fix(artifacts): accept dict-shaped artifacts in InMemoryArtifactService (#3495) #3622

Fangmbeng · 2025-11-19T23:36:44Z

Fixes #3495 — make InMemoryArtifactService robust to dict-shaped artifacts commonly produced by AgentSpace uploads. Adds unit test that covers dict and object artifact shapes.\n\nChanges: \n- in_memory_artifact_service: safe field extraction for dict/object artifacts; fallback mime-type.\n- artifact_util: is_artifact_ref handles dict shapes.\n- tests: focused unit test for dict/object artifact flows.\n

…ce (google#3495)

google-cla · 2025-11-19T23:36:49Z

Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

View this failed invocation of the CLA check for more information.

For the most up to date status, view the checks section at the bottom of the pull request.

gemini-code-assist · 2025-11-19T23:37:04Z

Summary of Changes

Hello @Fangmbeng, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request addresses an issue where the InMemoryArtifactService was not fully compatible with dictionary-shaped artifacts, particularly those originating from AgentSpace uploads. The changes enhance the service's ability to process and manage artifacts consistently, regardless of their underlying data structure (object or dictionary), thereby improving the overall reliability and flexibility of artifact handling within the system.

Highlights

Robust Artifact Handling: The InMemoryArtifactService has been updated to robustly accept and process artifacts that are provided as dictionary-shaped payloads, which are commonly produced by AgentSpace uploads. This ensures the service can handle both object-like types.Part and plain dict formats.
Flexible Field Extraction: A new helper function _get_field was introduced to safely extract fields (like inline_data, text, file_data) from artifacts, regardless of whether they are objects or dictionaries, improving code resilience.
Improved Mime-Type Determination: The logic for determining the mime-type during artifact saving now correctly handles both object and dictionary structures. A fallback mime-type of application/octet-stream is now used for unknown artifact shapes instead of raising an error, preventing service interruptions.
Enhanced Artifact Reference Resolution: The is_artifact_ref utility function and the load_artifact method now correctly parse artifact reference URIs from both object-like and dictionary-shaped artifacts, ensuring proper resolution of referenced artifacts.
Comprehensive Unit Testing: A new dedicated unit test file has been added to specifically cover scenarios involving dict-shaped and object-like artifacts, including their saving and loading, ensuring the new functionality works as expected.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	`/gemini review`	Performs a code review for the current pull request in its current state.
Pull Request Summary	`/gemini summary`	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	`/gemini help`	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

gemini-code-assist

Code Review

This pull request makes InMemoryArtifactService more robust by adding support for dictionary-shaped artifacts, which is a great improvement for interoperability. The changes in artifact_util.py and in_memory_artifact_service.py correctly handle both object and dict forms. A new focused unit test is also added to cover these new cases.

My review includes suggestions to improve code consistency and maintainability by reusing helper functions, removing duplicated code in tests, and improving error handling. I've also suggested an enhancement to the new tests to cover more edge cases related to empty artifacts.

gemini-code-assist · 2025-11-19T23:38:56Z

src/google/adk/artifacts/artifact_util.py

+  # Support both object-like `types.Part` and plain `dict` shapes.
+  file_data = None
+  if isinstance(artifact, dict):
+    file_data = artifact.get("file_data")
+  else:
+    file_data = getattr(artifact, "file_data", None)
+
+  if not file_data:
+    return False
+
+  file_uri = file_data.get("file_uri") if isinstance(file_data, dict) else getattr(file_data, "file_uri", None)
+  return bool(file_uri and isinstance(file_uri, str) and file_uri.startswith("artifact://"))


The logic to safely access fields from both dict and object shapes is a bit verbose and repeats the access pattern. To improve readability and reduce duplication, you could define a local helper function.

