Docin document downloader

docin-dl.py

###############################################################################
#
# Docin document downloader
#
# Valid as of 2022-16-08
#
###############################################################################
import argparse
import os
from types import SimpleNamespace
import urllib.request

conf = SimpleNamespace(
    docin_pid = 0,
    output_dir = ""
)

def download_image(pid):
    i = 0
    while True:
        i += 1
        try:
            urllib.request.urlretrieve(
                "http://211.147.220.164/index.jsp?file={}&pageno={}&width=1836&height=2376".format(pid, i),
                os.path.join(conf.output_dir, "{0:03d}.png".format(i))
            )
        except urllib.error.HTTPError:
            break
        else:
            print("Page", i, "saved.")

if __name__ == "__main__":
    # Parse the arguments
    parser = argparse.ArgumentParser()

    parser.add_argument("docin_pid", type=str, help="The number after \"p-\" in docin url")
    parser.add_argument("output_dir", type=str, help="The output directory")

    args = parser.parse_args()

    conf.docin_pid = args.docin_pid
    conf.output_dir = args.output_dir

    # Do the work
    download_image(conf.docin_pid)

Wish we could find a way to auto-remove the watermark!

If it's just a simple B&W image (white background & black-ish text), then a very primitive levels adjustment can be made to offset the whites value (a brighter white point I believe) and subsequently crush those grey values. This obviously isn't ideal though...

You could try to level adjust curves in photoshop, until the watermark disappears and then save profile into acv file Then apply this curves in python

https://gist.github.com/nazarovsky/5f972c415c0dfaa4dba103f439ba206f