bc-code-intelligence-mcp

# Code Analysis Phase - Systematic Discovery Methodology ## Overview **Objective**: Comprehensive code analysis with systematic discovery to ensure complete coverage of all business modules and prevent missed optimization opportunities. **Key Principle**: Different developers and AI agents often miss critical areas due to unsystematic exploration. This methodology ensures 100% coverage through structured analysis. ## 🔥 COMPACTION-RESISTANT CORE PROCESS (ALWAYS REMEMBER) **For EVERY procedure in EVERY module:** 1. **Query ALL patterns**: `find_bc_topics` domain="all" → get complete pattern inventory 2. **Check EVERY pattern**: For each pattern found, run `analyze_code_patterns` on current procedure 3. **Document coverage**: Record WHICH patterns were checked (proves completeness) 4. **Validate before next**: Confirm you checked ALL patterns before moving to next procedure **This 4-step loop is the core systematic process that must survive context compression.** --- ## Phase Execution Process ### Step 1: Workspace Assessment & Inventory Management **CRITICAL**: Establish systematic coverage to prevent missing critical files in large codebases. #### 🔍 Phase 1A: Workspace Assessment - [ ] **Complete File Inventory** - [ ] Scan entire project structure for all .al files - [ ] Count total files and categorize by type (codeunits, tables, pages, reports) - [ ] Assess codebase complexity: - Small: <50 files (standard analysis) - Medium: 50-200 files (focused analysis recommended) - Large: 200+ files (user scoping required) - Enterprise: 500+ files (significant time and token investment) #### 🚨 Phase 1B: User Feedback Gate (Large Codebases) **IF codebase contains 200+ files:** - [ ] **Present scope options to user:** - A) Comprehensive analysis (all files - significant time/token cost) - B) Module-focused analysis (specify business domains) - C) Performance-critical files only (reports, analytics, data processing) - D) Custom file selection (user-specified scope) - [ ] **Wait for user scope selection before proceeding** - [ ] **Document scope decision** in analysis log #### 📋 Phase 1C: Coverage Checklist Creation Based on scope selection: - [ ] **File Inventory Checklist**: Create comprehensive list of files to analyze - [ ] **Pattern Detection Checklist**: List all optimization patterns to check per file - [ ] **Expected Coverage Documentation**: "Analyzing [X] files across [Y] modules for [Z] patterns" - [ ] **Progress Tracking**: Update checklists as analysis proceeds #### 🔒 MANDATORY ENFORCEMENT GATES **CRITICAL**: These gates CANNOT be bypassed - failing any gate BLOCKS progression to next phase. - [ ] **Gate 1: Complete File Inventory** - [ ] Generate actual file list using `find` or `glob` commands (if within a Workspace, use the workspace folders. Otherwise, use the root folder.) - [ ] Cross-reference with planned analysis scope - [ ] **BLOCK ANALYSIS**: Cannot proceed until file inventory is verified complete - [ ] **Gate 2: Systematic Coverage Validation** - [ ] Document EVERY file that will be analyzed - [ ] Verify EVERY pattern will be checked per file - [ ] **FAILURE CONDITION**: Missing ANY files from scope = analysis restart required - [ ] **Gate 3: File-by-File Progress Tracking** - [ ] Create explicit checklist: `[ ] filename.al - ANALYZED` - [ ] Mark each file as complete ONLY after analysis documented - [ ] **BLOCKING CONDITION**: Cannot claim module complete until ALL files marked analyzed - [ ] **EXAMPLE**: Analytics Module (4/4): ✅ T4_File1.al ✅ T4_File2.al ✅ T4_File3.al ✅ T4_File4.al #### 🔍 Phase 1D: Business Domain Discovery - [ ] **Scan Project Structure** (within defined scope) - [ ] Review all codeunit files in analysis scope - [ ] Identify distinct business domains by naming patterns - [ ] **CRITICAL**: Examine all table objects (.al files) to understand data model - [ ] **Include related objects**: If codeunits reference tables, analyze those table definitions too - [ ] **Cross-reference shared infrastructure**: Review any shared/common objects across project tiers - [ ] **Categorize Business Functions** - [ ] Financial Operations (payments, invoicing, accounting) - [ ] Data Processing (calculations, aggregations, transformations) - [ ] Reporting & Analytics (reports, dashboards, summaries) - [ ] Core Business Logic (workflows, validations, business rules) - [ ] Integration & API (external systems, data exchange) - [ ] User Interface (pages, actions, user interactions) - [ ] Background Processing (jobs, scheduled tasks, batch operations) - [ ] Data Maintenance (cleanup, archival, migration) - [ ] Security & Permissions (access control, audit trails) - [ ] **Cross-Reference Analysis** - [ ] Map codeunits to business functions they serve - [ ] Identify shared utilities and common services - [ ] Document interdependencies between modules ### Step 2: Complete Procedure-Level Analysis **CRITICAL**: For each discovered business module, analyze EVERY procedure systematically: #### 🔍 Systematic Procedure Discovery Process - [ ] **Enumerate All Procedures** - List EVERY procedure in each