[CT-352] catch and retry malformed json

core/dbt/clients/registry.py

@@ -1,7 +1,12 @@
 import functools
+import json
 import requests
 from dbt.events.functions import fire_event
-from dbt.events.types import RegistryProgressMakingGETRequest, RegistryProgressGETResponse
+from dbt.events.types import (
+ RegistryProgressMakingGETRequest,
+ RegistryProgressGETResponse,
+ RegistryMalformedResponse,
+)
 from dbt.utils import memoized, _connection_exception_retry as connection_exception_retry
 from dbt import deprecations
 import os
@@ -31,18 +36,34 @@ def _get(path, registry_base_url=None):
  fire_event(RegistryProgressGETResponse(url=url, resp_code=resp.status_code))
  resp.raise_for_status()
 
- # It is unexpected for the content of the response to be None so if it is, raising this error
- # will cause this function to retry (if called within _get_with_retries) and hopefully get
- # a response. This seems to happen when there's an issue with the Hub.
+ # It is unexpected for the content of the response to be None or malformed. If it is,
+ # raising this error will cause this function to retry (if called within _get_with_retries)
+ # and hopefully get a valid response. This seems to happen when there's an issue with the Hub.
  # See https:/dbt-labs/dbt-core/issues/4577
- if resp.json() is None:
+ # and https:/dbt-labs/dbt-core/issues/4849
+ if not validJson(resp.json()):
  raise requests.exceptions.ContentDecodingError(
- "Request error: The response is None", response=resp
+ "Request error: The response is not valid json", response=resp
  )
  return resp.json()
 
 
+def validJson(jsonData):
+ try:
+ # the quotes in the response were converted to single quotes which is not
+ # valid json. json.dumps() will fix that.
+ json.loads(json.dumps(jsonData))
+ # We need to catch both malformed json and json = None
+ except (ValueError, TypeError) as exc:
+ # This event will allow us to see the actual json causing issues if more deps
+ # issues crop up. Otherwise we're flying a little blind.
+ fire_event(RegistryMalformedResponse(jsonData=jsonData, exc=exc))
+ return False
+ return True
+
+
 def index(registry_base_url=None):
+ # this returns a list of all packages on the Hub
  return _get_with_retries("api/v1/index.json", registry_base_url)
 
 

core/dbt/events/types.py

@@ -2393,6 +2393,18 @@ def message(self) -> str:
  return f"External call exception: {self.exc}"
 
 
+@dataclass
+class RegistryMalformedResponse(DebugLevel):
+ jsonData: str
+ exc: Exception
+ code: str = "M022"
+
+ def message(self) -> str:
+ prefix = f"Malformed json: {self.jsonData}"
+ error = f"Exception: {self.exc}"
+ return f"{prefix}\n{error}"
+
+
 # since mypy doesn't run on every file we need to suggest to mypy that every
 # class gets instantiated. But we don't actually want to run this code.
 # making the conditional `if False` causes mypy to skip it as dead code so
@@ -2747,3 +2759,4 @@ def message(self) -> str:
  GeneralWarningException(exc=Exception(""), log_fmt="")
  EventBufferFull()
  RecordRetryException(exc=Exception(""))
+ RegistryMalformedResponse(jsonData="", exc=Exception(""))

test/unit/test_events.py

@@ -416,6 +416,7 @@ def MockNode():
  EventBufferFull(),
  RecordRetryException(exc=Exception('')),
  UnitTestInfo(msg=''),
+ RegistryMalformedResponse(jsonData="",exc=Exception('')),
 ]