sqlc architecture: modules, API surface, dependency graph

At a Glance

---

What This System Does

sqlc is a general-purpose command-line tool built with Go, SQL. It operates as a command-line interface, enabling users to interact with the system through terminal commands. This architecture is optimized for automation, scripting, and integration into CI/CD pipelines.

Business Capabilities

The system is organized around 6 functional domains, each representing a distinct business capability:

Technical Foundation

Module Composition

The codebase follows a layered architecture with clear separation of concerns:

Codebase Health

Overall Health: 🟢 95/100

Key Insights

• ✅ Clean module boundaries — No circular dependencies detected, enabling safe refactoring and testing.
• ✅ Healthy coupling — Most-imported module (internal/sql/format/format.go) serves 73 files without excessive centralization.

Key Data Flows

The system processes information through 6 identified flows:

• resolve.go Request Flow (critical) — Import chain starting from internal/compiler/resolve.go — traces how this entry point depends on 2 downstream modules
• output_columns.go Request Flow (critical) — Import chain starting from internal/compiler/output_columns.go — traces how this entry point depends on 2 downstream modules
• parse_core.go Request Flow (critical) — Import chain starting from internal/compiler/parse_core.go — traces how this entry point depends on 2 downstream modules
• query_catalog.go Request Flow (critical) — Import chain starting from internal/compiler/query_catalog.go — traces how this entry point depends on 2 downstream modules
• parse.go Request Flow (critical) — Import chain starting from internal/compiler/parse.go — traces how this entry point depends on 2 downstream modules

Strategic Observations

• The codebase follows Core/kernel architecture — `internal/compiler/catalog_core.go`, `internal/compiler/expand_core.go`, `internal/compiler/parse_core.go`, Multi-output publishing — `internal/compiler/output_columns.go`, Analysis pipeline — `internal/compiler/analyze.go`, Acyclic dependency graph — clean layering architectural patterns, providing a familiar structure for developers.
• With 6 distinct domains, the system has clear separation of concerns suitable for team-based development.
• As a CLI tool, this system is automation-friendly and can integrate into existing workflows and pipelines.

---

This executive summary was generated from comprehensive repository analysis including dependency graphs, data flow detection, and architectural pattern recognition. Enable AI enhancement with `GITHUB_TOKEN` for strategic recommendations, risk assessments, and plain-language explanations tailored for non-technical stakeholders.

<!-- repolens:docs-badge -->  · Examples

The Big Picture

sqlc is a medium-sized codebase with clear structure, containing 35 modules spread across 401 files.

---

Technology Stack

The system is built on a modern technology foundation:

Architecture Patterns

The codebase follows 4 identified architectural patterns:

• Core/kernel architecture — `internal/compiler/catalog_core.go`, `internal/compiler/expand_core.go`, `internal/compiler/parse_core.go` — A structured approach to organizing code and managing complexity.
• Multi-output publishing — `internal/compiler/output_columns.go` — A structured approach to organizing code and managing complexity.
• Analysis pipeline — `internal/compiler/analyze.go` — A structured approach to organizing code and managing complexity.
• Acyclic dependency graph — clean layering — The system operates as a command-line interface, accepting user input through terminal commands and producing structured output.

Module Architecture

The system is organized into 3 architectural layers:

Layer Details

Other Modules — 13 modules

• internal/sql/ast — 315 files — Core application module
• examples/ondeck — 15 files — Core application module
• examples/authors — 14 files — Authentication and authorization logic
• internal/sql/catalog — 9 files — Core application module
• +9 more modules

analyzer — 1 module

• internal/compiler/analyze.go — 1 files — Code analysis and inference

core — 1 module

• internal/compiler/catalog_core.go — 1 files — Core application module

Functional Domains

The following domains represent the largest functional areas by file count:

Dependency Graph

Codebase Health: 🟢 95/100

Hub modules (most imported):

• internal/sql/format/format.go — imported by 73 files

---

This system overview was generated from comprehensive repository analysis. Enable AI enhancement with `GITHUB_TOKEN` for narrative context, architectural rationale, technology trade-off analysis, and scalability recommendations.

