binaryai.utils

Module Contents

Classes

QCloudHttpxAuth

Auth class that allows us to connect to QCloud services

Functions

sha256sum(→ str)	Computes sha256 hash sum of a file.
sign(key, msg)	Modified from https://docs.aws.amazon.com/general/latest/gr/sigv4-signed-request-examples.html
getSignatureKey(key, dateStamp, serviceName)	Modified from https://docs.aws.amazon.com/general/latest/gr/sigv4-signed-request-examples.html

binaryai.utils.sha256sum(path: str) → str

Computes sha256 hash sum of a file.

Parameters:

path – path of file

Returns:

hex digest of sha256

binaryai.utils.sign(key, msg)

Modified from https://docs.aws.amazon.com/general/latest/gr/sigv4-signed-request-examples.html

binaryai.utils.getSignatureKey(key, dateStamp, serviceName)

Modified from https://docs.aws.amazon.com/general/latest/gr/sigv4-signed-request-examples.html

class binaryai.utils.QCloudHttpxAuth(qcloud_secret_id, qcloud_secret_key, qcloud_host, qcloud_region, qcloud_service, qcloud_action, qcloud_apiversion)

Bases: httpx.Auth

Auth class that allows us to connect to QCloud services

requires_request_body = True

auth_flow(r: httpx.Request) → Generator[httpx.Request, httpx.Response, None]

Adds the authorization headers required by QCloud Signature v3.

get_qcloud_request_headers_handler(r)

Override get_qcloud_request_headers_handler() if you have a subclass that needs to call get_qcloud_request_headers() with an arbitrary set of QCloud credentials. The default implementation calls get_qcloud_request_headers() with self.qcloud_access_key, and self.qcloud_secret_access_key

get_qcloud_request_headers(r: httpx.Request, qcloud_secret_id, qcloud_secret_key)

Returns a dictionary containing the necessary headers for Amazon’s signature version 4 signing process. An example return value might look like

{

‘Authorization’: ‘…’, ‘…’,

}

classmethod get_canonical_path(r: httpx.Request)

Create canonical path. According to QCloud, this should always be “/”

classmethod get_canonical_querystring(r: httpx.Request)

Create the canonical query string. According to QCloud, by the end of this function our query string values must be URL-encoded (space=%20) and the parameters must be sorted by name.

This method assumes that the query params in r are already url encoded. If they are not url encoded by the time they make it to this function, QCloud may complain that the signature for your request is incorrect.

It appears elasticsearc-py url encodes query paramaters on its own:

https://github.com/elastic/elasticsearch-py/blob/5dfd6985e5d32ea353d2b37d01c2521b2089ac2b/elasticsearch/connection/http_requests.py#L64

If you are using a different client than elasticsearch-py, it will be your responsibility to urleconde your query params before this method is called.