Suggested change

# Support both object-like `types.Part` and plain `dict` shapes.

file_data = None

if isinstance(artifact, dict):

file_data = artifact.get("file_data")

else:

file_data = getattr(artifact, "file_data", None)

if not file_data:

return False

file_uri = file_data.get("file_uri") if isinstance(file_data, dict) else getattr(file_data, "file_uri", None)

return bool(file_uri and isinstance(file_uri, str) and file_uri.startswith("artifact://"))

# Support both object-like `types.Part` and plain `dict` shapes.

def _get_field(obj, name):

if isinstance(obj, dict):

return obj.get(name)

return getattr(obj, name, None)

file_data = _get_field(artifact, "file_data")

if not file_data:

return False

file_uri = _get_field(file_data, "file_uri")

return bool(file_uri and isinstance(file_uri, str) and file_uri.startswith("artifact://"))

gemini-code-assist · 2025-11-19T23:38:57Z

src/google/adk/artifacts/in_memory_artifact_service.py

+    if inline_data is not None:
+      # inline_data may be a dict or an object
+      artifact_version.mime_type = (
+          inline_data.get("mime_type")
+          if isinstance(inline_data, dict)
+          else inline_data.mime_type
+      )
+    elif text is not None:
      artifact_version.mime_type = "text/plain"
-    elif artifact.file_data is not None:
+    elif file_data is not None:
+      # If the artifact is an artifact-ref we validate the referenced URI.
      if artifact_util.is_artifact_ref(artifact):
-        if not artifact_util.parse_artifact_uri(artifact.file_data.file_uri):
-          raise ValueError(
-              f"Invalid artifact reference URI: {artifact.file_data.file_uri}"
-          )
-        # If it's a valid artifact URI, we store the artifact part as-is.
-        # And we don't know the mime type until we load it.
+        file_uri = (
+            file_data.get("file_uri") if isinstance(file_data, dict) else file_data.file_uri
+        )
+        if not artifact_util.parse_artifact_uri(file_uri):
+          raise ValueError(f"Invalid artifact reference URI: {file_uri}")
+        # Valid artifact URI: keep part as-is; mime type may be resolved later.
      else:
-        artifact_version.mime_type = artifact.file_data.mime_type
+        artifact_version.mime_type = (
+            file_data.get("mime_type") if isinstance(file_data, dict) else file_data.mime_type
+        )


You can simplify the extraction of mime_type and file_uri by consistently using the _get_field helper function you've defined earlier. This will make the code more concise and less repetitive.

if inline_data is not None: # inline_data may be a dict or an object artifact_version.mime_type = _get_field(inline_data, "mime_type") elif text is not None: artifact_version.mime_type = "text/plain" elif file_data is not None: # If the artifact is an artifact-ref we validate the referenced URI. if artifact_util.is_artifact_ref(artifact): file_uri = _get_field(file_data, "file_uri") if not artifact_util.parse_artifact_uri(file_uri): raise ValueError(f"Invalid artifact reference URI: {file_uri}") # Valid artifact URI: keep part as-is; mime type may be resolved later. else: artifact_version.mime_type = _get_field(file_data, "mime_type")

gemini-code-assist · 2025-11-19T23:38:57Z

src/google/adk/artifacts/in_memory_artifact_service.py

+      try:
+        if a == types.Part() or a == types.Part(text=""):
+          return True
+        inline = getattr(a, "inline_data", None)
+        if inline and not getattr(inline, "data", None):
+          return True
+      except Exception:
+        return False


Using except Exception: is very broad and can hide unexpected bugs. It's better to catch more specific exceptions that you anticipate might be raised from the operations in the try block, such as TypeError or AttributeError, to avoid accidentally silencing unrelated issues.

gemini-code-assist · 2025-11-19T23:38:57Z

tests/unittests/artifacts/test_in_memory_artifact_dict_handling.py

+# Stub google.genai.types if not available
+try:
+  from google.genai import types as types_mod  # type: ignore
+except Exception:
+  genai = _types.ModuleType("google.genai")
+  types_mod = _types.ModuleType("google.genai.types")
+  class Part:
+    def __init__(self, inline_data=None, text=None, file_data=None):
+      self.inline_data = inline_data
+      self.text = text
+      self.file_data = file_data
+
+    def __eq__(self, other):
+      if not isinstance(other, Part):
+        return False
+      return (
+        (self.inline_data == other.inline_data)
+        and (self.text == other.text)
+        and (self.file_data == other.file_data)
+      )
+  class Content:
+    pass
+  types_mod.Part = Part
+  types_mod.Content = Content
+  sys.modules["google.genai"] = genai
+  sys.modules["google.genai.types"] = types_mod