<!-- repolens:docs-badge -->  · Examples

Overview

sqlc is organized into 6 functional domains, each representing a distinct area of business capability. This document maps the technical structure to business-oriented functional areas.

Quick Reference:

---

1. Data Layer

Business Capability

path internal/sql/ast matches sql

This is a major domain with 11 modules and 335 source files.

Key Modules

Business Impact

• Change Frequency Impact: High — Changes here affect 335 files
• Domain Complexity: Complex (multiple modules)
• Recommendation: Ensure comprehensive testing before modifying this domain

---

2. Other

Business Capability

Uncategorized module

This is a significant domain with 14 modules and 48 source files.

Key Modules

Business Impact

• Change Frequency Impact: Medium — Changes here affect 48 files
• Domain Complexity: Complex (multiple modules)

---

3. Search & Discovery

Business Capability

path internal/sql/catalog matches catalog

This is a focused domain with 4 modules and 12 source files.

Key Modules

Business Impact

• Change Frequency Impact: Low — Changes here affect 12 files
• Domain Complexity: Moderate

---

4. Core Engine

Business Capability

path internal/compiler/engine.go matches engine

This is a focused domain with 3 modules and 3 source files.

Key Modules

Business Impact

• Change Frequency Impact: Low — Changes here affect 3 files
• Domain Complexity: Moderate

---

5. Analysis & Detection

Business Capability

path internal/compiler/analyze.go matches analyz

This is a focused domain with 2 modules and 2 source files.

Key Modules

Business Impact

• Change Frequency Impact: Low — Changes here affect 2 files
• Domain Complexity: Simple

---

6. Rendering & Output

Business Capability

path internal/compiler/output_columns.go matches output

This is a focused domain with 1 module and 1 source file.

Key Modules

Business Impact

• Change Frequency Impact: Low — Changes here affect 1 file
• Domain Complexity: Simple

---

Cross-Domain Dependencies

The following modules serve as shared infrastructure across multiple domains:

Cross-Domain Integration Points

These modules facilitate communication and data sharing between domains:

• `internal/sql/format/format.go` serves as a central integration point, imported by 73 different modules. Changes to this module have wide-ranging impact and should be carefully coordinated.

Domain Health Summary

---

Domain mapping is based on directory naming conventions and code organization patterns. Enable AI enhancement with `GITHUB_TOKEN` for business-language descriptions, stakeholder impact analysis, strategic recommendations, and cross-domain relationship narratives.

<!-- repolens:docs-badge -->  · Examples

Architectural Philosophy

sqlc follows a Core/kernel architecture — `internal/compiler/catalog_core.go`, `internal/compiler/expand_core.go`, `internal/compiler/parse_core.go` architecture with elements of Multi-output publishing — `internal/compiler/output_columns.go`, Analysis pipeline — `internal/compiler/analyze.go`, Acyclic dependency graph — clean layering.

Pattern Analysis

Core/kernel architecture — `internal/compiler/catalog_core.go`, `internal/compiler/expand_core.go`, `internal/compiler/parse_core.go`

A structured approach to organizing code and managing complexity.

Benefits: Structured organization, maintainability, clarity

Multi-output publishing — `internal/compiler/output_columns.go`

A structured approach to organizing code and managing complexity.

Benefits: Structured organization, maintainability, clarity

Analysis pipeline — `internal/compiler/analyze.go`

A structured approach to organizing code and managing complexity.

Benefits: Structured organization, maintainability, clarity

Acyclic dependency graph — clean layering

The system operates as a command-line interface, accepting user input through terminal commands and producing structured output.