codeunit - Do not skip any procedures - even small ones may have optimization opportunities - Create complete procedure inventory before analysis begins - [ ] **Individual Procedure Analysis** (COMPACTION-RESISTANT PROCESS) - **Step A**: For EACH procedure, run `find_bc_topics` with domain="all" to discover complete pattern inventory available - **Step B**: For EACH pattern from Step A, run `analyze_code_patterns` to check current procedure against that pattern - **Step C**: Document WHICH patterns were checked (not just what was found) - this proves systematic coverage - **Step D**: Before moving to next procedure, validate you checked ALL patterns from Step A inventory - **Step E**: Cross-check completeness: Ensure no procedures were skipped in any module - **Step F**: If patterns checked require checking related objects (tables, other codeunits), do so and document findings #### 🚨 Critical Performance Anti-Patterns to Find (Per Procedure) - [ ] **Manual Loop Processing** - Look for `repeat...until` loops processing records - Identify manual summation or aggregation operations - Find sequential record processing that could be optimized - [ ] **Inefficient Database Access** - Multiple database calls in loops - Missing field loading optimizations - Inefficient filtering or sorting operations - [ ] **Nested Loop Anti-Patterns** - **CRITICAL**: Look for procedures with nested record processing - Identify outer loop → inner loop patterns that create N+1 query problems - Find procedures that process parent records, then child records for each parent - [ ] **Complete Pattern Library Coverage** (COMPACTION-RESISTANT) - **Pattern Discovery Query**: Use `find_bc_topics` with broad search terms to enumerate ALL available optimization patterns - **Systematic Cross-Reference**: For each procedure, check it against EVERY pattern in the discovered inventory - **Coverage Documentation**: Record which pattern categories were checked for each procedure (proves no patterns were missed) - **Validation Loop**: Before analysis completion, verify every procedure was checked against every available pattern ### Step 3: Prioritization Framework Rank discovered optimization opportunities: #### 🎯 Priority Classification - [ ] **🔴 Critical (Red)**: Performance killers affecting core operations - [ ] **🟡 High (Yellow)**: Significant improvements with moderate effort - [ ] **🟢 Medium (Green)**: Good improvements requiring analysis - [ ] **⚪ Low (White)**: Minor optimizations or already optimized #### 📊 Impact Assessment For each optimization opportunity: - [ ] Estimate performance impact (execution time reduction) - [ ] Assess implementation complexity - [ ] Consider business criticality of the affected process - [ ] Evaluate risk level of proposed changes ### Step 4: Documentation Requirements (COMPACTION-RESISTANT VALIDATION) - [ ] **Module Coverage Report**: List all business modules discovered - [ ] **Complete Procedure Inventory**: Document EVERY procedure analyzed in EVERY module - [ ] **Pattern Coverage Matrix**: Document WHICH pattern categories were checked for EACH procedure (proves systematic coverage) - [ ] **Table Structure Analysis**: Document key table objects, SIFT keys, and FlowFields found - [ ] **Procedure-Level Pattern Analysis**: Document anti-patterns found in EACH procedure - [ ] **Coverage Validation Evidence**: Provide proof that no procedures were missed AND no patterns were missed - [ ] **Cross-Infrastructure Analysis**: Note shared tables/infrastructure that could benefit multiple modules - [ ] **Optimization Roadmap**: Prioritized list of improvement opportunities from ALL procedures - [ ] **Knowledge Gaps**: Areas requiring additional platform expertise --- ## Success Criteria ✅ **Complete Coverage**: All business modules in codebase identified and analyzed ✅ **Procedure-Level Completeness**: EVERY procedure in EVERY module has been individually analyzed ✅ **Pattern Detection**: Performance anti-patterns documented with specific examples from each procedure ✅ **No Gaps**: Verification that no procedures were skipped during analysis ✅ **Prioritization**: Clear ranking of optimization opportunities by impact/effort ✅ **Actionable Output**: Concrete next steps for performance optimization phase ## Quality Validation (COMPACTION-RESISTANT CHECKS) - **Module Thoroughness**: Can you explain the business purpose of every codeunit? - **Procedure Completeness**: Have you analyzed every single procedure in every module? - **Pattern Library Coverage**: Can you list WHICH pattern categories you checked for each procedure? - **Systematic Validation**: Did you run `find_bc_topics` to discover ALL available patterns before analysis? - **Coverage Evidence**: Can you prove no procedures were missed AND no patterns were missed? - **Pattern Verification**: Have you found actual optimization opportunities, not just cosmetic issues? - **Impact Assessment**: Are your priority rankings based on measurable performance impact? --- ## Next Phase Upon completion, proceed to **Performance Optimization Phase** with prioritized list of improvements.