I have a Local PDF document that contains various structure of tables (meaning that it's not a static structure table and I can't predict what does the table look like) and I need to do document processing to extract the tables from the PDF file in power automate, so I used the layout model service from document intelligence, this is details of the model: Microsoft pre-built layout model and to use this model via Power automate, I need to do Rest API call, this is my flow look like:
1- Get file content using path action (from file system)
2- POST HTTP request to call layout model
3- GET HTTP request to get result from layout model
4- parse JSON action to parse the output from GET HTTP request
so, this model returns JSON response with the extracted table in this structure:
"tables": [
{
"rowCount": 1,
"columnCount": 2,
"cells": [
{
"kind": "columnHeader",
"rowIndex": 0,
"columnIndex": 0,
"content": "Name",
"boundingRegions": [],
"spans": []
},
{
"kind": "columnHeader",
"rowIndex": 0,
"columnIndex": 1,
"content": "Age",
"boundingRegions": [],
"spans": []
},
{
"rowIndex": 1,
"columnIndex": 0,
"content": "Muhammed",
"boundingRegions": []
},
{
"rowIndex": 1,
"columnIndex": 1,
"content": "15",
"boundingRegions": []
}
}
]
]
the cell with "kind" columnHeader is a header cell, and the cell without "kind" key is a row cell. I need to convert the structure of table above into CSV in Power automate.