cachebox 3.0.0

The fastest memoizing and caching Python library written in Rust

cachebox

Changelog . Releases . API Reference

The fastest caching Python library written in Rust

Did you find any bugs?

[!NOTE]
The new version v3 has some incompatible with v2, for more info please see Incompatible changes

What does it do?
You can easily and powerfully perform caching operations in Python as fast as possible.

Features:

• 🚀 5-20x faster than other caching libraries ...
• 📊 Very low memory usage (1/3 of dictionary) ...
• (R) written in Rust
• 🤝 Support Python 3.8 and above
• 📦 Over 7 cache algorithms are supported
• 🧶 Completely thread-safe (uses RwLock)

Installing:
Install it from PyPi:

pip3 install -U cachebox

Page Contents

• ⁉️ When i need caching?
• 🤷‍♂️ Why cachebox?
• 🎓 Examples
• 💡 Incompatible changes
• ⁉️ Frequently Asked Questions
• ⏱️ BENCHMARK

When i need caching?

There are some situations that you may need caching to imprve your application speed:

1. Sometimes you have functions that take a long time to execute, and you need to call them each time.

2. Sometimes you need to temporarily store data in memory for a short period.

3. When dealing with remote APIs, Instead of making frequent API calls, store the responses in a cache.

4. Caching query results from databases can enhance performance.

5. and ...

Why cachebox?

Rust - It library uses Rust language to has high-performance.

SwissTable - It uses Google's high-performance SwissTable hash map.

Low memory usage - It has very low memory usage.

Zero-Dependecy - As we said, cachebox written in Rust so you don't have to install any other dependecies.

Thread-safe - It's completely thread-safe and uses read-writer locks to prevent problems.

Easy-To-Use - You only need to import it and choice your implementation to use and behave with it like a dictionary.

Examples

[!TIP]
See API Reference for more examples and references

decorators example:

import cachebox

@cachebox.cached(cachebox.FIFOCache(maxsize=128))
def factorial(n):
    return n * factorial(n-1) if n else 1

# Like functools.lru_cache, If maxsize is set to 0, the cache can grow without bound and limit.
@cachebox.cached(cachebox.LRUCache(maxsize=0))
def count_vowels(sentence):
    return sum(sentence.count(vowel) for vowel in 'AEIOUaeiou')

# Async are also supported
@cachebox.cached(cachebox.TTLCache(maxsize=20, ttl=5))
async def get_coin_price(coin):
    return await client.get_coin_price(coin)

class Application:
    # Use cachedmethod for methods
    @cachebox.cachedmethod(cachebox.LFUCache(maxsize=20))
    def send(self, request):
        self._send_request(request)

Implementations example:

import cachebox
import time

cache = cachebox.TTLCache(maxsize=5, ttl=3)
cache.insert("key", "value")
print(cache["key"]) # Output: value

time.sleep(3)
print(cache.get("key")) # Output: None

Incompatible changes

These are changes that are not compatible with the previous version:

[!NOTE]
You can see more info about changes in Changelog.

Maxsize default value changed!

The change applied is that when you pass 0 as maxsize, the value of sys.maxsize is automatically used.

import cachebox, sys
c = cachebox.Cache(0)

# In previous version:
assert c.maxsize == 0

# In new version:
assert c.maxsize == sys.maxsize

Iterators changed!

The change applied is that, in previous version you may make changes in cache after abtaining an iterator from it and that did not cause an error, but now you cannot make changes in cache while using the iterator.

import cachebox

c = cachebox.Cache(0, {i:i for i in range(100)})

for (key, value) in c.items():
    # This will raise RuntimeError, don't make changes
    del c[key]

Type-hint is now better!

In previous version, we couldn't use type-hints as possible as dictionary; now we can:

import cachebox

# In previous version this will raises an exception; but now is OK.
c: cachebox.Cache[int, str] = cachebox.Cache(0)

Cache iterators are not ordered!

In previous versions, some caches such as FIFOCache can return ordered iterators, but now all of them only can return unordered iterators.

import cachebox

c = cachebox.FIFOCache(20)
for i in range(10):
    c.insert(i, i)

for key in c:
    print(key)
# (5, 5)
# (9, 9)
# (0, 0)
# ...

__repr__ changed to __str__

We changed the __repr__ method to __str__:

import cachebox
c = cachebox.Cache(0)

print(c)
# Output: Cache(0 / 9223372036854775807, capacity=0)

print(repr(c))
# Output: <cachebox._cachebox.Cache object at 0x7f96938f06a0>

FAQ

Can we set maxsize to zero?

Yes, if you pass zero to maxsize, means there's no limit for items.

How to migrate from cachetools to cachebox?

cachebox syntax is very similar to cachetools. Just change these:

# If you pass infinity to a cache implementation, change it to zero.
cachetools.Cache(math.inf) -> cachebox.Cache(0)
# If you use `isinstance` for cachetools classes, change those.
isinstance(cache, cachetools.Cache) -> isinstance(cache, cachebox.BaseCacheImpl)

License

cachebox is provided under the MIT license. See LICENSE.

Future Plans

TODO List:

• Rewrite all cache algorithms and use low-level API hashmap
• Change hashing system
• Improve tests
• Rewrite stub-file (.pyi)
• Rewrite README.md
• Write an API referenece
• Add new functions such as cached_property.
• Add possible methods to implementations.
• Make better type-hint for cached and cachedmethod if possible.