_ctx.environment, attribute), value)
# no automatic escaping? joining is a lot easier then
if not eval_ctx.autoescape:
return text_type(d).join(imap(text_type, value))
# if the delimiter doesn't have an html representation we check
# if any of the items has. If yes we do a coercion to Markup
if not hasattr(d, "__html__"):
value = list(value)
do_escape = False
for idx, item in enumerate(value):
if hasattr(item, "__html__"):
do_escape = True
else:
value[idx] = text_type(item)
if do_escape:
d = escape(d)
else:
d = text_type(d)
return d.join(value)
# no html involved, to normal joining
return soft_unicode(d).join(imap(soft_unicode, value))
def do_center(value, width=80):
"""Centers the value in a field of a given width."""
return text_type(value).center(width)
@environmentfilter
def do_first(environment, seq):
"""Return the first item of a sequence."""
try:
return next(iter(seq))
except StopIteration:
return environment.undefined("No first item, sequence was empty.")
@environmentfilter
def do_last(environment, seq):
"""
Return the last item of a sequence.
Note: Does not work with generators. You may want to explicitly
convert it to a list:
.. sourcecode:: jinja
{{ data | selectattr('name', '==', 'Jinja') | list | last }}
"""
try:
return next(iter(reversed(seq)))
except StopIteration:
return environment.undefined("No last item, sequence was empty.")
@contextfilter
def do_random(context, seq):
"""Return a random item from the sequence."""
try:
return random.choice(seq)
except IndexError:
return context.environment.undefined("No random item, sequence was empty.")
def do_filesizeformat(value, binary=False):
"""Format the value like a 'human-readable' file size (i.e. 13 kB,
4.1 MB, 102 Bytes, etc). Per default decimal prefixes are used (Mega,
Giga, etc.), if the second parameter is set to `True` the binary
prefixes are used (Mebi, Gibi).
"""
bytes = float(value)
base = binary and 1024 or 1000
prefixes = [
(binary and "KiB" or "kB"),
(binary and "MiB" or "MB"),
(binary and "GiB" or "GB"),
(binary and "TiB" or "TB"),
(binary and "PiB" or "PB"),
(binary and "EiB" or "EB"),
(binary and "ZiB" or "ZB"),
(binary and "YiB" or "YB"),
]
if bytes == 1:
return "1 Byte"
elif bytes < base:
return "%d Bytes" % bytes
else:
for i, prefix in enumerate(prefixes):
unit = base ** (i + 2)
if bytes < unit:
return "%.1f %s" % ((base * bytes / unit), prefix)
return "%.1f %s" % ((base * bytes / unit), prefix)
def do_pprint(value, verbose=False):
"""Pretty print a variable. Useful for debugging.
With Jinja 1.2 onwards you can pass it a parameter. If this parameter
is truthy the output will be more verbose (this requires `pretty`)
"""
return pformat(value, verbose=verbose)
@evalcontextfilter
def do_urlize(
eval_ctx, value, trim_url_limit=None, nofollow=False, target=None, rel=None
):
"""Converts URLs in plain text into clickable links.
If you pass the filter an additional integer it will shorten the urls
to that number. Also a third argument exists that makes the urls
"nofollow":
.. sourcecode:: jinja
{{ mytext|urlize(40, true) }}
links are shortened to 40 chars and defined with rel="nofollow"
If *target* is specified, the ``target`` attribute will be added to the
``<a>`` tag:
.. sourcecode:: jinja
{{ mytext|urlize(40, target='_blank') }}
.. versionchanged:: 2.8+
The *target* parameter was added.
"""
policies = eval_ctx.environment.policies
rel = set((rel or "").split() or [])
if nofollow:
rel.add("nofollow")
rel.update((policies["urlize.rel"] or "").split())
if target is None:
target = policies["urlize.target"]
rel = " ".join(sorted(rel)) or None
rv = urlize(value, trim_url_limit, rel=rel, target=target)
if eval_ctx.autoescape:
rv = Markup(rv)
return rv
def do_indent(s, width=4, first=False, blank=False, indentfirst=None):
"""Return a copy of the string with each line indented by 4 spaces. The
first line and blank lines are not indented by default.
:param width: Number of spaces to indent by.
:param first: Don't skip indenting the first line.
:param blank: Don't skip indenting empty lines.
.. versionchanged:: 2.10
Blank lines are not indented by default.
Rename the ``indentfirst`` argument to ``first``.
"""
if indentfirst is not None:
warnings.warn(
"The 'indentfirst' argument is renamed to 'first' and will"
" be removed in version 3.0.",
DeprecationWarning,
stacklevel=2,
)
first = indentfirst
indention = u" " * width
newline = u"\n"
if isinstance(s, Markup):
indention = Markup(indention)
newline = Markup(newline)
s += newline # this quirk is necessary for splitlines method
if blank:
rv = (newline + indention).join(s.splitlines())
else:
lines = s.splitlines()
rv = lines.pop(0)
if lines:
rv += newline + newline.join(
indention + line if line else line for line in lines
)
if first:
rv = indention + rv
return rv
@environmentfilter
def do_truncate(env, s, length=255, killwords=False, end="...", leeway=None):
"""Return a truncated copy of the string. The length is specified
with the first parameter which defaults to ``255``. If the second
parameter is ``true`` the filter will cut the text at length. Otherwise
it will discard the last word. If the text was in fact
truncated it will append an ellipsis sign (``"..."``). If you want a
different ellipsis sign than ``"..."`` you can specify it using the
third parameter. Strings that only exceed the length by the tolerance
margin given in the fourth parameter will not be truncated.
.. sourcecode:: jinja
{{ "foo bar baz qux"|truncate(9) }}
-> "foo..."
{{ "foo bar baz qux"|truncate(9, True) }}
-> "foo ba..."
{{ "foo bar baz qux"|truncate(11) }}
-> "foo bar baz qux"
{{ "foo bar baz qux"|truncate(11, False, '...', 0) }}
-> "foo bar..."
The default leeway on newer Jinja versions is 5 and was 0 before but
can be reconfigured globally.
"""
if leeway is None:
leeway = env.policies["truncate.leeway"]
assert length >= len(end), "expected length >= %s, got %s" % (len(end), length)
assert leeway >= 0, "expected leeway >= 0, got %s" % leeway
if len(s) <= length + leeway:
return s
if killwords:
return s[: length - len(end)] + end
result = s[: length - len(end)].rsplit(" ", 1)[0]
return result + end
@environmentfilter
def do_wordwrap(
environment,
s,
width=79,
break_long_words=True,
wrapstring=None,
break_on_hyphens=True,
):
"""Wrap a string to the given width. Existing newlines are treated
as paragraphs to be wrapped separately.
:param s: Original text to wrap.
:param width: Maximum length of wrapped lines.
:param break_long_words: If a word is longer than ``width``, break
it across lines.
:param break_on_hyphens: If a word contains hyphens, it may be split
across lines.
:param wrapstring: String to join each wrapped line. Defaults to
:attr:`Environment.newline_sequence`.
.. versionchanged:: 2.11
Existing newlines are treated as paragraphs wrapped separately.
.. versionchanged:: 2.11
Added the ``break_on_hyphens`` parameter.
.. versionchanged:: 2.7
Added the ``wrapstring`` parameter.
"""
import textwrap
if not wrapstring:
wrapstring = environment.newline_sequence
# textwrap.wrap doesn't consider existing newlines when wrapping.
# If the string has a newline before width, wrap will still insert
# a newline at width, resulting in a short line. Instead, split and
# wrap each paragraph individually.
return wrapstring.join(
[
wrapstring.join(
textwrap.wrap(
line,
width=width,
expand_tabs=False,
replace_whitespace=False,
break_long_words=break_long_words,
break_on_hyphens=break_on_hyphens,
)
)
for line in s.splitlines()
]
)
def do_wordcount(s):
"""Count the words in that string."""
return len(_word_re.findall(soft_unicode(s)))
def do_int(value, default=0, base=10):
"""Convert the value into an integer. If the
conversion doesn't work it will return ``0``. You can
override this default using the first parameter. You
can also override the default base (10) in the second
parameter, which handles input with prefixes such as
0b, 0o and 0x for bases 2, 8 and 16 respectively.
The base is ignored for decimal numbers and non-string values.
"""
try:
if isinstance(value, string_types):
return int(value, base)
return int(value)
except (TypeError, ValueError):
# this quirk is necessary so that "42.23"|int gives 42.
try:
return int(float(value))
except (TypeError, ValueError):
return default
def do_float(value, default=0.0):
"""Convert the value into a floating point number. If the
conversion doesn't work it will return ``0.0``. You can
override this default using the first parameter.
"""
try:
return float(value)
except (TypeError, ValueError):
return default
def do_format(value, *args, **kwargs):
"""Apply the given values to a `printf-style`_ format string, like
``string % values``.
.. sourcecode:: jinja
{{ "%s, %s!"|format(greeting, name) }}
Hello, World!
In most cases it should be more convenient and efficient to use the
``%`` operator or :meth:`str.format`.
.. code-block:: text
{{ "%s, %s!" % (greeting, name) }}
{{ "{}, {}!".format(greeting, name) }}
.. _printf-style: https://docs.python.org/library/stdtypes.html
#printf-style-string-formatting
"""
if args and kwargs:
raise FilterArgumentError(
"can't handle positional and keyword arguments at the same time"
)
return soft_unicode(value) % (kwargs or args)
def do_trim(value, chars=None):
"""Strip leading and trailing characters, by default whitespace."""
return soft_unicode(value).strip(chars)
def do_striptags(value):
"""Strip SGML/XML tags and replace adjacent whitespace by one space."""
if hasattr(value, "__html__"):
value = value.__html__()
return Markup(text_type(value)).striptags()
def do_slice(value, slices, fill_with=None):
"""Slice an iterator and return a list of lists containing
those items. Useful if you want to create a div containing
three ul tags that represent columns:
.. sourcecode:: html+jinja
<div class="columnwrapper">
{%- for column in items|slice(3) %}
<ul class="column-{{ loop.index }}">
{%- for item in column %}
<li>{{ item }}</li>
{%- endfor %}
</ul>
{%- endfor %}
</div>
If you pass it a second argument it's used to fill missing
values on the last iteration.
"""
seq = list(value)
length = len(seq)
items_per_slice = length // slices
slices_with_extra = length % slices
offset = 0
for slice_number in range(slices):
start = offset + slice_number * items_per_slice
if slice_number < slices_with_extra:
offset += 1
end = offset + (slice_number + 1) * items_per_slice
tmp = seq[start:end]
if fill_with is not None and slice_number >= slices_with_extra:
tmp.append(fill_with)
yield tmp
def do_batch(value, linecount, fill_with=None):
"""
A filter that batches items. It works pretty much like `slice`
just the other way round. It returns a list of lists with the
given number of items. If you provide a second parameter this
is used to fill up missing items. See this example:
.. sourcecode:: html+jinja
<table>
{%- for row in items|batch(3, '&nbsp;') %}
<tr>
{%- for column in row %}
<td>{{ column }}</td>
{%- endfor %}
</tr>
{%- endfor %}
</table>
"""
tmp = []
for item in value:
if len(tmp) == linecount:
yield tmp
tmp = []
tmp.append(item)
if tmp:
if fill_with is not None and len(tmp) < linecount:
tmp += [fill_with] * (linecount - len(tmp))
yield tmp
def do_round(value, precision=0, method="common"):
"""Round the number to a given precision. The first
parameter specifies the precision (default is ``0``), the
second the rounding method:
- ``'common'`` rounds either up or down
- ``'ceil'`` always rounds up
- ``'floor'`` always rounds down
If you don't specify a method ``'common'`` is used.
.. sourcecode:: jinja
{{ 42.55|round }}
-> 43.0
{{ 42.55|round(1, 'floor') }}
-> 42.5
Note that even if rounded to 0 precision, a float is returned. If
you need a real integer, pipe it through `int`:
.. sourcecode:: jinja
{{ 42.55|round|int }}
-> 43
"""
if method not in {"common", "ceil", "floor"}:
raise FilterArgumentError("method must be common, ceil or floor")
if method == "common":
return round(value, precision)
func = getattr(math, method)
return func(value * (10 ** precision)) / (10 ** precision)
# Use a regular tuple repr here. This is what we did in the past and we
# really want to hide this custom type as much as possible. In particular
# we do not want to accidentally expose an auto generated repr in case
# people start to print this out in comments or something similar for
# debugging.
_GroupTuple = namedtuple("_GroupTuple", ["grouper", "list"])
_GroupTuple.__repr__ = tuple.__repr__
_GroupTuple.__str__ = tuple.__str__
@environmentfilter
def do_groupby(environment, value, attribute):
"""Group a sequence of objects by an attribute using Python's
:func:`itertools.groupby`. The attribute can use dot notation for
nested access, like ``"address.city"``. Unlike Python's ``groupby``,
the values are sorted first so only one group is returned for each
unique value.
For example, a list of ``User`` objects with a ``city`` attribute
can be rendered in groups. In this example, ``grouper`` refers to
the ``city`` value of the group.
.. sourcecode:: html+jinja
<ul>{% for city, items in users|groupby("city") %}
<li>{{ city }}
<ul>{% for user in items %}
<li>{{ user.name }}
{% endfor %}</ul>
</li>
{% endfor %}</ul>
``groupby`` yields namedtuples of ``(grouper, list)``, which
can be used instead of the tuple unpacking above. ``grouper`` is the
value of the attribute, and ``list`` is the items with that value.
.. sourcecode:: html+jinja
<ul>{% for group in users|groupby("city") %}
<li>{{ group.grouper }}: {{ group.list|join(", ") }}
{% endfor %}</ul>
.. versionchanged:: 2.6
The attribute supports dot notation for nested access.
"""
expr = make_attrgetter(environment, attribute)
return [
_GroupTuple(key, list(values))
for key, values in groupby(sorted(value, key=expr), expr)
]
@environmentfilter
def do_sum(environment, iterable, attribute=None, start=0):
"""Returns the sum of a sequence of numbers plus the value of parameter
'start' (which defaults to 0). When the sequence is empty it returns
start.
It is also possible to sum up only certain attributes:
.. sourcecode:: jinja
Total: {{ items|sum(attribute='price') }}
.. versionchanged:: 2.6
The `attribute` parameter was added to allow suming up over
attributes. Also the `start` parameter was moved on to the right.
"""
if attribute is not None:
iterable = imap(make_attrgetter(environment, attribute), iterable)
return sum(iterable, start)
def do_list(value):
"""Convert the value into a list. If it was a string the returned list
will be a list of characters.
"""
return list(value)
def do_mark_safe(value):
"""Mark the value as safe which means that in an environment with automatic
escaping enabled this variable will not be escaped.
"""
return Markup(value)
def do_mark_unsafe(value):
"""Mark a value as unsafe. This is the reverse operation for :func:`safe`."""
return text_type(value)
def do_reverse(value):
"""Reverse the object or return an iterator that iterates over it the other
way round.
"""
if isinstance(value, string_types):
return value[::-1]
try:
return reversed(value)
except TypeError:
try:
rv = list(value)
rv.reverse()
return rv
except TypeError:
raise FilterArgumentError("argument must be iterable")
@environmentfilter
def do_attr(environment, obj, name):
"""Get an attribute of an object. ``foo|attr("bar")`` works like
``foo.bar`` just that always an attribute is returned and items are not
looked up.
See :ref:`Notes on subscriptions <notes-on-subscriptions>` for more details.
"""
try:
name = str(name)
except UnicodeError:
pass
else:
try:
value = getattr(obj, name)
except AttributeError:
pass
else:
if environment.sandboxed and not environment.is_safe_attribute(
obj, name, value
):
return environment.unsafe_undefined(obj, name)
return value
return environment.undefined(obj=obj, name=name)
@contextfilter
def do_map(*args, **kwargs):
"""Applies a filter on a sequence of objects or looks up an attribute.
This is useful when dealing with lists of objects but you are really
only interested in a certain value of it.
The basic usage is mapping on an attribute. Imagine you have a list
of users but you are only interested in a list of usernames:
.. sourcecode:: jinja
Users on this page: {{ users|map(attribute='username')|join(', ') }}
You can specify a ``default`` value to use if an object in the list
does not have the given attribute.
.. sourcecode:: jinja
{{ users|map(attribute="username", default="Anonymous")|join(", ") }}
Alternatively you can let it invoke a filter by passing the name of the
filter and the arguments afterwards. A good example would be applying a
text conversion filter on a sequence:
.. sourcecode:: jinja
Users on this page: {{ titles|map('lower')|join(', ') }}
Similar to a generator comprehension such as:
.. code-block:: python
(u.username for u in users)
(u.username or "Anonymous" for u in users)
(do_lower(x) for x in titles)
.. versionchanged:: 2.11.0
Added the ``default`` parameter.
.. versionadded:: 2.7
"""
seq, func = prepare_map(args, kwargs)
if seq:
for item in seq:
yield func(item)
@contextfilter
def do_select(*args, **kwargs):
"""Filters a sequence of objects by applying a test to each object,
and only selecting the objects with the test succeeding.
If no test is specified, each object will be evaluated as a boolean.
Example usage:
.. sourcecode:: jinja
{{ numbers|select("odd") }}
{{ numbers|select("odd") }}
{{ numbers|select("divisibleby", 3) }}
{{ numbers|select("lessthan", 42) }}
{{ strings|select("equalto", "mystring") }}
Similar to a generator comprehension such as:
.. code-block:: python
(n for n in numbers if test_odd(n))
(n for n in numbers if test_divisibleby(n, 3))
.. versionadded:: 2.7
"""
return select_or_reject(args, kwargs, lambda x: x, False)
@contextfilter
def do_reject(*args, **kwargs):
"""Filters a sequence of objects by applying a test to each object,
and rejecting the objects with the test succeeding.
If no test is specified, each object will be evaluated as a boolean.
Example usage:
.. sourcecode:: jinja
{{ numbers|reject("odd") }}
Similar to a generator comprehension such as:
.. code-block:: python
(n for n in numbers if not test_odd(n))
.. versionadded:: 2.7
"""
return select_or_reject(args, kwargs, lambda x: not x, False)
@contextfilter
def do_selectattr(*args, **kwargs):
"""Filters a sequence of objects by applying a test to the specified
attribute of each object, and only selecting the objects with the
test succeeding.
If no test is specified, the attribute's value will be evaluated as
a boolean.
Example usage:
.. sourcecode:: jinja
{{ users|selectattr("is_active") }}
{{ users|selectattr("email", "none") }}
Similar to a generator comprehension such as:
.. code-block:: python
(u for user in users if user.is_active)
(u for user in users if test_none(user.email))
.. versionadded:: 2.7
"""
return select_or_reject(args, kwargs, lambda x: x, True)
@contextfilter
def do_rejectattr(*args, **kwargs):
"""Filters a sequence of objects by applying a test to the specified
attribute of each object, and rejecting the objects with the test
succeeding.
If no test is specified, the attribute's value will be evaluated as
a boolean.
.. sourcecode:: jinja
{{ users|rejectattr("is_active") }}
{{ users|rejectattr("email", "none") }}
Similar to a generator comprehension such as:
.. code-block:: python
(u for user in users if not user.is_active)
(u for user in users if not test_none(user.email))
.. versionadded:: 2.7
"""
return select_or_reject(args, kwargs, lambda x: not x, True)
@evalcontextfilter
def do_tojson(eval_ctx, value, indent=None):
"""Dumps a structure to JSON so that it's safe to use in ``<script>``
tags. It accepts the same arguments and returns a JSON string. Note that
this is available in templates through the ``|tojson`` filter which will
also mark the result as safe. Due to how this function escapes certain
characters this is safe even if used outside of ``<script>`` tags.
The following characters are escaped in strings:
- ``<``
- ``>``
- ``&``
- ``'``
This makes it safe to embed such strings in any place in HTML with the
notable exception of double quoted attributes. In that case single
quote your attributes or HTML escape it in addition.
The indent parameter can be used to enable pretty printing. Set it to
the number of spaces that the structures should be indented with.
Note that this filter is for use in HTML contexts only.
