Utility functions.
- :copyright: 2008 by Armin Ronacher.
+ :copyright: (c) 2010 by the Jinja Team.
:license: BSD, see LICENSE for more details.
"""
import re
import sys
-import string
+import errno
try:
from thread import allocate_lock
except ImportError:
from dummy_thread import allocate_lock
-from htmlentitydefs import name2codepoint
from collections import deque
-from copy import deepcopy
from itertools import imap
_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'&([^;]+);')
-_entities = name2codepoint.copy()
-_entities['apos'] = 39
+_letters = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ'
+_digits = '0123456789'
# special singleton representing missing values for the runtime
missing = type('MissingType', (), {'__repr__': lambda x: 'missing'})()
+# internal code
+internal_code = set()
+
# concatenate a list of strings and convert them to unicode.
# unfortunately there is a bug in python 2.4 and lower that causes
def concat(gen):
try:
return _concat(list(gen))
- except:
+ except Exception:
# this hack is needed so that the current frame
# does not show up in the traceback.
exc_type, exc_value, tb = sys.exc_info()
del _test_gen_bug, _error
+# for python 2.x we create outselves a next() function that does the
+# basics without exception catching.
+try:
+ next = next
+except NameError:
+ def next(x):
+ return x.next()
+
+
+# if this python version is unable to deal with unicode filenames
+# when passed to encode we let this function encode it properly.
+# This is used in a couple of places. As far as Jinja is concerned
+# filenames are unicode *or* bytestrings in 2.x and unicode only in
+# 3.x because compile cannot handle bytes
+if sys.version_info < (3, 0):
+ def _encode_filename(filename):
+ if isinstance(filename, unicode):
+ return filename.encode('utf-8')
+ return filename
+else:
+ def _encode_filename(filename):
+ assert filename is None or isinstance(filename, str), \
+ 'filenames must be strings'
+ return filename
+
+from keyword import iskeyword as is_python_keyword
+
+
+# common types. These do exist in the special types module too which however
+# does not exist in IronPython out of the box. Also that way we don't have
+# to deal with implementation specific stuff here
+class _C(object):
+ def method(self): pass
+def _func():
+ yield None
+FunctionType = type(_func)
+GeneratorType = type(_func())
+MethodType = type(_C.method)
+CodeType = type(_C.method.func_code)
+try:
+ raise TypeError()
+except TypeError:
+ _tb = sys.exc_info()[2]
+ TracebackType = type(_tb)
+ FrameType = type(_tb.tb_frame)
+del _C, _tb, _func
+
+
def contextfunction(f):
- """This decorator can be used to mark a callable as context callable. A
- context callable is passed the active context as first argument if it
- was directly stored in the context.
+ """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 decoraotr 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 callable as environment callable.
- A environment callable is passed the current environment as first argument
- if it was directly stored in the context.
+ """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.func_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 jinja2.runtime import Undefined
+ return isinstance(obj, Undefined)
+
+
+def consume(iterable):
+ """Consumes an iterable without doing anything with it."""
+ for event in iterable:
+ pass
+
+
def clear_caches():
"""Jinja2 keeps internal caches for environments and lexers. These are
used so that Jinja2 doesn't have to recreate environments and lexers all
raise
+def open_if_exists(filename, mode='rb'):
+ """Returns a file descriptor for the filename if that file exists,
+ otherwise `None`.
+ """
+ try:
+ return open(filename, mode)
+ except IOError, e:
+ if e.errno not in (errno.ENOENT, errno.EISDIR):
+ raise
+
+
+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'
+ # __builtin__ in 2.x, builtins in 3.x
+ if obj.__class__.__module__ in ('__builtin__', 'builtins'):
+ name = obj.__class__.__name__
+ else:
+ name = obj.__class__.__module__ + '.' + obj.__class__.__name__
+ return '%s object' % name
+
+
def pformat(obj, verbose=False):
"""Prettyprint an object. Either use the `pretty` library or the
builtin `pprint`.
