Textract extracts different text accroding to the OS it operates on

[bennnym]

Describe the bug I operate locally on a Mac and a simple test from a sample pdf passes locally, but fails in a Docker container.

To Reproduce Steps to reproduce the behaviour:

1. Download pdf from here to root of your folder.
2. name the pdf "sample.pdf"
3. add the following dependancies to a requirements.txt file ( in the root ) and pip install them with pip install -r requirements.txt

   textract==1.6.5
   pytest==6.2.4

4. Create a test in a file also placed in the root of your project @ test_textract.py

   
   import textract

class TestTextract: def test_occorunces_of_string(self): path = "sample_sectionals.pdf" extracted_text = textract.process(path, method="pdfminer").decode("utf-8") while "\n" in extracted_text or "\n" in extracted_text: extracted_text = extracted_text.replace("\n", " ") extracted_text = extracted_text.replace("\n", " ")

5. Run test with command: `pytest -s -vv`
6.  Add a Dockerfile in the root with the following content:
```docker
FROM        python:3.8.13-slim-buster

RUN         pip install --upgrade pip

COPY        . /opt/code

RUN         pip install textract==1.6.5 pytest==6.2.4

WORKDIR     /opt/code

ENTRYPOINT  [ "/bin/sh" ]

8. docker build . -t test
9. docker run test -c "pytest -s -vv test_textract.py"

Expected behaviour This test passes locally, and fails in Docker container ( with zero occurrences )

Screenshots If applicable, add screenshots to help explain your problem.

Desktop (please complete the following information):

• OS: [MacOS Monterey]
• Textract version [e.g. 1.6.5]
• Python version [e.g. 3.8]
• Virtual environment (yes)