.. versionadded:: 2.9
"""
policies = eval_ctx.environment.policies
dumper = policies["json.dumps_function"]
options = policies["json.dumps_kwargs"]
if indent is not None:
options = dict(options)
options["indent"] = indent
return htmlsafe_json_dumps(value, dumper=dumper, **options)
def prepare_map(args, kwargs):
context = args[0]
seq = args[1]
default = None
if len(args) == 2 and "attribute" in kwargs:
attribute = kwargs.pop("attribute")
default = kwargs.pop("default", None)
if kwargs:
raise FilterArgumentError(
"Unexpected keyword argument %r" % next(iter(kwargs))
)
func = make_attrgetter(context.environment, attribute, default=default)
else:
try:
name = args[2]
args = args[3:]
except LookupError:
raise FilterArgumentError("map requires a filter argument")
def func(item):
return context.environment.call_filter(
name, item, args, kwargs, context=context
)
return seq, func
def prepare_select_or_reject(args, kwargs, modfunc, lookup_attr):
context = args[0]
seq = args[1]
if lookup_attr:
try:
attr = args[2]
except LookupError:
raise FilterArgumentError("Missing parameter for attribute name")
transfunc = make_attrgetter(context.environment, attr)
off = 1
else:
off = 0
def transfunc(x):
return x
try:
name = args[2 + off]
args = args[3 + off :]
def func(item):
return context.environment.call_test(name, item, args, kwargs)
except LookupError:
func = bool
return seq, lambda item: modfunc(func(transfunc(item)))
def select_or_reject(args, kwargs, modfunc, lookup_attr):
seq, func = prepare_select_or_reject(args, kwargs, modfunc, lookup_attr)
if seq:
for item in seq:
if func(item):
yield item
FILTERS = {
"abs": abs,
"attr": do_attr,
"batch": do_batch,
"capitalize": do_capitalize,
"center": do_center,
"count": len,
"d": do_default,
"default": do_default,
"dictsort": do_dictsort,
"e": escape,
"escape": escape,
"filesizeformat": do_filesizeformat,
"first": do_first,
"float": do_float,
"forceescape": do_forceescape,
"format": do_format,
"groupby": do_groupby,
"indent": do_indent,
"int": do_int,
"join": do_join,
"last": do_last,
"length": len,
"list": do_list,
"lower": do_lower,
"map": do_map,
"min": do_min,
"max": do_max,
"pprint": do_pprint,
"random": do_random,
"reject": do_reject,
"rejectattr": do_rejectattr,
"replace": do_replace,
"reverse": do_reverse,
"round": do_round,
"safe": do_mark_safe,
"select": do_select,
"selectattr": do_selectattr,
"slice": do_slice,
"sort": do_sort,
"string": soft_unicode,
"striptags": do_striptags,
"sum": do_sum,
"title": do_title,
"trim": do_trim,
"truncate": do_truncate,
"unique": do_unique,
"upper": do_upper,
"urlencode": do_urlencode,
"urlize": do_urlize,
"wordcount": do_wordcount,
"wordwrap": do_wordwrap,
"xmlattr": do_xmlattr,
"tojson": do_tojson,
}
from ._compat import iteritems
from .visitor import NodeVisitor
VAR_LOAD_PARAMETER = "param"
VAR_LOAD_RESOLVE = "resolve"
VAR_LOAD_ALIAS = "alias"
VAR_LOAD_UNDEFINED = "undefined"
def find_symbols(nodes, parent_symbols=None):
sym = Symbols(parent=parent_symbols)
visitor = FrameSymbolVisitor(sym)
for node in nodes:
visitor.visit(node)
return sym
def symbols_for_node(node, parent_symbols=None):
sym = Symbols(parent=parent_symbols)
sym.analyze_node(node)
return sym
class Symbols(object):
def __init__(self, parent=None, level=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):
visitor = RootVisitor(self)
visitor.visit(node, **kwargs)
def _define_ref(self, name, load=None):
ident = "l_%d_%s" % (self.level, name)
self.refs[name] = ident
if load is not None:
self.loads[ident] = load
return ident
def find_load(self, target):
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):
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):
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
)
return rv
def copy(self):
rv = object.__new__(self.__class__)
rv.__dict__.update(self.__dict__)
rv.refs = self.refs.copy()
rv.loads = self.loads.copy()
rv.stores = self.stores.copy()
return rv
def store(self, name):
self.stores.add(name)
# If we have not see the name referenced yet, we need to figure
# out what to set it to.
if name not in self.refs:
# If there is a parent scope we check if the name has a
# reference there. If it does it means we might have to alias
# to a variable there.
if self.parent is not None:
outer_ref = self.parent.find_ref(name)
if outer_ref is not None:
self._define_ref(name, load=(VAR_LOAD_ALIAS, outer_ref))
return
# Otherwise we can just set it to undefined.
self._define_ref(name, load=(VAR_LOAD_UNDEFINED, None))
def declare_parameter(self, name):
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:
self._define_ref(name, load=(VAR_LOAD_RESOLVE, name))
def branch_update(self, branch_symbols):
stores = {}
for branch in branch_symbols:
for target in branch.stores:
if target in self.stores:
continue
stores[target] = stores.get(target, 0) + 1
for sym in branch_symbols:
self.refs.update(sym.refs)
self.loads.update(sym.loads)
self.stores.update(sym.stores)
for name, branch_count in iteritems(stores):
if branch_count == len(branch_symbols):
continue
target = self.find_ref(name)
assert target is not None, "should not happen"
if self.parent is not None:
outer_target = self.parent.find_ref(name)
if outer_target is not None:
self.loads[target] = (VAR_LOAD_ALIAS, outer_target)
continue
self.loads[target] = (VAR_LOAD_RESOLVE, name)
def dump_stores(self):
rv = {}
node = self
while node is not None:
for name in node.stores:
if name not in rv:
rv[name] = self.find_ref(name)
node = node.parent
return rv
def dump_param_targets(self):
rv = set()
node = self
while node is not None:
for target, (instr, _) in iteritems(self.loads):
if instr == VAR_LOAD_PARAMETER:
rv.add(target)
node = node.parent
return rv
class RootVisitor(NodeVisitor):
def __init__(self, symbols):
self.sym_visitor = FrameSymbolVisitor(symbols)
def _simple_visit(self, node, **kwargs):
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
def visit_AssignBlock(self, node, **kwargs):
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",)):
self.sym_visitor.visit(child)
def visit_OverlayScope(self, node, **kwargs):
for child in node.body:
self.sym_visitor.visit(child)
def visit_For(self, node, for_branch="body", **kwargs):
if for_branch == "body":
self.sym_visitor.visit(node.target, store_as_param=True)
branch = node.body
elif for_branch == "else":
branch = node.else_
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)
def visit_With(self, node, **kwargs):
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__
)
class FrameSymbolVisitor(NodeVisitor):
"""A visitor for `Frame.inspect`."""
def __init__(self, symbols):
self.symbols = symbols
def visit_Name(self, node, store_as_param=False, **kwargs):
"""All assignments to names go through this function."""
if store_as_param or node.ctx == "param":
self.symbols.declare_parameter(node.name)
elif node.ctx == "store":
self.symbols.store(node.name)
elif node.ctx == "load":
self.symbols.load(node.name)
def visit_NSRef(self, node, **kwargs):
self.symbols.load(node.name)
def visit_If(self, node, **kwargs):
self.visit(node.test, **kwargs)
original_symbols = self.symbols
def inner_visit(nodes):
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):
self.symbols.store(node.name)
def visit_Import(self, node, **kwargs):
self.generic_visit(node, **kwargs)
self.symbols.store(node.target)
def visit_FromImport(self, node, **kwargs):
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):
"""Visit assignments in the correct order."""
self.visit(node.node, **kwargs)
self.visit(node.target, **kwargs)
def visit_For(self, node, **kwargs):
"""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):
self.visit(node.call, **kwargs)
def visit_FilterBlock(self, node, **kwargs):
self.visit(node.filter, **kwargs)
def visit_With(self, node, **kwargs):
for target in node.values:
self.visit(target)
def visit_AssignBlock(self, node, **kwargs):
"""Stop visiting at block assigns."""
self.visit(node.target, **kwargs)
def visit_Scope(self, node, **kwargs):
"""Stop visiting at scopes."""
def visit_Block(self, node, **kwargs):
"""Stop visiting at blocks."""
def visit_OverlayScope(self, node, **kwargs):
"""Do not visit into overlay scopes."""
# -*- coding: utf-8 -*-
"""Implements a Jinja / Python combination lexer. The ``Lexer`` class
is used to do some preprocessing. It filters out invalid operators like
the bitshift operators we don't allow in templates. It separates
template code and python code in expressions.
"""
import re
from ast import literal_eval
from collections import deque
from operator import itemgetter
from ._compat import implements_iterator
from ._compat import intern
from ._compat import iteritems
from ._compat import text_type
from .exceptions import TemplateSyntaxError
from .utils import LRUCache
# cache for the lexers. Exists in order to be able to have multiple
# environments with the same lexer
_lexer_cache = LRUCache(50)
# static regular expressions
whitespace_re = re.compile(r"\s+", re.U)
newline_re = re.compile(r"(\r\n|\r|\n)")
string_re = re.compile(
r"('([^'\\]*(?:\\.[^'\\]*)*)'" r'|"([^"\\]*(?:\\.[^"\\]*)*)")', re.S
)
integer_re = re.compile(r"(\d+_)*\d+")
float_re = re.compile(
r"""
(?<!\.) # doesn't start with a .
(\d+_)*\d+ # digits, possibly _ separated
(
(\.(\d+_)*\d+)? # optional fractional part
e[+\-]?(\d+_)*\d+ # exponent part
|
\.(\d+_)*\d+ # required fractional part
)
""",
re.IGNORECASE | re.VERBOSE,
)
try:
# check if this Python supports Unicode identifiers
compile("föö", "<unknown>", "eval")
except SyntaxError:
# Python 2, no Unicode support, use ASCII identifiers
name_re = re.compile(r"[a-zA-Z_][a-zA-Z0-9_]*")
check_ident = False
else:
# Unicode support, import generated re pattern and set flag to use
# str.isidentifier to validate during lexing.
from ._identifier import pattern as name_re
check_ident = True
# internal the tokens and keep references to them
TOKEN_ADD = intern("add")
TOKEN_ASSIGN = intern("assign")
TOKEN_COLON = intern("colon")
TOKEN_COMMA = intern("comma")
TOKEN_DIV = intern("div")
TOKEN_DOT = intern("dot")
TOKEN_EQ = intern("eq")
TOKEN_FLOORDIV = intern("floordiv")
TOKEN_GT = intern("gt")
TOKEN_GTEQ = intern("gteq")
TOKEN_LBRACE = intern("lbrace")
TOKEN_LBRACKET = intern("lbracket")
TOKEN_LPAREN = intern("lparen")
TOKEN_LT = intern("lt")
TOKEN_LTEQ = intern("lteq")
TOKEN_MOD = intern("mod")
TOKEN_MUL = intern("mul")
TOKEN_NE = intern("ne")
TOKEN_PIPE = intern("pipe")
TOKEN_POW = intern("pow")
TOKEN_RBRACE = intern("rbrace")
TOKEN_RBRACKET = intern("rbracket")
TOKEN_RPAREN = intern("rparen")
TOKEN_SEMICOLON = intern("semicolon")
TOKEN_SUB = intern("sub")
TOKEN_TILDE = intern("tilde")
TOKEN_WHITESPACE = intern("whitespace")
TOKEN_FLOAT = intern("float")
TOKEN_INTEGER = intern("integer")
TOKEN_NAME = intern("name")
TOKEN_STRING = intern("string")
TOKEN_OPERATOR = intern("operator")
TOKEN_BLOCK_BEGIN = intern("block_begin")
TOKEN_BLOCK_END = intern("block_end")
TOKEN_VARIABLE_BEGIN = intern("variable_begin")
TOKEN_VARIABLE_END = intern("variable_end")
TOKEN_RAW_BEGIN = intern("raw_begin")
TOKEN_RAW_END = intern("raw_end")
TOKEN_COMMENT_BEGIN = intern("comment_begin")
TOKEN_COMMENT_END = intern("comment_end")
TOKEN_COMMENT = intern("comment")
TOKEN_LINESTATEMENT_BEGIN = intern("linestatement_begin")
TOKEN_LINESTATEMENT_END = intern("linestatement_end")
TOKEN_LINECOMMENT_BEGIN = intern("linecomment_begin")
TOKEN_LINECOMMENT_END = intern("linecomment_end")
TOKEN_LINECOMMENT = intern("linecomment")
TOKEN_DATA = intern("data")
TOKEN_INITIAL = intern("initial")
TOKEN_EOF = intern("eof")
# bind operators to token types
operators = {
"+": TOKEN_ADD,
"-": TOKEN_SUB,
"/": TOKEN_DIV,
"//": TOKEN_FLOORDIV,
"*": TOKEN_MUL,
"%": TOKEN_MOD,
"**": TOKEN_POW,
"~": TOKEN_TILDE,
"[": TOKEN_LBRACKET,
"]": TOKEN_RBRACKET,
"(": TOKEN_LPAREN,
")": TOKEN_RPAREN,
"{": TOKEN_LBRACE,
"}": TOKEN_RBRACE,
"==": TOKEN_EQ,
"!=": TOKEN_NE,
">": TOKEN_GT,
">=": TOKEN_GTEQ,
"<": TOKEN_LT,
"<=": TOKEN_LTEQ,
"=": TOKEN_ASSIGN,
".": TOKEN_DOT,
":": TOKEN_COLON,
"|": TOKEN_PIPE,
",": TOKEN_COMMA,
";": TOKEN_SEMICOLON,
}
reverse_operators = dict([(v, k) for k, v in iteritems(operators)])
assert len(operators) == len(reverse_operators), "operators dropped"
operator_re = re.compile(
"(%s)" % "|".join(re.escape(x) for x in sorted(operators, key=lambda x: -len(x)))
)
ignored_tokens = frozenset(
[
TOKEN_COMMENT_BEGIN,
TOKEN_COMMENT,
TOKEN_COMMENT_END,
TOKEN_WHITESPACE,
TOKEN_LINECOMMENT_BEGIN,
TOKEN_LINECOMMENT_END,
TOKEN_LINECOMMENT,
]
)
ignore_if_empty = frozenset(
[TOKEN_WHITESPACE, TOKEN_DATA, TOKEN_COMMENT, TOKEN_LINECOMMENT]
)
def _describe_token_type(token_type):
if token_type in reverse_operators:
return reverse_operators[token_type]
return {
TOKEN_COMMENT_BEGIN: "begin of comment",
TOKEN_COMMENT_END: "end of comment",
TOKEN_COMMENT: "comment",
TOKEN_LINECOMMENT: "comment",
TOKEN_BLOCK_BEGIN: "begin of statement block",
TOKEN_BLOCK_END: "end of statement block",
TOKEN_VARIABLE_BEGIN: "begin of print statement",
TOKEN_VARIABLE_END: "end of print statement",
TOKEN_LINESTATEMENT_BEGIN: "begin of line statement",
TOKEN_LINESTATEMENT_END: "end of line statement",
TOKEN_DATA: "template data / text",
TOKEN_EOF: "end of template",
}.get(token_type, token_type)
def describe_token(token):
"""Returns a description of the token."""
if token.type == TOKEN_NAME:
return token.value
return _describe_token_type(token.type)
def describe_token_expr(expr):
"""Like `describe_token` but for token expressions."""
if ":" in expr:
type, value = expr.split(":", 1)
if type == TOKEN_NAME:
return value
else:
type = expr
return _describe_token_type(type)
def count_newlines(value):
"""Count the number of newline characters in the string. This is
useful for extensions that filter a stream.
"""
return len(newline_re.findall(value))
def compile_rules(environment):
"""Compiles all the rules from the environment into a list of rules."""
e = re.escape
rules = [
(
len(environment.comment_start_string),
TOKEN_COMMENT_BEGIN,
e(environment.comment_start_string),
),
(
len(environment.block_start_string),
TOKEN_BLOCK_BEGIN,
e(environment.block_start_string),
),
(
len(environment.variable_start_string),
TOKEN_VARIABLE_BEGIN,
e(environment.variable_start_string),
),
]
if environment.line_statement_prefix is not None:
rules.append(
(
len(environment.line_statement_prefix),
TOKEN_LINESTATEMENT_BEGIN,
r"^[ \t\v]*" + e(environment.line_statement_prefix),
)
)
if environment.line_comment_prefix is not None:
rules.append(
(
len(environment.line_comment_prefix),
TOKEN_LINECOMMENT_BEGIN,
r"(?:^|(?<=\S))[^\S\r\n]*" + e(environment.line_comment_prefix),
)
)
return [x[1:] for x in sorted(rules, reverse=True)]
class Failure(object):
"""Class that raises a `TemplateSyntaxError` if called.
Used by the `Lexer` to specify known errors.
"""
def __init__(self, message, cls=TemplateSyntaxError):
self.message = message
self.error_class = cls
def __call__(self, lineno, filename):
raise self.error_class(self.message, lineno, filename)
class Token(tuple):
"""Token class."""
__slots__ = ()
lineno, type, value = (property(itemgetter(x)) for x in range(3))
def __new__(cls, lineno, type, value):
return tuple.__new__(cls, (lineno, intern(str(type)), value))
def __str__(self):
if self.type in reverse_operators:
return reverse_operators[self.type]
elif self.type == "name":
return self.value
return self.type
def test(self, expr):
"""Test a token against a token expression. This can either be a
token type or ``'token_type:token_value'``. This can only test
against string values and types.
"""
# here we do a regular string equality check as test_any is usually
# passed an iterable of not interned strings.
if self.type == expr:
return True
elif ":" in expr:
return expr.split(":", 1) == [self.type, self.value]
return False
def test_any(self, *iterable):
"""Test against multiple token expressions."""
for expr in iterable:
if self.test(expr):
return True
return False
def __repr__(self):
return "Token(%r, %r, %r)" % (self.lineno, self.type, self.value)
@implements_iterator
class TokenStreamIterator(object):
"""The iterator for tokenstreams. Iterate over the stream
until the eof token is reached.
"""
def __init__(self, stream):
self.stream = stream
def __iter__(self):
return self
def __next__(self):
token = self.stream.current
if token.type is TOKEN_EOF:
self.stream.close()
raise StopIteration()
next(self.stream)
return token
@implements_iterator
class TokenStream(object):
"""A token stream is an iterable that yields :class:`Token`\\s. The
parser however does not iterate over it but calls :meth:`next` to go
one token ahead. The current active token is stored as :attr:`current`.
"""
def __init__(self, generator, name, filename):
self._iter = iter(generator)
self._pushed = deque()
self.name = name
self.filename = filename
self.closed = False
self.current = Token(1, TOKEN_INITIAL, "")
next(self)
def __iter__(self):
return TokenStreamIterator(self)
def __bool__(self):
return bool(self._pushed) or self.current.type is not TOKEN_EOF
__nonzero__ = __bool__ # py2
@property
def eos(self):
"""Are we at the end of the stream?"""
return not self
def push(self, token):
"""Push a token back to the stream."""
self._pushed.append(token)
def look(self):
"""Look at the next token."""
old_token = next(self)
result = self.current
self.push(result)
self.current = old_token
return result
def skip(self, n=1):
"""Got n tokens ahead."""
for _ in range(n):
next(self)
def next_if(self, expr):
"""Perform the token test and return the token if it matched.
Otherwise the return value is `None`.
"""
if self.current.test(expr):
return next(self)
def skip_if(self, expr):
"""Like :meth:`next_if` but only returns `True` or `False`."""
return self.next_if(expr) is not None
def __next__(self):
"""Go one token ahead and return the old one.
Use the built-in :func:`next` instead of calling this directly.
"""
rv = self.current
if self._pushed:
self.current = self._pushed.popleft()
elif self.current.type is not TOKEN_EOF:
try:
self.current = next(self._iter)
except StopIteration:
self.close()
return rv
def close(self):
"""Close the stream."""
self.current = Token(self.current.lineno, TOKEN_EOF, "")
self._iter = None
self.closed = True
def expect(self, expr):
"""Expect a given token type and return it. This accepts the same
argument as :meth:`jinja2.lexer.Token.test`.
"""
if not self.current.test(expr):
expr = describe_token_expr(expr)
if self.current.type is TOKEN_EOF:
raise TemplateSyntaxError(
"unexpected end of template, expected %r." % expr,
self.current.lineno,
self.name,
self.filename,
)
raise TemplateSyntaxError(
"expected token %r, got %r" % (expr, describe_token(self.current)),
self.current.lineno,
self.name,
self.filename,
)
try:
return self.current
finally:
next(self)
def get_lexer(environment):
"""Return a lexer which is probably cached."""
key = (
environment.block_start_string,
environment.block_end_string,
environment.variable_start_string,
environment.variable_end_string,
environment.comment_start_string,
environment.comment_end_string,
environment.line_statement_prefix,
environment.line_comment_prefix,
environment.trim_blocks,
environment.lstrip_blocks,
environment.newline_sequence,
environment.keep_trailing_newline,
)
lexer = _lexer_cache.get(key)
if lexer is None:
lexer = Lexer(environment)
_lexer_cache[key] = lexer
return lexer
class OptionalLStrip(tuple):
"""A special tuple for marking a point in the state that can have
lstrip applied.
"""
__slots__ = ()
# Even though it looks like a no-op, creating instances fails
# without this.
def __new__(cls, *members, **kwargs):
return super(OptionalLStrip, cls).__new__(cls, members)
class Lexer(object):
"""Class that implements a lexer for a given environment. Automatically
created by the environment class, usually you don't have to do that.
Note that the lexer is not automatically bound to an environment.
Multiple environments can share the same lexer.
"""
def __init__(self, environment):
# shortcuts
e = re.escape
def c(x):
return re.compile(x, re.M | re.S)
# lexing rules for tags
tag_rules = [
(whitespace_re, TOKEN_WHITESPACE, None),
(float_re, TOKEN_FLOAT, None),
(integer_re, TOKEN_INTEGER, None),
(name_re, TOKEN_NAME, None),
(string_re, TOKEN_STRING, None),
(operator_re, TOKEN_OPERATOR, None),
]
# assemble the root lexing rule. because "|" is ungreedy
# we have to sort by length so that the lexer continues working
# as expected when we have parsing rules like <% for block and
# <%= for variables. (if someone wants asp like syntax)
# variables are just part of the rules if variable processing
# is required.
root_tag_rules = compile_rules(environment)
# block suffix if trimming is enabled
block_suffix_re = environment.trim_blocks and "\\n?" or ""
# If lstrip is enabled, it should not be applied if there is any
# non-whitespace between the newline and block.
self.lstrip_unless_re = c(r"[^ \t]") if environment.lstrip_blocks else None
self.newline_sequence = environment.newline_sequence
self.keep_trailing_newline = environment.keep_trailing_newline
# global lexing rules
self.rules = {
"root": [
# directives
(
c(
"(.*?)(?:%s)"
% "|".join(
[
r"(?P<raw_begin>%s(\-|\+|)\s*raw\s*(?:\-%s\s*|%s))"
% (
e(environment.block_start_string),
e(environment.block_end_string),
e(environment.block_end_string),
)
]
+ [
r"(?P<%s>%s(\-|\+|))" % (n, r)
for n, r in root_tag_rules
]
)
),
OptionalLStrip(TOKEN_DATA, "#bygroup"),
"#bygroup",
),
# data
(c(".+"), TOKEN_DATA, None),
],
# comments
TOKEN_COMMENT_BEGIN: [
(
c(
r"(.*?)((?:\-%s\s*|%s)%s)"
% (
e(environment.comment_end_string),
e(environment.comment_end_string),
block_suffix_re,
)
),
(TOKEN_COMMENT, TOKEN_COMMENT_END),
"#pop",
),
(c("(.)"), (Failure("Missing end of comment tag"),), None),
],
# blocks
TOKEN_BLOCK_BEGIN: [
(
c(
r"(?:\-%s\s*|%s)%s"
% (
e(environment.block_end_string),
e(environment.block_end_string),
block_suffix_re,
)
),
TOKEN_BLOCK_END,
"#pop",
),
]
+ tag_rules,
# variables
TOKEN_VARIABLE_BEGIN: [
(
c(
r"\-%s\s*|%s"
% (
e(environment.variable_end_string),
e(environment.variable_end_string),
)
),
TOKEN_VARIABLE_END,
"#pop",
)
]
+ tag_rules,
# raw block
TOKEN_RAW_BEGIN: [
(
c(
r"(.*?)((?:%s(\-|\+|))\s*endraw\s*(?:\-%s\s*|%s%s))"
% (
e(environment.block_start_string),
e(environment.block_end_string),
e(environment.block_end_string),
block_suffix_re,
)
),
OptionalLStrip(TOKEN_DATA, TOKEN_RAW_END),
"#pop",
),
(c("(.)"), (Failure("Missing end of raw directive"),), None),
],
# line statements
TOKEN_LINESTATEMENT_BEGIN: [
(c(r"\s*(\n|$)"), TOKEN_LINESTATEMENT_END, "#pop")
]
+ tag_rules,
# line comments
TOKEN_LINECOMMENT_BEGIN: [
(
c(r"(.*?)()(?=\n|$)"),
(TOKEN_LINECOMMENT, TOKEN_LINECOMMENT_END),
"#pop",
)
],
}
def _normalize_newlines(self, value):
"""Called for strings and template data to normalize it to unicode."""
return newline_re.sub(self.newline_sequence, value)
def tokenize(self, source, name=None, filename=None, state=None):
"""Calls tokeniter + tokenize and wraps it in a token stream."""
stream = self.tokeniter(source, name, filename, state)
return TokenStream(self.wrap(stream, name, filename), name, filename)
def wrap(self, stream, name=None, filename=None):
"""This is called with the stream as returned by `tokenize` and wraps
every token in a :class:`Token` and converts the value.
