DUDE_imdb_loader.py

# coding=utf-8
# Copyright 2020 The HuggingFace Datasets Authors and the current dataset script contributor.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""DUDE dataset loader"""

import os
from pathlib import Path
import time
import copy
import json
import numpy as np
import pandas as pd
from tqdm import tqdm
from io import BytesIO

tqdm.pandas()
from joblib import Parallel, delayed

# import pdf2image
import PyPDF2

from datasets import load_dataset_builder, load_dataset
from PIL import Image as PIL_Image


MAX_PAGES = 50
MAX_PDF_SIZE = 100000000  # almost 100MB
MIN_WIDTH, MIN_HEIGHT = 150, 150


def load_json(json_path):
    return json.load(open(json_path, "r"))


def save_json(json_path, data):
    with open(json_path, "w") as f:
        json.dump(data, f)


# 0. PDF to images

# 1. OCR metadata


def pdf_to_images(document_filepath):
    def images_to_pagenames(images, document_filepath, page_image_dir):
        "/home/jordy/Downloads/DUDE_train-val-test_binaries/PDF/val/001d6f557c342ef5a67cd38a29da9e83.pdf"

        page_image_names = []
        for page_idx, page_image in enumerate(images):
            page_image_name = document_filepath.replace("PDF", "images").replace(
                ".pdf", f"_{page_idx}.jpg"
            )

            # page_image_names.append(page_image_name.replace(page_images_dir, "")) #without dir

            if not os.path.exists(page_image_name):
                page_image.save(page_image_name)
        return page_image_names

    example = {}
    example["document"] = document_filepath
    example["pages"] = 0
    example["page_image_names"] = []
    images = []

    page_image_dir = "/".join(example["document"].split("/")[:-1]).replace("PDF", "images")
    if not os.path.exists(page_image_dir):
        os.makedirs(page_image_dir)

    # if len(example["document"]) > MAX_PDF_SIZE:
    #     logger.warning(f"too large document {len(example['document'])}")
    #     return example
    try:
        reader = PyPDF2.PdfReader(example["document"])
    except Exception as e:
        logger.warning(f"read_pdf {e}")
        return example

    reached_page_limit = False
    page_iterator = reader.pages

    for p, page in enumerate(page_iterator):
        if reached_page_limit:
            break
        for image in page.images:
            # try:
            # except Exception as e:
            #     logger.warning(f"get_images {e}")
            if len(images) == MAX_PAGES:
                reached_page_limit = True
                break

            im = PIL_Image.open(BytesIO(image.data))
            if im.width < MIN_WIDTH and im.height < MIN_HEIGHT:
                continue
            images.append(im)
    example["pages"] = len(images)
    if len(images) == 0:
        return example

    example["page_image_names"] = images_to_pagenames(images, example["document"], page_image_dir)

    return example


def pdf_to_images_block(document_paths_blocks):
    new_doc_metadata = {}
    for document_filepath in document_paths_blocks:
        docId = document_filepath.split("/")[-1].replace(".pdf", "")
        new_doc_metadata[docId] = pdf_to_images(document_filepath)
    return new_doc_metadata


"""
def get_document_metadata(docs_metadata, doc_id, document_filepath):

    if doc_id in docs_metadata and docs_metadata[doc_id]["num_pages"] != -1:
        num_pages = docs_metadata[doc_id]["num_pages"]
        page_image_names = docs_metadata[doc_id]["page_image_names"]

    else:
        try:
            images = pdf2image.convert_from_path(document_filepath)
        except:
            print(doc_id)
            return -1, -1
        num_pages = len(images)
        page_image_dir = ("/").join(
            document_filepath.replace(documents_dir, page_images_dir).split("/")[:-1]
        )
        if not os.path.exists(page_image_dir):
            os.makedirs(page_image_dir)

        page_image_names = []
        for page_idx, page_image in enumerate(images):
            page_image_name = document_filepath.replace(documents_dir, page_images_dir).replace(
                ".pdf", f"_{page_idx}.jpg"
            )
            page_image_names.append(page_image_name.replace(page_images_dir, ""))

