feat(records): Records API by andersfylling · Pull Request #2616 · cognitedata/cognite-sdk-python

andersfylling · 2026-05-08T11:56:29Z

Implements all 6 endpoints (ingest, upsert, delete, list, sync,
aggregate) for cognite.client.data_modeling.records with full
typed DTOs, async generator for sync(), chunked writes, and the
complete 16-step wiring (testing mocks, docs, doctest registration,
retry classification, sync codegen).

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

Implements all 6 endpoints (ingest, upsert, delete, list, sync, aggregate) for cognite.client.data_modeling.records with full typed DTOs, async generator for sync(), chunked writes, and the complete 16-step wiring (testing mocks, docs, doctest registration, retry classification, sync codegen). Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

andersfylling · 2026-05-08T12:00:57Z

Any commit after this message are follow ups to correct claude code first iteration.

Sphinx autosummary requires the class to be importable via BUILD_COGNITE_SDK_DOCS block so it can resolve AsyncCogniteClient.data_modeling.records in the RST directives. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

codecov · 2026-05-08T12:08:56Z

Codecov Report

❌ Patch coverage is 95.57070% with 26 lines in your changes missing coverage. Please review.
✅ Project coverage is 93.06%. Comparing base (d8ba283) to head (cebcbf0).

Files with missing lines	Patch %	Lines
cognite/client/_api/data_modeling/records.py	86.45%	13 Missing ⚠️
cognite/client/utils/_concurrency.py	81.96%	11 Missing ⚠️
...gnite/client/data_classes/data_modeling/records.py	99.20%	2 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #2616      +/-   ##
==========================================
+ Coverage   93.04%   93.06%   +0.01%     
==========================================
  Files         486      490       +4     
  Lines       49577    50162     +585     
==========================================
+ Hits        46130    46683     +553     
- Misses       3447     3479      +32

Files with missing lines	Coverage Δ
cognite/client/_api/data_modeling/__init__.py	`100.00% <100.00%> (ø)`
cognite/client/_cognite_client.py	`95.68% <100.00%> (+0.03%)`	⬆️
cognite/client/_sync_api/data_modeling/__init__.py	`100.00% <100.00%> (ø)`
cognite/client/_sync_api/data_modeling/records.py	`100.00% <100.00%> (ø)`
...nite/client/data_classes/data_modeling/__init__.py	`100.00% <100.00%> (ø)`
cognite/client/testing.py	`100.00% <100.00%> (ø)`
cognite/client/utils/_url.py	`100.00% <ø> (ø)`
...s_unit/test_api/test_data_modeling/test_records.py	`100.00% <100.00%> (ø)`
tests/tests_unit/test_api_client.py	`99.71% <ø> (ø)`
tests/tests_unit/test_docstring_examples.py	`100.00% <100.00%> (ø)`
... and 3 more

... and 6 files with indirect coverage changes

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

- Replace en-dash (–) with hyphen (-) in 4 docstrings (RUF002) - Remove unused type: ignore [misc] comment in sync records API (mypy) - Fix docstring quote style from double to single in Literal types (custom-checks) - Add RecordsAPI to _make_accessors_for_building_docs() so Sphinx can discover data_modeling.records Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

ruff requires imports sorted alphabetically; records must come before space_statistics. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

haakonvt · 2026-05-11T09:59:24Z

@gemini-code-assist why are you commenting about CogniteObject? That is not a class in the SDK.

gemini-code-assist · 2026-05-11T09:59:38Z

Apologies for the confusion. You are correct that CogniteObject is not a public class in the SDK. It appears I made a mistake in my previous reference. Thank you for pointing that out.

haakonvt · 2026-05-11T10:00:35Z

Apologies for the confusion. You are correct that CogniteObject is not a public class in the SDK. It appears I made a mistake in my previous reference. Thank you for pointing that out.

@gemini-code-assist do we have outdated instructions about this?

gemini-code-assist · 2026-05-11T10:00:50Z

Apologies for the confusion in our previous interaction. Regarding your question, we do not have outdated instructions in the codebase or the provided context. The mention of CogniteObject was an error on my part during the conversation, and I appreciate you correcting me. The implementation of the Records API follows the standard SDK patterns as outlined in PLAN.md and the implemented files.

haakonvt · 2026-05-11T17:51:00Z

Just mentioning this (likely close candidate) #2535

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

andersfylling · 2026-05-18T07:56:26Z

/gemini review

gemini-code-assist

Code Review

This pull request introduces the RecordsAPI to the Data Modeling module, enabling ingestion, upsertion, deletion, listing, syncing, and aggregation of records along with their associated data classes and concurrency configurations. Feedback identifies a bug in the sync method where the initialization cursor is not cleared between pages, a potential StopIteration error when handling empty aggregate specifications, and the incorrect classification of idempotent record operations as non-retryable.

gemini-code-assist · 2026-05-18T09:28:08Z

+                yield SyncRecord._load(item)
+            if not data["hasNext"]:
+                break
+            body["cursor"] = data["nextCursor"]


The sync loop does not remove initializeCursor from the request body when updating the cursor for subsequent pages. While some APIs might ignore the initialization parameter if a cursor is present, sending both is incorrect and can lead to unexpected behavior or infinite loops if the API prioritizes the initialization parameter.