"""
for lineno, token, value in stream:
if token in ignored_tokens:
continue
elif token == TOKEN_LINESTATEMENT_BEGIN:
token = TOKEN_BLOCK_BEGIN
elif token == TOKEN_LINESTATEMENT_END:
token = TOKEN_BLOCK_END
# we are not interested in those tokens in the parser
elif token in (TOKEN_RAW_BEGIN, TOKEN_RAW_END):
continue
elif token == TOKEN_DATA:
value = self._normalize_newlines(value)
elif token == "keyword":
token = value
elif token == TOKEN_NAME:
value = str(value)
if check_ident and not value.isidentifier():
raise TemplateSyntaxError(
"Invalid character in identifier", lineno, name, filename
)
elif token == TOKEN_STRING:
# try to unescape string
try:
value = (
self._normalize_newlines(value[1:-1])
.encode("ascii", "backslashreplace")
.decode("unicode-escape")
)
except Exception as e:
msg = str(e).split(":")[-1].strip()
raise TemplateSyntaxError(msg, lineno, name, filename)
elif token == TOKEN_INTEGER:
value = int(value.replace("_", ""))
elif token == TOKEN_FLOAT:
# remove all "_" first to support more Python versions
value = literal_eval(value.replace("_", ""))
elif token == TOKEN_OPERATOR:
token = operators[value]
yield Token(lineno, token, value)
def tokeniter(self, source, name, filename=None, state=None):
"""This method tokenizes the text and returns the tokens in a
generator. Use this method if you just want to tokenize a template.
"""
source = text_type(source)
lines = source.splitlines()
if self.keep_trailing_newline and source:
for newline in ("\r\n", "\r", "\n"):
if source.endswith(newline):
lines.append("")
break
source = "\n".join(lines)
pos = 0
lineno = 1
stack = ["root"]
if state is not None and state != "root":
assert state in ("variable", "block"), "invalid state"
stack.append(state + "_begin")
statetokens = self.rules[stack[-1]]
source_length = len(source)
balancing_stack = []
lstrip_unless_re = self.lstrip_unless_re
newlines_stripped = 0
line_starting = True
while 1:
# tokenizer loop
for regex, tokens, new_state in statetokens:
m = regex.match(source, pos)
# if no match we try again with the next rule
if m is None:
continue
# we only match blocks and variables if braces / parentheses
# are balanced. continue parsing with the lower rule which
# is the operator rule. do this only if the end tags look
# like operators
if balancing_stack and tokens in (
TOKEN_VARIABLE_END,
TOKEN_BLOCK_END,
TOKEN_LINESTATEMENT_END,
):
continue
# tuples support more options
if isinstance(tokens, tuple):
groups = m.groups()
if isinstance(tokens, OptionalLStrip):
# Rule supports lstrip. Match will look like
# text, block type, whitespace control, type, control, ...
text = groups[0]
# Skipping the text and first type, every other group is the
# whitespace control for each type. One of the groups will be
# -, +, or empty string instead of None.
strip_sign = next(g for g in groups[2::2] if g is not None)
if strip_sign == "-":
# Strip all whitespace between the text and the tag.
stripped = text.rstrip()
newlines_stripped = text[len(stripped) :].count("\n")
groups = (stripped,) + groups[1:]
elif (
# Not marked for preserving whitespace.
strip_sign != "+"
# lstrip is enabled.
and lstrip_unless_re is not None
# Not a variable expression.
and not m.groupdict().get(TOKEN_VARIABLE_BEGIN)
):
# The start of text between the last newline and the tag.
l_pos = text.rfind("\n") + 1
if l_pos > 0 or line_starting:
# If there's only whitespace between the newline and the
# tag, strip it.
if not lstrip_unless_re.search(text, l_pos):
groups = (text[:l_pos],) + groups[1:]
for idx, token in enumerate(tokens):
# failure group
if token.__class__ is Failure:
raise token(lineno, filename)
# bygroup is a bit more complex, in that case we
# yield for the current token the first named
# group that matched
elif token == "#bygroup":
for key, value in iteritems(m.groupdict()):
if value is not None:
yield lineno, key, value
lineno += value.count("\n")
break
else:
raise RuntimeError(
"%r wanted to resolve "
"the token dynamically"
" but no group matched" % regex
)
# normal group
else:
data = groups[idx]
if data or token not in ignore_if_empty:
yield lineno, token, data
lineno += data.count("\n") + newlines_stripped
newlines_stripped = 0
# strings as token just are yielded as it.
else:
data = m.group()
# update brace/parentheses balance
if tokens == TOKEN_OPERATOR:
if data == "{":
balancing_stack.append("}")
elif data == "(":
balancing_stack.append(")")
elif data == "[":
balancing_stack.append("]")
elif data in ("}", ")", "]"):
if not balancing_stack:
raise TemplateSyntaxError(
"unexpected '%s'" % data, lineno, name, filename
)
expected_op = balancing_stack.pop()
if expected_op != data:
raise TemplateSyntaxError(
"unexpected '%s', "
"expected '%s'" % (data, expected_op),
lineno,
name,
filename,
)
# yield items
if data or tokens not in ignore_if_empty:
yield lineno, tokens, data
lineno += data.count("\n")
line_starting = m.group()[-1:] == "\n"
# fetch new position into new variable so that we can check
# if there is a internal parsing error which would result
# in an infinite loop
pos2 = m.end()
# handle state changes
if new_state is not None:
# remove the uppermost state
if new_state == "#pop":
stack.pop()
# resolve the new state by group checking
elif new_state == "#bygroup":
for key, value in iteritems(m.groupdict()):
if value is not None:
stack.append(key)
break
else:
raise RuntimeError(
"%r wanted to resolve the "
"new state dynamically but"
" no group matched" % regex
)
# direct state name given
else:
stack.append(new_state)
statetokens = self.rules[stack[-1]]
# we are still at the same position and no stack change.
# this means a loop without break condition, avoid that and
# raise error
elif pos2 == pos:
raise RuntimeError(
"%r yielded empty string without stack change" % regex
)
# publish new function and start again
pos = pos2
break
# if loop terminated without break we haven't found a single match
# either we are at the end of the file or we have a problem
else:
# end of text
if pos >= source_length:
return
# something went wrong
raise TemplateSyntaxError(
"unexpected char %r at %d" % (source[pos], pos),
lineno,
name,
filename,
)
# -*- coding: utf-8 -*-
"""API and implementations for loading templates from different data
sources.
"""
import os
import sys
import weakref
from hashlib import sha1
from os import path
from types import ModuleType
from ._compat import abc
from ._compat import fspath
from ._compat import iteritems
from ._compat import string_types
from .exceptions import TemplateNotFound
from .utils import internalcode
from .utils import open_if_exists
def split_template_path(template):
"""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
):
raise TemplateNotFound(template)
elif piece and piece != ".":
pieces.append(piece)
return pieces
class BaseLoader(object):
"""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
:class:`Template` object.
A very basic example for a loader that looks up templates on the file
system could look like this::
from jinja2 import BaseLoader, TemplateNotFound
from os.path import join, exists, getmtime
class MyLoader(BaseLoader):
def __init__(self, path):
self.path = path
def get_source(self, environment, template):
path = join(self.path, template)
if not exists(path):
raise TemplateNotFound(template)
mtime = getmtime(path)
with file(path) as f:
source = f.read().decode('utf-8')
return source, path, lambda: mtime == getmtime(path)
"""
#: if set to `False` it indicates that the loader cannot provide access
#: to the source of templates.
#:
#: .. versionadded:: 2.4
has_source_access = True
def get_source(self, environment, template):
"""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
if no loader extension is used.
The last item in the tuple is the `uptodate` function. If auto
reloading is enabled it's always called to check if the template
changed. No arguments are passed so the function must store the
old state somewhere (for example in a closure). If it returns `False`
the template will be reloaded.
"""
if not self.has_source_access:
raise RuntimeError(
"%s cannot provide access to the source" % self.__class__.__name__
)
raise TemplateNotFound(template)
def list_templates(self):
"""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")
@internalcode
def load(self, environment, name, globals=None):
"""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
loaders (such as :class:`PrefixLoader` or :class:`ChoiceLoader`)
will not call this method but `get_source` directly.
"""
code = None
if globals is None:
globals = {}
# first we try to get the source for this template together
# with the filename and the uptodate function.
source, filename, uptodate = self.get_source(environment, name)
# try to load the code from the bytecode cache if there is a
# bytecode cache configured.
bcc = environment.bytecode_cache
if bcc is not None:
bucket = bcc.get_bucket(environment, name, filename, source)
code = bucket.code
# if we don't have code so far (not cached, no longer up to
# date) etc. we compile the template
if code is None:
code = environment.compile(source, name, filename)
# if the bytecode cache is available and the bucket doesn't
# have a code so far, we give the bucket the new code and put
# it back to the bytecode cache.
if bcc is not None and bucket.code is None:
bucket.code = code
bcc.set_bucket(bucket)
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.
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::
>>> loader = FileSystemLoader('/path/to/templates')
>>> loader = FileSystemLoader(['/path/to/templates', '/other/path'])
Per default the template encoding is ``'utf-8'`` which can be changed
by setting the `encoding` parameter to something else.
To follow symbolic links, set the *followlinks* parameter to ``True``::
>>> loader = FileSystemLoader('/path/to/templates', followlinks=True)
.. versionchanged:: 2.8
The ``followlinks`` parameter was added.
"""
def __init__(self, searchpath, encoding="utf-8", followlinks=False):
if not isinstance(searchpath, abc.Iterable) or isinstance(
searchpath, string_types
):
searchpath = [searchpath]
# In Python 3.5, os.path.join doesn't support Path. This can be
# simplified to list(searchpath) when Python 3.5 is dropped.
self.searchpath = [fspath(p) for p in searchpath]
self.encoding = encoding
self.followlinks = followlinks
def get_source(self, environment, template):
pieces = split_template_path(template)
for searchpath in self.searchpath:
filename = path.join(searchpath, *pieces)
f = open_if_exists(filename)
if f is None:
continue
try:
contents = f.read().decode(self.encoding)
finally:
f.close()
mtime = path.getmtime(filename)
def uptodate():
try:
return path.getmtime(filename) == mtime
except OSError:
return False
return contents, filename, uptodate
raise TemplateNotFound(template)
def list_templates(self):
found = set()
for searchpath in self.searchpath:
walk_dir = os.walk(searchpath, followlinks=self.followlinks)
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 = template[2:]
if template not in found:
found.add(template)
return sorted(found)
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::
loader = PackageLoader('mypackage', 'views')
If the package path is not given, ``'templates'`` is assumed.
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.
"""
def __init__(self, package_name, package_path="templates", encoding="utf-8"):
from pkg_resources import DefaultProvider
from pkg_resources import get_provider
from pkg_resources import ResourceManager
provider = get_provider(package_name)
self.encoding = encoding
self.manager = ResourceManager()
self.filesystem_bound = isinstance(provider, DefaultProvider)
self.provider = provider
self.package_path = package_path
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)
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
source = self.provider.get_resource_string(self.manager, p)
return source.decode(self.encoding), filename, uptodate
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:
>>> loader = DictLoader({'index.html': 'source here'})
Because auto reloading is rarely useful this is disabled per default.
"""
def __init__(self, mapping):
self.mapping = mapping
def get_source(self, environment, template):
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):
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,
filename, uptodatefunc)`` or `None` if the template does not exist.
>>> def load_template(name):
... if name == 'index.html':
... return '...'
...
>>> loader = FunctionLoader(load_template)
The `uptodatefunc` is a function that is called if autoreload is enabled
and has to return `True` if the template is still up to date. For more
details have a look at :meth:`BaseLoader.get_source` which has the same
return value.
"""
def __init__(self, load_func):
self.load_func = load_func
def get_source(self, environment, template):
rv = self.load_func(template)
if rv is None:
raise TemplateNotFound(template)
elif isinstance(rv, string_types):
return rv, None, None
return rv
class PrefixLoader(BaseLoader):
"""A loader that is passed a dict of loaders where each loader is bound
to a prefix. The prefix is delimited from the template by a slash per
default, which can be changed by setting the `delimiter` argument to
something else::
loader = PrefixLoader({
'app1': PackageLoader('mypackage.app1'),
'app2': PackageLoader('mypackage.app2')
})
By loading ``'app1/index.html'`` the file from the app1 package is loaded,
by loading ``'app2/index.html'`` the file from the second.
"""
def __init__(self, mapping, delimiter="/"):
self.mapping = mapping
self.delimiter = delimiter
def get_loader(self, template):
try:
prefix, name = template.split(self.delimiter, 1)
loader = self.mapping[prefix]
except (ValueError, KeyError):
raise TemplateNotFound(template)
return loader, name
def get_source(self, environment, template):
loader, name = self.get_loader(template)
try:
return loader.get_source(environment, name)
except TemplateNotFound:
# re-raise the exception with the correct filename here.
# (the one that includes the prefix)
raise TemplateNotFound(template)
@internalcode
def load(self, environment, name, globals=None):
loader, local_name = self.get_loader(name)
try:
return loader.load(environment, local_name, globals)
except TemplateNotFound:
# re-raise the exception with the correct filename here.
# (the one that includes the prefix)
raise TemplateNotFound(name)
def list_templates(self):
result = []
for prefix, loader in iteritems(self.mapping):
for template in loader.list_templates():
result.append(prefix + self.delimiter + template)
return result
class ChoiceLoader(BaseLoader):
"""This loader works like the `PrefixLoader` just that no prefix is
specified. If a template could not be found by one loader the next one
is tried.
>>> loader = ChoiceLoader([
... FileSystemLoader('/path/to/user/templates'),
... FileSystemLoader('/path/to/system/templates')
... ])
This is useful if you want to allow users to override builtin templates
from a different location.
"""
def __init__(self, loaders):
self.loaders = loaders
def get_source(self, environment, template):
for loader in self.loaders:
try:
return loader.get_source(environment, template)
except TemplateNotFound:
pass
raise TemplateNotFound(template)
@internalcode
def load(self, environment, name, globals=None):
for loader in self.loaders:
try:
return loader.load(environment, name, globals)
except TemplateNotFound:
pass
raise TemplateNotFound(name)
def list_templates(self):
found = set()
for loader in self.loaders:
found.update(loader.list_templates())
return sorted(found)
class _TemplateModule(ModuleType):
"""Like a normal module but with support for weak references"""
class ModuleLoader(BaseLoader):
"""This loader loads templates from precompiled templates.
Example usage:
>>> loader = ChoiceLoader([
... ModuleLoader('/path/to/compiled/templates'),
... FileSystemLoader('/path/to/templates')
... ])
Templates can be precompiled with :meth:`Environment.compile_templates`.
"""
has_source_access = False
def __init__(self, path):
package_name = "_jinja2_module_templates_%x" % id(self)
# create a fake module that looks for the templates in the
# path given.
mod = _TemplateModule(package_name)
if not isinstance(path, abc.Iterable) or isinstance(path, string_types):
path = [path]
mod.__path__ = [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
# loader that created it goes out of business.
self.module = mod
self.package_name = package_name
@staticmethod
def get_template_key(name):
return "tmpl_" + sha1(name.encode("utf-8")).hexdigest()
@staticmethod
def get_module_filename(name):
return ModuleLoader.get_template_key(name) + ".py"
@internalcode
def load(self, environment, name, globals=None):
key = self.get_template_key(name)
module = "%s.%s" % (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)
# 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)
return environment.template_class.from_module_dict(
environment, mod.__dict__, globals
)
# -*- coding: utf-8 -*-
"""Functions that expose information about templates that might be
interesting for introspection.
"""
from . import nodes
from ._compat import iteritems
from ._compat import string_types
from .compiler import CodeGenerator
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 write(self, x):
"""Don't write."""
def enter_frame(self, frame):
"""Remember all undeclared identifiers."""
CodeGenerator.enter_frame(self, frame)
for _, (action, param) in iteritems(frame.symbols.loads):
if action == "resolve" and param not in self.environment.globals:
self.undeclared_identifiers.add(param)
def find_undeclared_variables(ast):
"""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
runtime, all variables are returned.
>>> from jinja2 import Environment, meta
>>> env = Environment()
>>> ast = env.parse('{% set foo = 42 %}{{ bar + foo }}')
>>> meta.find_undeclared_variables(ast) == set(['bar'])
True
.. admonition:: Implementation
Internally the code generator is used for finding undeclared variables.
This is good to know because the code generator might raise a
:exc:`TemplateAssertionError` during compilation and as a matter of
fact this function can currently raise that exception as well.
"""
codegen = TrackingCodeGenerator(ast.environment)
codegen.visit(ast)
return codegen.undeclared_identifiers
def find_referenced_templates(ast):
"""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
yielded.
>>> from jinja2 import Environment, meta
>>> env = Environment()
>>> ast = env.parse('{% extends "layout.html" %}{% include helper %}')
>>> list(meta.find_referenced_templates(ast))
['layout.html', None]
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):
# a tuple with some non consts in there
if isinstance(node.template, (nodes.Tuple, nodes.List)):
for template_name in node.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):
yield template_name.value
# something dynamic in there
else:
yield None
# something dynamic we don't know about here
else:
yield None
continue
# constant is a basestring, direct template name
if isinstance(node.template.value, string_types):
yield node.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):
yield template_name
# something else we don't care about, we could warn here
else:
yield None
from ast import literal_eval
from itertools import chain
from itertools import islice
from . import nodes
from ._compat import text_type
from .compiler import CodeGenerator
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.
:param nodes: Iterable of nodes to concatenate.
"""
head = list(islice(nodes, 2))
if not head:
return None
if len(head) == 1:
raw = head[0]
else:
raw = u"".join([text_type(v) for v in chain(head, nodes)])
try:
return literal_eval(raw)
except (ValueError, SyntaxError, MemoryError):
return raw
class NativeCodeGenerator(CodeGenerator):
"""A code generator which renders Python types by not adding
``to_string()`` around output nodes.
"""
@staticmethod
def _default_finalize(value):
return value
def _output_const_repr(self, group):
return repr(u"".join([text_type(v) for v in group]))
def _output_child_to_const(self, node, frame, finalize):
const = node.as_const(frame.eval_ctx)
if not has_safe_repr(const):
raise nodes.Impossible()
if isinstance(node, nodes.TemplateData):
return const
return finalize.const(const)
def _output_child_pre(self, node, frame, finalize):
if finalize.src is not None:
self.write(finalize.src)
def _output_child_post(self, node, frame, finalize):
if finalize.src is not None:
self.write(")")
class NativeEnvironment(Environment):
"""An environment that renders templates to native Python types."""
code_generator_class = NativeCodeGenerator
class NativeTemplate(Template):
environment_class = NativeEnvironment
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:
return self.environment.handle_exception()
NativeEnvironment.template_class = NativeTemplate
# -*- coding: utf-8 -*-
"""AST nodes generated by the parser for the compiler. Also provides
some node tree helper functions used by the parser and compiler in order
to normalize nodes.
"""
import operator
from collections import deque
from markupsafe import Markup
from ._compat import izip
from ._compat import PY2
from ._compat import text_type
from ._compat import with_metaclass
_binop_to_func = {
"*": operator.mul,
"/": operator.truediv,
"//": operator.floordiv,
"**": operator.pow,
"%": operator.mod,
"+": operator.add,
"-": operator.sub,
}
_uaop_to_func = {"not": operator.not_, "+": operator.pos, "-": operator.neg}
_cmpop_to_func = {
"eq": operator.eq,
"ne": operator.ne,
"gt": operator.gt,
"gteq": operator.ge,
"lt": operator.lt,
"lteq": operator.le,
"in": lambda a, b: a in b,
"notin": lambda a, b: a not in b,
}
class Impossible(Exception):
"""Raised if the node could not perform a requested action."""
class NodeType(type):
"""A metaclass for nodes that handles the field and attribute
inheritance. fields and attributes from the parent class are
automatically forwarded to the child."""
def __new__(mcs, name, bases, d):
for attr in "fields", "attributes":
storage = []
storage.extend(getattr(bases[0], attr, ()))
storage.extend(d.get(attr, ()))
assert len(bases) == 1, "multiple inheritance not allowed"
assert len(storage) == len(set(storage)), "layout conflict"
d[attr] = tuple(storage)
d.setdefault("abstract", False)
return type.__new__(mcs, name, bases, d)
class EvalContext(object):
"""Holds evaluation time information. Custom attributes can be attached
to it in extensions.
