refactor(commandxlat): Split evaluateContextCommand into per-command handlers

[bobtista]

Replaces #2714 by @RikuAnt (tracking #1192), rebased onto current main.

evaluateContextCommand was ~830 lines. This splits it into per-command handlers, leaving the function as the dispatch chain alone (~180 lines), with no change to behavior:

• resolveContextTarget / resolveGuiCommandTarget — null out draw/obj so evaluation falls back to a position interaction (masked, mine, force-move/attack; shrubbery/mine/relationship filtering).
• 21 handleXCommand helpers — one per arm of the else-if chain.

Since #2765 (merge ZH CommandXlat) and #2805 (move CommandXlat to Core), this lives in Core/, which is what #2714 was blocked on.

Notes

• Handler bodies are statement-for-statement moves of the arms they replace, verified mechanically against the pre-refactor revision.
• Two exceptions, both merges of arms that were identical apart from one constant: handleAttackObjectCommand takes the hint type (ATTACKRESULT_POSSIBLE / ATTACKRESULT_POSSIBLE_AFTER_MOVING), and handleHackCommand takes the special power (disable vehicle / steal cash / disable building).
• handleSetRallyPointCommand and handleImpossibleAttackHint drop a dead msgType = MSG_INVALID that was immediately overwritten.
• The result assignment stays inside the ATTACKRESULT_POSSIBLE condition — two later arms read it, so hoisting it into a helper would have changed behavior.
• Legacy space/mixed indentation in the moved bodies is normalized to tabs; without it those lines land at column 0 once the enclosing scopes are gone.
• Incorporates the refactor: Refactor evaluateContextCommand phase 1 (#1192) #2714 feedback: verbose refactor comments removed, mixed tab/space indentation fixed, braces on their own lines.

[greptile-apps]

Greptile Summary

Refactors context-command evaluation without changing its intended behavior.

• Extracts target normalization, GUI-command handling, waypoint handling, and individual action dispatch paths into private helpers.
• Replaces the nested action guard and dispatch chain with early returns.
• Adds the corresponding helper declarations to CommandXlat.h.

Confidence Score: 5/5

The PR appears safe to merge.

No blocking failures remain.

Important Files Changed

Filename	Overview
Core/GameEngine/Include/GameClient/CommandXlat.h	Declares the private helper methods used to decompose context-command evaluation.
Core/GameEngine/Source/GameClient/MessageStream/CommandXlat.cpp	Extracts the existing target normalization and ordered command-dispatch logic into helpers while preserving branch order and outcomes.

Flowchart

%%{init: {'theme': 'neutral'}}%%
flowchart TD
    A[evaluateContextCommand] --> B[Resolve context target]
    B --> C{Selection or action guard exits?}
    C -->|Yes| D[Return message type]
    C -->|No| E{Waypoint mode?}
    E -->|Yes| F[Handle waypoint command]
    E -->|No| G{GUI command active?}
    G -->|Yes| H[Normalize GUI target and dispatch]
    G -->|No| I[Evaluate ordered context actions]
    I --> J[Delegate to matching helper]
    I --> K[Handle default movement]
    F --> D
    H --> D
    J --> D
    K --> D

Loading

_{Reviews (2): Last reviewed commit: "refactor(commandxlat): Split evaluateCon..." | Re-trigger Greptile}

[xezon]

I think this refactor does not go far enough. Function can be split into many small handle functions.

[xezon]

Is normalizeContextInputs a fitting name for this?

[bobtista]

no, renamed to resolveContextTarget. Same for normalizeGuiCommandTarget while I'm at it.

[bobtista]

I think this refactor does not go far enough. Function can be split into many small handle functions.

Agreed. I took another pass and split evaluateContextCommand into 21 per-command handlers, plus separate target-resolution, GUI-command, waypoint, and default-move helpers. The main function is now only the ordered dispatch chain, with its behavior and branch order preserved.