from typing import List

import pytesseract
from PIL import Image

import gradio as gr

def tesseract_ocr(filepath: str, languages: List[str]):
    image = Image.open(filepath)
    oem_psm_config = '--oem 3 --psm 11 --tessdata-dir ./tessdata'
    return pytesseract.image_to_string(image=image, lang='+'.join(languages), config=oem_psm_config)

title = "Shan Tesseract OCR"
description = "Gradio demo for Tesseract-OCR Shan. Tesseract is an open source text recognition (OCR) Engine."
article = "<p style='text-align: center'><a href='https://tesseract-ocr.github.io/' target='_blank'>Tesseract documentation</a> | <a href='https://github.com/tesseract-ocr/tesseract' target='_blank'>Github Repo</a></p>"

examples = [
    ["examples/example2.png", ["eng", "shn"]],
    ["examples/example3.png", ["eng", "shn"]],
    ["examples/example4.png", ["eng", "shn"]],
    ["examples/example1.png", ["eng", "shn"]],
]

language_choices = pytesseract.get_languages(config='--tessdata-dir ./tessdata')

demo = gr.Interface(
    fn=tesseract_ocr, 
    inputs=[
        gr.Image(type="filepath", label="Input"), 
        gr.CheckboxGroup(language_choices, type="value", value=['eng'], label='language')
        ],
    outputs='text',
    title=title,
    description=description,
    article=article,
    examples=examples,
)

if __name__ == '__main__':
    demo.launch()
    print("Finished running")