##// END OF EJS Templates
caches: use repo.lru based Dict cache. This LRUDict uses Timing Algo to not have to use locking...
caches: use repo.lru based Dict cache. This LRUDict uses Timing Algo to not have to use locking for the LRU implementation, this it's safer to use for dogpile. We used it before with beaker, so it's generally more stable.

File last commit:

r2939:913f92bf default
r2945:ec5716e4 default
Show More
utils.py
320 lines | 11.2 KiB | text/x-python | PythonLexer
# -*- coding: utf-8 -*-
# Copyright (C) 2015-2018 RhodeCode GmbH
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU Affero General Public License, version 3
# (only), as published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU Affero General Public License
# along with this program. If not, see <http://www.gnu.org/licenses/>.
#
# This program is dual-licensed. If you wish to learn more about the
# RhodeCode Enterprise Edition, including its added features, Support services,
# and proprietary license terms, please see https://rhodecode.com/licenses/
import os
import time
import logging
import functools
import threading
from dogpile.cache import CacheRegion
from dogpile.cache.util import compat
import rhodecode
from rhodecode.lib.utils import safe_str, sha1
from rhodecode.lib.utils2 import safe_unicode, str2bool
from rhodecode.model.db import Session, CacheKey, IntegrityError
from . import region_meta
log = logging.getLogger(__name__)
class RhodeCodeCacheRegion(CacheRegion):
def conditional_cache_on_arguments(
self, namespace=None,
expiration_time=None,
should_cache_fn=None,
to_str=compat.string_type,
function_key_generator=None,
condition=True):
"""
Custom conditional decorator, that will not touch any dogpile internals if
condition isn't meet. This works a bit different than should_cache_fn
And it's faster in cases we don't ever want to compute cached values
"""
expiration_time_is_callable = compat.callable(expiration_time)
if function_key_generator is None:
function_key_generator = self.function_key_generator
def decorator(fn):
if to_str is compat.string_type:
# backwards compatible
key_generator = function_key_generator(namespace, fn)
else:
key_generator = function_key_generator(namespace, fn, to_str=to_str)
@functools.wraps(fn)
def decorate(*arg, **kw):
key = key_generator(*arg, **kw)
@functools.wraps(fn)
def creator():
return fn(*arg, **kw)
if not condition:
return creator()
timeout = expiration_time() if expiration_time_is_callable \
else expiration_time
return self.get_or_create(key, creator, timeout, should_cache_fn)
def invalidate(*arg, **kw):
key = key_generator(*arg, **kw)
self.delete(key)
def set_(value, *arg, **kw):
key = key_generator(*arg, **kw)
self.set(key, value)
def get(*arg, **kw):
key = key_generator(*arg, **kw)
return self.get(key)
def refresh(*arg, **kw):
key = key_generator(*arg, **kw)
value = fn(*arg, **kw)
self.set(key, value)
return value
decorate.set = set_
decorate.invalidate = invalidate
decorate.refresh = refresh
decorate.get = get
decorate.original = fn
decorate.key_generator = key_generator
return decorate
return decorator
def make_region(*arg, **kw):
return RhodeCodeCacheRegion(*arg, **kw)
def get_default_cache_settings(settings, prefixes=None):
prefixes = prefixes or []
cache_settings = {}
for key in settings.keys():
for prefix in prefixes:
if key.startswith(prefix):
name = key.split(prefix)[1].strip()
val = settings[key]
if isinstance(val, basestring):
val = val.strip()
cache_settings[name] = val
return cache_settings
def compute_key_from_params(*args):
"""
Helper to compute key from given params to be used in cache manager
"""
return sha1("_".join(map(safe_str, args)))
def key_generator(namespace, fn):
fname = fn.__name__
def generate_key(*args):
namespace_pref = namespace or 'default'
arg_key = compute_key_from_params(*args)
final_key = "{}:{}_{}".format(namespace_pref, fname, arg_key)
return final_key
return generate_key
def get_or_create_region(region_name, region_namespace=None):
from rhodecode.lib.rc_cache.backends import FileNamespaceBackend
region_obj = region_meta.dogpile_cache_regions.get(region_name)
if not region_obj:
raise EnvironmentError(
'Region `{}` not in configured: {}.'.format(
region_name, region_meta.dogpile_cache_regions.keys()))
region_uid_name = '{}:{}'.format(region_name, region_namespace)
if isinstance(region_obj.actual_backend, FileNamespaceBackend):
region_exist = region_meta.dogpile_cache_regions.get(region_namespace)
if region_exist:
log.debug('Using already configured region: %s', region_namespace)
return region_exist
cache_dir = region_meta.dogpile_config_defaults['cache_dir']
expiration_time = region_obj.expiration_time
if not os.path.isdir(cache_dir):
os.makedirs(cache_dir)
new_region = make_region(