Benefits: Automation-friendly, scriptable, integrates with CI/CD pipelines

System Layers

The system is organized into 6 functional layers, each encapsulating a distinct area of responsibility:

Module Layers

Modules are classified into 3 architectural roles:

Other Modules

Role: General purpose functionality

analyzer

Role: Supporting application functionality

core

Role: Supporting application functionality

Technology Foundation

Scale & Complexity Analysis

Dependency Health

Architecture Health Score: 🟢 95/100

Architectural Strengths

• Clean module boundaries — No circular dependencies detected. This enables safe refactoring, isolated testing, and clear ownership boundaries.
• Balanced coupling — No single module dominates the import graph. This reduces change risk and enables parallel development.

Hub modules (most imported):

• internal/sql/format/format.go — imported by 73 files

---

This architecture overview was generated from comprehensive dependency graph analysis and pattern detection. Enable AI enhancement with `GITHUB_TOKEN` for architectural narratives, design rationale explanations, scalability analysis, and refactoring recommendations.

<!-- repolens:docs-badge -->  · Examples

A complete inventory of all code modules in the repository, organized by location and size. Each module represents a distinct area of responsibility within the codebase.

Total modules: 35

---

Module Inventory

---

Module detection is based on the `module_roots` setting in your RepoLens configuration. Adjust this setting to change how modules are grouped.

<!-- repolens:docs-badge -->  · Examples

A complete listing of user-facing pages and backend API endpoints. This map shows every URL a user or client can interact with, along with the source file that handles each route.

---

Route Detection

This project is a CLI tool — it does not expose HTTP routes or serve web pages. This is expected behavior, not a detection failure.

CLI tools interact through terminal commands rather than URLs. See the System Overview or Developer Onboarding documents for command documentation.

<!-- repolens:docs-badge -->  · Examples

This page documents both the API endpoints the application exposes and the external services it integrates with. Use it as a reference for understanding the system's interfaces.

---

Internal API Endpoints

These are the backend services the application provides to handle incoming requests.

No API routes were detected in this scan. If your project uses a routing framework (Express, Fastify, NestJS, etc.), ensure the relevant directories are included in your scan.include patterns.

---

External API Integrations

Third-party services the application connects to, grouped by category.

No external API integrations were detected. This section populates automatically when RepoLens identifies SDK imports or API client usage in your codebase.

---

HTTP method reference — GET: retrieve data, POST: create, PUT/PATCH: update, DELETE: remove.

<!-- repolens:docs-badge -->  · Examples

Understanding Data Movement

Data flows describe how information moves through sqlc — from external inputs through processing layers to storage or presentation. Understanding these flows is essential for debugging, performance optimization, and architectural decisions.

Flow Summary

6 data flows identified in the codebase:

---

1. resolve.go Request Flow 🔴 CRITICAL FLOW

Description

Import chain starting from internal/compiler/resolve.go — traces how this entry point depends on 2 downstream modules

Characteristics

Processing Steps

Flow Diagram

┌─ START ─┐
│ `resolve.go` (entry point, imports 334 modules)    │
│         ↓
│ `a_array_expr.go` (shared infrastructure, used ... │
│         ↓
│ `format.go` (shared infrastructure, used by 73 ... │
└─ END ───┘

Involved Modules

Dependencies

Shared libraries used: internal/sql/astutils

Insights

• ⚠️ Critical path — Failures here have significant business impact
• 📊 Consider adding monitoring and alerting
• 🧪 Ensure comprehensive test coverage

---

2. output_columns.go Request Flow 🔴 CRITICAL FLOW

Description

Import chain starting from internal/compiler/output_columns.go — traces how this entry point depends on 2 downstream modules

Characteristics

Processing Steps

Flow Diagram

┌─ START ─┐
│ `output_columns.go` (entry point, imports 329 m... │
│         ↓
│ `a_array_expr.go` (shared infrastructure, used ... │
│         ↓
│ `format.go` (shared infrastructure, used by 73 ... │
└─ END ───┘

