feat(adapter): render typed history frames

[isaacbmiller]

Summary

• render typed history frames as prior user, assistant, and observation messages
• avoid duplicating current inputs when a history episode is already open
• keep legacy dict history formatting working

Stack

• Base PR: feat(history): add typed history frames #51
• Next PR: feat(tool): preserve native tool call ids #53

Validation

• uv run --extra dev pytest -q tests/adapters/test_history.py tests/adapters/test_history_formatting.py
• Full stack focused validation command listed in feat(history): add typed history frames #51.

[greptile-apps]

Greptile Summary

This PR moves typed History rendering into History.to_lm_messages, replacing a single flat blob with structured user → assistant → user(observations) chat turns that are stable across subsequent LM calls, and suppresses duplicate inputs when an episode is still open.

• History.to_lm_messages classifies each frame's inputs/outputs/observations into correctly-roled messages; legacy dict entries are mapped via _entry_to_frame which preserves unknown fields in the assistant turn.
• Adapter.plan_fields and Adapter.format are updated to call to_lm_messages and clear current-turn inputs when has_open_episode() is true, preventing the earlier duplicate-input problem.
• Two focused adapter formatting tests cover the open-episode path for ChatAdapter and JSONAdapter.

Confidence Score: 3/5

Safe to merge for the common HistoryFrame-based agentic path; the _format_outputs fallback for legacy dict entries with mixed known/unknown output fields generates a malformed assistant prompt turn.

The new to_lm_messages path works correctly for the primary use case. The _format_outputs fallback branch calls format_assistant_message_content (which already ends with [[ ## completed ## ]]) and then appends a second marker, producing a structurally malformed assistant history turn for legacy dict entries with mixed fields — the exact case the PR promises to handle correctly.

dspy/adapters/types/history.py — specifically the _format_outputs fallback path (lines 159–171)

Important Files Changed

Filename	Overview
dspy/adapters/types/history.py	Adds to_lm_messages to History, converting typed frames into structured user/assistant/user message triples; the _format_outputs fallback for frames with mixed signature and non-signature output fields produces a double [[ ## completed ## ]] marker.
dspy/adapters/base.py	Updates both plan_fields and format to call history.to_lm_messages and suppress duplicate inputs when an episode is open; also replaces {field: None for ...} with dict.fromkeys.
tests/adapters/test_history_formatting.py	New focused tests for ChatAdapter and JSONAdapter open-episode formatting; no coverage for the closed-episode multi-turn scenario central to the PR's motivation.

Sequence Diagram

sequenceDiagram
    participant Caller
    participant Adapter
    participant History

    Caller->>Adapter: format(signature, demos, inputs)
    Adapter->>History: has_open_episode()
    History-->>Adapter: true / false
    Adapter->>History: to_lm_messages(adapter, signature_without_history)
    loop each frame
        History->>History: _entry_to_frame(signature, entry)
        alt frame.inputs
            History->>Adapter: format_user_message_content(inputs)
            Adapter-->>History: user message
        end
        alt frame.outputs
            History->>History: _format_outputs(adapter, signature, outputs)
            History->>Adapter: format_assistant_message_content(signature_outputs)
            Adapter-->>History: assistant content
            History-->>History: assistant message
        end
        alt frame.observations
            History->>History: _format_observations(observations)
            History-->>History: user message (observations)
        end
    end
    History-->>Adapter: list[LMMessage]
    alt has_open_episode
        Adapter->>Adapter: clear current inputs_copy
    end
    Adapter->>Adapter: "format_user_message_content({}, main_request=True)"
    Note over Adapter: Respond with... appended as final user message
    Adapter-->>Caller: [system, ...history turns..., user(Respond with...)]

Loading

_{Reviews (3): Last reviewed commit: "feat(adapter): render typed history fram..." | Re-trigger Greptile}

[greptile-apps]

Double [[ ## completed ## ]] marker in mixed-output fallback

When outputs contains both signature-recognized fields (signature_outputs) and unrecognized fields (unknown_outputs), the fallback branch calls format_assistant_message_content — which itself appends [[ ## completed ## ]] — strips it, then appends the unknown-field sections and another [[ ## completed ## ]]. The resulting assistant history turn carries two end-of-turn sentinels, which can mislead the LM about the expected format or cause the DSPy response parser to stop early on any future response that contains a first completed marker before all fields.

The fast path (if signature_outputs and not unknown_outputs) is clean; the fix is to also strip the trailing [[ ## completed ## ]] from the format_assistant_message_content output before joining, so the final explicit sections.append("[[ ## completed ## ]]") is the only one.