chat-with-your-data-solution-accelerator icon indicating copy to clipboard operation
chat-with-your-data-solution-accelerator copied to clipboard

Tables from word documents are not read

Open andrewwiebe opened this issue 1 year ago • 1 comments

https://github.com/Azure-Samples/chat-with-your-data-solution-accelerator/blob/a9997b075f4bf8b228afa79c4b2fbc552846afd4/code/backend/batch/utilities/document_loading/word_document.py#L35

We are just starting to roll out a production version of this application. A user uploaded a word doc that has all their text in tables. The batch was not picking up any text at all. I realized that the code to get the text from word documents is only loading the document.paragraphs objects from word documents which means it is skipping anything that is in a table in a word document.

andrewwiebe avatar Oct 24 '24 15:10 andrewwiebe