name=region_uid_name, function_key_generator=key_generator
)
namespace_filename = os.path.join(
cache_dir, "{}.cache.dbm".format(region_namespace))
# special type that allows 1db per namespace
new_region.configure(
backend='dogpile.cache.rc.file_namespace',
expiration_time=expiration_time,
arguments={"filename": namespace_filename}
)
# create and save in region caches
log.debug('configuring new region: %s',region_uid_name)
region_obj = region_meta.dogpile_cache_regions[region_namespace] = new_region
return region_obj
def clear_cache_namespace(cache_region, cache_namespace_uid):
region = get_or_create_region(cache_region, cache_namespace_uid)
cache_keys = region.backend.list_keys(prefix=cache_namespace_uid)
region.delete_multi(cache_keys)
return len(cache_keys)
class ActiveRegionCache(object):
def __init__(self, context):
self.context = context
def should_invalidate(self):
return False
class FreshRegionCache(object):
def __init__(self, context):
self.context = context
def should_invalidate(self):
return True
class InvalidationContext(object):
"""
usage::
from rhodecode.lib import rc_cache
cache_namespace_uid = CacheKey.SOME_NAMESPACE.format(1)
region = rc_cache.get_or_create_region('cache_perms', cache_namespace_uid)
@region.conditional_cache_on_arguments(namespace=cache_namespace_uid, condition=True)
def heavy_compute(cache_name, param1, param2):
print('COMPUTE {}, {}, {}'.format(cache_name, param1, param2))
# invalidation namespace is shared namespace key for all process caches
# we use it to send a global signal
invalidation_namespace = 'repo_cache:1'
inv_context_manager = rc_cache.InvalidationContext(
uid=cache_namespace_uid, invalidation_namespace=invalidation_namespace)
with inv_context_manager as invalidation_context:
args = ('one', 'two')
# re-compute and store cache if we get invalidate signal
if invalidation_context.should_invalidate():
result = heavy_compute.refresh(*args)
else:
result = heavy_compute(*args)
compute_time = inv_context_manager.compute_time
log.debug('result computed in %.3fs' ,compute_time)
# To send global invalidation signal, simply run
CacheKey.set_invalidate(invalidation_namespace)
"""
def __repr__(self):
return '<InvalidationContext:{}[{}]>'.format(
safe_str(self.cache_key), safe_str(self.uid))
def __init__(self, uid, invalidation_namespace='',
raise_exception=False, thread_scoped=None):
self.uid = uid
self.invalidation_namespace = invalidation_namespace
self.raise_exception = raise_exception
self.proc_id = safe_unicode(rhodecode.CONFIG.get('instance_id') or 'DEFAULT')
self.thread_id = 'global'
if thread_scoped is None:
# if we set "default" we can override this via .ini settings
thread_scoped = str2bool(rhodecode.CONFIG.get('cache_thread_scoped'))
# Append the thread id to the cache key if this invalidation context
# should be scoped to the current thread.
if thread_scoped is True:
self.thread_id = threading.current_thread().ident
self.cache_key = compute_key_from_params(uid)
self.cache_key = 'proc:{}_thread:{}_{}'.format(
self.proc_id, self.thread_id, self.cache_key)
self.compute_time = 0
def get_or_create_cache_obj(self, uid, invalidation_namespace=''):
cache_obj = CacheKey.get_active_cache(self.cache_key)
log.debug('Fetched cache obj %s using %s cache key.', cache_obj, self.cache_key)
invalidation_namespace = invalidation_namespace or self.invalidation_namespace
if not cache_obj:
cache_obj = CacheKey(self.cache_key, cache_args=invalidation_namespace)
return cache_obj
def __enter__(self):
"""
Test if current object is valid, and return CacheRegion function
that does invalidation and calculation
"""
# register or get a new key based on uid
self.cache_obj = self.get_or_create_cache_obj(uid=self.uid)
self._start_time = time.time()
if self.cache_obj.cache_active:
# means our cache obj is existing and marked as it's
# cache is not outdated, we return ActiveRegionCache
self.skip_cache_active_change = True
return ActiveRegionCache(context=self)
# the key is either not existing or set to False, we return
# the real invalidator which re-computes value. We additionally set
# the flag to actually update the Database objects
self.skip_cache_active_change = False
return FreshRegionCache(context=self)
def __exit__(self, exc_type, exc_val, exc_tb):
# save compute time
self.compute_time = time.time() - self._start_time
if self.skip_cache_active_change:
return
try:
self.cache_obj.cache_active = True
Session().add(self.cache_obj)
Session().commit()
except IntegrityError:
# if we catch integrity error, it means we inserted this object
# assumption is that's really an edge race-condition case and
# it's safe is to skip it
Session().rollback()
except Exception:
log.exception('Failed to commit on cache key update')
Session().rollback()
if self.raise_exception:
raise