Involved Modules

Dependencies

Shared libraries used: internal/sql/astutils

Insights

• ⚠️ Critical path — Failures here have significant business impact
• 📊 Consider adding monitoring and alerting
• 🧪 Ensure comprehensive test coverage

---

3. parse_core.go Request Flow 🔴 CRITICAL FLOW

Description

Import chain starting from internal/compiler/parse_core.go — traces how this entry point depends on 2 downstream modules

Characteristics

Processing Steps

Flow Diagram

┌─ START ─┐
│ `parse_core.go` (entry point, imports 329 modules) │
│         ↓
│ `a_array_expr.go` (shared infrastructure, used ... │
│         ↓
│ `format.go` (shared infrastructure, used by 73 ... │
└─ END ───┘

Involved Modules

Dependencies

Shared libraries used: internal/sql/astutils

Insights

• ⚠️ Critical path — Failures here have significant business impact
• 📊 Consider adding monitoring and alerting
• 🧪 Ensure comprehensive test coverage

---

4. query_catalog.go Request Flow 🔴 CRITICAL FLOW

Description

Import chain starting from internal/compiler/query_catalog.go — traces how this entry point depends on 2 downstream modules

Characteristics

Processing Steps

Flow Diagram

┌─ START ─┐
│ `query_catalog.go` (entry point, imports 327 mo... │
│         ↓
│ `a_array_expr.go` (shared infrastructure, used ... │
│         ↓
│ `format.go` (shared infrastructure, used by 73 ... │
└─ END ───┘

Involved Modules

Dependencies

Shared libraries used: internal/sql/astutils

Insights

• ⚠️ Critical path — Failures here have significant business impact
• 📊 Consider adding monitoring and alerting
• 🧪 Ensure comprehensive test coverage

---

5. parse.go Request Flow 🔴 CRITICAL FLOW

Description

Import chain starting from internal/compiler/parse.go — traces how this entry point depends on 2 downstream modules

Characteristics

Processing Steps

Flow Diagram

┌─ START ─┐
│ `parse.go` (entry point, imports 326 modules)      │
│         ↓
│ `a_array_expr.go` (shared infrastructure, used ... │
│         ↓
│ `format.go` (shared infrastructure, used by 73 ... │
└─ END ───┘

Involved Modules

Dependencies

Shared libraries used: internal/sql/astutils

Insights

• ⚠️ Critical path — Failures here have significant business impact
• 📊 Consider adding monitoring and alerting
• 🧪 Ensure comprehensive test coverage

---

6. Authentication Flow 🔴 CRITICAL FLOW

Description

User authentication and session management

Characteristics

Processing Steps

Flow Diagram

┌─ START ─┐
│ User submits login credentials                     │
│         ↓
│ Auth API validates credentials                     │
│         ↓
│ Session is created and stored                      │
│         ↓
│ User is redirected to authenticated area           │
│         ↓
│ Subsequent requests include auth token             │
└─ END ───┘

Involved Modules

Dependencies

Shared libraries used: internal/sql/astutils

External services: Authentication service

Insights

• ⚠️ Critical path — Failures here have significant business impact
• 📊 Consider adding monitoring and alerting
• 🧪 Ensure comprehensive test coverage

---

Import Network

The system has 10632 internal import edges connecting 362 source files.

Network Topology

Key Integration Points

These modules aggregate data from multiple sources:

---

Flow detection uses naming conventions, import patterns, and dependency graph analysis. Enable AI enhancement with `GITHUB_TOKEN` for end-to-end flow narratives, failure mode analysis, data transformation descriptions, and performance bottleneck identification.

<!-- repolens:docs-badge -->  · Examples

No changes detected.

<!-- repolens:docs-badge -->  · Examples

What is this? This visual diagram shows how different parts of your codebase connect to each other. Arrows indicate dependencies (which modules use which).

