UNPKG

@autobe/agent

Version:

AI backend server code generator

145 lines (136 loc) 6.45 kB
// import { tags } from "typia"; // export interface ICheck { // /** The title or description of the validation rule/check item */ // title: string; // /** The validation state (true: passed/satisfied, false: failed/violated) */ // state: boolean; // } export interface IAutoBeTestCorrectApplication { /** * Main entry point for AI Function Call - analyzes compilation errors and * generates corrected E2E test code. * * The AI executes this function to perform the complete error correction * workflow: compilation error analysis → draft correction → code review → * final corrected implementation. This multi-step process ensures systematic * error resolution while preserving original test functionality and * maintaining code quality. * * The corrector analyzes compilation diagnostics to identify specific issues, * develops correction strategies, and produces corrected code through * iterative refinement with comprehensive review and validation. * * @param props Complete specification for error correction workflow including * analysis steps, draft implementation, review process, and final code * generation */ rewrite(props: IAutoBeTestCorrectApplication.IProps): void; } export namespace IAutoBeTestCorrectApplication { export interface IProps { /** * Step 1: Deep compilation error analysis and correction strategy. * * AI performs comprehensive analysis of compilation errors to develop * targeted correction strategies. This step involves deep examination of * error messages, identification of error patterns, understanding root * causes, and planning systematic corrections. * * The AI examines each compilation diagnostic to understand where the * implementation diverged from correct TypeScript usage, identifies the * business logic intent behind the failed code, and formulates strategies * to fix errors while preserving the original test purpose. This analysis * correlates error patterns with code structure to ensure corrections * address root causes rather than symptoms. * * This deep analysis forms the foundation for all subsequent correction * efforts, ensuring a methodical approach to resolving compilation issues. * * Workflow: Error diagnostic analysis → Root cause identification → * Correction strategy planning → Business logic preservation strategy */ think: string; /** * Step 2: Draft corrected TypeScript E2E test code implementation. * * AI generates the first corrected version of the test code based on error * analysis and correction strategies. This draft addresses all identified * compilation errors while preserving the original business logic and test * workflow. The code is compilation-error-free and follows all established * conventions. * * The implementation incorporates lessons learned from error analysis to * produce properly typed, syntactically correct code that maintains the * intended test functionality. All type safety requirements and framework * conventions are followed in this corrected implementation. * * Workflow: Error correction → TypeScript implementation → Functional * preservation * * DO: Resolve all compilation errors while maintaining original test intent */ draft: string; /** * Step 3-4: Review and finalization process. * * Encapsulates the review and final implementation phases into a single * revision process. This structured approach ensures systematic validation * and refinement of the corrected code through comprehensive review * followed by production-ready implementation. * * The revision process maintains clear separation between review feedback * and final deliverable while ensuring all corrections are properly * validated and integrated. */ revise: IReviseProps; } /** * Revision properties for the final review and implementation phases. * * This interface encapsulates the final two steps of the error correction * workflow, ensuring systematic review and production-ready code delivery. */ export interface IReviseProps { /** * Step 3: Code review and correction validation. * * AI performs a comprehensive review of the corrected draft implementation, * validating that all compilation errors have been resolved and that the * code maintains the original functionality. This review examines both * technical correctness and business logic preservation. * * The review process includes verification of TypeScript compilation * compatibility, API integration correctness, test workflow completeness, * and adherence to all quality standards. Any remaining issues or potential * improvements are identified for incorporation into the final * implementation. * * Workflow: Draft validation → Compilation verification → Functionality * review → Quality assessment */ review: string; /** * Step 4: Final production-ready corrected test code. * * AI produces the final, polished version of the corrected test code * incorporating all review feedback and validation results. This code * represents the completed error correction, guaranteed to compile * successfully while preserving all original test functionality and * business logic. When the draft correction already perfectly resolves all * issues with no problems found during review, this value can be null, * indicating no further refinement was necessary. * * The final implementation resolves all compilation issues, maintains * strict type safety, follows all established conventions, and delivers a * production-ready test that accurately validates the intended API * behaviors and user workflows. A null value signifies the draft correction * was already optimal and requires no modifications. * * Workflow: Review integration → Final refinement → Production-ready * implementation (or null if draft needs no changes). This is the ultimate * deliverable that will replace the compilation-failed code when provided, * otherwise the draft correction is used as-is. */ final: string | null; } }