# Parser

**Parser** is responsible for defining element structures based on output from **Loader**.

To give you an idea what a Parser does, this is a snippet of a sample JSON output:

```json
{
    "text": "[Header] This is the Header of the Document",
    "structure": "header", // 👈 parser defines this as "header" (previous value from Loader is "uncategorized")
    "metadata": {
        "source": "pdf-example.pdf",
        "source_type": "pdf",
        ...
    }
}
```

You can compare with the output from Loader [here](https://gdplabs.gitbook.io/sdk/~/revisions/w6A7tUKJGDYFXuci5HcW/tutorials/document-processing-orchestrator/loader).

Possible `structure` values:

1. `PAGE`
2. `HEADER` 
3. `TITLE`
4. `HEADING` (`HEADING 1` through `HEADING 6` )
5. `PARAGRAPH` 
6. `FOOTER`
7. `FOOTNOTE`
8. `TABLE`
9. `IMAGE`
10. `AUDIO`
11. `VIDEO`
12. `UNCATEGORIZED` 

Our Parser has the following sub components:

1. Document
   1. DOCX Parser
   2. PDF Parser
   3. PPTX Parser
   4. TXT Parser
   5. XLSX Parser
2. HTML parser
   1. HTML Flat Paser
3. Image parser
   1. Image MIME Normalizer Parser
   2. Image Plain Small Filter Parser
4. Table Parser
   1. Table Caption Parser