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)
+ words = _word_split_re.split(unicode(escape(text)))
nofollow_attr = nofollow and ' rel="nofollow"' or ''
for i, word in enumerate(words):
match = _punctuation_re.match(word)
'@' not in middle and
not middle.startswith('http://') and
len(middle) > 0 and
- middle[0] in string.letters + string.digits and (
+ middle[0] in _letters + _digits and (
middle.endswith('.org') or
middle.endswith('.net') or
middle.endswith('.com')
def generate_lorem_ipsum(n=5, html=True, min=20, max=100):
"""Generate some lorem impsum for the template."""
from jinja2.constants import LOREM_IPSUM_WORDS
- from random import choice, random, randrange
+ from random import choice, randrange
words = LOREM_IPSUM_WORDS.split()
result = []
return Markup(u'\n'.join(u'<p>%s</p>' % escape(x) for x in result))
-class Markup(unicode):
- """Marks a string as being safe for inclusion in HTML/XML output without
- needing to be escaped. This implements the `__html__` interface a couple
- of frameworks and web applications use.
-
- The `escape` function returns markup objects so that double escaping can't
- happen. If you want to use autoescaping in Jinja just set the finalizer
- of the environment to `escape`.
- """
- __slots__ = ()
-
- def __html__(self):
- return self
-
- def __add__(self, other):
- if hasattr(other, '__html__') or isinstance(other, basestring):
- return self.__class__(unicode(self) + unicode(escape(other)))
- return NotImplemented
-
- def __radd__(self, other):
- if hasattr(other, '__html__') or isinstance(other, basestring):
- return self.__class__(unicode(escape(other)) + unicode(self))
- return NotImplemented
-
- def __mul__(self, num):
- if not isinstance(num, (int, long)):
- return NotImplemented
- return self.__class__(unicode.__mul__(self, num))
- __rmul__ = __mul__
-
- def __mod__(self, arg):
- if isinstance(arg, tuple):
- arg = tuple(imap(_MarkupEscapeHelper, arg))
- else:
- arg = _MarkupEscapeHelper(arg)
- return self.__class__(unicode.__mod__(self, arg))
-
- def __repr__(self):
- return '%s(%s)' % (
- self.__class__.__name__,
- unicode.__repr__(self)
- )
-
- def join(self, seq):
- return self.__class__(unicode.join(self, imap(escape, seq)))
- join.__doc__ = unicode.join.__doc__
-
- def split(self, *args, **kwargs):
- return map(self.__class__, unicode.split(self, *args, **kwargs))
- split.__doc__ = unicode.split.__doc__
-
- def rsplit(self, *args, **kwargs):
- return map(self.__class__, unicode.rsplit(self, *args, **kwargs))
- rsplit.__doc__ = unicode.rsplit.__doc__
-
- def splitlines(self, *args, **kwargs):
- return map(self.__class__, unicode.splitlines(self, *args, **kwargs))
- splitlines.__doc__ = unicode.splitlines.__doc__
-
- def unescape(self):
- """Unescape markup."""
- def handle_match(m):
- name = m.group(1)
- if name in _entities:
- return unichr(_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
- return u''
- return _entity_re.sub(handle_match, unicode(self))
-
- def striptags(self):
- """Strip tags and resolve enities."""
- stripped = u' '.join(_striptags_re.sub('', self).split())
- return Markup(stripped).unescape()
-
- def make_wrapper(name):
- orig = getattr(unicode, name)
- def func(self, *args, **kwargs):
- args = list(args)
- for idx, arg in enumerate(args):
- if hasattr(arg, '__html__') or isinstance(arg, basestring):
- args[idx] = escape(arg)
- for name, arg in kwargs.iteritems():
- if hasattr(arg, '__html__') or isinstance(arg, basestring):
- kwargs[name] = escape(arg)
- return self.__class__(orig(self, *args, **kwargs))
- func.__name__ = orig.__name__
- func.__doc__ = orig.__doc__
- return func
- for method in '__getitem__', '__getslice__', 'capitalize', \
- 'title', 'lower', 'upper', 'replace', 'ljust', \
- 'rjust', 'lstrip', 'rstrip', 'center', 'strip', \
- 'translate', 'expandtabs', 'swapcase', 'zfill':
- locals()[method] = make_wrapper(method)
-
- # new in python 2.5
- if hasattr(unicode, 'partition'):
- locals().update(
- partition=make_wrapper('partition'),
- rpartition=make_wrapper('rpartition')
- )
- del method, make_wrapper
-
-
-class _MarkupEscapeHelper(object):
- """Helper for Markup.__mod__"""
-
- def __init__(self, obj):
- self.obj = obj
-
- __getitem__ = lambda s, x: _MarkupEscapeHelper(s.obj[x])
- __unicode__ = lambda s: unicode(escape(s.obj))
- __str__ = lambda s: str(escape(s.obj))
- __repr__ = lambda s: str(repr(escape(s.obj)))
- __int__ = lambda s: int(s.obj)
- __float__ = lambda s: float(s.obj)
-
-
class LRUCache(object):
"""A simple LRU Cache implementation."""
