Be sure to include all the following modules in the requirements.txt file and install them into your virtual environment:

beautifulsoup4==4.6.0
Pillow==5.1.0
PyPDF2==1.26.0
python-docx==0.8.6

Check that the directory to search has the following files (all are available in GitHub in the Chapter04/documents directory). Note that file5.pdf and file6.pdf are copies of document-1.pdf, for simplicity. file1.txt to file4.txt are empty files:

├── dir
│   ├── file1.txt
│   ├── file2.txt
│   ├── file6.pdf
│   └── subdir
│       ├── file3.txt
│       ├── file4.txt
│       └── file5.pdf
├── document-1.docx
├── document-1.pdf
├── document-2-1.pdf
├── document-2.pdf
├── example_iso.txt
├── example_output_iso.txt
├── example_utf8.txt
├── top_films.csv
└── zen_of_python.txt

We've prepared a script, scan.py, that will search for a word in all the .txt, .csv, .pdf, and .docx files. The script is available in the Chapter04 directory of the GitHub repository.