            if not os.path.exists(page_image_name):
                page_image.save(page_image_name)

    return num_pages, page_image_names


def get_document_metadata_block(docs_metadata, documents_path_dict, documents_blocks):
    new_doc_metadata = {}
    for doc_id in documents_blocks:
        document_filepath = documents_path_dict[doc_id]
        num_pages, page_image_names = get_document_metadata(
            docs_metadata, doc_id, document_filepath
        )
        new_doc_metadata[doc_id] = {"num_pages": num_pages, "page_image_names": page_image_names}

    return new_doc_metadata
"""


def parse_textract_bbox(box):
    # 0.47840896, 0.12897822, 0.5341576 , 0.14347914 # x,w,y,h
    return np.array([box["Left"], box["Width"], box["Top"], box["Height"]])


def parse_azure_box(box, page_width, page_height):
    # Box in Azure are in format X top left, Y top left, X top right, Y top right, X bottom right, Y bottom right, X bottom left, Y bottom left
    # [14.1592, 3.8494, 28.668, 3.8494, 28.668, 8.0487, 13.9844, 7.8738]
    left = min(box[0], box[6])
    right = max(box[2], box[4])
    top = min(box[1], box[3])
    bottom = max(box[5], box[7])
    width = right - left
    height = bottom - top

    # Normalize
    left = left / page_width
    top = top / page_height
    width = width / page_width
    height = height / page_height

    return [left, width, top, height]


def get_ocr_information(doc_id):
    ocr_info = load_json(os.path.join(ocr_dir, doc_id + "_original.json"))
    num_pages, _ = get_document_metadata(documents_metadata, doc_id, documents[doc_id])

    page_ocr_tokens = [[] for page_ix in range(num_pages)]
    page_ocr_boxes = [[] for page_ix in range(num_pages)]
    for ocr_block in ocr_info:
        for ocr_extraction in ocr_block["Blocks"]:
            if ocr_extraction["BlockType"] == "WORD":
                text = ocr_extraction["Text"].lower()
                bounding_box = parse_textract_bbox(ocr_extraction["Geometry"]["BoundingBox"])
                page = ocr_extraction["Page"] - 1

                page_ocr_tokens[page].append(text)
                page_ocr_boxes[page].append(bounding_box)

    for page in range(num_pages):
        page_ocr_boxes[page] = np.array(page_ocr_boxes[page])

    page_ocr_boxes = page_ocr_boxes
    return page_ocr_tokens, page_ocr_boxes


def create_header(split, version, has_answer):
    header = {
        "creation_time": time.time(),
        "version": version,
        "dataset_type": split,
        "has_answer": has_answer,
    }

    return header


def get_document_info(documents_metadata, doc_id):
    doc_metadata = documents_metadata[doc_id]
    num_pages = doc_metadata["num_pages"]
    page_image_names = doc_metadata["page_image_names"]
    return num_pages, page_image_names


def format_answers(answers_list):
    answers_list = list(set([answer.lower() for answer in answers_list]))
    return answers_list


def create_imdb_record_from_json(
    record, documents_metadata, documents_ocr_information, split, include_answers
):

    doc_id = record["docId"]
    # document_filepath = documents_dict[doc_id]
    num_pages, page_image_names = get_document_info(documents_metadata, doc_id)
    document_ocr_info = documents_ocr_information[doc_id]

    if include_answers:
        answers = format_answers(record["answers"])
    else:
        answers = None

    imdb_record = {
        "question_id": record["questionId"],
        "question": record["question"],
        "docId": doc_id,
        "image_name": page_image_names,
        "num_pages": num_pages,
        "ocr_tokens": document_ocr_info["ocr_tokens"],
        "ocr_normalized_boxes": document_ocr_info["ocr_boxes"],
        "set_name": split,
        "answers": answers,
        "answer_page": None,
        "extra": {
            # 'question_type': record['qtype'],
            # 'industry': record['industry'],
            # 'category': record['category'],
            "answer_type": record["answer_type"],
        },
    }