- # this is fast for small capacities (something around 200) but doesn't
- # scale. But as long as it's only used for the database connections in
- # a non request fallback it's fine.
+
+ # 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
del self._queue[idx]
break
+ 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 an shallow copy of the instance."""
rv = self.__class__(self.capacity)
"""
rv = self._mapping[key]
if self._queue[-1] != key:
- self._remove(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
self._wlock.acquire()
try:
if key in self._mapping:
- self._remove(key)
+ try:
+ self._remove(key)
+ except ValueError:
+ # __getitem__ is not locked, it might happen
+ pass
elif len(self._mapping) == self.capacity:
del self._mapping[self._popleft()]
self._append(key)
self._wlock.acquire()
try:
del self._mapping[key]
- self._remove(key)
+ try:
+ self._remove(key)
+ except ValueError:
+ # __getitem__ is not locked, it might happen
+ pass
finally:
self._wlock.release()
- def __iter__(self):
- """Iterate over all values in the cache dict, ordered by
+ 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."""
+ 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."""
+ 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.
"""
- return reversed(self._queue)
+ return reversed(tuple(self._queue))
+
+ __iter__ = iterkeys
def __reversed__(self):
"""Iterate over the values in the cache dict, oldest items
coming first.
"""
- return iter(self._queue)
+ return iter(tuple(self._queue))
__copy__ = copy
-# we have to import it down here as the speedups module imports the
-# markup type which is define above.
+# register the LRU cache as mutable mapping if possible
try:
- from jinja2._speedups import escape, soft_unicode
+ from collections import MutableMapping
+ MutableMapping.register(LRUCache)
except ImportError:
- def escape(s):
- """Convert the characters &, <, >, and " in string s to HTML-safe
- sequences. Use this if you need to display text that might contain
- such characters in HTML. Marks return value as markup string.
- """
- if hasattr(s, '__html__'):
- return s.__html__()
- return Markup(unicode(s)
- .replace('&', '&')
- .replace('>', '>')
- .replace('<', '<')
- .replace('"', '"')
- )
+ pass
- def soft_unicode(s):
- """Make a string unicode if it isn't already. That way a markup
- string is not converted back to unicode.
- """
- if not isinstance(s, unicode):
- s = unicode(s)
- return s
+
+class Cycler(object):
+ """A cycle helper for templates."""
+
+ def __init__(self, *items):
+ if not items:
+ raise RuntimeError('at least one item has to be provided')
+ self.items = items
+ self.reset()
+
+ def reset(self):
+ """Resets the cycle."""
+ self.pos = 0
+
+ @property
+ def current(self):
+ """Returns the current item."""
+ return self.items[self.pos]
+
+ def next(self):
+ """Goes one item ahead and returns it."""
+ rv = self.current
+ self.pos = (self.pos + 1) % len(self.items)
+ return rv
+
+
+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
+
+
+# try markupsafe first, if that fails go with Jinja2's bundled version
+# of markupsafe. Markupsafe was previously Jinja2's implementation of
+# the Markup object but was moved into a separate package in a patchleve
+# release
+try:
+ from markupsafe import Markup, escape, soft_unicode
+except ImportError:
+ from jinja2._markupsafe import Markup, escape, soft_unicode
# partials