Suggested change

body["cursor"] = data["nextCursor"]

body["cursor"] = data["nextCursor"]

body.pop("initializeCursor", None)

gemini-code-assist · 2026-05-18T09:28:08Z

+
+
+def _load_aggregate_spec(data: dict[str, Any]) -> AggregateSpec:
+    key = next(iter(data))


Using next(iter(data)) on a dictionary without checking if it is empty will raise a StopIteration exception. This results in a poor error message when parsing malformed or empty aggregate specifications, violating the style guide principle on providing meaningful error messages.

if not data: raise ValueError("Empty aggregate specification") key = next(iter(data))

References

The style guide requires providing meaningful error messages for fallible operations and ensuring IO/data parsing safety. ^(link)

gemini-code-assist · 2026-05-18T09:28:08Z

+    "streams/[^/]+/records",
+    "streams/[^/]+/records/upsert",


Adding upsert and the base records path to NON_RETRYABLE_CREATE_DELETE_RESOURCE_PATHS prevents the SDK from retrying these operations on transient network or server errors (5xx).

Upsert is idempotent by definition and should always be retryable.

Delete for records is also idempotent (as unknown IDs are ignored) but is made non-retryable here because the regex automatically appends (/delete)? to paths in this list.

Ingest (base path) is safe to retry even if mutable, as a successful first attempt followed by a retry would simply return a 422 (which is not retried further), whereas a transient failure on the first attempt would be correctly recovered.

Following the pattern of the sibling instances API, these should be removed from the non-retryable list.

Suggested change

"streams/[^/]+/records",

"streams/[^/]+/records/upsert",

"sequences",

"streams",

"simulators",

andersfylling and others added 2 commits May 8, 2026 13:18

docs(records): add implementation plan for Records API

816f123

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

andersfylling changed the title ~~feat/records api~~ feat(records): Records API May 8, 2026

andersfylling mentioned this pull request May 8, 2026

docs: add CLAUDE.md and Claude Code skill for new sub-API authoring #2612

Draft

andersfylling and others added 2 commits May 8, 2026 14:01

run pre-commit

4766c89

andersfylling and others added 2 commits May 8, 2026 14:14

run pre-commit

cf4b9c6

andersfylling marked this pull request as ready for review May 11, 2026 08:59

andersfylling requested review from a team as code owners May 11, 2026 08:59

This comment was marked as outdated.

Sign in to view

fix(records): fix import order for RecordsAPI in _cognite_client.py

5324157

ruff requires imports sorted alphabetically; records must come before space_statistics. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

andersfylling marked this pull request as draft May 11, 2026 10:12

andersfylling marked this pull request as ready for review May 18, 2026 06:47

Merge branch 'master' into feat/records-api

84a42e0

andersfylling marked this pull request as draft May 18, 2026 06:47

This comment was marked as outdated.

Sign in to view

andersfylling and others added 3 commits May 18, 2026 09:14

remove plan

e12e9dc

improve concurreny support

6c25e44

chore: revert pyproject.toml changes to separate PR #2625

2ccc6b1

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

This comment was marked as resolved.

Sign in to view

andersfylling added 2 commits May 18, 2026 11:00

fix up review issues

32d529a

dont downgrade version

58e7da6

andersfylling marked this pull request as ready for review May 18, 2026 09:25

Merge branch 'master' into feat/records-api

63c8be4

gemini-code-assist Bot reviewed May 18, 2026

View reviewed changes

improve concurrency limits

cebcbf0

	body["cursor"] = data["nextCursor"]
	body["cursor"] = data["nextCursor"]
	body.pop("initializeCursor", None)



		def _load_aggregate_spec(data: dict[str, Any]) -> AggregateSpec:
		key = next(iter(data))

-    "streams/[^/]+/records",
-    "streams/[^/]+/records/upsert",
+    "sequences",
+    "streams",
+    "simulators",

Conversation

andersfylling commented May 8, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

andersfylling commented May 8, 2026

Uh oh!

codecov Bot commented May 8, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

This comment was marked as outdated.

Uh oh!

haakonvt commented May 11, 2026

Uh oh!

gemini-code-assist Bot commented May 11, 2026

Uh oh!

haakonvt commented May 11, 2026

Uh oh!

gemini-code-assist Bot commented May 11, 2026

Uh oh!

haakonvt commented May 11, 2026

Uh oh!

This comment was marked as outdated.

Uh oh!

andersfylling commented May 18, 2026

Uh oh!

This comment was marked as resolved.

Uh oh!

gemini-code-assist Bot left a comment

Choose a reason for hiding this comment

Code Review

Uh oh!

gemini-code-assist Bot May 18, 2026

Choose a reason for hiding this comment

Uh oh!

gemini-code-assist Bot May 18, 2026

Choose a reason for hiding this comment

Uh oh!

gemini-code-assist Bot May 18, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

andersfylling commented May 8, 2026 •

edited

Loading

codecov Bot commented May 8, 2026 •

edited

Loading