"""
def __init__(self, environment, template_name=None):
self.environment = environment
if callable(environment.autoescape):
self.autoescape = environment.autoescape(template_name)
else:
self.autoescape = environment.autoescape
self.volatile = False
def save(self):
return self.__dict__.copy()
def revert(self, old):
self.__dict__.clear()
self.__dict__.update(old)
def get_eval_context(node, ctx):
if ctx is None:
if node.environment is None:
raise RuntimeError(
"if no eval context is passed, the "
"node must have an attached "
"environment."
)
return EvalContext(node.environment)
return ctx
class Node(with_metaclass(NodeType, object)):
"""Baseclass for all Jinja nodes. There are a number of nodes available
of different types. There are four major types:
- :class:`Stmt`: statements
- :class:`Expr`: expressions
- :class:`Helper`: helper nodes
- :class:`Template`: the outermost wrapper node
All nodes have fields and attributes. Fields may be other nodes, lists,
or arbitrary values. Fields are passed to the constructor as regular
positional arguments, attributes as keyword arguments. Each node has
two attributes: `lineno` (the line number of the node) and `environment`.
The `environment` attribute is set at the end of the parsing process for
all nodes automatically.
"""
fields = ()
attributes = ("lineno", "environment")
abstract = True
def __init__(self, *fields, **attributes):
if self.abstract:
raise TypeError("abstract nodes are not instantiable")
if fields:
if len(fields) != len(self.fields):
if not self.fields:
raise TypeError("%r takes 0 arguments" % self.__class__.__name__)
raise TypeError(
"%r takes 0 or %d argument%s"
% (
self.__class__.__name__,
len(self.fields),
len(self.fields) != 1 and "s" or "",
)
)
for name, arg in izip(self.fields, fields):
setattr(self, name, arg)
for attr in self.attributes:
setattr(self, attr, attributes.pop(attr, None))
if attributes:
raise TypeError("unknown attribute %r" % next(iter(attributes)))
def iter_fields(self, exclude=None, only=None):
"""This method iterates over all fields that are defined and yields
``(key, value)`` tuples. Per default all fields are returned, but
it's possible to limit that to some fields by providing the `only`
parameter or to exclude some using the `exclude` parameter. Both
should be sets or tuples of field names.
"""
for name in self.fields:
if (
(exclude is only is None)
or (exclude is not None and name not in exclude)
or (only is not None and name in only)
):
try:
yield name, getattr(self, name)
except AttributeError:
pass
def iter_child_nodes(self, exclude=None, only=None):
"""Iterates over all direct child nodes of the node. This iterates
over all fields and yields the values of they are nodes. If the value
of a field is a list all the nodes in that list are returned.
"""
for _, item in self.iter_fields(exclude, only):
if isinstance(item, list):
for n in item:
if isinstance(n, Node):
yield n
elif isinstance(item, Node):
yield item
def find(self, node_type):
"""Find the first node of a given type. If no such node exists the
return value is `None`.
"""
for result in self.find_all(node_type):
return result
def find_all(self, node_type):
"""Find all the nodes of a given type. If the type is a tuple,
the check is performed for any of the tuple items.
"""
for child in self.iter_child_nodes():
if isinstance(child, node_type):
yield child
for result in child.find_all(node_type):
yield result
def set_ctx(self, ctx):
"""Reset the context of a node and all child nodes. Per default the
parser will all generate nodes that have a 'load' context as it's the
most common one. This method is used in the parser to set assignment
targets and other nodes to a store context.
"""
todo = deque([self])
while todo:
node = todo.popleft()
if "ctx" in node.fields:
node.ctx = ctx
todo.extend(node.iter_child_nodes())
return self
def set_lineno(self, lineno, override=False):
"""Set the line numbers of the node and children."""
todo = deque([self])
while todo:
node = todo.popleft()
if "lineno" in node.attributes:
if node.lineno is None or override:
node.lineno = lineno
todo.extend(node.iter_child_nodes())
return self
def set_environment(self, environment):
"""Set the environment for all nodes."""
todo = deque([self])
while todo:
node = todo.popleft()
node.environment = environment
todo.extend(node.iter_child_nodes())
return self
def __eq__(self, other):
return type(self) is type(other) and tuple(self.iter_fields()) == tuple(
other.iter_fields()
)
def __ne__(self, other):
return not self.__eq__(other)
# Restore Python 2 hashing behavior on Python 3
__hash__ = object.__hash__
def __repr__(self):
return "%s(%s)" % (
self.__class__.__name__,
", ".join("%s=%r" % (arg, getattr(self, arg, None)) for arg in self.fields),
)
def dump(self):
def _dump(node):
if not isinstance(node, Node):
buf.append(repr(node))
return
buf.append("nodes.%s(" % node.__class__.__name__)
if not node.fields:
buf.append(")")
return
for idx, field in enumerate(node.fields):
if idx:
buf.append(", ")
value = getattr(node, field)
if isinstance(value, list):
buf.append("[")
for idx, item in enumerate(value):
if idx:
buf.append(", ")
_dump(item)
buf.append("]")
else:
_dump(value)
buf.append(")")
buf = []
_dump(self)
return "".join(buf)
class Stmt(Node):
"""Base node for all statements."""
abstract = True
class Helper(Node):
"""Nodes that exist in a specific context only."""
abstract = True
class Template(Node):
"""Node that represents a template. This must be the outermost node that
is passed to the compiler.
"""
fields = ("body",)
class Output(Stmt):
"""A node that holds multiple expressions which are then printed out.
This is used both for the `print` statement and the regular template data.
"""
fields = ("nodes",)
class Extends(Stmt):
"""Represents an extends statement."""
fields = ("template",)
class For(Stmt):
"""The for loop. `target` is the target for the iteration (usually a
:class:`Name` or :class:`Tuple`), `iter` the iterable. `body` is a list
of nodes that are used as loop-body, and `else_` a list of nodes for the
`else` block. If no else node exists it has to be an empty list.
For filtered nodes an expression can be stored as `test`, otherwise `None`.
"""
fields = ("target", "iter", "body", "else_", "test", "recursive")
class If(Stmt):
"""If `test` is true, `body` is rendered, else `else_`."""
fields = ("test", "body", "elif_", "else_")
class Macro(Stmt):
"""A macro definition. `name` is the name of the macro, `args` a list of
arguments and `defaults` a list of defaults if there are any. `body` is
a list of nodes for the macro body.
"""
fields = ("name", "args", "defaults", "body")
class CallBlock(Stmt):
"""Like a macro without a name but a call instead. `call` is called with
the unnamed macro as `caller` argument this node holds.
"""
fields = ("call", "args", "defaults", "body")
class FilterBlock(Stmt):
"""Node for filter sections."""
fields = ("body", "filter")
class With(Stmt):
"""Specific node for with statements. In older versions of Jinja the
with statement was implemented on the base of the `Scope` node instead.
.. versionadded:: 2.9.3
"""
fields = ("targets", "values", "body")
class Block(Stmt):
"""A node that represents a block."""
fields = ("name", "body", "scoped")
class Include(Stmt):
"""A node that represents the include tag."""
fields = ("template", "with_context", "ignore_missing")
class Import(Stmt):
"""A node that represents the import tag."""
fields = ("template", "target", "with_context")
class FromImport(Stmt):
"""A node that represents the from import tag. It's important to not
pass unsafe names to the name attribute. The compiler translates the
attribute lookups directly into getattr calls and does *not* use the
subscript callback of the interface. As exported variables may not
start with double underscores (which the parser asserts) this is not a
problem for regular Jinja code, but if this node is used in an extension
extra care must be taken.
The list of names may contain tuples if aliases are wanted.
"""
fields = ("template", "names", "with_context")
class ExprStmt(Stmt):
"""A statement that evaluates an expression and discards the result."""
fields = ("node",)
class Assign(Stmt):
"""Assigns an expression to a target."""
fields = ("target", "node")
class AssignBlock(Stmt):
"""Assigns a block to a target."""
fields = ("target", "filter", "body")
class Expr(Node):
"""Baseclass for all expressions."""
abstract = True
def as_const(self, eval_ctx=None):
"""Return the value of the expression as constant or raise
:exc:`Impossible` if this was not possible.
An :class:`EvalContext` can be provided, if none is given
a default context is created which requires the nodes to have
an attached environment.
.. versionchanged:: 2.4
the `eval_ctx` parameter was added.
"""
raise Impossible()
def can_assign(self):
"""Check if it's possible to assign something to this node."""
return False
class BinExpr(Expr):
"""Baseclass for all binary expressions."""
fields = ("left", "right")
operator = None
abstract = True
def as_const(self, eval_ctx=None):
eval_ctx = get_eval_context(self, eval_ctx)
# intercepted operators cannot be folded at compile time
if (
self.environment.sandboxed
and self.operator in self.environment.intercepted_binops
):
raise Impossible()
f = _binop_to_func[self.operator]
try:
return f(self.left.as_const(eval_ctx), self.right.as_const(eval_ctx))
except Exception:
raise Impossible()
class UnaryExpr(Expr):
"""Baseclass for all unary expressions."""
fields = ("node",)
operator = None
abstract = True
def as_const(self, eval_ctx=None):
eval_ctx = get_eval_context(self, eval_ctx)
# intercepted operators cannot be folded at compile time
if (
self.environment.sandboxed
and self.operator in self.environment.intercepted_unops
):
raise Impossible()
f = _uaop_to_func[self.operator]
try:
return f(self.node.as_const(eval_ctx))
except Exception:
raise Impossible()
class Name(Expr):
"""Looks up a name or stores a value in a name.
The `ctx` of the node can be one of the following values:
- `store`: store a value in the name
- `load`: load that name
- `param`: like `store` but if the name was defined as function parameter.
"""
fields = ("name", "ctx")
def can_assign(self):
return self.name not in ("true", "false", "none", "True", "False", "None")
class NSRef(Expr):
"""Reference to a namespace value assignment"""
fields = ("name", "attr")
def can_assign(self):
# We don't need any special checks here; NSRef assignments have a
# runtime check to ensure the target is a namespace object which will
# have been checked already as it is created using a normal assignment
# which goes through a `Name` node.
return True
class Literal(Expr):
"""Baseclass for literals."""
abstract = True
class Const(Literal):
"""All constant values. The parser will return this node for simple
constants such as ``42`` or ``"foo"`` but it can be used to store more
complex values such as lists too. Only constants with a safe
representation (objects where ``eval(repr(x)) == x`` is true).
"""
fields = ("value",)
def as_const(self, eval_ctx=None):
rv = self.value
if (
PY2
and type(rv) is text_type
and self.environment.policies["compiler.ascii_str"]
):
try:
rv = rv.encode("ascii")
except UnicodeError:
pass
return rv
@classmethod
def from_untrusted(cls, value, lineno=None, environment=None):
"""Return a const object if the value is representable as
constant value in the generated code, otherwise it will raise
an `Impossible` exception.
"""
from .compiler import has_safe_repr
if not has_safe_repr(value):
raise Impossible()
return cls(value, lineno=lineno, environment=environment)
class TemplateData(Literal):
"""A constant template string."""
fields = ("data",)
def as_const(self, eval_ctx=None):
eval_ctx = get_eval_context(self, eval_ctx)
if eval_ctx.volatile:
raise Impossible()
if eval_ctx.autoescape:
return Markup(self.data)
return self.data
class Tuple(Literal):
"""For loop unpacking and some other things like multiple arguments
for subscripts. Like for :class:`Name` `ctx` specifies if the tuple
is used for loading the names or storing.
"""
fields = ("items", "ctx")
def as_const(self, eval_ctx=None):
eval_ctx = get_eval_context(self, eval_ctx)
return tuple(x.as_const(eval_ctx) for x in self.items)
def can_assign(self):
for item in self.items:
if not item.can_assign():
return False
return True
class List(Literal):
"""Any list literal such as ``[1, 2, 3]``"""
fields = ("items",)
def as_const(self, eval_ctx=None):
eval_ctx = get_eval_context(self, eval_ctx)
return [x.as_const(eval_ctx) for x in self.items]
class Dict(Literal):
"""Any dict literal such as ``{1: 2, 3: 4}``. The items must be a list of
:class:`Pair` nodes.
"""
fields = ("items",)
def as_const(self, eval_ctx=None):
eval_ctx = get_eval_context(self, eval_ctx)
return dict(x.as_const(eval_ctx) for x in self.items)
class Pair(Helper):
"""A key, value pair for dicts."""
fields = ("key", "value")
def as_const(self, eval_ctx=None):
eval_ctx = get_eval_context(self, eval_ctx)
return self.key.as_const(eval_ctx), self.value.as_const(eval_ctx)
class Keyword(Helper):
"""A key, value pair for keyword arguments where key is a string."""
fields = ("key", "value")
def as_const(self, eval_ctx=None):
eval_ctx = get_eval_context(self, eval_ctx)
return self.key, self.value.as_const(eval_ctx)
class CondExpr(Expr):
"""A conditional expression (inline if expression). (``{{
foo if bar else baz }}``)
"""
fields = ("test", "expr1", "expr2")
def as_const(self, eval_ctx=None):
eval_ctx = get_eval_context(self, eval_ctx)
if self.test.as_const(eval_ctx):
return self.expr1.as_const(eval_ctx)
# if we evaluate to an undefined object, we better do that at runtime
if self.expr2 is None:
raise Impossible()
return self.expr2.as_const(eval_ctx)
def args_as_const(node, eval_ctx):
args = [x.as_const(eval_ctx) for x in node.args]
kwargs = dict(x.as_const(eval_ctx) for x in node.kwargs)
if node.dyn_args is not None:
try:
args.extend(node.dyn_args.as_const(eval_ctx))
except Exception:
raise Impossible()
if node.dyn_kwargs is not None:
try:
kwargs.update(node.dyn_kwargs.as_const(eval_ctx))
except Exception:
raise Impossible()
return args, kwargs
class Filter(Expr):
"""This node applies a filter on an expression. `name` is the name of
the filter, the rest of the fields are the same as for :class:`Call`.
If the `node` of a filter is `None` the contents of the last buffer are
filtered. Buffers are created by macros and filter blocks.
"""
fields = ("node", "name", "args", "kwargs", "dyn_args", "dyn_kwargs")
def as_const(self, eval_ctx=None):
eval_ctx = get_eval_context(self, eval_ctx)
if eval_ctx.volatile or self.node is None:
raise Impossible()
# we have to be careful here because we call filter_ below.
# if this variable would be called filter, 2to3 would wrap the
# call in a list because it is assuming we are talking about the
# builtin filter function here which no longer returns a list in
# python 3. because of that, do not rename filter_ to filter!
filter_ = self.environment.filters.get(self.name)
if filter_ is None or getattr(filter_, "contextfilter", False) is True:
raise Impossible()
# We cannot constant handle async filters, so we need to make sure
# to not go down this path.
if eval_ctx.environment.is_async and getattr(
filter_, "asyncfiltervariant", False
):
raise Impossible()
args, kwargs = args_as_const(self, eval_ctx)
args.insert(0, self.node.as_const(eval_ctx))
if getattr(filter_, "evalcontextfilter", False) is True:
args.insert(0, eval_ctx)
elif getattr(filter_, "environmentfilter", False) is True:
args.insert(0, self.environment)
try:
return filter_(*args, **kwargs)
except Exception:
raise Impossible()
class Test(Expr):
"""Applies a test on an expression. `name` is the name of the test, the
rest of the fields are the same as for :class:`Call`.
"""
fields = ("node", "name", "args", "kwargs", "dyn_args", "dyn_kwargs")
def as_const(self, eval_ctx=None):
test = self.environment.tests.get(self.name)
if test is None:
raise Impossible()
eval_ctx = get_eval_context(self, eval_ctx)
args, kwargs = args_as_const(self, eval_ctx)
args.insert(0, self.node.as_const(eval_ctx))
try:
return test(*args, **kwargs)
except Exception:
raise Impossible()
class Call(Expr):
"""Calls an expression. `args` is a list of arguments, `kwargs` a list
of keyword arguments (list of :class:`Keyword` nodes), and `dyn_args`
and `dyn_kwargs` has to be either `None` or a node that is used as
node for dynamic positional (``*args``) or keyword (``**kwargs``)
arguments.
"""
fields = ("node", "args", "kwargs", "dyn_args", "dyn_kwargs")
class Getitem(Expr):
"""Get an attribute or item from an expression and prefer the item."""
fields = ("node", "arg", "ctx")
def as_const(self, eval_ctx=None):
eval_ctx = get_eval_context(self, eval_ctx)
if self.ctx != "load":
raise Impossible()
try:
return self.environment.getitem(
self.node.as_const(eval_ctx), self.arg.as_const(eval_ctx)
)
except Exception:
raise Impossible()
def can_assign(self):
return False
class Getattr(Expr):
"""Get an attribute or item from an expression that is a ascii-only
bytestring and prefer the attribute.
"""
fields = ("node", "attr", "ctx")
def as_const(self, eval_ctx=None):
if self.ctx != "load":
raise Impossible()
try:
eval_ctx = get_eval_context(self, eval_ctx)
return self.environment.getattr(self.node.as_const(eval_ctx), self.attr)
except Exception:
raise Impossible()
def can_assign(self):
return False
class Slice(Expr):
"""Represents a slice object. This must only be used as argument for
:class:`Subscript`.
"""
fields = ("start", "stop", "step")
def as_const(self, eval_ctx=None):
eval_ctx = get_eval_context(self, eval_ctx)
def const(obj):
if obj is None:
return None
return obj.as_const(eval_ctx)
return slice(const(self.start), const(self.stop), const(self.step))
class Concat(Expr):
"""Concatenates the list of expressions provided after converting them to
unicode.
"""
fields = ("nodes",)
def as_const(self, eval_ctx=None):
eval_ctx = get_eval_context(self, eval_ctx)
return "".join(text_type(x.as_const(eval_ctx)) for x in self.nodes)
class Compare(Expr):
"""Compares an expression with some other expressions. `ops` must be a
list of :class:`Operand`\\s.
"""
fields = ("expr", "ops")
def as_const(self, eval_ctx=None):
eval_ctx = get_eval_context(self, eval_ctx)
result = value = self.expr.as_const(eval_ctx)
try:
for op in self.ops:
new_value = op.expr.as_const(eval_ctx)
result = _cmpop_to_func[op.op](value, new_value)
if not result:
return False
value = new_value
except Exception:
raise Impossible()
return result
class Operand(Helper):
"""Holds an operator and an expression."""
fields = ("op", "expr")
if __debug__:
Operand.__doc__ += "\nThe following operators are available: " + ", ".join(
sorted(
"``%s``" % x
for x in set(_binop_to_func) | set(_uaop_to_func) | set(_cmpop_to_func)
)
)
class Mul(BinExpr):
"""Multiplies the left with the right node."""
operator = "*"
class Div(BinExpr):
"""Divides the left by the right node."""
operator = "/"
class FloorDiv(BinExpr):
"""Divides the left by the right node and truncates conver the
result into an integer by truncating.
"""
operator = "//"
class Add(BinExpr):
"""Add the left to the right node."""
operator = "+"
class Sub(BinExpr):
"""Subtract the right from the left node."""
operator = "-"
class Mod(BinExpr):
"""Left modulo right."""
operator = "%"
class Pow(BinExpr):
"""Left to the power of right."""
operator = "**"
class And(BinExpr):
"""Short circuited AND."""
operator = "and"
def as_const(self, eval_ctx=None):
eval_ctx = get_eval_context(self, eval_ctx)
return self.left.as_const(eval_ctx) and self.right.as_const(eval_ctx)
class Or(BinExpr):
"""Short circuited OR."""
operator = "or"
def as_const(self, eval_ctx=None):
eval_ctx = get_eval_context(self, eval_ctx)
return self.left.as_const(eval_ctx) or self.right.as_const(eval_ctx)
class Not(UnaryExpr):
"""Negate the expression."""
operator = "not"
class Neg(UnaryExpr):
"""Make the expression negative."""
operator = "-"
class Pos(UnaryExpr):
"""Make the expression positive (noop for most expressions)"""
operator = "+"
# Helpers for extensions
class EnvironmentAttribute(Expr):
"""Loads an attribute from the environment object. This is useful for
extensions that want to call a callback stored on the environment.
"""
fields = ("name",)
class ExtensionAttribute(Expr):
"""Returns the attribute of an extension bound to the environment.
The identifier is the identifier of the :class:`Extension`.
This node is usually constructed by calling the
:meth:`~jinja2.ext.Extension.attr` method on an extension.
"""
fields = ("identifier", "name")
class ImportedName(Expr):
"""If created with an import name the import name is returned on node
access. For example ``ImportedName('cgi.escape')`` returns the `escape`
function from the cgi module on evaluation. Imports are optimized by the
compiler so there is no need to assign them to local variables.
"""
fields = ("importname",)
class InternalName(Expr):
"""An internal name in the compiler. You cannot create these nodes
yourself but the parser provides a
:meth:`~jinja2.parser.Parser.free_identifier` method that creates
a new identifier for you. This identifier is not available from the
template and is not threated specially by the compiler.
"""
fields = ("name",)
def __init__(self):
raise TypeError(
"Can't create internal names. Use the "
"`free_identifier` method on a parser."
)
class MarkSafe(Expr):
"""Mark the wrapped expression as safe (wrap it as `Markup`)."""
fields = ("expr",)
def as_const(self, eval_ctx=None):
eval_ctx = get_eval_context(self, eval_ctx)
return Markup(self.expr.as_const(eval_ctx))
class MarkSafeIfAutoescape(Expr):
"""Mark the wrapped expression as safe (wrap it as `Markup`) but
only if autoescaping is active.
.. versionadded:: 2.5
"""
fields = ("expr",)
def as_const(self, eval_ctx=None):
eval_ctx = get_eval_context(self, eval_ctx)
if eval_ctx.volatile:
raise Impossible()
expr = self.expr.as_const(eval_ctx)
if eval_ctx.autoescape:
return Markup(expr)
return expr
class ContextReference(Expr):
"""Returns the current template context. It can be used like a
:class:`Name` node, with a ``'load'`` ctx and will return the
current :class:`~jinja2.runtime.Context` object.
Here an example that assigns the current template name to a
variable named `foo`::
Assign(Name('foo', ctx='store'),
Getattr(ContextReference(), 'name'))
This is basically equivalent to using the
:func:`~jinja2.contextfunction` decorator when using the
high-level API, which causes a reference to the context to be passed
as the first argument to a function.
"""
class DerivedContextReference(Expr):
"""Return the current template context including locals. Behaves
exactly like :class:`ContextReference`, but includes local
variables, such as from a ``for`` loop.
.. versionadded:: 2.11
"""
class Continue(Stmt):
"""Continue a loop."""
class Break(Stmt):
"""Break a loop."""
class Scope(Stmt):
"""An artificial scope."""
fields = ("body",)
class OverlayScope(Stmt):
"""An overlay scope for extensions. This is a largely unoptimized scope
that however can be used to introduce completely arbitrary variables into
a sub scope from a dictionary or dictionary like object. The `context`
field has to evaluate to a dictionary object.
Example usage::
OverlayScope(context=self.call_method('get_context'),
body=[...])
.. versionadded:: 2.10
"""
fields = ("context", "body")
class EvalContextModifier(Stmt):
"""Modifies the eval context. For each option that should be modified,
a :class:`Keyword` has to be added to the :attr:`options` list.
Example to change the `autoescape` setting::
EvalContextModifier(options=[Keyword('autoescape', Const(True))])
"""
fields = ("options",)
class ScopedEvalContextModifier(EvalContextModifier):
"""Modifies the eval context and reverts it later. Works exactly like
:class:`EvalContextModifier` but will only modify the
:class:`~jinja2.nodes.EvalContext` for nodes in the :attr:`body`.
"""
fields = ("body",)
# make sure nobody creates custom nodes
def _failing_new(*args, **kwargs):
raise TypeError("can't create custom node types")
NodeType.__new__ = staticmethod(_failing_new)
del _failing_new
# -*- 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.
"""
from . import nodes
from .visitor import NodeTransformer
def optimize(node, environment):
"""The context hint can be used to perform an static optimization
based on the context given."""
optimizer = Optimizer(environment)
return optimizer.visit(node)
class Optimizer(NodeTransformer):
def __init__(self, environment):
self.environment = environment
def generic_visit(self, node, *args, **kwargs):
node = super(Optimizer, self).generic_visit(node, *args, **kwargs)
# 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
# -*- coding: utf-8 -*-
"""Parse tokens from the lexer into nodes for the compiler."""
from . import nodes
from ._compat import imap
from .exceptions import TemplateAssertionError
from .exceptions import TemplateSyntaxError
from .lexer import describe_token
from .lexer import describe_token_expr
_statement_keywords = frozenset(
[
"for",
"if",
"block",
"extends",
"print",
"macro",
"include",
"from",
"import",
"set",
"with",
"autoescape",
]
)
_compare_operators = frozenset(["eq", "ne", "lt", "lteq", "gt", "gteq"])
_math_nodes = {
"add": nodes.Add,
"sub": nodes.Sub,
"mul": nodes.Mul,
"div": nodes.Div,
"floordiv": nodes.FloorDiv,
"mod": nodes.Mod,
}
class Parser(object):
"""This is the central parsing class Jinja uses. It's passed to
extensions and can be used to parse expressions or statements.
"""
def __init__(self, environment, source, name=None, filename=None, state=None):
self.environment = environment
self.stream = environment._tokenize(source, name, filename, state)
self.name = name
self.filename = filename
self.closed = False
self.extensions = {}
for extension in environment.iter_extensions():
for tag in extension.tags:
self.extensions[tag] = extension.parse
self._last_identifier = 0
self._tag_stack = []
self._end_token_stack = []
def fail(self, msg, lineno=None, exc=TemplateSyntaxError):
"""Convenience method that raises `exc` with the message, passed
line number or last line number as well as the current name and
filename.
"""
if lineno is None:
lineno = self.stream.current.lineno
raise exc(msg, lineno, self.name, self.filename)
def _fail_ut_eof(self, name, end_token_stack, lineno):
expected = []
for exprs in end_token_stack:
expected.extend(imap(describe_token_expr, exprs))
if end_token_stack:
currently_looking = " or ".join(
"'%s'" % describe_token_expr(expr) for expr in end_token_stack[-1]
)
else:
currently_looking = None
if name is None:
message = ["Unexpected end of template."]
else:
message = ["Encountered unknown tag '%s'." % name]
if currently_looking:
if name is not None and name in expected:
message.append(
"You probably made a nesting mistake. Jinja "
"is expecting this tag, but currently looking "
"for %s." % currently_looking
)
else:
message.append(
"Jinja was looking for the following tags: "
"%s." % currently_looking
)
if self._tag_stack:
message.append(
"The innermost block that needs to be "
"closed is '%s'." % self._tag_stack[-1]
)
self.fail(" ".join(message), lineno)
def fail_unknown_tag(self, name, lineno=None):
"""Called if the parser encounters an unknown tag. Tries to fail
with a human readable error message that could help to identify
the problem.
"""
return self._fail_ut_eof(name, self._end_token_stack, lineno)
def fail_eof(self, end_tokens=None, lineno=None):
"""Like fail_unknown_tag but for end of template situations."""
stack = list(self._end_token_stack)
if end_tokens is not None:
stack.append(end_tokens)
return self._fail_ut_eof(None, stack, lineno)
def is_tuple_end(self, extra_end_rules=None):
"""Are we at the end of a tuple?"""
if self.stream.current.type in ("variable_end", "block_end", "rparen"):
return True
elif extra_end_rules is not None:
return self.stream.current.test_any(extra_end_rules)
return False
def free_identifier(self, lineno=None):
"""Return a new free identifier as :class:`~jinja2.nodes.InternalName`."""
self._last_identifier += 1
rv = object.__new__(nodes.InternalName)
nodes.Node.__init__(rv, "fi%d" % self._last_identifier, lineno=lineno)
return rv
def parse_statement(self):
"""Parse a single statement."""
token = self.stream.current
if token.type != "name":
self.fail("tag name expected", token.lineno)
self._tag_stack.append(token.value)
pop_tag = True
try:
if token.value in _statement_keywords:
return getattr(self, "parse_" + self.stream.current.value)()
if token.value == "call":
return self.parse_call_block()
if token.value == "filter":
return self.parse_filter_block()
ext = self.extensions.get(token.value)
if ext is not None:
return ext(self)
# did not work out, remove the token we pushed by accident
# from the stack so that the unknown tag fail function can
# produce a proper error message.
self._tag_stack.pop()
pop_tag = False
self.fail_unknown_tag(token.value, token.lineno)
finally:
if pop_tag:
self._tag_stack.pop()
def parse_statements(self, end_tokens, drop_needle=False):
"""Parse multiple statements into a list until one of the end tokens
is reached. This is used to parse the body of statements as it also
parses template data if appropriate. The parser checks first if the
current token is a colon and skips it if there is one. Then it checks
for the block end and parses until if one of the `end_tokens` is
reached. Per default the active token in the stream at the end of
the call is the matched end token. If this is not wanted `drop_needle`
can be set to `True` and the end token is removed.
"""
# the first token may be a colon for python compatibility
self.stream.skip_if("colon")
# in the future it would be possible to add whole code sections
# by adding some sort of end of statement token and parsing those here.
self.stream.expect("block_end")
result = self.subparse(end_tokens)
# we reached the end of the template too early, the subparser
# does not check for this, so we do that now
if self.stream.current.type == "eof":
self.fail_eof(end_tokens)
if drop_needle:
next(self.stream)
return result
def parse_set(self):
"""Parse an assign statement."""
lineno = next(self.stream).lineno
target = self.parse_assign_target(with_namespace=True)
if self.stream.skip_if("assign"):
expr = self.parse_tuple()
return nodes.Assign(target, expr, lineno=lineno)
filter_node = self.parse_filter(None)
body = self.parse_statements(("name:endset",), drop_needle=True)
return nodes.AssignBlock(target, filter_node, body, lineno=lineno)
def parse_for(self):
"""Parse a for loop."""
lineno = self.stream.expect("name:for").lineno
target = self.parse_assign_target(extra_end_rules=("name:in",))
self.stream.expect("name:in")
iter = self.parse_tuple(
with_condexpr=False, extra_end_rules=("name:recursive",)
)
test = None
if self.stream.skip_if("name:if"):
test = self.parse_expression()
recursive = self.stream.skip_if("name:recursive")
body = self.parse_statements(("name:endfor", "name:else"))
if next(self.stream).value == "endfor":
else_ = []
else:
else_ = self.parse_statements(("name:endfor",), drop_needle=True)
return nodes.For(target, iter, body, else_, test, recursive, lineno=lineno)
def parse_if(self):
"""Parse an if construct."""
node = result = nodes.If(lineno=self.stream.expect("name:if").lineno)
while 1:
node.test = self.parse_tuple(with_condexpr=False)
node.body = self.parse_statements(("name:elif", "name:else", "name:endif"))
node.elif_ = []
node.else_ = []
token = next(self.stream)
if token.test("name:elif"):
node = nodes.If(lineno=self.stream.current.lineno)
result.elif_.append(node)
continue
elif token.test("name:else"):
result.else_ = self.parse_statements(("name:endif",), drop_needle=True)
break
return result
def parse_with(self):
node = nodes.With(lineno=next(self.stream).lineno)
targets = []
values = []
while self.stream.current.type != "block_end":
if targets:
self.stream.expect("comma")
target = self.parse_assign_target()
target.set_ctx("param")
targets.append(target)
self.stream.expect("assign")
values.append(self.parse_expression())
node.targets = targets
node.values = values
node.body = self.parse_statements(("name:endwith",), drop_needle=True)
return node
def parse_autoescape(self):
node = nodes.ScopedEvalContextModifier(lineno=next(self.stream).lineno)
node.options = [nodes.Keyword("autoescape", self.parse_expression())]
node.body = self.parse_statements(("name:endautoescape",), drop_needle=True)
return nodes.Scope([node])
def parse_block(self):
node = nodes.Block(lineno=next(self.stream).lineno)
node.name = self.stream.expect("name").value
node.scoped = self.stream.skip_if("name:scoped")
# common problem people encounter when switching from django
# to jinja. we do not support hyphens in block names, so let's
# raise a nicer error message in that case.
if self.stream.current.type == "sub":
self.fail(
"Block names in Jinja have to be valid Python "
"identifiers and may not contain hyphens, use an "
"underscore instead."
)
node.body = self.parse_statements(("name:endblock",), drop_needle=True)
self.stream.skip_if("name:" + node.name)
return node
def parse_extends(self):
node = nodes.Extends(lineno=next(self.stream).lineno)
node.template = self.parse_expression()
return node
def parse_import_context(self, node, default):
if self.stream.current.test_any(
"name:with", "name:without"
) and self.stream.look().test("name:context"):
node.with_context = next(self.stream).value == "with"
self.stream.skip()
else:
node.with_context = default
return node
def parse_include(self):
node = nodes.Include(lineno=next(self.stream).lineno)
node.template = self.parse_expression()
if self.stream.current.test("name:ignore") and self.stream.look().test(
"name:missing"
):
node.ignore_missing = True
self.stream.skip(2)
else:
node.ignore_missing = False
return self.parse_import_context(node, True)
def parse_import(self):
node = nodes.Import(lineno=next(self.stream).lineno)
node.template = self.parse_expression()
self.stream.expect("name:as")
node.target = self.parse_assign_target(name_only=True).name
return self.parse_import_context(node, False)
def parse_from(self):
node = nodes.FromImport(lineno=next(self.stream).lineno)
node.template = self.parse_expression()
self.stream.expect("name:import")
node.names = []
def parse_context():
if self.stream.current.value in (
"with",
"without",
) and self.stream.look().test("name:context"):
node.with_context = next(self.stream).value == "with"
self.stream.skip()
return True
return False
while 1:
if node.names:
self.stream.expect("comma")
if self.stream.current.type == "name":
if parse_context():
break
target = self.parse_assign_target(name_only=True)
if target.name.startswith("_"):
self.fail(
"names starting with an underline can not be imported",
target.lineno,
exc=TemplateAssertionError,
)
if self.stream.skip_if("name:as"):
alias = self.parse_assign_target(name_only=True)
node.names.append((target.name, alias.name))
else:
node.names.append(target.name)
if parse_context() or self.stream.current.type != "comma":
break
else:
self.stream.expect("name")
if not hasattr(node, "with_context"):
node.with_context = False
return node
def parse_signature(self, node):
node.args = args = []
node.defaults = defaults = []
self.stream.expect("lparen")
while self.stream.current.type != "rparen":
if args:
self.stream.expect("comma")
arg = self.parse_assign_target(name_only=True)
arg.set_ctx("param")
if self.stream.skip_if("assign"):
defaults.append(self.parse_expression())
elif defaults:
self.fail("non-default argument follows default argument")
args.append(arg)
self.stream.expect("rparen")
def parse_call_block(self):
node = nodes.CallBlock(lineno=next(self.stream).lineno)
if self.stream.current.type == "lparen":
self.parse_signature(node)
else:
node.args = []
node.defaults = []
node.call = self.parse_expression()
if not isinstance(node.call, nodes.Call):
self.fail("expected call", node.lineno)
node.body = self.parse_statements(("name:endcall",), drop_needle=True)
return node
def parse_filter_block(self):
node = nodes.FilterBlock(lineno=next(self.stream).lineno)
node.filter = self.parse_filter(None, start_inline=True)
node.body = self.parse_statements(("name:endfilter",), drop_needle=True)
return node
def parse_macro(self):
node = nodes.Macro(lineno=next(self.stream).lineno)
node.name = self.parse_assign_target(name_only=True).name
self.parse_signature(node)
node.body = self.parse_statements(("name:endmacro",), drop_needle=True)
return node
def parse_print(self):
node = nodes.Output(lineno=next(self.stream).lineno)
node.nodes = []
while self.stream.current.type != "block_end":
if node.nodes:
self.stream.expect("comma")
node.nodes.append(self.parse_expression())
return node
def parse_assign_target(
self,
with_tuple=True,
name_only=False,
extra_end_rules=None,
with_namespace=False,
):
"""Parse an assignment target. As Jinja allows assignments to
tuples, this function can parse all allowed assignment targets. Per
default assignments to tuples are parsed, that can be disable however
by setting `with_tuple` to `False`. If only assignments to names are
wanted `name_only` can be set to `True`. The `extra_end_rules`
parameter is forwarded to the tuple parsing function. If
`with_namespace` is enabled, a namespace assignment may be parsed.
"""
if with_namespace and self.stream.look().type == "dot":
token = self.stream.expect("name")
next(self.stream) # dot
attr = self.stream.expect("name")
target = nodes.NSRef(token.value, attr.value, lineno=token.lineno)
elif name_only:
token = self.stream.expect("name")
target = nodes.Name(token.value, "store", lineno=token.lineno)
else:
if with_tuple:
target = self.parse_tuple(
simplified=True, extra_end_rules=extra_end_rules
)
else:
target = self.parse_primary()
target.set_ctx("store")
if not target.can_assign():
self.fail(
"can't assign to %r" % target.__class__.__name__.lower(), target.lineno
)
return target
def parse_expression(self, with_condexpr=True):
"""Parse an expression. Per default all expressions are parsed, if
the optional `with_condexpr` parameter is set to `False` conditional
expressions are not parsed.
"""
if with_condexpr:
return self.parse_condexpr()
return self.parse_or()
def parse_condexpr(self):
lineno = self.stream.current.lineno
expr1 = self.parse_or()
while self.stream.skip_if("name:if"):
expr2 = self.parse_or()
if self.stream.skip_if("name:else"):
expr3 = self.parse_condexpr()
else:
expr3 = None
expr1 = nodes.CondExpr(expr2, expr1, expr3, lineno=lineno)
lineno = self.stream.current.lineno
return expr1
def parse_or(self):
lineno = self.stream.current.lineno
left = self.parse_and()
while self.stream.skip_if("name:or"):
right = self.parse_and()
left = nodes.Or(left, right, lineno=lineno)
lineno = self.stream.current.lineno
return left
def parse_and(self):
lineno = self.stream.current.lineno
left = self.parse_not()
while self.stream.skip_if("name:and"):
right = self.parse_not()
left = nodes.And(left, right, lineno=lineno)
lineno = self.stream.current.lineno
return left
def parse_not(self):
if self.stream.current.test("name:not"):
lineno = next(self.stream).lineno
return nodes.Not(self.parse_not(), lineno=lineno)
return self.parse_compare()
def parse_compare(self):
lineno = self.stream.current.lineno
expr = self.parse_math1()
ops = []
while 1:
token_type = self.stream.current.type
if token_type in _compare_operators:
next(self.stream)
ops.append(nodes.Operand(token_type, self.parse_math1()))
elif self.stream.skip_if("name:in"):
ops.append(nodes.Operand("in", self.parse_math1()))
elif self.stream.current.test("name:not") and self.stream.look().test(
"name:in"
):
self.stream.skip(2)
ops.append(nodes.Operand("notin", self.parse_math1()))
else:
break
lineno = self.stream.current.lineno
if not ops:
return expr
return nodes.Compare(expr, ops, lineno=lineno)
def parse_math1(self):
lineno = self.stream.current.lineno
left = self.parse_concat()
while self.stream.current.type in ("add", "sub"):
cls = _math_nodes[self.stream.current.type]
next(self.stream)
right = self.parse_concat()
left = cls(left, right, lineno=lineno)
lineno = self.stream.current.lineno
return left
def parse_concat(self):
lineno = self.stream.current.lineno
args = [self.parse_math2()]
while self.stream.current.type == "tilde":
next(self.stream)
args.append(self.parse_math2())
if len(args) == 1:
return args[0]
return nodes.Concat(args, lineno=lineno)
def parse_math2(self):
lineno = self.stream.current.lineno
left = self.parse_pow()
while self.stream.current.type in ("mul", "div", "floordiv", "mod"):
cls = _math_nodes[self.stream.current.type]
next(self.stream)
right = self.parse_pow()
left = cls(left, right, lineno=lineno)
lineno = self.stream.current.lineno
return left
def parse_pow(self):
lineno = self.stream.current.lineno
left = self.parse_unary()
while self.stream.current.type == "pow":
next(self.stream)
right = self.parse_unary()
left = nodes.Pow(left, right, lineno=lineno)
lineno = self.stream.current.lineno
return left
def parse_unary(self, with_filter=True):
token_type = self.stream.current.type
lineno = self.stream.current.lineno
if token_type == "sub":
next(self.stream)
node = nodes.Neg(self.parse_unary(False), lineno=lineno)
elif token_type == "add":
next(self.stream)
node = nodes.Pos(self.parse_unary(False), lineno=lineno)
else:
node = self.parse_primary()
node = self.parse_postfix(node)
if with_filter:
node = self.parse_filter_expr(node)
return node
def parse_primary(self):
token = self.stream.current
if token.type == "name":
if token.value in ("true", "false", "True", "False"):
node = nodes.Const(token.value in ("true", "True"), lineno=token.lineno)
elif token.value in ("none", "None"):
node = nodes.Const(None, lineno=token.lineno)
else:
node = nodes.Name(token.value, "load", lineno=token.lineno)
next(self.stream)
elif token.type == "string":
next(self.stream)
buf = [token.value]
lineno = token.lineno
while self.stream.current.type == "string":
buf.append(self.stream.current.value)
next(self.stream)
node = nodes.Const("".join(buf), lineno=lineno)
elif token.type in ("integer", "float"):
next(self.stream)
node = nodes.Const(token.value, lineno=token.lineno)
elif token.type == "lparen":
next(self.stream)
node = self.parse_tuple(explicit_parentheses=True)
self.stream.expect("rparen")
elif token.type == "lbracket":
node = self.parse_list()
elif token.type == "lbrace":
node = self.parse_dict()
else:
self.fail("unexpected '%s'" % describe_token(token), token.lineno)
return node
def parse_tuple(
self,
simplified=False,
with_condexpr=True,
extra_end_rules=None,
explicit_parentheses=False,
):
"""Works like `parse_expression` but if multiple expressions are
delimited by a comma a :class:`~jinja2.nodes.Tuple` node is created.
This method could also return a regular expression instead of a tuple
if no commas where found.
The default parsing mode is a full tuple. If `simplified` is `True`
only names and literals are parsed. The `no_condexpr` parameter is
forwarded to :meth:`parse_expression`.
Because tuples do not require delimiters and may end in a bogus comma
an extra hint is needed that marks the end of a tuple. For example
for loops support tuples between `for` and `in`. In that case the
`extra_end_rules` is set to ``['name:in']``.
`explicit_parentheses` is true if the parsing was triggered by an
expression in parentheses. This is used to figure out if an empty
tuple is a valid expression or not.
"""
lineno = self.stream.current.lineno
if simplified:
parse = self.parse_primary
elif with_condexpr:
parse = self.parse_expression
else:
def parse():
return self.parse_expression(with_condexpr=False)
args = []
is_tuple = False
while 1:
if args:
self.stream.expect("comma")
if self.is_tuple_end(extra_end_rules):
break
args.append(parse())
if self.stream.current.type == "comma":
is_tuple = True
else:
break
lineno = self.stream.current.lineno
if not is_tuple:
if args:
return args[0]
# if we don't have explicit parentheses, an empty tuple is
# not a valid expression. This would mean nothing (literally
# nothing) in the spot of an expression would be an empty
# tuple.
if not explicit_parentheses:
self.fail(
"Expected an expression, got '%s'"
% describe_token(self.stream.current)
)
return nodes.Tuple(args, "load", lineno=lineno)
def parse_list(self):
token = self.stream.expect("lbracket")
items = []
while self.stream.current.type != "rbracket":
if items:
self.stream.expect("comma")
if self.stream.current.type == "rbracket":
break
items.append(self.parse_expression())
self.stream.expect("rbracket")
return nodes.List(items, lineno=token.lineno)
def parse_dict(self):
token = self.stream.expect("lbrace")
items = []
while self.stream.current.type != "rbrace":
if items:
self.stream.expect("comma")
if self.stream.current.type == "rbrace":
break
key = self.parse_expression()
self.stream.expect("colon")
value = self.parse_expression()
items.append(nodes.Pair(key, value, lineno=key.lineno))
self.stream.expect("rbrace")
return nodes.Dict(items, lineno=token.lineno)
def parse_postfix(self, node):
while 1:
token_type = self.stream.current.type
if token_type == "dot" or token_type == "lbracket":
node = self.parse_subscript(node)
# calls are valid both after postfix expressions (getattr
# and getitem) as well as filters and tests
elif token_type == "lparen":
node = self.parse_call(node)
else:
break
return node
def parse_filter_expr(self, node):
while 1:
token_type = self.stream.current.type
if token_type == "pipe":
node = self.parse_filter(node)
elif token_type == "name" and self.stream.current.value == "is":
node = self.parse_test(node)
# calls are valid both after postfix expressions (getattr
# and getitem) as well as filters and tests
elif token_type == "lparen":
node = self.parse_call(node)
else:
break
return node
def parse_subscript(self, node):
token = next(self.stream)
if token.type == "dot":
attr_token = self.stream.current
next(self.stream)
if attr_token.type == "name":
return nodes.Getattr(
node, attr_token.value, "load", lineno=token.lineno
)
elif attr_token.type != "integer":
self.fail("expected name or number", attr_token.lineno)
arg = nodes.Const(attr_token.value, lineno=attr_token.lineno)
return nodes.Getitem(node, arg, "load", lineno=token.lineno)
if token.type == "lbracket":
args = []
while self.stream.current.type != "rbracket":
if args:
self.stream.expect("comma")
args.append(self.parse_subscribed())
self.stream.expect("rbracket")
if len(args) == 1:
arg = args[0]
else:
arg = nodes.Tuple(args, "load", lineno=token.lineno)
return nodes.Getitem(node, arg, "load", lineno=token.lineno)
self.fail("expected subscript expression", token.lineno)
def parse_subscribed(self):
lineno = self.stream.current.lineno
if self.stream.current.type == "colon":
next(self.stream)
args = [None]
else:
node = self.parse_expression()
if self.stream.current.type != "colon":
return node
next(self.stream)
args = [node]
if self.stream.current.type == "colon":
args.append(None)
elif self.stream.current.type not in ("rbracket", "comma"):
args.append(self.parse_expression())
else:
args.append(None)
if self.stream.current.type == "colon":
next(self.stream)
if self.stream.current.type not in ("rbracket", "comma"):
args.append(self.parse_expression())
else:
args.append(None)
else:
args.append(None)
return nodes.Slice(lineno=lineno, *args)
def parse_call(self, node):
token = self.stream.expect("lparen")
args = []
kwargs = []
dyn_args = dyn_kwargs = None
require_comma = False
def ensure(expr):
if not expr:
self.fail("invalid syntax for function call expression", token.lineno)
while self.stream.current.type != "rparen":
if require_comma:
self.stream.expect("comma")
# support for trailing comma
if self.stream.current.type == "rparen":
break
if self.stream.current.type == "mul":
ensure(dyn_args is None and dyn_kwargs is None)
next(self.stream)
dyn_args = self.parse_expression()
elif self.stream.current.type == "pow":
ensure(dyn_kwargs is None)
next(self.stream)
dyn_kwargs = self.parse_expression()
else:
if (
self.stream.current.type == "name"
and self.stream.look().type == "assign"
):
# Parsing a kwarg
ensure(dyn_kwargs is None)
key = self.stream.current.value
self.stream.skip(2)
value = self.parse_expression()
kwargs.append(nodes.Keyword(key, value, lineno=value.lineno))
else:
# Parsing an arg
ensure(dyn_args is None and dyn_kwargs is None and not kwargs)
args.append(self.parse_expression())
require_comma = True
self.stream.expect("rparen")
if node is None:
return args, kwargs, dyn_args, dyn_kwargs
return nodes.Call(node, args, kwargs, dyn_args, dyn_kwargs, lineno=token.lineno)
def parse_filter(self, node, start_inline=False):
while self.stream.current.type == "pipe" or start_inline:
if not start_inline:
next(self.stream)
token = self.stream.expect("name")
name = token.value
while self.stream.current.type == "dot":
next(self.stream)
name += "." + self.stream.expect("name").value
if self.stream.current.type == "lparen":
args, kwargs, dyn_args, dyn_kwargs = self.parse_call(None)
else:
args = []
kwargs = []
dyn_args = dyn_kwargs = None
node = nodes.Filter(
node, name, args, kwargs, dyn_args, dyn_kwargs, lineno=token.lineno
)
start_inline = False
return node
def parse_test(self, node):
token = next(self.stream)
if self.stream.current.test("name:not"):
next(self.stream)
negated = True
else:
negated = False
name = self.stream.expect("name").value
while self.stream.current.type == "dot":
next(self.stream)
name += "." + self.stream.expect("name").value
dyn_args = dyn_kwargs = None
kwargs = []
if self.stream.current.type == "lparen":
args, kwargs, dyn_args, dyn_kwargs = self.parse_call(None)
elif self.stream.current.type in (
"name",
"string",
"integer",
"float",
"lparen",
"lbracket",
"lbrace",
) and not self.stream.current.test_any("name:else", "name:or", "name:and"):
if self.stream.current.test("name:is"):
self.fail("You cannot chain multiple tests with is")
arg_node = self.parse_primary()
arg_node = self.parse_postfix(arg_node)
args = [arg_node]
else:
args = []
node = nodes.Test(
node, name, args, kwargs, dyn_args, dyn_kwargs, lineno=token.lineno
)
if negated:
node = nodes.Not(node, lineno=token.lineno)
return node
def subparse(self, end_tokens=None):
body = []
data_buffer = []
add_data = data_buffer.append
if end_tokens is not None:
self._end_token_stack.append(end_tokens)
def flush_data():
if data_buffer:
lineno = data_buffer[0].lineno
body.append(nodes.Output(data_buffer[:], lineno=lineno))
del data_buffer[:]
try:
while self.stream:
token = self.stream.current
if token.type == "data":
if token.value:
add_data(nodes.TemplateData(token.value, lineno=token.lineno))
next(self.stream)
elif token.type == "variable_begin":
next(self.stream)
add_data(self.parse_tuple(with_condexpr=True))
self.stream.expect("variable_end")
elif token.type == "block_begin":
flush_data()
next(self.stream)
if end_tokens is not None and self.stream.current.test_any(
*end_tokens
):
return body
rv = self.parse_statement()
if isinstance(rv, list):
body.extend(rv)
else:
body.append(rv)
self.stream.expect("block_end")
else:
raise AssertionError("internal parsing error")
flush_data()
finally:
if end_tokens is not None:
self._end_token_stack.pop()
return body
def parse(self):
"""Parse the whole template into a `Template` node."""
result = nodes.Template(self.subparse(), lineno=1)
result.set_environment(self.environment)
return result
# -*- coding: utf-8 -*-
"""The runtime functions and state used by compiled templates."""
import sys
from itertools import chain
from types import MethodType
from markupsafe import escape # noqa: F401
from markupsafe import Markup
from markupsafe import soft_unicode
from ._compat import abc
from ._compat import imap
from ._compat import implements_iterator
from ._compat import implements_to_string
from ._compat import iteritems
from ._compat import PY2
from ._compat import string_types
from ._compat import text_type
from ._compat import with_metaclass
from .exceptions import TemplateNotFound # noqa: F401
from .exceptions import TemplateRuntimeError # noqa: F401
from .exceptions import UndefinedError
from .nodes import EvalContext
from .utils import concat
from .utils import evalcontextfunction
from .utils import internalcode
from .utils import missing
from .utils import Namespace # noqa: F401
from .utils import object_type_repr
# these variables are exported to the template runtime
exported = [
"LoopContext",
"TemplateReference",
"Macro",
"Markup",
"TemplateRuntimeError",
"missing",
"concat",
"escape",
"markup_join",
"unicode_join",
"to_string",
"identity",
"TemplateNotFound",
"Namespace",
"Undefined",
]
#: the name of the function that is used to convert something into
#: a string. We can just use the text type here.
to_string = text_type
def identity(x):
"""Returns its argument. Useful for certain things in the
environment.
"""
return x
def markup_join(seq):
"""Concatenation that escapes if necessary and converts to unicode."""
buf = []
iterator = imap(soft_unicode, seq)
for arg in iterator:
buf.append(arg)
if hasattr(arg, "__html__"):
return Markup(u"").join(chain(buf, iterator))
return concat(buf)
def unicode_join(seq):
"""Simple args to unicode conversion and concatenation."""
return concat(imap(text_type, seq))
def new_context(
environment,
template_name,
blocks,
vars=None,
shared=None,
globals=None,
locals=None,
):
"""Internal helper for context creation."""
if vars is None:
vars = {}
if shared:
parent = vars
else:
parent = dict(globals or (), **vars)
if locals:
# if the parent is shared a copy should be created because
# we don't want to modify the dict passed
if shared:
parent = dict(parent)
for key, value in iteritems(locals):
if value is not missing:
parent[key] = value
return environment.context_class(environment, parent, template_name, blocks)
class TemplateReference(object):
"""The `self` in templates."""
def __init__(self, context):
self.__context = context
def __getitem__(self, name):
blocks = self.__context.blocks[name]
return BlockReference(name, self.__context, blocks, 0)
def __repr__(self):
return "<%s %r>" % (self.__class__.__name__, self.__context.name)
def _get_func(x):
return getattr(x, "__func__", x)
class ContextMeta(type):
def __new__(mcs, name, bases, d):
rv = type.__new__(mcs, name, bases, d)
if bases == ():
return rv
resolve = _get_func(rv.resolve)
default_resolve = _get_func(Context.resolve)
resolve_or_missing = _get_func(rv.resolve_or_missing)
default_resolve_or_missing = _get_func(Context.resolve_or_missing)
# If we have a changed resolve but no changed default or missing
# resolve we invert the call logic.
if (
resolve is not default_resolve
and resolve_or_missing is default_resolve_or_missing
):
rv._legacy_resolve_mode = True
elif (
resolve is default_resolve
and resolve_or_missing is default_resolve_or_missing
):
rv._fast_resolve_mode = True
return rv
def resolve_or_missing(context, key, missing=missing):
if key in context.vars:
return context.vars[key]
if key in context.parent:
return context.parent[key]
return missing
class Context(with_metaclass(ContextMeta)):
"""The template context holds the variables of a template. It stores the
values passed to the template and also the names the template exports.
Creating instances is neither supported nor useful as it's created
automatically at various stages of the template evaluation and should not
be created by hand.
The context is immutable. Modifications on :attr:`parent` **must not**
happen and modifications on :attr:`vars` are allowed from generated
template code only. Template filters and global functions marked as
:func:`contextfunction`\\s get the active context passed as first argument
and are allowed to access the context read-only.
The template context supports read only dict operations (`get`,
`keys`, `values`, `items`, `iterkeys`, `itervalues`, `iteritems`,
`__getitem__`, `__contains__`). Additionally there is a :meth:`resolve`
method that doesn't fail with a `KeyError` but returns an
:class:`Undefined` object for missing variables.
"""
# XXX: we want to eventually make this be a deprecation warning and
# remove it.
_legacy_resolve_mode = False
_fast_resolve_mode = False
def __init__(self, environment, parent, name, blocks):
self.parent = parent
self.vars = {}
self.environment = environment
self.eval_ctx = EvalContext(self.environment, name)
self.exported_vars = set()
self.name = name
# create the initial mapping of blocks. Whenever template inheritance
# takes place the runtime will update this mapping with the new blocks
# from the template.
self.blocks = dict((k, [v]) for k, v in iteritems(blocks))
# In case we detect the fast resolve mode we can set up an alias
# here that bypasses the legacy code logic.
if self._fast_resolve_mode:
self.resolve_or_missing = MethodType(resolve_or_missing, self)
def super(self, name, current):
"""Render a parent block."""
try:
blocks = self.blocks[name]
index = blocks.index(current) + 1
blocks[index]
except LookupError:
return self.environment.undefined(
"there is no parent block called %r." % name, name="super"
)
return BlockReference(name, self, blocks, index)
def get(self, key, default=None):
"""Returns an item from the template context, if it doesn't exist
`default` is returned.
"""
try:
return self[key]
except KeyError:
return default
def resolve(self, key):
"""Looks up a variable like `__getitem__` or `get` but returns an
:class:`Undefined` object with the name of the name looked up.
"""
if self._legacy_resolve_mode:
rv = resolve_or_missing(self, key)
else:
rv = self.resolve_or_missing(key)
if rv is missing:
return self.environment.undefined(name=key)
return rv
def resolve_or_missing(self, key):
"""Resolves a variable like :meth:`resolve` but returns the
special `missing` value if it cannot be found.
"""
if self._legacy_resolve_mode:
rv = self.resolve(key)
if isinstance(rv, Undefined):
rv = missing
return rv
return resolve_or_missing(self, key)
def get_exported(self):
"""Get a new dict with the exported variables."""
return dict((k, self.vars[k]) for k in self.exported_vars)
def get_all(self):
"""Return the complete context as dict including the exported
variables. For optimizations reasons this might not return an
actual copy so be careful with using it.
"""
if not self.vars:
return self.parent
if not self.parent:
return self.vars
return dict(self.parent, **self.vars)
@internalcode
def call(__self, __obj, *args, **kwargs): # noqa: B902
"""Call the callable with the arguments and keyword arguments
provided but inject the active context or environment as first
argument if the callable is a :func:`contextfunction` or
:func:`environmentfunction`.
"""
if __debug__:
__traceback_hide__ = True # noqa
# Allow callable classes to take a context
if hasattr(__obj, "__call__"): # noqa: B004
fn = __obj.__call__
for fn_type in (
"contextfunction",
"evalcontextfunction",
"environmentfunction",
):
if hasattr(fn, fn_type):
__obj = fn
break
if callable(__obj):
if getattr(__obj, "contextfunction", False) is True:
args = (__self,) + args
elif getattr(__obj, "evalcontextfunction", False) is True:
args = (__self.eval_ctx,) + args
elif getattr(__obj, "environmentfunction", False) is True:
args = (__self.environment,) + args
try:
return __obj(*args, **kwargs)
except StopIteration:
return __self.environment.undefined(
"value was undefined because "
"a callable raised a "
"StopIteration exception"
)
def derived(self, locals=None):
"""Internal helper function to create a derived context. This is
used in situations where the system needs a new context in the same
template that is independent.
"""
context = new_context(
self.environment, self.name, {}, self.get_all(), True, None, locals
)
context.eval_ctx = self.eval_ctx
context.blocks.update((k, list(v)) for k, v in iteritems(self.blocks))
return context
def _all(meth): # noqa: B902
def proxy(self):
return getattr(self.get_all(), meth)()
proxy.__doc__ = getattr(dict, meth).__doc__
proxy.__name__ = meth
return proxy
keys = _all("keys")
values = _all("values")
items = _all("items")
# not available on python 3
if PY2:
iterkeys = _all("iterkeys")
itervalues = _all("itervalues")
iteritems = _all("iteritems")
del _all
def __contains__(self, name):
return name in self.vars or name in self.parent
def __getitem__(self, key):
"""Lookup a variable or raise `KeyError` if the variable is
undefined.
"""
item = self.resolve_or_missing(key)
if item is missing:
raise KeyError(key)
return item
def __repr__(self):
return "<%s %s of %r>" % (
self.__class__.__name__,
repr(self.get_all()),
self.name,
)
abc.Mapping.register(Context)
class BlockReference(object):
"""One block on a template reference."""
def __init__(self, name, context, stack, depth):
self.name = name
self._context = context
self._stack = stack
self._depth = depth
@property
def super(self):
"""Super the block."""
if self._depth + 1 >= len(self._stack):
return self._context.environment.undefined(
"there is no parent block called %r." % self.name, name="super"
)
return BlockReference(self.name, self._context, self._stack, self._depth + 1)
@internalcode
def __call__(self):
rv = concat(self._stack[self._depth](self._context))
if self._context.eval_ctx.autoescape:
rv = Markup(rv)
return rv
@implements_iterator
class LoopContext:
"""A wrapper iterable for dynamic ``for`` loops, with information
about the loop and iteration.
"""
#: Current iteration of the loop, starting at 0.
index0 = -1
_length = None
_after = missing
_current = missing
_before = missing
_last_changed_value = missing
def __init__(self, iterable, undefined, recurse=None, depth0=0):
"""
:param iterable: Iterable to wrap.
:param undefined: :class:`Undefined` class to use for next and
previous items.
:param recurse: The function to render the loop body when the
loop is marked recursive.
:param depth0: Incremented when looping recursively.
"""
self._iterable = iterable
self._iterator = self._to_iterator(iterable)
self._undefined = undefined
self._recurse = recurse
#: How many levels deep a recursive loop currently is, starting at 0.
self.depth0 = depth0
@staticmethod
def _to_iterator(iterable):
return iter(iterable)
@property
def length(self):
"""Length of the iterable.
If the iterable is a generator or otherwise does not have a
size, it is eagerly evaluated to get a size.
"""
if self._length is not None:
return self._length
try:
self._length = len(self._iterable)
except TypeError:
iterable = list(self._iterator)
self._iterator = self._to_iterator(iterable)
self._length = len(iterable) + self.index + (self._after is not missing)
return self._length
def __len__(self):
return self.length
@property
def depth(self):
"""How many levels deep a recursive loop currently is, starting at 1."""
return self.depth0 + 1
@property
def index(self):
"""Current iteration of the loop, starting at 1."""
return self.index0 + 1
@property
def revindex0(self):
"""Number of iterations from the end of the loop, ending at 0.
Requires calculating :attr:`length`.
"""
return self.length - self.index
@property
def revindex(self):
"""Number of iterations from the end of the loop, ending at 1.
Requires calculating :attr:`length`.
"""
return self.length - self.index0
@property
def first(self):
"""Whether this is the first iteration of the loop."""
return self.index0 == 0
def _peek_next(self):
"""Return the next element in the iterable, or :data:`missing`
if the iterable is exhausted. Only peeks one item ahead, caching
the result in :attr:`_last` for use in subsequent checks. The
cache is reset when :meth:`__next__` is called.
"""
if self._after is not missing:
return self._after
self._after = next(self._iterator, missing)
return self._after
@property
def last(self):
"""Whether this is the last iteration of the loop.
Causes the iterable to advance early. See
:func:`itertools.groupby` for issues this can cause.
The :func:`groupby` filter avoids that issue.
"""
return self._peek_next() is missing
@property
def previtem(self):
"""The item in the previous iteration. Undefined during the
first iteration.
"""
if self.first:
return self._undefined("there is no previous item")
return self._before
@property
def nextitem(self):
"""The item in the next iteration. Undefined during the last
iteration.
Causes the iterable to advance early. See
:func:`itertools.groupby` for issues this can cause.
The :func:`groupby` filter avoids that issue.
"""
rv = self._peek_next()
if rv is missing:
return self._undefined("there is no next item")
return rv
def cycle(self, *args):
"""Return a value from the given args, cycling through based on
the current :attr:`index0`.
:param args: One or more values to cycle through.
"""
if not args:
raise TypeError("no items for cycling given")
return args[self.index0 % len(args)]
def changed(self, *value):
"""Return ``True`` if previously called with a different value
(including when called for the first time).
:param value: One or more values to compare to the last call.
"""
if self._last_changed_value != value:
self._last_changed_value = value
return True
return False
def __iter__(self):
return self
def __next__(self):
if self._after is not missing:
rv = self._after
self._after = missing
else:
rv = next(self._iterator)
self.index0 += 1
self._before = self._current
self._current = rv
return rv, self
@internalcode
def __call__(self, iterable):
"""When iterating over nested data, render the body of the loop
recursively with the given inner iterable data.
The loop must have the ``recursive`` marker for this to work.
"""
if self._recurse is None:
raise TypeError(
"The loop must have the 'recursive' marker to be called recursively."
