cmyk-magic.ipynb

Kernel: Python 3

import pdf2image

---------------------------------------------------------------------------
ModuleNotFoundError                       Traceback (most recent call last)
<ipython-input-1-bd56f4c59bef> in <module>()
----> 1 import pdf2image

ModuleNotFoundError: No module named 'pdf2image'

---------------------------------------------------------------------------
NOTE: If your import is failing due to a missing package, you can
manually install dependencies using either !pip or !apt.

To view examples of installing some common dependencies, click the
"Open Examples" button below.
---------------------------------------------------------------------------

!pip install pdf2image

Collecting pdf2image
  Downloading pdf2image-1.16.0-py3-none-any.whl (10 kB)
Requirement already satisfied: pillow in /usr/local/lib/python3.7/dist-packages (from pdf2image) (7.1.2)
Installing collected packages: pdf2image
Successfully installed pdf2image-1.16.0

from google.colab import drive

drive.mount("/content/drive")

Mounted at /content/drive

!ls /content/drive/MyDrive/MLAPA

 cnn-vis-modified.gslides
'Flaxifying (V)AE MLAPP examples.gdoc'
 GANs.gdraw
 GenModelsTree.gdraw
 hier-vae.pdf
 InfTree.gdraw
'Introduction to Bayesian Machine Learning.gslides'
'Intro to Prob Stats for ML.gslides'
'LSTM (1).gdraw'
 LSTM.gdraw
'Machine learning: a probabilistic perspective MLAPA.gdoc'
 mlapa-30jan17.pdf
'MLAPA figures.gslides'
'MLAPA todo.gsheet'
'MLAPA v2.gdoc'
 MLtriad.gdraw
 nnet-8jan16.pdf
 pmlSol-camera-22may12.pdf
'sRGB Color Space Profile.icm'
 survey-results-google-RG-dec2015.png
 Text-200416.zip
 USWebCoatedSWOP.icc

from pdf2image import convert_from_path

!ls

2dgridDAGa.pdf	sample_data

!sudo apt-get install poppler-utils

Reading package lists... Done
Building dependency tree       
Reading state information... Done
The following NEW packages will be installed:
  poppler-utils
0 upgraded, 1 newly installed, 0 to remove and 40 not upgraded.
Need to get 154 kB of archives.
After this operation, 613 kB of additional disk space will be used.
Get:1 http://archive.ubuntu.com/ubuntu bionic-updates/main amd64 poppler-utils amd64 0.62.0-2ubuntu2.12 [154 kB]
Fetched 154 kB in 1s (304 kB/s)
debconf: unable to initialize frontend: Dialog
debconf: (No usable dialog-like program is installed, so the dialog based frontend cannot be used. at /usr/share/perl5/Debconf/FrontEnd/Dialog.pm line 76, <> line 1.)
debconf: falling back to frontend: Readline
debconf: unable to initialize frontend: Readline
debconf: (This frontend requires a controlling tty.)
debconf: falling back to frontend: Teletype
dpkg-preconfigure: unable to re-open stdin: 
Selecting previously unselected package poppler-utils.
(Reading database ... 160837 files and directories currently installed.)
Preparing to unpack .../poppler-utils_0.62.0-2ubuntu2.12_amd64.deb ...
Unpacking poppler-utils (0.62.0-2ubuntu2.12) ...
Setting up poppler-utils (0.62.0-2ubuntu2.12) ...
Processing triggers for man-db (2.8.3-2ubuntu0.1) ...

input_path = "/content/2dgridDAGa.pdf"
temp_file_path = "/content/temp2dgridDAGa"
out = convert_from_path(input_path, output_file=temp_file_path, use_pdftocairo=True, fmt="png", single_file=True)