This block of code to stub google.genai.types appears to be a duplicate of the block at lines 8-39. You can remove this redundant section.

gemini-code-assist · 2025-11-19T23:38:57Z

tests/unittests/artifacts/test_in_memory_artifact_dict_handling.py

+
+  # list keys should include the user-scoped filename
+  keys = await svc.list_artifact_keys(app_name="app", user_id="user")
+  assert "user:avatar.png" in keys


This is a good start for testing dict-shaped artifacts. To improve test coverage, consider adding tests for the new _is_empty logic in load_artifact. Specifically, you could add test cases that save various forms of empty artifacts (e.g., {}, {'text': ''}, {'inline_data': {'data': b''}}) and then assert that load_artifact correctly returns None for them.

…ack mime type

…andling-3495

ryanaiagent · 2025-11-21T01:16:22Z

Hi @Fangmbeng , Thank you for your contribution!
Can you please fix the failing unit tests and lint errors before we proceed with the review.

…andling-3495

fix(artifacts): accept dict-shaped artifacts in InMemoryArtifactServi…

dd2bb04

…ce (google#3495)

adk-bot added the services [Component] This issue is related to runtime services, e.g. sessions, memory, artifacts, etc label Nov 19, 2025

gemini-code-assist bot reviewed Nov 19, 2025

View reviewed changes

Fangmbeng and others added 3 commits November 20, 2025 00:07

test(artifacts): add artifact-ref and empty-artifact tests; log fallb…

044872c

…ack mime type

chore(repo): commit local changes before merging upstream/main

78d3f77

Merge remote-tracking branch 'upstream/main' into fix/artifact-dict-h…

ee87bc4

…andling-3495

ryanaiagent self-assigned this Nov 21, 2025

ryanaiagent added the request clarification [Status] The maintainer need clarification or more information from the author label Nov 21, 2025

Fangmbeng added 2 commits November 21, 2025 02:17

style: run pyink/isort formatting fixes

9fe506e

Merge remote-tracking branch 'upstream/main' into fix/artifact-dict-h…

6d7cbd9

…andling-3495

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

fix(artifacts): accept dict-shaped artifacts in InMemoryArtifactService (#3495) #3622

fix(artifacts): accept dict-shaped artifacts in InMemoryArtifactService (#3495) #3622

Fangmbeng commented Nov 19, 2025

Uh oh!

google-cla bot commented Nov 19, 2025

Uh oh!

gemini-code-assist bot commented Nov 19, 2025

Uh oh!

gemini-code-assist bot left a comment

Uh oh!

gemini-code-assist bot Nov 19, 2025

Uh oh!

gemini-code-assist bot Nov 19, 2025

Uh oh!

gemini-code-assist bot Nov 19, 2025

Uh oh!

gemini-code-assist bot Nov 19, 2025

Uh oh!

gemini-code-assist bot Nov 19, 2025

Uh oh!

ryanaiagent commented Nov 21, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

fix(artifacts): accept dict-shaped artifacts in InMemoryArtifactService (#3495) #3622

Are you sure you want to change the base?

fix(artifacts): accept dict-shaped artifacts in InMemoryArtifactService (#3495) #3622

Conversation

Fangmbeng commented Nov 19, 2025

Uh oh!

google-cla bot commented Nov 19, 2025

Uh oh!

gemini-code-assist bot commented Nov 19, 2025

Summary of Changes

Highlights

Footnotes

Uh oh!

gemini-code-assist bot left a comment

Choose a reason for hiding this comment

Code Review

Uh oh!

gemini-code-assist bot Nov 19, 2025

Choose a reason for hiding this comment

Uh oh!

gemini-code-assist bot Nov 19, 2025

Choose a reason for hiding this comment

Uh oh!

gemini-code-assist bot Nov 19, 2025

Choose a reason for hiding this comment

Uh oh!

gemini-code-assist bot Nov 19, 2025

Choose a reason for hiding this comment

Uh oh!

gemini-code-assist bot Nov 19, 2025

Choose a reason for hiding this comment

Uh oh!

ryanaiagent commented Nov 21, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants