orange-bioinformatics / orangecontrib / bio / obiKEGG /

Caching framework for cached kegg api calls.

import os
import UserDict
import sqlite3
import cPickle as pickle

from contextlib import closing

from datetime import datetime, date, timedelta
from . import conf

class Store(object):
    def __init__(self):
        self.timestamp = 0

    def open(self):
        raise NotImplementedError

    def __enter__(self):
        return self

    def __exit__(self, *args):

class Sqlite3Store(Store, UserDict.DictMixin):
    def __init__(self, filename):
        self.filename = filename
        self.con = sqlite3.connect(filename)
            CREATE TABLE IF NOT EXISTS cache
                (key TEXT UNIQUE,
                 value TEXT
            CREATE INDEX IF NOT EXISTS cache_index
            ON cache (key)

    def __getitem__(self, key):
        cur = self.con.execute("""
            SELECT value
            FROM cache
            WHERE key=?
        """, (key,))
        r = cur.fetchall()

        if not r:
            raise KeyError(key)
            pickle_str = str(r[0][0])
                return pickle.loads(pickle_str)
            except Exception:
                raise KeyError(key)

    def __setitem__(self, key, value):
        value = pickle.dumps(value)
            INSERT OR REPLACE INTO cache
            VALUES (?, ?)
        """, (key, value))

    def __delitem__(self, key):
            DELETE FROM cache
            WHERE key=?
        """, (key,))

    def keys(self):
        cur = self.con.execute("""
            SELECT key
            FROM cache
        return [str(r[0]) for r in cur.fetchall()]

    def close(self):

class DictStore(Store, UserDict.DictMixin):
    def __init__(self):

    def close(self):

class cache_entry(object):
    def __init__(self, value, mtime=None, expires=None):
        self.value = value
        self.mtime = mtime
        self.expires = expires


class cached_wrapper(object):
    TODO: needs documentation
    def __init__(self, function, instance, class_, cache_store,
        self.function = function
        self.instance = instance
        self.class_ = class_
        self.cache_store = cache_store
        self.last_modified = last_modified

    def has_key(self, key):
        with closing(self.cache_store()) as store:
            return key in store

    def key_from_args(self, args, kwargs=None):
        key = self.function.__name__ + repr(args)
        if kwargs is not None:
            raise NotImplementedError("kwargs are not supported yet")
        return key

    def invalidate_key(self, key):
        with closing(self.cache_store()) as store:
            del store[key]

    def last_modified_from_args(self, args, kwargs=None):
        if self.instance is not None:
            return self.instance.last_modified(args)

    def invalidate_args(self, args):
        return self.invalidate_key(self.key_from_args(args))

    def invalidate_all(self):
        prefix = self.key_from_args(()).rstrip(",)")
        with self.cache_store() as store:
            for key in store.keys():
                if key.startswith(prefix):
                    del store[key]

    def memoize(self, args, kwargs, value, timestamp=None):
        key = self.key_from_args(args, kwargs)
        if timestamp is None:
            timestamp =

        with closing(self.cache_store()) as store:
            store[key] = cache_entry(value, mtime=timestamp)

    def __call__(self, *args):
        key = self.key_from_args(args)
        with closing(self.cache_store()) as store:
            if self.key_has_valid_cache(key, store):
                rval = store[key].value
                rval = self.function(self.instance, *args)
                store[key] = cache_entry(rval,, None)

        return rval

    def key_has_valid_cache(self, key, store):
        if key not in store:
            return False
            entry = store[key]
            return self.is_entry_valid(entry, None)

    def is_entry_valid(self, entry, args):
        # For now always return True, the caching architecture needs to be
        # completely reworked
        return True

        # Need to check datetime first (it subclasses date)
        if isinstance(entry.mtime, datetime):
            mtime = entry.mtime
        elif isinstance(entry.mtime, date):
            mtime = datetime(entry.mtime.year, entry.mtime.month,
                   , 1, 1, 1)
            return False

        last_modified = self.last_modified_from_args(args)

        if isinstance(last_modified, date):
            last_modified = datetime(last_modified.year, last_modified.month,
                           , 1, 1, 1)
        elif isinstance(last_modified, basestring):
            # Could have different format
            mtime = mtime.strftime("%Y %m %d %H %M %S")

        elif last_modified is None:
            if conf.params["cache.invalidate"] == "always":
                return False
            elif conf.params["cache.invalidate"] == "session":
                last_modified = _SESSION_START
            elif conf.params["cache.invalidate"] == "daily":
                last_modified =, minute=0,
                                                       second=0, microsecond=0)
            elif conf.params["cache.invalidate"] == "weekly":
                last_modified = - timedelta(7)
            else:  # ???
        return last_modified <= mtime

class cached_method(object):
    def __init__(self, function):
        self.function = function

    def __get__(self, instance, owner):
        if instance is not None:
            return cached_wrapper(self.function, instance, owner,
                                  self.get_cache_store(instance, owner))
        return self

    def get_cache_store(self, instance, owner):
        if hasattr(instance, "cache_store"):
            return instance.cache_store
        elif not hasattr("_cached_method_cache"):
            instance._cached_method_cache = DictStore()
        return instance._cached_method_cache

class bget_cached_method(cached_method):
    def __get__(self, instance, owner):
        if instance is not None:
            return cached_wrapper(self.function, instance, owner,
                                  self.get_cache_store(instance, owner),
                                  self.get_last_modified(instance, owner))
        return self

    def get_last_modified(self, instance, owner):
        if hasattr(instance, "last_modified"):
            return instance.last_modified

def touch_dir(path):
    path = os.path.expanduser(path)
    if not os.path.exists(path):

def clear_cache():
    """Clear all locally cached KEGG data.
    import glob
    path = conf.params["cache.path"]
    if os.path.realpath(path) != os.path.realpath(conf.kegg_dir):
        raise Exception("Non default cache path. Please remove the contents "
                        "of %r manually." % path)

    for cache_filename in glob.glob(os.path.join(path, "*.sqlite3")):

    for ko_filename in glob.glob(os.path.join(path, "*.keg")):

    for kgml_filename in glob.glob(os.path.join(path, "*.xml")):

    for png_filename in glob.glob(os.path.join(path, "*.png")):
Tip: Filter by directory path e.g. /media app.js to search for public/media/app.js.
Tip: Use camelCasing e.g. ProjME to search for
Tip: Filter by extension type e.g. /repo .js to search for all .js files in the /repo directory.
Tip: Separate your search with spaces e.g. /ssh pom.xml to search for src/ssh/pom.xml.
Tip: Use ↑ and ↓ arrow keys to navigate and return to view the file.
Tip: You can also navigate files with Ctrl+j (next) and Ctrl+k (previous) and view the file with Ctrl+o.
Tip: You can also navigate files with Alt+j (next) and Alt+k (previous) and view the file with Alt+o.