# Document

**Kind:** Database Model

**Source:** [`atloria-monorepo/libs/database/prisma/schema.prisma`](https://github.com/sherkety/atloria/blob/main/atloria-monorepo/libs/database/prisma/schema.prisma#L375)

`Document` stores structured document content and its rendered HTML representation. It associates a document with a URL identifier and tracks classification, origin, lifecycle state, and page type for use across content ingestion, rendering, and publishing workflows.

## Fields

| Field | Type | Required | Key |
|---|---|---|---|
| `id` | `String` | ✓ | PK |
| `urlId` | `String` | - | unique |
| `title` | `String` | ✓ |  |
| `slug` | `String` | ✓ |  |
| `content` | `String` | ✓ |  |
| `contentHtml` | `String` | - |  |
| `type` | `DocumentType` | ✓ |  |
| `source` | `DocumentSource` | ✓ |  |
| `state` | `DocumentState` | ✓ |  |
| `pageType` | `String` | - |  |
| `version` | `Int` | ✓ |  |
| `parentVersionId` | `String` | - |  |
| `canRegenerate` | `Boolean` | ✓ |  |
| `kgEntityId` | `String` | - |  |
| `kgEntityType` | `String` | - |  |
| `isPublic` | `Boolean` | ✓ |  |
| `publishedAt` | `DateTime` | - |  |
| `scheduledPublishAt` | `DateTime` | - |  |
| `parentId` | `String` | - |  |
| `collectionId` | `String` | - |  |
| `toc` | `Json` | - |  |

## Diagram

```mermaid
erDiagram
  DOCUMENT {
    String id
    String urlId
    String title
    String slug
    String content
    String contentHtml
    DocumentType type
    DocumentSource source
    DocumentState state
    String pageType
  }
```

## Usage

```ts
import { PrismaClient, type DocumentSource, type DocumentState, type DocumentType } from "@prisma/client";

const prisma = new PrismaClient();

async function createDocument(
  type: DocumentType,
  source: DocumentSource,
  state: DocumentState,
) {
  const document = await prisma.document.create({
    data: {
      id: crypto.randomUUID(),
      urlId: "url_123",
      title: "Getting Started",
      slug: "getting-started",
      content: "# Getting Started\n\nDocument source content.",
      contentHtml: "<h1>Getting Started</h1><p>Document source content.</p>",
      type,
      source,
      state,
      pageType: "documentation",
    },
  });

  return document;
}
```

## AI Coding Instructions

- Keep `content` and `contentHtml` synchronized: store source content in `content` and the rendered output in `contentHtml`.
- Use the generated Prisma client (`prisma.document`) for reads and writes rather than constructing database queries manually.
- Provide valid `DocumentType`, `DocumentSource`, and `DocumentState` enum values defined in the Prisma schema; do not persist arbitrary strings for these fields.
- Treat `slug` and `urlId` as important lookup and routing identifiers; validate their values before creating or updating documents.
- Update document state deliberately as part of ingestion and publishing workflows so consumers can distinguish draft, processed, and published content.

## Relationships

- HAS_ONE → `Document`
- HAS_ONE → `Document`
- HAS_ONE → `Collection`

## Referenced By

- `User` (HAS_ONE)
- `User` (HAS_ONE)
- `Document` (HAS_ONE)
- `Document` (HAS_ONE)
- `Comment` (HAS_ONE)
- `Suggestion` (HAS_ONE)
- `CodeSymbol` (HAS_ONE)
- `CollaborationSession` (HAS_ONE)
- `DocumentVersion` (HAS_ONE)
- `DocumentRevision` (HAS_ONE)