import fitz

doc = fitz.open(r'C:\git\spark-lesson\reference\sources\liu-discharge-transitions-thesis.pdf')
print(f'Total pages: {len(doc)}')

output = []
for i in range(len(doc)):
    text = doc[i].get_text()
    if text.strip():
        output.append(f'=== PAGE {i+1} ===')
        output.append(text)

full_text = '\n'.join(output)

with open(r'C:\git\spark-lesson\reference\sources\liu-discharge-transitions-thesis.txt', 'w', encoding='utf-8') as f:
    f.write(full_text)

print(f'Extracted {len(output)//2} pages')
print(f'Total characters: {len(full_text)}')
print(f'Total lines: {full_text.count(chr(10))}')