pip install git+https://github.com/the-deep/deepex
from deep_parser import TextFromFile
PDF_DOCUMENT = [YOUR_DOCUMENT_PATH]
with open(PDF_DOCUMENT_PATH,'rb') as f:
binary = base64.b64encode(f.read())
document = TextFromFile(stream=binary, ext="pdf")
text, other = document.extract_text()
TextFromFile class also accepts a PDF document URL from a website:
from deep_parser import TextFromFile
PDF_URL = [DOCUMENT_URL]
document = TextFromFile(stream=None, ext="pdf", from_web=True, url=PDF_URL)
text, other = document.extract_text()
extract_text() method returns a tupla with the extracted text and a Results class instance.
Output format can be selected with the output_format parameter:
text, images = document.extract_text(output_format="list")
return list-formatted text. Results instance can be used for document images processing or saving, for example:
text, other = document.extract_text()
other.save_images(directory_path = DIRECTORY_PATH)
extract_text() method can be replaced with extract_text_multi() for a multi-processing management of extraction.
You can also extract text from webpages/html:
from deep_parser import TextFromWeb
URL = WEBSITE_URL
webpage = TextFromWeb(url=URL)
text = webpage.extract_text(output_format="list", url=URL)