Showing: 30 modules and 50 relationships

Source: Real import analysis

---

Architecture Diagram

┌─────────────────────────────────────────────────────────────┐
│              🏗️  SYSTEM ARCHITECTURE MAP                    │
└─────────────────────────────────────────────────────────────┘

⚙️  CORE LOGIC
│
├── catalog_core.go (1 file)
│      → ast
│      → catalog
├── expand_core.go (1 file)
│      → ast
└── parse_core.go (1 file)
       → ast
       → astutils
       → named
       ... +2 more
│
🔍  ANALYSIS
│
└── analyze.go (1 file)
       → ast
       → named
       → preprocess
       ... +1 more
│
📋  RENDERING
│
└── format (1 file)
│
📤  PUBLISHING
│
└── output_columns.go (1 file)
       → ast
       → astutils
       → catalog
│
🛠️  UTILITIES
│
└── astutils (4 files)
       → ast
│
✅  TESTING
│
└── booktest (6 files)
│
📦  OTHER
│
├── ast (315 files)
│      → format
├── ondeck (15 files)
├── authors (14 files)
├── catalog (9 files)
│      → ast
├── preprocess (4 files)
│      → ast
│      → named
├── validate (4 files)
│      → ast
│      → astutils
│      → catalog
├── batch (2 files)
├── jets (2 files)
├── named (2 files)
├── compat.go (1 file)
│      → ast
│      → astutils
├── compile.go (1 file)
│      → ast
│      → preprocess
├── engine.go (1 file)
│      → catalog
├── expand.go (1 file)
│      → ast
│      → astutils
├── find_params.go (1 file)
│      → ast
│      → astutils
├── parse.go (1 file)
│      → ast
│      → astutils
│      → preprocess
│      ... +1 more
├── query.go (1 file)
│      → ast
│      → catalog
├── query_catalog.go (1 file)
│      → ast
│      → catalog
│      → preprocess
├── resolve.go (1 file)
│      → ast
│      → astutils
│      → catalog
│      ... +2 more
├── result.go (1 file)
│      → catalog
├── returning.go (1 file)
│      → ast
├── selector.go (1 file)
└── to_column.go (1 file)
       → ast
       → astutils
│

Legend:
  →  imports / depends on
  ╌→ tests

---

Key Connections

Most depended-on modules (highest import count):

Most dependent modules (highest dependency count):

<!-- repolens:docs-badge -->  · Examples

Welcome to sqlc! 👋

This guide will help you get up to speed quickly. sqlc is a general-purpose tools project built with Go, SQL, containing 35 modules across 401 files.

What You'll Learn

1. ✅ Repository structure and organization
2. ✅ Technology stack and tools
3. ✅ Key modules to understand first
4. ✅ How data flows through the system
5. ✅ How to start contributing

---

Quick Reference Card

---

Repository Structure

Understanding the directory layout is your first step to navigating the codebase:

Where to Begin

Start by exploring the main src/ directory or look for entry point files like index.js, main.js, or app.js.

---

Technology Stack

---

Core Modules to Understand

These are the most important modules by size and centrality. Understanding them will unlock the rest of the codebase:

Key Integration Points

These modules are imported by many other files. Changes here have wide impact:

• `internal/sql/format/format.go` — Used by 73 files — Core application module

⚠️ Caution: Be extra careful when modifying these files. Consider the ripple effects.

---

How Data Flows

Understanding these flows helps you see how the system processes information end-to-end:

Debugging tip: When investigating issues, trace the data flow from input to output to identify where problems occur.

