need to analyze the docx files using the Azure AI Document Intelligence client library for Python - version 1.0.0b4

S,A 20

Hi Team,

I need to analyze the docx files using the Azure AI Document Intelligence client library for Python - version 1.0.0b4.

What I have observed is that when I convert the docx file to a pdf I am able to get better metadata of that pdf file including page numbers.
Below attached image is result of a pdf file.

User's image

This is not the same for the docx files where I get only one document element which includes entire docx file information.
I am looking for an API that can convert MS-DOCX file to PDF on linux machines(Azure app service).
I tried libreoffice(soffice) but there is mismatch in page format for MS-DOCX and libreoffice converted documents.
In the Azure AI | Document Intelligence Studio there is an option "print to PDF" for non pdf files. I was wondering if I could get any Azure API service to convert docx file to pdf.
User's image

Please help with any solution.

YutongTie-MSFT 52,776 Reputation points

2024-10-04T17:20:00.9366667+00:00

Hello @S,A

Thanks for reaching out to us, in addition to the answer below, I am happy to answer your last question, there is currently no direct API as part of Azure Document Intelligence to Print to PDF as the Azure Document Intelligence Studio UI, I would suggest you to try Graph API as well.

Please let us know if there is any question.

Regards,

Yutong
YutongTie-MSFT 52,776 Reputation points

2024-10-07T15:56:50.53+00:00

Hello @S,A

Just checking in to see if the below answer helped. If this answers your query, do click Accept Answer and Yes for was this answer helpful. And, if you have any further query do let us know. Thanks a lot.

Regards,

Yutong
S,A 20 Reputation points

2024-10-15T14:36:16.0066667+00:00
Hi Yutong,

Thanks for your response, I tried Graph API (REST APIs via Postman) to convert DOCX files in my personal One Drive to PDF and it worked.

However, there are more things that did not work.

I have tried using the msgraph-sdk Python SDK to convert the docx file. The PIP install itself failed as msgraph-sdk has super long filepaths. I had to enable longfilepath in the Windows Registry to complete the installation.

I wrote Python code to create msgraph client and convert docx file to PDF. Looks like I need some drive-id, file-id, etc to find and convert the files. I was unable to get these details on my personal one-drive files. While using the REST APIs, I used the GET API on URL: https://graph.microsoft.com/v1.0/me/drive/root:/Documents/my_file.docx:/content?format=pdf with Bearer token. Is there a similar Python API that can fetch and convert files using the file path instead of file_id?

I am unable to find a APIs to get refresh token for using graph REST APIs via Python code. Also a programmatic way to give consent to read files from one-drive via Graph service.

I am unable to understand how to request creation of a separate One-Drive, as the solution I am trying build is at enterprise level (highly confidential) and I can't use my personal one drive account and files in it.

Like I said, I was able to use the Graph REST APIs via Postman application to download docx files as PDF, I wrote a simple Python API to do the same using request library, GET API and Authorization token in the headers. It worked in my local machine. But, when I deployed the API in my Azure App Service, it gives a NULL response. I am thinking this could be due to Azure App Service Network settings where Graph API should be marked as allowed. Could you please help me with how to do that?

I understand my questions are too many. Please bear with me as I am confused connecting Azure and Office365 services together in one solution.
YutongTie-MSFT 52,776 Reputation points

2024-10-15T19:51:40.89+00:00

@S,A Thanks for your response, I can understand that there are a lot of collaboration need from Microsoft Graph and Azure Service, but I do recommend you raising a ticket on Microsoft Graph side - https://developer.microsoft.com/en-us/graph/support since Microsoft Graph is not supported here, experts here for Graph API is very limited.

Regards,

Yutong

Accepted answer

Vahid Ghafarpour 21,725 Reputation points

2024-10-04T15:41:12.0066667+00:00

Thanks for posting your question in the Microsoft Q&A forum.

I'd try MS Graph API:

https://learn.microsoft.com/en-us/graph/api/driveitem-get-content-format?view=graph-rest-1.0&tabs=http

https://learn.microsoft.com/en-us/archive/msdn-magazine/2016/june/azure-app-services-using-azure-app-services-to-convert-a-web-page-to-pdf

https://help.syncfusion.com/document-processing/word/conversions/word-to-pdf/net/convert-word-document-to-pdf-in-azure-functions-v1

** Please don't forget to close up the thread here by upvoting and accept it as an answer if it is helpful **
Please sign in to rate this answer.
S,A 20 Reputation points

2024-10-15T14:37:36.69+00:00

Hi Vahid,

Thanks for your response, I tried Graph API (REST APIs via Postman) to convert DOCX files in my personal One Drive to PDF and it worked.

However, there are more things that did not work.

I have tried using the msgraph-sdk Python SDK to convert the docx file. The PIP install itself failed as msgraph-sdk has super long filepaths. I had to enable longfilepath in the Windows Registry to complete the installation.

I wrote Python code to create msgraph client and convert docx file to PDF. Looks like I need some drive-id, file-id, etc to find and convert the files. I was unable to get these details on my personal one-drive files. While using the REST APIs, I used the GET API on URL: https://graph.microsoft.com/v1.0/me/drive/root:/Documents/my_file.docx:/content?format=pdf with Bearer token. Is there a similar Python API that can fetch and convert files using the file path instead of file_id?

I am unable to find a APIs to get refresh token for using graph REST APIs via Python code. Also a programmatic way to give consent to read files from one-drive via Graph service.

I am unable to understand how to request creation of a separate One-Drive, as the solution I am trying build is at enterprise level (highly confidential) and I can't use my personal one drive account and files in it.

Like I said, I was able to use the Graph REST APIs via Postman application to download docx files as PDF, I wrote a simple Python API to do the same using request library, GET API and Authorization token in the headers. It worked in my local machine. But, when I deployed the API in my Azure App Service, it gives a NULL response. I am thinking this could be due to Azure App Service Network settings where Graph API should be marked as allowed. Could you please help me with how to do that?

I understand my questions are too many. Please bear with me as I am confused connecting Azure and Office365 services together in one solution.
Sign in to comment

Use comments to ask for clarification, additional information, or improvements to the question.

Share via

need to analyze the docx files using the Azure AI Document Intelligence client library for Python - version 1.0.0b4

0 additional answers

Your answer