Add DocuForge API — Document processing (Excel, CSV, PDF)

[rnhowcla]

Add DocuForge API OpenAPI 3.0 spec.

Document processing REST API for Excel, CSV, and PDF files.
Free tier: 50 calls/month.

GitHub: https://github.com/rnhowcla/docuforge-api

[gemini-code-assist]

Code Review

This pull request introduces the initial OpenAPI specification for the DocuForge API, which includes endpoints for Excel, CSV, and PDF processing. The review feedback highlights a critical security issue regarding the use of HTTP and a raw IPv6 address for the production server, which should be updated to HTTPS and a domain name. Additionally, several endpoints are missing 'content' definitions in their responses, which is necessary for proper documentation and client generation. Finally, the 'created_at' field in the account endpoint should include a 'date-time' format for better schema precision.

[gemini-code-assist]

The production server is configured with a raw IPv6 address and uses the insecure http protocol. For a production API that requires authentication via X-API-Key, it is critical to use https to prevent credentials from being transmitted in cleartext. Additionally, using a domain name instead of a static IP is standard practice for API stability and discoverability.

[gemini-code-assist]

The response for a successful CSV conversion is missing a content definition. Without specifying the media type (e.g., text/csv) and schema, client generators and documentation tools cannot correctly handle the output.

          description: CSV file
          content:
            text/csv:
              schema:
                type: string
                format: binary

[gemini-code-assist]

The response for the auto-format endpoint is missing the content definition. It should specify the Excel media type to ensure proper handling by clients.

          description: Formatted Excel file
          content:
            application/vnd.openxmlformats-officedocument.spreadsheetml.sheet:
              schema:
                type: string
                format: binary

[gemini-code-assist]

The response for the CSV-to-Excel conversion is missing the content definition. It should specify the Excel media type as seen in the /excel/clean endpoint.

          description: Excel file
          content:
            application/vnd.openxmlformats-officedocument.spreadsheetml.sheet:
              schema:
                type: string
                format: binary

[gemini-code-assist]

The response for the CSV cleaning endpoint is missing the content definition. It should specify the text/csv media type.

          description: Cleaned CSV file
          content:
            text/csv:
              schema:
                type: string
                format: binary

[gemini-code-assist]

The text extraction endpoint is missing a content definition for the response. It should likely return text/plain with a string schema.

          description: Extracted text
          content:
            text/plain:
              schema:
                type: string

[gemini-code-assist]

The PDF metadata endpoint is missing the content definition. Since the description mentions JSON, it should define application/json along with the expected schema for the metadata object (e.g., properties for pages, author, title, etc.).

[gemini-code-assist]

The PDF merge endpoint is missing the content definition for the resulting PDF file. This is necessary for tools to recognize the binary output as a PDF.

          description: Merged PDF
          content:
            application/pdf:
              schema:
                type: string
                format: binary

[gemini-code-assist]

The created_at property should include a format: date-time to provide better information for client generators and ensure consistent parsing of the timestamp.

                    type: string
                    format: date-time