---

Getting Started Checklist

Follow these steps to set up your development environment:

1. Clone and Install

git clone <repository-url>
cd sqlc
npm install  # or yarn, pnpm

2. Explore the Structure

• Review the directory structure above
• Open the README.md for project-specific instructions
• Look for CONTRIBUTING.md for contribution guidelines

3. Understand Core Modules

Start with the high-priority modules listed above. Read the code and comments to understand:

• What problem each module solves
• How modules interact with each other
• What data structures are used

4. Make Your First Change

Start small:

1. Find a small bug or typo to fix
2. Make the change in a feature branch
3. Run tests to ensure nothing breaks
4. Submit a pull request for review

---

Common Pitfalls

New contributors often run into these issues:

• Environment setup — Check for .env.example files and ensure you have all required environment variables.
• Branch naming — Follow the team's branch naming conventions (usually feature/, fix/, docs/).

---

Additional Resources

---

This onboarding guide was generated from comprehensive repository analysis. Enable AI enhancement with `GITHUB_TOKEN` for narrative walkthroughs, common pitfall documentation, debugging tips, architecture deep-dives, and personalized contribution recommendations.

Welcome to the team! Happy coding! 🚀

<!-- repolens:docs-badge -->  · Examples

No GraphQL schema was detected in this repository.

RepoLens scans for .graphql and .gql schema files, inline SDL via tagged template literals (e.g. gql\...\``), resolver patterns, and popular GraphQL libraries including Apollo, Yoga, Nexus, Pothos, Mercurius, type-graphql, Relay, and urql.

If your project uses GraphQL, ensure the relevant source directories are included in your scan.include configuration.

<!-- repolens:docs-badge -->  · Examples

No TypeScript type declarations were detected in this repository.

RepoLens analyzes .ts and .tsx files for interface declarations, type aliases, classes, and enums. Ensure your TypeScript source directories are included in the scan.include configuration.

<!-- repolens:docs-badge -->  · Examples

362 files · 10632 imports · 1 external packages · ✓ no circular dependencies · 2 orphan file(s)

This document maps every import relationship in the codebase, identifies the most-connected modules, and flags circular dependencies that may complicate refactoring or increase build times.

Overview

Hub Modules

These are the most-imported modules in the codebase. Changes to hub modules have the widest blast radius and should be reviewed carefully:

External Dependencies

The codebase imports 1 external package. These are third-party modules (not the language standard library):

---

Generated by RepoLens import analysis. Dependency detection covers JavaScript and TypeScript (ES modules, dynamic import, CommonJS, re-exports; comments, strings, and type-only imports are skipped), Python (`import` / `from ... import`, including relative imports), and Go (`go.mod` module paths). Go `_test.go` files are not part of the graph.*

<!-- repolens:docs-badge -->  · Examples

No baseline found. Run repolens publish once to establish a baseline.

A baseline snapshot of the current architecture has been saved. On subsequent runs, this report will track structural changes across the following dimensions:

<!-- repolens:docs-badge -->  · Examples

No security anti-patterns were detected in the scanned source files.

RepoLens scans JavaScript and TypeScript source files for common security risks including code injection (eval), XSS (innerHTML), SQL injection, command injection, prototype pollution, hardcoded credentials, and insecure randomness.

Test files, configs, and build outputs are excluded from analysis to reduce false positives.

<!-- repolens:docs-badge -->  · Examples

362 files analyzed · Average health: 94/100 · 325 A, 13 B, 19 C, 5 D, 0 F

This report synthesizes complexity, coupling, documentation, and security signals into a single health score per module. Lower scores indicate higher risk and maintenance burden.

Overview

Grade Distribution

Hotspots — Priority Refactoring Targets

These modules have the lowest health scores and should be prioritized for improvement:

Most Complex Functions

These functions have the highest cyclomatic complexity and are the most likely sources of bugs:

All Modules by Health Score

Showing 50 of 362 modules (sorted worst-first)

Scoring Methodology

The health score (0–100) is computed from four dimensions:

---

Generated by RepoLens code health analysis. Complexity covers JavaScript/TypeScript, Python, and Go. It is regex-based and may differ from AST-based tools. Scores are relative indicators, not absolute quality measures.

<!-- repolens:docs-badge -->  · Examples