Problem Description

I have a directory containing hundreds of PDF files. I want to convert them to text files using Google's Cloud Vision API. I would like to have a script that does the following:

1. Take two directory names as input - input-dir, output-dir
2. Crawls the input-dir directory to read the pdf files
3. Converts the pdf files into the same structure in the output-dir.
4. It should use google json auth through the standard GOOGLE_APPLICATION_CREDENTIALS environment variable.

Acceptance Criteria

the script uses python 3.10+
Any libraries must be specified in a requirements.txt file.
It should create a one-to-one mapping between the pdf file and the text file using Google's Cloud Vision API.