!ls

2dgridDAGa.pdf	sample_data  temp2dgridDAGa.png

import os
import shutil
from pdf2image import convert_from_path
from PIL import Image
from PIL import ImageCms
import argparse
from glob import glob
from tqdm import tqdm
import functools
import multiprocessing
import concurrent.futures


def split_file_name(input_path):
    base_name, dir_name = os.path.basename(input_path), os.path.dirname(input_path)
    file_name, ext = os.path.splitext(os.path.basename(base_name))
    return base_name, dir_name, file_name, ext


def convert(
    input_path,
    output_path,
    color_space="CMYK",
    input_profile_path=None,
    output_profile_path=None,
    quality=100,
    verbose=False,
    overwrite=False,
):
    """converts an image or pdf into a color space of choice
    for CMYK the default output format is JPG
    Keyword arguments:
    input_path -- the input path of the file
    output_path -- the output path for the result to be written.
    color_space -- the color space to convert to , default value is CMYK
    input_profile_path -- the path to the input profile
    output_profile_path -- the path to the output profile
    """
    try:
        if not overwrite and os.path.exists(output_path):
            return True

        if input_path.endswith(".pdf") or input_path.endswith(".PDF"):
            _, dir_name, file_name, _ = split_file_name(output_path)
            temp_file_name = "temp" + file_name
            temp_file_path = os.path.join(dir_name, temp_file_name)

            print("input", input_path)
            print("output", temp_file_path)
            print("call convert ")

            # convert_from_path(input_path,output_file=temp_file_path,fmt="png",use_pdftocairo=True,single_file=True)
            convert_from_path(input_path, output_file=temp_file_path, fmt="png", single_file=True)
            temp_file_path += ".png"

            print(temp_file_path)

            _convert_profiles(
                temp_file_path,
                output_path,
                color_space=color_space,
                input_profile_path=input_profile_path,
                output_profile_path=output_profile_path,
                quality=quality,
            )
            # os.remove(temp_file_path)
            print("done")
            return True
        elif (
            input_path.endswith(".png")
            or input_path.endswith(".PNG")
            or input_path.endswith(".jpg")
            or input_path.endswith(".JPG")
            or input_path.endswith(".jpeg")
            or input_path.endswith(".JPEG")
        ):

            print("else block")

            return _convert_profiles(
                input_path,
                output_path,
                color_space=color_space,
                input_profile_path=input_profile_path,
                output_profile_path=output_profile_path,
                quality=quality,
            )
        else:
            print(f"{input_path} is not a valid image file, copying it instead to {output_path}.")
            shutil.copy(input_path, output_path)
            return False
    except Exception as e:
        print("exception")
        if verbose:
            print(f"Error in file: {input_path}\n", e)
        return False


def _convert_profiles(
    input_path=None,
    output_path=None,
    color_space="CMYK",
    input_profile_path=None,
    output_profile_path=None,
    quality="100",
):
    try:
        with Image.open(input_path) as im:
            img_cmyk = ImageCms.profileToProfile(
                im, input_profile_path, output_profile_path, renderingIntent=0, outputMode=color_space
            )
            quality = int(quality)
            img_cmyk.save(output_path, quality=quality)

            return True
    except Exception as e:
        print(e)
        print(f"cannot convert{input_path}, copying it instead.")
        shutil.copy(input_path, output_path)
        return False

import os

input_path = "/content/2dgridDAGa.pdf"
temp_file_path = "/content/2dgridDAGa"
convert(
    input_path,
    temp_file_path,
    color_space="RGB",
    quality=80,
    verbose=True,
    input_profile_path="/content/drive/MyDrive/MLAPA/sRGB Color Space Profile.icm",
    output_profile_path="/content/drive/MyDrive/MLAPA/sRGB Color Space Profile.icm",
)

input /content/2dgridDAGa.pdf
output /content/temp2dgridDAGa
call convert 
/content/temp2dgridDAGa.png
[Errno 2] No such file or directory: '/content/temp2dgridDAGa.png'
cannot convert/content/temp2dgridDAGa.png, copying it instead.
exception
Error in file: /content/2dgridDAGa.pdf
 [Errno 2] No such file or directory: '/content/temp2dgridDAGa.png'

False

!ls

2dgridDAGa.pdf	drive  sample_data