)
return self._recurse(iterable, self._recurse, depth=self.depth)
def __repr__(self):
return "<%s %d/%d>" % (self.__class__.__name__, self.index, self.length)
class Macro(object):
"""Wraps a macro function."""
def __init__(
self,
environment,
func,
name,
arguments,
catch_kwargs,
catch_varargs,
caller,
default_autoescape=None,
):
self._environment = environment
self._func = func
self._argument_count = len(arguments)
self.name = name
self.arguments = arguments
self.catch_kwargs = catch_kwargs
self.catch_varargs = catch_varargs
self.caller = caller
self.explicit_caller = "caller" in arguments
if default_autoescape is None:
default_autoescape = environment.autoescape
self._default_autoescape = default_autoescape
@internalcode
@evalcontextfunction
def __call__(self, *args, **kwargs):
# This requires a bit of explanation, In the past we used to
# decide largely based on compile-time information if a macro is
# safe or unsafe. While there was a volatile mode it was largely
# unused for deciding on escaping. This turns out to be
# problematic for macros because whether a macro is safe depends not
# on the escape mode when it was defined, but rather when it was used.
#
# Because however we export macros from the module system and
# there are historic callers that do not pass an eval context (and
# will continue to not pass one), we need to perform an instance
# check here.
#
# This is considered safe because an eval context is not a valid
# argument to callables otherwise anyway. Worst case here is
# that if no eval context is passed we fall back to the compile
# time autoescape flag.
if args and isinstance(args[0], EvalContext):
autoescape = args[0].autoescape
args = args[1:]
else:
autoescape = self._default_autoescape
# try to consume the positional arguments
arguments = list(args[: self._argument_count])
off = len(arguments)
# For information why this is necessary refer to the handling
# of caller in the `macro_body` handler in the compiler.
found_caller = False
# if the number of arguments consumed is not the number of
# arguments expected we start filling in keyword arguments
# and defaults.
if off != self._argument_count:
for name in self.arguments[len(arguments) :]:
try:
value = kwargs.pop(name)
except KeyError:
value = missing
if name == "caller":
found_caller = True
arguments.append(value)
else:
found_caller = self.explicit_caller
# it's important that the order of these arguments does not change
# if not also changed in the compiler's `function_scoping` method.
# the order is caller, keyword arguments, positional arguments!
if self.caller and not found_caller:
caller = kwargs.pop("caller", None)
if caller is None:
caller = self._environment.undefined("No caller defined", name="caller")
arguments.append(caller)
if self.catch_kwargs:
arguments.append(kwargs)
elif kwargs:
if "caller" in kwargs:
raise TypeError(
"macro %r was invoked with two values for "
"the special caller argument. This is "
"most likely a bug." % self.name
)
raise TypeError(
"macro %r takes no keyword argument %r"
% (self.name, next(iter(kwargs)))
)
if self.catch_varargs:
arguments.append(args[self._argument_count :])
elif len(args) > self._argument_count:
raise TypeError(
"macro %r takes not more than %d argument(s)"
% (self.name, len(self.arguments))
)
return self._invoke(arguments, autoescape)
def _invoke(self, arguments, autoescape):
"""This method is being swapped out by the async implementation."""
rv = self._func(*arguments)
if autoescape:
rv = Markup(rv)
return rv
def __repr__(self):
return "<%s %s>" % (
self.__class__.__name__,
self.name is None and "anonymous" or repr(self.name),
)
@implements_to_string
class Undefined(object):
"""The default undefined type. This undefined type can be printed and
iterated over, but every other access will raise an :exc:`UndefinedError`:
>>> foo = Undefined(name='foo')
>>> str(foo)
''
>>> not foo
True
>>> foo + 42
Traceback (most recent call last):
...
jinja2.exceptions.UndefinedError: 'foo' is undefined
"""
__slots__ = (
"_undefined_hint",
"_undefined_obj",
"_undefined_name",
"_undefined_exception",
)
def __init__(self, hint=None, obj=missing, name=None, exc=UndefinedError):
self._undefined_hint = hint
self._undefined_obj = obj
self._undefined_name = name
self._undefined_exception = exc
@property
def _undefined_message(self):
"""Build a message about the undefined value based on how it was
accessed.
"""
if self._undefined_hint:
return self._undefined_hint
if self._undefined_obj is missing:
return "%r is undefined" % self._undefined_name
if not isinstance(self._undefined_name, string_types):
return "%s has no element %r" % (
object_type_repr(self._undefined_obj),
self._undefined_name,
)
return "%r has no attribute %r" % (
object_type_repr(self._undefined_obj),
self._undefined_name,
)
@internalcode
def _fail_with_undefined_error(self, *args, **kwargs):
"""Raise an :exc:`UndefinedError` when operations are performed
on the undefined value.
"""
raise self._undefined_exception(self._undefined_message)
@internalcode
def __getattr__(self, name):
if name[:2] == "__":
raise AttributeError(name)
return self._fail_with_undefined_error()
__add__ = (
__radd__
) = (
__mul__
) = (
__rmul__
) = (
__div__
) = (
__rdiv__
) = (
__truediv__
) = (
__rtruediv__
) = (
__floordiv__
) = (
__rfloordiv__
) = (
__mod__
) = (
__rmod__
) = (
__pos__
) = (
__neg__
) = (
__call__
) = (
__getitem__
) = (
__lt__
) = (
__le__
) = (
__gt__
) = (
__ge__
) = (
__int__
) = (
__float__
) = (
__complex__
) = __pow__ = __rpow__ = __sub__ = __rsub__ = _fail_with_undefined_error
def __eq__(self, other):
return type(self) is type(other)
def __ne__(self, other):
return not self.__eq__(other)
def __hash__(self):
return id(type(self))
def __str__(self):
return u""
def __len__(self):
return 0
def __iter__(self):
if 0:
yield None
def __nonzero__(self):
return False
__bool__ = __nonzero__
def __repr__(self):
return "Undefined"
def make_logging_undefined(logger=None, base=None):
"""Given a logger object this returns a new undefined class that will
log certain failures. It will log iterations and printing. If no
logger is given a default logger is created.
Example::
logger = logging.getLogger(__name__)
LoggingUndefined = make_logging_undefined(
logger=logger,
base=Undefined
)
.. versionadded:: 2.8
:param logger: the logger to use. If not provided, a default logger
is created.
:param base: the base class to add logging functionality to. This
defaults to :class:`Undefined`.
"""
if logger is None:
import logging
logger = logging.getLogger(__name__)
logger.addHandler(logging.StreamHandler(sys.stderr))
if base is None:
base = Undefined
def _log_message(undef):
if undef._undefined_hint is None:
if undef._undefined_obj is missing:
hint = "%s is undefined" % undef._undefined_name
elif not isinstance(undef._undefined_name, string_types):
hint = "%s has no element %s" % (
object_type_repr(undef._undefined_obj),
undef._undefined_name,
)
else:
hint = "%s has no attribute %s" % (
object_type_repr(undef._undefined_obj),
undef._undefined_name,
)
else:
hint = undef._undefined_hint
logger.warning("Template variable warning: %s", hint)
class LoggingUndefined(base):
def _fail_with_undefined_error(self, *args, **kwargs):
try:
return base._fail_with_undefined_error(self, *args, **kwargs)
except self._undefined_exception as e:
logger.error("Template variable error: %s", str(e))
raise e
def __str__(self):
rv = base.__str__(self)
_log_message(self)
return rv
def __iter__(self):
rv = base.__iter__(self)
_log_message(self)
return rv
if PY2:
def __nonzero__(self):
rv = base.__nonzero__(self)
_log_message(self)
return rv
def __unicode__(self):
rv = base.__unicode__(self)
_log_message(self)
return rv
else:
def __bool__(self):
rv = base.__bool__(self)
_log_message(self)
return rv
return LoggingUndefined
# No @implements_to_string decorator here because __str__
# is not overwritten from Undefined in this class.
# This would cause a recursion error in Python 2.
class ChainableUndefined(Undefined):
"""An undefined that is chainable, where both ``__getattr__`` and
``__getitem__`` return itself rather than raising an
:exc:`UndefinedError`.
>>> foo = ChainableUndefined(name='foo')
>>> str(foo.bar['baz'])
''
>>> foo.bar['baz'] + 42
Traceback (most recent call last):
...
jinja2.exceptions.UndefinedError: 'foo' is undefined
.. versionadded:: 2.11.0
"""
__slots__ = ()
def __html__(self):
return self.__str__()
def __getattr__(self, _):
return self
__getitem__ = __getattr__
@implements_to_string
class DebugUndefined(Undefined):
"""An undefined that returns the debug info when printed.
>>> foo = DebugUndefined(name='foo')
>>> str(foo)
'{{ foo }}'
>>> not foo
True
>>> foo + 42
Traceback (most recent call last):
...
jinja2.exceptions.UndefinedError: 'foo' is undefined
"""
__slots__ = ()
def __str__(self):
if self._undefined_hint is None:
if self._undefined_obj is missing:
return u"{{ %s }}" % self._undefined_name
return "{{ no such element: %s[%r] }}" % (
object_type_repr(self._undefined_obj),
self._undefined_name,
)
return u"{{ undefined value printed: %s }}" % self._undefined_hint
@implements_to_string
class StrictUndefined(Undefined):
"""An undefined that barks on print and iteration as well as boolean
tests and all kinds of comparisons. In other words: you can do nothing
with it except checking if it's defined using the `defined` test.
>>> foo = StrictUndefined(name='foo')
>>> str(foo)
Traceback (most recent call last):
...
jinja2.exceptions.UndefinedError: 'foo' is undefined
>>> not foo
Traceback (most recent call last):
...
jinja2.exceptions.UndefinedError: 'foo' is undefined
>>> foo + 42
Traceback (most recent call last):
...
jinja2.exceptions.UndefinedError: 'foo' is undefined
"""
__slots__ = ()
__iter__ = (
__str__
) = (
__len__
) = (
__nonzero__
) = __eq__ = __ne__ = __bool__ = __hash__ = Undefined._fail_with_undefined_error
# remove remaining slots attributes, after the metaclass did the magic they
# are unneeded and irritating as they contain wrong data for the subclasses.
del (
Undefined.__slots__,
ChainableUndefined.__slots__,
DebugUndefined.__slots__,
StrictUndefined.__slots__,
)
# -*- coding: utf-8 -*-
"""A sandbox layer that ensures unsafe operations cannot be performed.
Useful when the template itself comes from an untrusted source.
"""
import operator
import types
import warnings
from collections import deque
from string import Formatter
from markupsafe import EscapeFormatter
from markupsafe import Markup
from ._compat import abc
from ._compat import PY2
from ._compat import range_type
from ._compat import string_types
from .environment import Environment
from .exceptions import SecurityError
#: maximum number of items a range may produce
MAX_RANGE = 100000
#: attributes of function objects that are considered unsafe.
if PY2:
UNSAFE_FUNCTION_ATTRIBUTES = {
"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 method attributes. function attributes are unsafe for methods too
UNSAFE_METHOD_ATTRIBUTES = {"im_class", "im_func", "im_self"}
#: unsafe generator attributes.
UNSAFE_GENERATOR_ATTRIBUTES = {"gi_frame", "gi_code"}
#: unsafe attributes on coroutines
UNSAFE_COROUTINE_ATTRIBUTES = {"cr_frame", "cr_code"}
#: unsafe attributes on async generators
UNSAFE_ASYNC_GENERATOR_ATTRIBUTES = {"ag_code", "ag_frame"}
# 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=__name__
)
_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
_mutable_set_types += (abc.MutableSet,)
_mutable_mapping_types += (abc.MutableMapping,)
_mutable_sequence_types += (abc.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",
]
),
),
)
class _MagicFormatMapping(abc.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__ not in ("format", "format_map"):
return None
obj = callable.__self__
if isinstance(obj, string_types):
return obj
def safe_range(*args):
"""A range that can't generate ranges with a length of more than
MAX_RANGE items.
"""
rng = range_type(*args)
if len(rng) > MAX_RANGE:
raise OverflowError(
"Range too big. The sandbox blocks ranges larger than"
" MAX_RANGE (%d)." % MAX_RANGE
)
return rng
def unsafe(f):
"""Marks a function or method as unsafe.
::
@unsafe
def delete(self):
pass
"""
f.unsafe_callable = True
return f
def is_internal_attribute(obj, attr):
"""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
:meth:`~SandboxedEnvironment.is_safe_attribute` is overridden.
>>> from jinja2.sandbox import is_internal_attribute
>>> is_internal_attribute(str, "mro")
True
>>> is_internal_attribute(str, "upper")
False
"""
if isinstance(obj, types.FunctionType):
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:
return True
elif isinstance(obj, type):
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):
if attr in UNSAFE_COROUTINE_ATTRIBUTES:
return True
elif hasattr(types, "AsyncGeneratorType") and isinstance(
obj, types.AsyncGeneratorType
):
if attr in UNSAFE_ASYNC_GENERATOR_ATTRIBUTES:
return True
return attr.startswith("__")
def modifies_known_mutable(obj, attr):
"""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`.
>>> modifies_known_mutable({}, "clear")
True
>>> modifies_known_mutable({}, "keys")
False
>>> modifies_known_mutable([], "append")
True
>>> modifies_known_mutable([], "index")
False
If called with an unsupported object (such as unicode) `False` is
returned.
>>> modifies_known_mutable("foo", "upper")
False
"""
for typespec, unsafe in _mutable_spec:
if isinstance(obj, typespec):
return attr in unsafe
return False
class SandboxedEnvironment(Environment):
"""The sandboxed environment. It works like the regular environment but
tells the compiler to generate sandboxed code. Additionally subclasses of
this environment may override the methods that tell the runtime what
attributes or functions are safe to access.
If the template tries to access insecure code a :exc:`SecurityError` is
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 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}
#: a set of binary operators that should be intercepted. Each operator
#: that is added to this set (empty by default) is delegated to the
#: :meth:`call_binop` method that will perform the operator. The default
#: operator callback is specified by :attr:`binop_table`.
#:
#: The following binary operators are interceptable:
#: ``//``, ``%``, ``+``, ``*``, ``-``, ``/``, and ``**``
#:
#: The default operation form the operator table corresponds to the
#: builtin function. 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
intercepted_binops = 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
#: :meth:`call_unop` method that will perform the operator. The default
#: operator callback is specified by :attr:`unop_table`.
#:
#: The following unary operators are interceptable: ``+``, ``-``
#:
#: The default operation form the operator table corresponds to the
#: builtin function. 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
intercepted_unops = 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 executed 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
self.binop_table = self.default_binop_table.copy()
self.unop_table = self.default_unop_table.copy()
def is_safe_attribute(self, obj, attr, value):
"""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))
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.
"""
return not (
getattr(obj, "unsafe_callable", False) or getattr(obj, "alters_data", False)
)
def call_binop(self, context, operator, left, right):
"""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.
.. versionadded:: 2.6
"""
return self.binop_table[operator](left, right)
def call_unop(self, context, operator, arg):
"""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.
.. versionadded:: 2.6
"""
return self.unop_table[operator](arg)
def getitem(self, obj, argument):
"""Subscribe an object from sandboxed code."""
try:
return obj[argument]
except (TypeError, LookupError):
if isinstance(argument, string_types):
try:
attr = str(argument)
except Exception:
pass
else:
try:
value = getattr(obj, attr)
except AttributeError:
pass
else:
if self.is_safe_attribute(obj, argument, value):
return value
return self.unsafe_undefined(obj, argument)
return self.undefined(obj=obj, name=argument)
def getattr(self, obj, attribute):
"""Subscribe an object from sandboxed code and prefer the
attribute. The attribute passed *must* be a bytestring.
"""
try:
value = getattr(obj, attribute)
except AttributeError:
try:
return obj[attribute]
except (TypeError, LookupError):
pass
else:
if self.is_safe_attribute(obj, attribute, value):
return value
return self.unsafe_undefined(obj, attribute)
return self.undefined(obj=obj, name=attribute)
def unsafe_undefined(self, obj, attribute):
"""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,
)
def format_string(self, s, args, kwargs, format_func=None):
"""If a format call is detected, then this is routed through this
method so that our safety sandbox can be used for it.
"""
if isinstance(s, Markup):
formatter = SandboxedEscapeFormatter(self, s.escape)
else:
formatter = SandboxedFormatter(self)
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 %d given"
% (len(args) + (kwargs is not None))
)
kwargs = args[0]
args = None
kwargs = _MagicFormatMapping(args, kwargs)
rv = formatter.vformat(s, args, kwargs)
return type(s)(rv)
def call(__self, __context, __obj, *args, **kwargs): # noqa: B902
"""Call an object from sandboxed code."""
fmt = inspect_format_method(__obj)
if fmt is not None:
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,))
return __context.call(__obj, *args, **kwargs)
class ImmutableSandboxedEnvironment(SandboxedEnvironment):
"""Works exactly like the regular `SandboxedEnvironment` but does not
permit modifications on the builtin mutable objects `list`, `set`, and
`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):
return False
return not modifies_known_mutable(obj, attr)
# This really is not a public API apparently.
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):
self._env = env
def get_field(self, field_name, args, kwargs):
first, rest = formatter_field_name_split(field_name)
obj = self.get_value(first, args, kwargs)
for is_attr, i in rest:
if is_attr:
obj = self._env.getattr(obj, i)
else:
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)
# -*- coding: utf-8 -*-
"""Built-in template tests used with the ``is`` operator."""
import decimal
import operator
import re
from ._compat import abc
from ._compat import integer_types
from ._compat import string_types
from ._compat import text_type
from .runtime import Undefined
number_re = re.compile(r"^-?\d+(\.\d+)?$")
regex_type = type(number_re)
test_callable = callable
def test_odd(value):
"""Return true if the variable is odd."""
return value % 2 == 1
def test_even(value):
"""Return true if the variable is even."""
return value % 2 == 0
def test_divisibleby(value, num):
"""Check if a variable is divisible by a number."""
return value % num == 0
def test_defined(value):
"""Return true if the variable is defined:
.. sourcecode:: jinja
{% if variable is defined %}
value of variable: {{ variable }}
{% else %}
variable is not defined
{% endif %}
See the :func:`default` filter for a simple way to set undefined
variables.
"""
return not isinstance(value, Undefined)
def test_undefined(value):
"""Like :func:`defined` but the other way round."""
return isinstance(value, Undefined)
def test_none(value):
"""Return true if the variable is none."""
return value is None
def test_boolean(value):
"""Return true if the object is a boolean value.
.. versionadded:: 2.11
"""
return value is True or value is False
def test_false(value):
"""Return true if the object is False.
.. versionadded:: 2.11
"""
return value is False
def test_true(value):
"""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):
"""Return true if the object is an integer.
.. versionadded:: 2.11
"""
return isinstance(value, integer_types) and value is not True and value is not False
# NOTE: The existing 'number' test matches booleans and integers
def test_float(value):
"""Return true if the object is a float.
.. versionadded:: 2.11
"""
return isinstance(value, float)
def test_lower(value):
"""Return true if the variable is lowercased."""
return text_type(value).islower()
def test_upper(value):
"""Return true if the variable is uppercased."""
return text_type(value).isupper()
def test_string(value):
"""Return true if the object is a string."""
return isinstance(value, string_types)
def test_mapping(value):
"""Return true if the object is a mapping (dict etc.).
.. versionadded:: 2.6
"""
return isinstance(value, abc.Mapping)
def test_number(value):
"""Return true if the variable is a number."""
return isinstance(value, integer_types + (float, complex, decimal.Decimal))
def test_sequence(value):
"""Return true if the variable is a sequence. Sequences are variables
that are iterable.
"""
try:
len(value)
value.__getitem__
except Exception:
return False
return True
def test_sameas(value, other):
"""Check if an object points to the same memory address than another
object:
.. sourcecode:: jinja
{% if foo.attribute is sameas false %}
the foo attribute really is the `False` singleton
{% endif %}
"""
return value is other
def test_iterable(value):
"""Check if it's possible to iterate over an object."""
try:
iter(value)
except TypeError:
return False
return True
def test_escaped(value):
"""Check if the value is escaped."""
return hasattr(value, "__html__")
def test_in(value, seq):
"""Check if value is in seq.
.. versionadded:: 2.10
"""
return value in seq
TESTS = {
"odd": test_odd,
"even": test_even,
"divisibleby": test_divisibleby,
"defined": test_defined,
"undefined": test_undefined,
"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": 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,
}
# -*- coding: utf-8 -*-
import json
import os
import re
import warnings
from collections import deque
from random import choice
from random import randrange
from threading import Lock
from markupsafe import escape
from markupsafe import Markup
from ._compat import abc
from ._compat import string_types
from ._compat import text_type
from ._compat import url_quote
_word_split_re = re.compile(r"(\s+)")
_punctuation_re = re.compile(
"^(?P<lead>(?:%s)*)(?P<middle>.*?)(?P<trail>(?:%s)*)$"
% (
"|".join(map(re.escape, ("(", "<", "&lt;"))),
"|".join(map(re.escape, (".", ",", ")", ">", "\n", "&gt;"))),
)
)
_simple_email_re = re.compile(r"^\S+@[a-zA-Z0-9._-]+\.[a-zA-Z0-9._-]+$")
_striptags_re = re.compile(r"(<!--.*?-->|<[^>]*>)")
_entity_re = re.compile(r"&([^;]+);")
_letters = "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ"
_digits = "0123456789"
# special singleton representing missing values for the runtime
missing = type("MissingType", (), {"__repr__": lambda x: "missing"})()
# internal code
internal_code = set()
concat = u"".join
_slash_escape = "\\/" not in json.dumps("/")
def contextfunction(f):
"""This decorator can be used to mark a function or method context callable.
A context callable is passed the active :class:`Context` as first argument when
called from the template. This is useful if a function wants to get access
to the context or functions provided on the context object. For example
a function that returns a sorted list of template variables the current
template exports could look like this::
@contextfunction
def get_exported_names(context):
return sorted(context.exported_vars)
"""
f.contextfunction = True
return f
def evalcontextfunction(f):
"""This decorator can be used to mark a function or method as an eval
context callable. This is similar to the :func:`contextfunction`
but instead of passing the context, an evaluation context object is
passed. For more information about the eval context, see
:ref:`eval-context`.
.. versionadded:: 2.4
"""
f.evalcontextfunction = True
return f
def environmentfunction(f):
"""This decorator can be used to mark a function or method as environment
callable. This decorator works exactly like the :func:`contextfunction`
decorator just that the first argument is the active :class:`Environment`
and not context.
"""
f.environmentfunction = True
return f
def internalcode(f):
"""Marks the function as internally used"""
internal_code.add(f.__code__)
return f
def is_undefined(obj):
"""Check if the object passed is undefined. This does nothing more than
performing an instance check against :class:`Undefined` but looks nicer.
This can be used for custom filters or tests that want to react to
undefined variables. For example a custom default filter can look like
this::
def default(var, default=''):
if is_undefined(var):
return default
return var
"""
from .runtime import Undefined
return isinstance(obj, Undefined)
def consume(iterable):
"""Consumes an iterable without doing anything with it."""
for _ in iterable:
pass
def clear_caches():
"""Jinja keeps internal caches for environments and lexers. These are
used so that Jinja doesn't have to recreate environments and lexers all
the time. Normally you don't have to care about that but if you are
measuring memory consumption you may want to clean the caches.
