Document parsing guide

idp/documentParsing converts a document into structured JSON and Markdown. It is commonly used before extraction, search, RAG, or review workflows.

curl --location --request POST 'https://api-server.compdf.com/server/v2/process/idp/documentParsing' \
  --header 'x-api-key: <your-public-key>' \
  --form 'file=@/path/to/document.pdf' \
  --form 'pageRanges=1-3' \
  --form 'parseOptions={"applyDocumentTree":true,"mergeTables":true}'

What this endpoint returns

For a typical request, the response body contains:

Path	What it is used for
code, message, x_request_id	Request status and troubleshooting
file_type	Parsed input type, for example PDF
result	Main parse result object, including page output and summary counters
metrics	Page-level processing metadata such as dpi, angle, and duration
image_process	Extra image-processing output; usually empty for standard parsing requests

The most commonly used fields inside result are:

Path	What it is used for
result.pages	Per-page parse output, including structured and content
result.markdown	A merged Markdown view of the document
result.catalog	Document tree / TOC when hierarchy detection is enabled
result.valid_page_number	Number of successfully parsed pages
result.total_page_number	Total page count of the input file
result.success_count	Count of successfully processed pages or result units
result.detail	Flat paragraph-level list merging all pages in reading order
result.excel_base64	Base64-encoded Excel output when export format includes Excel

Recommended reading order

• Start with Parse options to understand request parameters such as image_type, content_filter, options_json, and ignore_labels.
• Then read Response structure for the top-level JSON contract.
• Use Page details when you need to map headings, tables, and footnotes back to page coordinates.
• Use Metrics when you are monitoring processing quality or performance.