We are facing problem when using Custom GPT actions to read a PDF file stored on Google Drive, Although the connection to the Google API works perfectly for Google Docs, it seems unable to read or extract the text from PDF files. I’m attaching the problem message given by Custom GPT for your reference and the schema we used (1st schema is accessing and downloading the pdf file but could not able to read the content where as in the 2nd schema we wanted to change its document’s format from pdf and then try to read it but we unable to do it) .
openapi: 3.1.0
info:
title: Google Drive PDF API
description: API to interact with Google Drive for handling PDF files, including listing, downloading, and converting PDF files.
version: 1.0.0
servers:
- url: https://www.googleapis.com/drive/v3
description: Google Drive API v3
paths:
/files:
get:
operationId: listPDFFiles
summary: List PDF files in Google Drive
description: Retrieves a list of PDF files in the user's Google Drive.
parameters:
- name: q
in: query
description: Query string for searching files. Only PDFs will be returned.
required: false
schema:
type: string
example: "mimeType='application/pdf'"
- name: fields
in: query
description: Selector specifying a subset of fields to include in the response.
required: false
schema:
type: string
example: "files(id,name,mimeType)"
responses:
'200':
description: A list of PDF files
content:
application/json:
schema:
type: object
properties:
files:
type: array
items:
type: object
properties:
id:
type: string
name:
type: string
mimeType:
type: string
/files/{fileId}:
get:
operationId: downloadPDFFile
summary: Download a PDF file from Google Drive
description: Downloads a PDF file from Google Drive by its file ID.
parameters:
- name: fileId
in: path
description: The ID of the PDF file to download.
required: true
schema:
type: string
responses:
'200':
description: PDF file downloaded successfully
content:
application/octet-stream:
schema:
type: string
format: binary
/files/{fileId}/export:
get:
operationId: convertPDFFile
summary: Convert a PDF file to another format
description: Converts a PDF file from Google Drive to a specified format (e.g., text or image).
parameters:
- name: fileId
in: path
description: The ID of the PDF file to convert.
required: true
schema:
type: string
- name: mimeType
in: query
description: The MIME type of the format to convert the PDF file to. Example formats could include "text/plain" or "image/png".
required: true
schema:
type: string
example: "text/plain"
responses:
'200':
description: PDF file converted successfully
content:
application/octet-stream:
schema:
type: string
format: binary
components:
securitySchemes:
apiKeyAuth:
type: apiKey
in: query
name: key
schemas:
File:
type: object
properties:
id:
type: string
name:
type: string
mimeType:
type: string
security:
- apiKeyAuth: []
schema 2:
openapi: 3.1.0
info:
title: Google Drive PDF API
description: API to interact with Google Drive for handling PDF files, including listing and downloading PDF files.
version: 1.0.0
servers:
- url: https://www.googleapis.com/drive/v3
description: Google Drive API v3
paths:
/files:
get:
operationId: listPDFFiles
summary: List PDF files in Google Drive
description: Retrieves a list of PDF files in the user's Google Drive.
parameters:
- name: q
in: query
description: Query string for searching files. Only PDFs will be returned.
required: false
schema:
type: string
example: "mimeType='application/pdf'"
- name: fields
in: query
description: Selector specifying a subset of fields to include in the response.
required: false
schema:
type: string
example: "files(id,name,mimeType)"
responses:
'200':
description: A list of PDF files
content:
application/json:
schema:
type: object
properties:
files:
type: array
items:
type: object
properties:
id:
type: string
name:
type: string
mimeType:
type: string
/files/{fileId}:
get:
operationId: downloadPDFFile
summary: Download a PDF file from Google Drive
description: Downloads a PDF file from Google Drive by its file ID.
parameters:
- name: fileId
in: path
description: The ID of the PDF file to download.
required: true
schema:
type: string
responses:
'200':
description: PDF file downloaded successfully
content:
application/octet-stream:
schema:
type: string
format: binary
/files/{fileId}/export:
get:
operationId: convertPDFFile
summary: Convert a PDF file to another format
description: Converts a PDF file from Google Drive to a specified format (e.g., text or image).
parameters:
- name: fileId
in: path
description: The ID of the PDF file to convert.
required: true
schema:
type: string
- name: mimeType
in: query
description: The MIME type of the format to convert the PDF file to. Example formats could include "text/plain" or "image/png".
required: true
schema:
type: string
example: "text/plain"
responses:
'200':
description: PDF file converted successfully
content:
application/octet-stream:
schema:
type: string
format: binary
components:
securitySchemes:
apiKeyAuth:
type: apiKey
in: query
name: key
schemas:
File:
type: object
properties:
id:
type: string
name:
type: string
mimeType:
type: string
security:
- apiKeyAuth: []
The goal is to use the Google Drive API to access and retrieve these PDF files, and then use a custom GPT (Generative Pre-trained Transformer) model to read and extract text from them. The extracted text will be used for further processing.
Expected Outcome: A solution that can access PDF files in Google Drive, extract the text from them, and integrate this text with a custom GPT model for various applications, such as content analysis or automated summarization.
