'''
This Python code uses OpenCV to convert coloured images to Grayscale and then
the PNG files are converted back into a Grayscale or BW PDF

User needs to specify threshold value of the pixel color to convert to black 
and white. This needs to be tried for each type of PDF file as the quality of 
colour varies for each image. The script finall converts the images into a PDF

syntax: py pdf-To-BW-Pdf.py Input.pdf 127 1 234 where 127 is threshold value, 1
is state page and 234 is last page. Note page number starts with 1 in Ghostview
'''
import cv2, PIL
import fitz
import glob, sys, os, subprocess
from PIL import Image

resolution = 100  
th = int(sys.argv[2]) 

i = int(sys.argv[3])  #Start page
j = int(sys.argv[4])  #Last page

pdf_name = str(sys.argv[1])
#------------------------------------------------------------------------------
# Make directory named by variable output_dir
output_dir = "PDF2PNG"
os.makedirs(output_dir, exist_ok=True)

file_name = os.path.basename(pdf_name)
file_name = file_name.split(".")[0]
png_name = output_dir + "/" + file_name + "-%04d.png"

gs = 'gswin32c' if (sys.platform == 'win32') else 'gswin64'

#f-Strings: syntax is similar to str.format() but less verbose
subprocess.run(["gswin64", "-dBATCH", "-dNOPAUSE", "-sDEVICE=png16m", 
   f"-r{resolution}", f"-dFirstPage={i}", f"-dLastPage={j}", 
   f"-sOutputFile={png_name}", f"{pdf_name}"], stdout=subprocess.PIPE)
#------------------------------------------------------------------------------
pdff = "PDF-Merged.pdf"
pdfx = "PDF-Tempor.pdf"
bwFile = "BW.png"

pdfm =  fitz.open()

w = 842
imgList = output_dir + "/" + "*.PNG"

img = Image.open(glob.glob(imgList)[0])
wc = (w/float(img.size[0]))
h = int((float(img.size[1])*float(wc))) 
#------------------------------------------------------------------------------
for f in glob.glob(imgList):
	#convert the image to grayscale format: both the approaches are identical
	imgGray = cv2.imread(f, 0) #imgGray = cv2.cvtColor(f, cv2.COLOR_BGR2GRAY)

	#Convert the image to black and white, thresholding turns the border of the 
	#object in the image completely white with all pixels having same intensity
	(thresh, bwImg) = cv2.threshold(imgGray, th, 255, cv2.THRESH_BINARY)

	cv2.imwrite(bwFile, bwImg)
	img2pdf = Image.open(bwFile)
	img2pdf.resize((w, h), Image.Resampling.LANCZOS)
	img2pdf.save(pdfx, "PDF", resolution=100.0)
	p = fitz.open(pdfx)
	pdfm.insertPDF(p)
	p.close()
	
pdfm.save(pdff)
pdfm.close()
os.remove(pdfx)
os.remove(bwFile)
img.close()

#------------------------------------------------------------------------------
#Check whether folder exists or not
if os.path.exists(output_dir):
    # checking whether the folder is empty or not
	if len(os.listdir(output_dir)) == 0:
		os.rmdir(folder_path)
	else:
		files = glob.glob(output_dir + "/" + "*.png")
		for f in files:
			os.remove(f)
		os.rmdir(output_dir)