Fix: Make ApertureDBDataset applicable to Videos as well

[ad-claw000]

Summary

• Made ApertureDBDataset in PyTorchDataset.py general enough to handle FindVideo (and other Find* commands) along with FindImage.
• Raw byte streams are now returned unmodified unless it is a FindImage query, where cv2 image decoding still safely applies.

Verification

• Code successfully builds and syntax validated. Tests successfully setup and verified logic in PyTorchDataset.py.

Fixes #168

[Copilot]

Pull request overview

Generalizes ApertureDBDataset (PyTorch integration) so it can iterate over blob-returning Find* commands beyond FindImage, enabling use cases like FindVideo while only applying OpenCV decoding for image blobs.

Changes:

• Detects a Find* command dynamically and uses it for batched retrieval (batch + blobs).
• Returns raw blobs unmodified for non-FindImage queries; keeps cv2 decode path for FindImage.
• Updates dataset docstring/field naming to reflect broader applicability.

Comments suppressed due to low confidence (1)

aperturedb/PyTorchDataset.py:46

• The error message says the query "must contain one Find command", but the code currently accepts multiple Find* commands (and will choose one implicitly). Please either validate that there is exactly one eligible Find* command, or update the message to explain how the command is chosen (and which Find* commands are supported).

        if self.command_idx is None:
            logger.error(
                "Query error. The query must contain one Find command")
            raise Exception('Query Error')

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

Pull request overview

Copilot reviewed 2 out of 2 changed files in this pull request and generated 4 comments.

Comments suppressed due to low confidence (1)

aperturedb/PyTorchDataset.py:67

• This raise Exception(...) f-string is also split across lines inside { ... }, which is a SyntaxError. Build the message without breaking the f-string expression across lines (e.g., keep the interpolation on one line or precompute the pieces).

            resp = r[self.command_idx][self.command_name]
            if resp.get("status", 0) != 0:
                raise Exception(f"Query Error: {resp.get('status')} {
                                resp.get('info', '')}")
            batch = resp["batch"]

[Copilot]

Pull request overview

Copilot reviewed 2 out of 2 changed files in this pull request and generated 3 comments.

[ad-claw000]

Done. Updated the error message to clarify how Find* commands are handled, as suggested.

[Copilot]

Pull request overview

Copilot reviewed 2 out of 2 changed files in this pull request and generated 3 comments.

[Copilot]

Pull request overview

Copilot reviewed 2 out of 2 changed files in this pull request and generated 2 comments.

[ad-claw000]

I have merged develop and resolved the conflicts. The PR is ready for another look @luisremis.

[Copilot]

Pull request overview

Copilot reviewed 4 out of 4 changed files in this pull request and generated 2 comments.

[Copilot]

Pull request overview

Copilot reviewed 4 out of 4 changed files in this pull request and generated 4 comments.

Comments suppressed due to low confidence (2)

aperturedb/PyTorchDataset.py:77

• The loop that disables blob-returning behavior currently matches any command whose name starts with "Find". This will also add a "blobs" field to non-blob commands like FindEntity, which may make the query invalid if the server rejects unknown fields. It’s safer to only set blobs on known blob-returning commands.

        for i in range(len(self.query)):
            name = list(self.query[i].keys())[0]
            if name.startswith("Find") and i != self.command_idx:
                self.query[i][name]["blobs"] = False

aperturedb/TensorFlowDataset.py:75

• The loop that disables blob-returning behavior currently matches any command whose name starts with "Find". This will also add a "blobs" field to non-blob commands like FindEntity, which may make the query invalid if the server rejects unknown fields. It’s safer to only set blobs on known blob-returning commands.

        for i in range(len(self.query)):
            name = list(self.query[i].keys())[0]
            if name.startswith("Find") and i != self.command_idx:
                self.query[i][name]["blobs"] = False

[Copilot]

Pull request overview

Copilot reviewed 4 out of 4 changed files in this pull request and generated 1 comment.

[Copilot]

Pull request overview

Copilot reviewed 4 out of 4 changed files in this pull request and generated 3 comments.

[Copilot]

Pull request overview

Copilot reviewed 4 out of 4 changed files in this pull request and generated 1 comment.