0
# Document Processing
1

2
Document representation and processing capabilities for content management in LangChain applications. Documents are the fundamental units of text data used throughout the framework.
3

4
## Capabilities
5

6
### Document
7

8
Core document representation containing text content and metadata.
9

10
```typescript { .api }
11
/**
12
 * Core document representation
13
 * @template Metadata - Type of metadata object
14
 */
15
class Document<Metadata = Record<string, unknown>> {
16
  /** Main text content of the document */
17
  pageContent: string;
18
  /** Associated metadata */
19
  metadata: Metadata;
20
  /** Optional unique identifier */
21
  id?: string;
22
  
23
  constructor(fields: DocumentInput<Metadata>);
24
  
25
  /** Convert to JSON representation */
26
  toJSON(): Serialized;
27
  
28
  /** Create document from JSON */
29
  static fromJSON(json: Serialized): Document;
30
}
31
```
32

33
**Usage Examples:**
34

35
```typescript
36
import { Document } from "@langchain/core/documents";
37

38
// Simple document
39
const doc1 = new Document({
40
  pageContent: "LangChain is a framework for building applications with LLMs.",
41
  metadata: {
42
    source: "documentation",
43
    category: "introduction"
44
  }
45
});
46

47
// Document with custom metadata type
48
interface ArticleMetadata {
49
  title: string;
50
  author: string;
51
  publishDate: Date;
52
  tags: string[];
53
}
54

55
const article = new Document<ArticleMetadata>({
56
  pageContent: "This is the article content...",
57
  metadata: {
58
    title: "Understanding LangChain",
59
    author: "John Doe", 
60
    publishDate: new Date("2024-01-15"),
61
    tags: ["langchain", "llm", "tutorial"]
62
  },
63
  id: "article-123"
64
});
65

66
console.log(article.pageContent); // "This is the article content..."
67
console.log(article.metadata.title); // "Understanding LangChain"
68
```
69

70
## Types
71

72
```typescript { .api }
73
interface DocumentInput<Metadata = Record<string, unknown>> {
74
  /** Main text content */
75
  pageContent: string;
76
  /** Document metadata */
77
  metadata?: Metadata;
78
  /** Optional document identifier */
79
  id?: string;
80
}
81

82
interface DocumentInterface<Metadata = Record<string, unknown>> {
83
  pageContent: string;
84
  metadata: Metadata;
85
  id?: string;
86
}
87
```