ajsya wrote:

So I'm getting an error when starting my program if I don't change the cloud variable in the first few seconds of the program starting up even though the project does have cloud data.

[code / errors hidden]

{"method":"handshake","user":"SansStudios","project_id":"622084628"}

{"method":"set","project_id":"622084628","name":"☁ REQUEST","value":"11182912241823"}

{"method":"set","project_id":"622084628","name":"☁ RESPONSE","value":"111829122418234612242829284660384140394055"}

client = CloudClient(username='user', project_id='123', ignore_missing_variables=True)

Breaking Changes

Changed CloudClient.connected to CloudClient.client_setup
Changed method CloudClient.start() to CloudClient.setup()
CloudClient.setup() no longer starts the cloud client's main loop.
CloudClient.setup() returns a task that starts the main loop instead of starting the main loop.
SegmentDump.dump() now has a default delay of 0 seconds due to rate limiting handling (see below)

Features

CloudClient.set_cloud() is now rate limited to 0.1 requests per second. Requests will be queued and fulfilled in the order they were sent.
CloudClient objects can now be stopped with the CloudClient.stop() method.
MissingCloudVariable exceptions can be ignored through the CloudClient via the argument ignore_missing_variables=True

Fixes

Some docs fixes

ajsya wrote:

can you encode lists?

docs wrote:

encoder (Callable[, str], optional) – a callable function that is used to encode all sent cloud data. The function must return a string of digits and take 1 argument.

decoder (Callable[, str], optional) – a callable function that is used to decode all received cloud data. The function must take 1 argument.

from scratchcloud.ext import BaseCodec
class ListCodec():
    def __init__(self, base: BaseCodec = BaseCodec(), separator: str = ','):
        self.separator = separator
        self.base = base
    # Decode method that takes data, decodes it using base codec, and splits the result into a list
    def decode(self, data):
        return self.base.decode(data).split(self.separator)
    # Encode method that takes data, and if it is a list, turns the list into a string with commas seperating.
    def encode(self, data: list | str):
        if isinstance(data, list):
            data = self.separator.join(data)
        return self.base.encode(data)

from scratchcloud import CloudClient, CloudChange
# make a new list codec to use in client
codec = ListCodec()
# add the ListCodec.encode and ListCodec.decode methods to the client
client = CloudClient('username', '622084628', encoder=codec.encode, decoder=codec.decode)
@client.event
async def on_message(c: CloudChange):
    print(f'{c.name} changed to {c.value}')
    # set the value of the RESPONSE variable to a list
    await client.set_cloud('RESPONSE', ['send', 'this', 'list!'])

coolcoder1213 wrote:

SansStudios wrote:

coolcoder1213 wrote:

Cool. I mostly use s2py though.

Well, if you ever decide you wanna try async python, this is the library for that!

I also use scratchconnect but maybe I will use this. I am also developing my own package like scratchcloud and s2py.

coolcoder1213 wrote:

coolcoder1213 wrote:

SansStudios wrote:

coolcoder1213 wrote:

Cool. I mostly use s2py though.

Well, if you ever decide you wanna try async python, this is the library for that!

I also use scratchconnect but maybe I will use this. I am also developing my own package like scratchcloud and s2py.

Actually, scratchattach is THE BEST!!! But I will use this too sometimes.

coolcoder1213 wrote:

Actually, scratchattach is THE BEST!!! But I will use this too sometimes.

ISTILLMAKESTUFF wrote:

coolcoder1213 wrote:

coolcoder1213 wrote:

SansStudios wrote:

coolcoder1213 wrote:

Cool. I mostly use s2py though.

Well, if you ever decide you wanna try async python, this is the library for that!

I also use scratchconnect but maybe I will use this. I am also developing my own package like scratchcloud and s2py.

Actually, scratchattach is THE BEST!!! But I will use this too sometimes.

Why is scratchattach the best, because it is made b a famous scratcher named timmccool?

NFlex23 wrote:

Bump. I kind of wish this worked with Python 3.8 since that's what Replit uses, but I understand your decision to use 3.10.

SansStudios wrote:

(#31)

NFlex23 wrote:

Bump. I kind of wish this worked with Python 3.8 since that's what Replit uses, but I understand your decision to use 3.10.

Thanks for the bump!

Unfortunately all of my typing relies on | union typing, and the API extension uses Dataclasses, both which are 3.10 features. It's kind of messed up that repl doesn't support 3.10

I could make a 3.8 version without API interaction if that would be useful

NFlex23 wrote:

Bump. I kind of wish this worked with Python 3.8 since that's what Replit uses, but I understand your decision to use 3.10.

SansStudios wrote:

TheCloudDev wrote:

Pretty cool!

Thanks! Means a lot from the creator of scratch2py

ajsya wrote:

I might consider using this for some of my projects since you say that it's faster than ScratchConnect. ScratchConnect is cool but can be a little dense for some projects.

Yay! scratchcloud is near instantaneous for reading changes in cloud variables which is where it excels.