Use text vertices of Google OCR to print text with its original indentation

Question

Is it possible to convert vertices of Google OCR vision into a text preserving it's original indentation?

For instance using the example here, for the following image

Google vision returns each detected text with it's 4-vertices (it's Cartesian coordinates).

After executing the python function "detect_text" we get:

Result :

"WAITING? PLEASE TURN OFF YOUR ENGINE" bounds: (52,137),(375,137),(375,330),(52,330)

"WAITING" bounds: (59,137),(342,151),(340,190),(57,177)

"?" bounds: (345,151),(375,152),(373,191),(343,190)

"PLEASE" bounds: (204,205),(318,210),(317,232),(203,227)

"TURN" bounds: (205,236),(288,239),(287,261),(204,258)

"OFF" bounds: (301,240),(356,242),(355,263),(300,261)

"YOUR" bounds: (205,269),(290,272),(289,293),(204,290)

"ENGINE" bounds: (205,302),(323,304),(323,326),(205,324)

I've found different ways to plot these coordinates, but how can I print the result with it's original indentation?

Expected output :

WAITING ?

       PLEASE
       TURN OFF
       YOUR
       ENGINE

Of course, this will not preserve text formatting, but at least we'll keep relatively original spaces and tabulations!

Thank you in advance!