"""
from .environment import _spontaneous_environments
from .lexer import _lexer_cache
_spontaneous_environments.clear()
_lexer_cache.clear()
def import_string(import_name, silent=False):
"""Imports an object based on a string. This is useful if you want to
use import paths as endpoints or something similar. An import path can
be specified either in dotted notation (``xml.sax.saxutils.escape``)
or with a colon as object delimiter (``xml.sax.saxutils:escape``).
If the `silent` is True the return value will be `None` if the import
fails.
:return: imported object
"""
try:
if ":" in import_name:
module, obj = import_name.split(":", 1)
elif "." in import_name:
module, _, obj = import_name.rpartition(".")
else:
return __import__(import_name)
return getattr(__import__(module, None, None, [obj]), obj)
except (ImportError, AttributeError):
if not silent:
raise
def open_if_exists(filename, mode="rb"):
"""Returns a file descriptor for the filename if that file exists,
otherwise ``None``.
"""
if not os.path.isfile(filename):
return None
return open(filename, mode)
def object_type_repr(obj):
"""Returns the name of the object's type. For some recognized
singletons the name of the object is returned instead. (For
example for `None` and `Ellipsis`).
"""
if obj is None:
return "None"
elif obj is Ellipsis:
return "Ellipsis"
cls = type(obj)
# __builtin__ in 2.x, builtins in 3.x
if cls.__module__ in ("__builtin__", "builtins"):
name = cls.__name__
else:
name = cls.__module__ + "." + cls.__name__
return "%s object" % name
def pformat(obj, verbose=False):
"""Prettyprint an object. Either use the `pretty` library or the
builtin `pprint`.
"""
try:
from pretty import pretty
return pretty(obj, verbose=verbose)
except ImportError:
from pprint import pformat
return pformat(obj)
def urlize(text, trim_url_limit=None, rel=None, target=None):
"""Converts any URLs in text into clickable links. Works on http://,
https:// and www. links. Links can have trailing punctuation (periods,
commas, close-parens) and leading punctuation (opening parens) and
it'll still do the right thing.
If trim_url_limit is not None, the URLs in link text will be limited
to trim_url_limit characters.
If nofollow is True, the URLs in link text will get a rel="nofollow"
attribute.
If target is not None, a target attribute will be added to the link.
"""
trim_url = (
lambda x, limit=trim_url_limit: limit is not None
and (x[:limit] + (len(x) >= limit and "..." or ""))
or x
)
words = _word_split_re.split(text_type(escape(text)))
rel_attr = rel and ' rel="%s"' % text_type(escape(rel)) or ""
target_attr = target and ' target="%s"' % escape(target) or ""
for i, word in enumerate(words):
match = _punctuation_re.match(word)
if match:
lead, middle, trail = match.groups()
if middle.startswith("www.") or (
"@" not in middle
and not middle.startswith("http://")
and not middle.startswith("https://")
and len(middle) > 0
and middle[0] in _letters + _digits
and (
middle.endswith(".org")
or middle.endswith(".net")
or middle.endswith(".com")
)
):
middle = '<a href="http://%s"%s%s>%s</a>' % (
middle,
rel_attr,
target_attr,
trim_url(middle),
)
if middle.startswith("http://") or middle.startswith("https://"):
middle = '<a href="%s"%s%s>%s</a>' % (
middle,
rel_attr,
target_attr,
trim_url(middle),
)
if (
"@" in middle
and not middle.startswith("www.")
and ":" not in middle
and _simple_email_re.match(middle)
):
middle = '<a href="mailto:%s">%s</a>' % (middle, middle)
if lead + middle + trail != word:
words[i] = lead + middle + trail
return u"".join(words)
def generate_lorem_ipsum(n=5, html=True, min=20, max=100):
"""Generate some lorem ipsum for the template."""
from .constants import LOREM_IPSUM_WORDS
words = LOREM_IPSUM_WORDS.split()
result = []
for _ in range(n):
next_capitalized = True
last_comma = last_fullstop = 0
word = None
last = None
p = []
# each paragraph contains out of 20 to 100 words.
for idx, _ in enumerate(range(randrange(min, max))):
while True:
word = choice(words)
if word != last:
last = word
break
if next_capitalized:
word = word.capitalize()
next_capitalized = False
# add commas
if idx - randrange(3, 8) > last_comma:
last_comma = idx
last_fullstop += 2
word += ","
# add end of sentences
if idx - randrange(10, 20) > last_fullstop:
last_comma = last_fullstop = idx
word += "."
next_capitalized = True
p.append(word)
# ensure that the paragraph ends with a dot.
p = u" ".join(p)
if p.endswith(","):
p = p[:-1] + "."
elif not p.endswith("."):
p += "."
result.append(p)
if not html:
return u"\n\n".join(result)
return Markup(u"\n".join(u"<p>%s</p>" % escape(x) for x in result))
def unicode_urlencode(obj, charset="utf-8", for_qs=False):
"""Quote a string for use in a URL using the given charset.
This function is misnamed, it is a wrapper around
:func:`urllib.parse.quote`.
:param obj: String or bytes to quote. Other types are converted to
string then encoded to bytes using the given charset.
:param charset: Encode text to bytes using this charset.
:param for_qs: Quote "/" and use "+" for spaces.
"""
if not isinstance(obj, string_types):
obj = text_type(obj)
if isinstance(obj, text_type):
obj = obj.encode(charset)
safe = b"" if for_qs else b"/"
rv = url_quote(obj, safe)
if not isinstance(rv, text_type):
rv = rv.decode("utf-8")
if for_qs:
rv = rv.replace("%20", "+")
return rv
class LRUCache(object):
"""A simple LRU Cache implementation."""
# this is fast for small capacities (something below 1000) but doesn't
# scale. But as long as it's only used as storage for templates this
# won't do any harm.
def __init__(self, capacity):
self.capacity = capacity
self._mapping = {}
self._queue = deque()
self._postinit()
def _postinit(self):
# alias all queue methods for faster lookup
self._popleft = self._queue.popleft
self._pop = self._queue.pop
self._remove = self._queue.remove
self._wlock = Lock()
self._append = self._queue.append
def __getstate__(self):
return {
"capacity": self.capacity,
"_mapping": self._mapping,
"_queue": self._queue,
}
def __setstate__(self, d):
self.__dict__.update(d)
self._postinit()
def __getnewargs__(self):
return (self.capacity,)
def copy(self):
"""Return a shallow copy of the instance."""
rv = self.__class__(self.capacity)
rv._mapping.update(self._mapping)
rv._queue.extend(self._queue)
return rv
def get(self, key, default=None):
"""Return an item from the cache dict or `default`"""
try:
return self[key]
except KeyError:
return default
def setdefault(self, key, default=None):
"""Set `default` if the key is not in the cache otherwise
leave unchanged. Return the value of this key.
"""
try:
return self[key]
except KeyError:
self[key] = default
return default
def clear(self):
"""Clear the cache."""
self._wlock.acquire()
try:
self._mapping.clear()
self._queue.clear()
finally:
self._wlock.release()
def __contains__(self, key):
"""Check if a key exists in this cache."""
return key in self._mapping
def __len__(self):
"""Return the current size of the cache."""
return len(self._mapping)
def __repr__(self):
return "<%s %r>" % (self.__class__.__name__, self._mapping)
def __getitem__(self, key):
"""Get an item from the cache. Moves the item up so that it has the
highest priority then.
Raise a `KeyError` if it does not exist.
"""
self._wlock.acquire()
try:
rv = self._mapping[key]
if self._queue[-1] != key:
try:
self._remove(key)
except ValueError:
# if something removed the key from the container
# when we read, ignore the ValueError that we would
# get otherwise.
pass
self._append(key)
return rv
finally:
self._wlock.release()
def __setitem__(self, key, value):
"""Sets the value for an item. Moves the item up so that it
has the highest priority then.
"""
self._wlock.acquire()
try:
if key in self._mapping:
self._remove(key)
elif len(self._mapping) == self.capacity:
del self._mapping[self._popleft()]
self._append(key)
self._mapping[key] = value
finally:
self._wlock.release()
def __delitem__(self, key):
"""Remove an item from the cache dict.
Raise a `KeyError` if it does not exist.
"""
self._wlock.acquire()
try:
del self._mapping[key]
try:
self._remove(key)
except ValueError:
pass
finally:
self._wlock.release()
def items(self):
"""Return a list of items."""
result = [(key, self._mapping[key]) for key in list(self._queue)]
result.reverse()
return result
def iteritems(self):
"""Iterate over all items."""
warnings.warn(
"'iteritems()' will be removed in version 3.0. Use"
" 'iter(cache.items())' instead.",
DeprecationWarning,
stacklevel=2,
)
return iter(self.items())
def values(self):
"""Return a list of all values."""
return [x[1] for x in self.items()]
def itervalue(self):
"""Iterate over all values."""
warnings.warn(
"'itervalue()' will be removed in version 3.0. Use"
" 'iter(cache.values())' instead.",
DeprecationWarning,
stacklevel=2,
)
return iter(self.values())
def itervalues(self):
"""Iterate over all values."""
warnings.warn(
"'itervalues()' will be removed in version 3.0. Use"
" 'iter(cache.values())' instead.",
DeprecationWarning,
stacklevel=2,
)
return iter(self.values())
def keys(self):
"""Return a list of all keys ordered by most recent usage."""
return list(self)
def iterkeys(self):
"""Iterate over all keys in the cache dict, ordered by
the most recent usage.
"""
warnings.warn(
"'iterkeys()' will be removed in version 3.0. Use"
" 'iter(cache.keys())' instead.",
DeprecationWarning,
stacklevel=2,
)
return iter(self)
def __iter__(self):
return reversed(tuple(self._queue))
def __reversed__(self):
"""Iterate over the keys in the cache dict, oldest items
coming first.
"""
return iter(tuple(self._queue))
__copy__ = copy
abc.MutableMapping.register(LRUCache)
def select_autoescape(
enabled_extensions=("html", "htm", "xml"),
disabled_extensions=(),
default_for_string=True,
default=False,
):
"""Intelligently sets the initial value of autoescaping based on the
filename of the template. This is the recommended way to configure
autoescaping if you do not want to write a custom function yourself.
If you want to enable it for all templates created from strings or
for all templates with `.html` and `.xml` extensions::
from jinja2 import Environment, select_autoescape
env = Environment(autoescape=select_autoescape(
enabled_extensions=('html', 'xml'),
default_for_string=True,
))
Example configuration to turn it on at all times except if the template
ends with `.txt`::
from jinja2 import Environment, select_autoescape
env = Environment(autoescape=select_autoescape(
disabled_extensions=('txt',),
default_for_string=True,
default=True,
))
The `enabled_extensions` is an iterable of all the extensions that
autoescaping should be enabled for. Likewise `disabled_extensions` is
a list of all templates it should be disabled for. If a template is
loaded from a string then the default from `default_for_string` is used.
If nothing matches then the initial value of autoescaping is set to the
value of `default`.
For security reasons this function operates case insensitive.
.. versionadded:: 2.9
"""
enabled_patterns = tuple("." + x.lstrip(".").lower() for x in enabled_extensions)
disabled_patterns = tuple("." + x.lstrip(".").lower() for x in disabled_extensions)
def autoescape(template_name):
if template_name is None:
return default_for_string
template_name = template_name.lower()
if template_name.endswith(enabled_patterns):
return True
if template_name.endswith(disabled_patterns):
return False
return default
return autoescape
def htmlsafe_json_dumps(obj, dumper=None, **kwargs):
"""Works exactly like :func:`dumps` but is safe for use in ``<script>``
tags. It accepts the same arguments and returns a JSON string. Note that
this is available in templates through the ``|tojson`` filter which will
also mark the result as safe. Due to how this function escapes certain
characters this is safe even if used outside of ``<script>`` tags.
The following characters are escaped in strings:
- ``<``
- ``>``
- ``&``
- ``'``
This makes it safe to embed such strings in any place in HTML with the
notable exception of double quoted attributes. In that case single
quote your attributes or HTML escape it in addition.
"""
if dumper is None:
dumper = json.dumps
rv = (
dumper(obj, **kwargs)
.replace(u"<", u"\\u003c")
.replace(u">", u"\\u003e")
.replace(u"&", u"\\u0026")
.replace(u"'", u"\\u0027")
)
return Markup(rv)
class Cycler(object):
"""Cycle through values by yield them one at a time, then restarting
once the end is reached. Available as ``cycler`` in templates.
Similar to ``loop.cycle``, but can be used outside loops or across
multiple loops. For example, render a list of folders and files in a
list, alternating giving them "odd" and "even" classes.
.. code-block:: html+jinja
{% set row_class = cycler("odd", "even") %}
<ul class="browser">
{% for folder in folders %}
<li class="folder {{ row_class.next() }}">{{ folder }}
{% endfor %}
{% for file in files %}
<li class="file {{ row_class.next() }}">{{ file }}
{% endfor %}
</ul>
:param items: Each positional argument will be yielded in the order
given for each cycle.
.. versionadded:: 2.1
"""
def __init__(self, *items):
if not items:
raise RuntimeError("at least one item has to be provided")
self.items = items
self.pos = 0
def reset(self):
"""Resets the current item to the first item."""
self.pos = 0
@property
def current(self):
"""Return the current item. Equivalent to the item that will be
returned next time :meth:`next` is called.
"""
return self.items[self.pos]
def next(self):
"""Return the current item, then advance :attr:`current` to the
next item.
"""
rv = self.current
self.pos = (self.pos + 1) % len(self.items)
return rv
__next__ = next
class Joiner(object):
"""A joining helper for templates."""
def __init__(self, sep=u", "):
self.sep = sep
self.used = False
def __call__(self):
if not self.used:
self.used = True
return u""
return self.sep
class Namespace(object):
"""A namespace object that can hold arbitrary attributes. It may be
initialized from a dictionary or with keyword arguments."""
def __init__(*args, **kwargs): # noqa: B902
self, args = args[0], args[1:]
self.__attrs = dict(*args, **kwargs)
def __getattribute__(self, name):
# __class__ is needed for the awaitable check in async mode
if name in {"_Namespace__attrs", "__class__"}:
return object.__getattribute__(self, name)
try:
return self.__attrs[name]
except KeyError:
raise AttributeError(name)
def __setitem__(self, name, value):
self.__attrs[name] = value
def __repr__(self):
return "<Namespace %r>" % self.__attrs
# does this python version support async for in and async generators?
try:
exec("async def _():\n async for _ in ():\n yield _")
have_async_gen = True
except SyntaxError:
have_async_gen = False
def soft_unicode(s):
from markupsafe import soft_unicode
warnings.warn(
"'jinja2.utils.soft_unicode' will be removed in version 3.0."
" Use 'markupsafe.soft_unicode' instead.",
DeprecationWarning,
stacklevel=2,
)
return soft_unicode(s)
# -*- coding: utf-8 -*-
"""API for traversing the AST nodes. Implemented by the compiler and
meta introspection.
"""
from .nodes import Node
class NodeVisitor(object):
"""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.
Per default the visitor functions for the nodes are ``'visit_'`` +
class name of the node. So a `TryFinally` node visit function would
be `visit_TryFinally`. This behavior can be changed by overriding
the `get_visitor` function. If no visitor function exists for a node
(return value `None`) the `generic_visit` visitor is used instead.
"""
def get_visitor(self, node):
"""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)
def visit(self, node, *args, **kwargs):
"""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):
"""Called if no explicit visitor function exists for a node."""
for node in node.iter_child_nodes():
self.visit(node, *args, **kwargs)
class NodeTransformer(NodeVisitor):
"""Walks the abstract syntax tree and allows modifications of nodes.
The `NodeTransformer` will walk the AST and use the return value of the
visitor functions to replace or remove the old node. If the return
value of the visitor function is `None` the node will be removed
from the previous location otherwise it's replaced with the return
value. The return value may be the original node in which case no
replacement takes place.
"""
def generic_visit(self, node, *args, **kwargs):
for field, old_value in node.iter_fields():
if isinstance(old_value, list):
new_values = []
for value in old_value:
if isinstance(value, Node):
value = self.visit(value, *args, **kwargs)
if value is None:
continue
elif not isinstance(value, Node):
new_values.extend(value)
continue
new_values.append(value)
old_value[:] = new_values
elif isinstance(old_value, Node):
new_node = self.visit(old_value, *args, **kwargs)
if new_node is None:
delattr(node, field)
else:
setattr(node, field, new_node)
return node
def visit_list(self, node, *args, **kwargs):
"""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
# -*- coding: utf-8 -*-
"""
markupsafe
~~~~~~~~~~
Implements an escape function and a Markup string to replace HTML
special characters with safe representations.
:copyright: 2010 Pallets
:license: BSD-3-Clause
"""
import re
import string
from ._compat import int_types
from ._compat import iteritems
from ._compat import Mapping
from ._compat import PY2
from ._compat import string_types
from ._compat import text_type
from ._compat import unichr
__version__ = "1.1.1"
__all__ = ["Markup", "soft_unicode", "escape", "escape_silent"]
_striptags_re = re.compile(r"(<!--.*?-->|<[^>]*>)")
_entity_re = re.compile(r"&([^& ;]+);")
class Markup(text_type):
"""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('<a href="/foo">foo</a>')
This is a subclass of the text type (``str`` in Python 3,
``unicode`` in Python 2). It has the same methods as that type, but
all methods escape their arguments and return a ``Markup`` instance.
>>> Markup('<em>%s</em>') % 'foo & bar'
Markup('<em>foo &amp; bar</em>')
>>> Markup('<em>Hello</em> ') + '<foo>'
Markup('<em>Hello</em> &lt;foo&gt;')
"""
__slots__ = ()
def __new__(cls, base=u"", encoding=None, errors="strict"):
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):
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)))
return NotImplemented
def __radd__(self, other):
if hasattr(other, "__html__") or isinstance(other, string_types):
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))
return NotImplemented
__rmul__ = __mul__
def __mod__(self, arg):
if isinstance(arg, tuple):
arg = tuple(_MarkupEscapeHelper(x, self.escape) for x in arg)
else:
arg = _MarkupEscapeHelper(arg, self.escape)
return self.__class__(text_type.__mod__(self, arg))
def __repr__(self):
return "%s(%s)" % (self.__class__.__name__, text_type.__repr__(self))
def join(self, seq):
return self.__class__(text_type.join(self, map(self.escape, seq)))
join.__doc__ = text_type.join.__doc__
def split(self, *args, **kwargs):
return list(map(self.__class__, text_type.split(self, *args, **kwargs)))
split.__doc__ = text_type.split.__doc__
def rsplit(self, *args, **kwargs):
return list(map(self.__class__, text_type.rsplit(self, *args, **kwargs)))
rsplit.__doc__ = text_type.rsplit.__doc__
def splitlines(self, *args, **kwargs):
return list(map(self.__class__, text_type.splitlines(self, *args, **kwargs)))
splitlines.__doc__ = text_type.splitlines.__doc__
def unescape(self):
"""Convert escaped markup back into a text string. This replaces
HTML entities with the characters they represent.
>>> Markup('Main &raquo; <em>About</em>').unescape()
'Main » <em>About</em>'
"""
from ._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))
def striptags(self):
""":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()
@classmethod
def escape(cls, s):
"""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): # noqa: B902
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] = make_simple_escaping_wrapper(method)
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 format(self, *args, **kwargs):
formatter = EscapeFormatter(self.escape)
kwargs = _MagicFormatMapping(args, kwargs)
return self.__class__(formatter.vformat(self, args, kwargs))
def __html_format__(self, format_spec):
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
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)
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(
"Format specifier {0} given, but {1} does not"
" define __html_format__. A class that defines"
" __html__ must define __html_format__ to work"
" with format specifiers.".format(format_spec, type(value))
)
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):
"""Helper for various string-wrapped functions."""
for key, value in iterable:
if hasattr(value, "__html__") or isinstance(value, string_types):
obj[key] = escape(value)
return obj
class _MarkupEscapeHelper(object):
"""Helper for Markup.__mod__"""
def __init__(self, obj, escape):
self.obj = obj
self.escape = escape
def __getitem__(self, item):
return _MarkupEscapeHelper(self.obj[item], self.escape)
def __str__(self):
return text_type(self.escape(self.obj))
__unicode__ = __str__
def __repr__(self):
return str(self.escape(repr(self.obj)))
def __int__(self):
return int(self.obj)
def __float__(self):
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.
try:
from ._speedups import escape, escape_silent, soft_unicode
except ImportError:
from ._native import escape, escape_silent, soft_unicode
if not PY2:
soft_str = soft_unicode
__all__.append("soft_str")
# -*- coding: utf-8 -*-
"""
markupsafe._compat
~~~~~~~~~~~~~~~~~~
:copyright: 2010 Pallets
:license: BSD-3-Clause
"""
import sys
PY2 = sys.version_info[0] == 2
if not PY2:
text_type = str
string_types = (str,)
unichr = chr
int_types = (int,)
def iteritems(x):
return iter(x.items())
from collections.abc import Mapping
else:
text_type = unicode
string_types = (str, unicode)
unichr = unichr
int_types = (int, long)
def iteritems(x):
return x.iteritems()
from collections import Mapping
# -*- coding: utf-8 -*-
"""
markupsafe._constants
~~~~~~~~~~~~~~~~~~~~~
:copyright: 2010 Pallets
:license: BSD-3-Clause
"""
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,
}
# -*- coding: utf-8 -*-
"""
markupsafe._native
~~~~~~~~~~~~~~~~~~
Native Python implementation used when the C module is not compiled.
:copyright: 2010 Pallets
:license: BSD-3-Clause
"""
from . import Markup
from ._compat import text_type
def escape(s):
"""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 Markup(s.__html__())
return Markup(
text_type(s)
.replace("&", "&amp;")
.replace(">", "&gt;")
.replace("<", "&lt;")
.replace("'", "&#39;")
.replace('"', "&#34;")
)
def escape_silent(s):
"""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):
"""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_unicode(value))
Markup('&lt;User 1&gt;')
"""
if not isinstance(s, text_type):
s = text_type(s)
return s
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment