{DNS} Migrate to track 2

[jsntcy]

Description
Changes made for the migration:

• Model: RecordSetPaged --> RecordSetListResult
• Method: delete ---> begin_delete
• Property: arecords ---> a_records
• Exception: CloudError ---> HttpResponseError
• Property in test checker: arecords ---> aRecords

Testing Guide

History Notes

[Network] BREAKING CHANGE: az network dns record-set a show: Property arecords in output will be changed to aRecords.

---

This checklist is used to make sure that common guidelines for a pull request are followed.

• The PR title and description has followed the guideline in Submitting Pull Requests.

• I adhere to the Command Guidelines.

• I adhere to the Error Handling Guidelines.

[jsntcy]

Why do we need change arecords to aRecords?

In track 2 SDK, the property arecords in model RecordSet is changed to a_records

• Track 2: https://github.com/Azure/azure-sdk-for-python/blob/master/sdk/network/azure-mgmt-dns/azure/mgmt/dns/v2018_05_01/models/_models_py3.py#L371 (a_records)
• Track 1: https://github.com/Azure/azure-sdk-for-python/blob/release/v3/sdk/network/azure-mgmt-dns/azure/mgmt/dns/v2018_05_01/models/_models_py3.py#L344 (arecords)

In Azure CLI, we have the following code to do some transformation for the result from SDK.

azure-cli/src/azure-cli-core/azure/cli/core/commands/__init__.py

Line 706 in 1ca05db

result = todict(result, AzCliCommandInvoker.remove_additional_prop_layer)

While code above will call the following code in knack which will get the different output for arecords and a_records.

• For arecords, the transformation result is arecords.
• For a_records, the transformation result is aRecords.

Not sure if it's by design or a bug and need further investigation.

KEYS_CAMELCASE_PATTERN = re.compile('(?!^)_([a-zA-Z])')

def to_camel_case(s):
    return re.sub(KEYS_CAMELCASE_PATTERN, lambda x: x.group(1).upper(), s)

https://github.com/microsoft/knack/blob/dev/knack/util.py#L126

@jiasli, any ideas for this behavior from knack?