Testing New Secure feature

[nealbgit]

@CaffeineLab -

Testing with code below, but getting an error.
Any ideas?

CODE:

emil = "NP4-34-CD"

# Make sure this folder exists, we'll create it in the future.
out_box = f'c:/Temp/test/reports'

cert_file = 'c:/Temp/test/02222025.cert'
key_file = 'c:/Temp/test/02222025.key'

ercot = Session(market='ercot_private', loglevel="INFO")
ercot.authorize(cert=(cert_file, key_file))
data, meta = ercot.fetch_listing(report_type_id=emil)
report = ercot.fetch_doc(data[0], out_box)

ERROR:

ercot.authorize(cert=(cert_file, key_file))
File "C:\Python312\Lib\site-packages\isodata\connector.py", line 22, in authorize
    self.token = self.get_token()
                 ^^^^^^^^^^^^^^^^
File "C:\Python312\Lib\site-packages\isodata\ercot\connector.py", line 139, in get_token
    if self.cert[0].exists() is False:
       ^^^^^^^^^^^^^^^^^^^
AttributeError: 'str' object has no attribute 'exists'

[CaffeineLab]

Just pushed up a version .20b that should allow you to use either style of passing a file - string or pathlib.Path()

Now that I know I have an audience I'll have to pay it more attention.

[nealbgit]

Thanks - will give it a test.

[CaffeineLab]

There is a sizeable update for tomorrow that cleans up a bunch of little things and makes use of sessions instead of one-off requests.

[nealbgit]

My latest test with the sample code gives the following error:

report = ercot.fetch_doc(data[0], out_box)
                             ~~~~^^^
IndexError: list index out of range

[CaffeineLab]

Internal error was: Did not recieve a numeric ReportTypeId. [reporttypeid] required.

Clearly one of us needs to work on his documentation skill. Change the EMIL to the numeric ReportTypeID, eg 12335 or 13089. getting the EMIL id working again is on the long list, and I've added support for faults and empty document lists.

Doc's will get updated with next push.

[nealbgit]

ReportTypeID worked! Thanks.

[nealbgit]

Not sure if you'd like to add something like the following, where users can add their .pfx and password, and convert it for them to cert and key. Might be nice, since Ercot provides the cert in .pfx format.

from cryptography.hazmat.primitives import serialization
from cryptography.hazmat.primitives.serialization import pkcs12
import os

def pfx_to_cert_key(pfx_file, pfx_password):
    """
    Convert .pfx certificate file to separate .cert and .key files.

    Args:
    - pfx_file (str): Path to the .pfx file.
    - pfx_password (str): Password for the .pfx file.
    """

    # Load the .pfx file
    with open(pfx_file, "rb") as f:
        pfx_data = f.read()

    # Extract private key and certificate from .pfx
    private_key, cert, additional_certificates = pkcs12.load_key_and_certificates(pfx_data, pfx_password.encode())

    # Get filename without extension
    filename = os.path.splitext(pfx_file)[0]

    # Save private key to .key file
    with open(f"{filename}.key", "wb") as f:
        f.write(
            private_key.private_bytes(
                encoding=serialization.Encoding.PEM,
                format=serialization.PrivateFormat.TraditionalOpenSSL,
                encryption_algorithm=serialization.NoEncryption(),
            )
        )

    # Save certificate to .cert file
    with open(f"{filename}.cert", "wb") as f:
        f.write(
            cert.public_bytes(serialization.Encoding.PEM)
        )

# Usage example:
pfx_to_cert_key("/path/to/pfx", "pfx_password")

[CaffeineLab]

Yeah - it's a good thing to have around. I'll sort out how to place it shortly.

[CaffeineLab]

Rewrote the majority of it and pushed as 0.0.22

pip install -v "python-isodata==0.0.22"

I'll probably start pushing this into my production workflow by next week because it certainly is better than using the old SOAP interface.

[nealbgit]

Great job!

[CaffeineLab]

I appreciate the push to get it cleaned up.