    return imdb_record


def create_imdb_from_json(
    data, documents_metadata, documents_ocr_information, split, version, include_answers
):
    imdb_header = create_header(split, version, include_answers)

    imdb_records = []
    for record in tqdm(data):
        imdb_records.append(
            create_imdb_record_from_json(
                record, documents_metadata, documents_ocr_information, split, include_answers
            )
        )

    imdb = [imdb_header] + imdb_records

    return imdb


if __name__ == "__main__":
    dataset = load_dataset(
        "../DUDE_loader/DUDE_loader.py",
        "DUDE",
        data_dir="/home/jordy/Downloads/DUDE_train-val-test_binaries",
    )
    splits = datasets.keys()

    split = "val"


    # 1. PDF to image dir and collect document metadata (num_pages, page_image_names)

    document_paths = sorted(
        set([x["document"] for x in dataset[split] if x["data_split"] == split])
    )

    document_paths = document_paths[:30]

    # document_ids = [document_filepath.split("/")[-1].replace(".pdf", "") for document_filepath in document_paths]
    documents_metadata_filename = "documents_metadata.json"
    if os.path.exists(documents_metadata_filename):
        documents_metadata = load_json(documents_metadata_filename)
    else:
        documents_metadata = {}
        # for document_filepath in tqdm(document_paths):
        #     documents_metadata[docId] = pdf_to_images(document_filepath)
        num_jobs = 6
        block_size = int(len(document_paths) / num_jobs) + 1
        print(f"{block_size} * {num_jobs} = {block_size*num_jobs} ({len(document_paths)})")
        documents_blocks = [
            document_paths[block_size * i : block_size * i + block_size] for i in range(num_jobs)
        ]
        print(
            "chunksize",
            len(set([doc_id for doc_block in documents_blocks for doc_id in doc_block])),
        )
        parallel_results = Parallel(n_jobs=num_jobs)(
            delayed(get_document_metadata_block)(documents_metadata, documents, document_paths[i])
            for i in range(num_jobs)
        )

        for block_result in parallel_results:
            for doc_id, metadata in tqdm(block_result.items()):
                if doc_id not in documents_metadata:
                    documents_metadata[doc_id] = metadata

        save_json(documents_metadata_filename, documents_metadata)

    #2. Process OCR to obtain doc_ocr_info 
    documents_ocr_filename = "documents_ocr.json"

    if os.path.exists(documents_ocr_filename):
        documents_ocr_info = load_json(documents_ocr_filename)
    else:
        documents_ocr_info = {}
        no_ocr = []
        error_ocr = []

        for document_filepath in document_paths:
            docId = document_filepath.split("/")[-1].replace(".pdf", "")
        try:
            ocr_tokens, ocr_boxes = get_ocr_information(docId)
            doc_ocr_info[docId] = {'ocr_tokens': ocr_tokens, 'ocr_boxes': ocr_boxes}
        except FileNotFoundError:
            no_ocr.append(docId)
        except KeyError:
            error_ocr.append(docId)
            
        save_json(documents_ocr_filename, documents_ocr_info)

    # page_image_dir = '/'.join(dataset['val']['document'][0].split("/")[:-1]).replace('PDF', 'images')
    # if not os.path.exists(page_image_dir):
    #     os.makedirs(page_image_dir)
    # dataset.info.features

    """
    Describe all steps that need to happen after loading HF DUDE dataset
    Change functions 
    

    page_images_dir


    
    2. Process OCR to obtain doc_ocr_info 
    """
    # update dataset with
    # for split in SPLITS

    # documents_metadata
    # doc_ocr_info

    # dict to unique docs
    # documents_metadata[doc_id] = {"num_pages": num_pages, "page_image_names": image_names}
    # doc_ocr_info[doc_id] = {"ocr_tokens": ocr_tokens, "ocr_boxes": ocr_boxes}

    """
    train_imdb = create_imdb_from_json(
        train_data,
        documents_metadata=documents_metadata,
        documents_ocr_information=doc_ocr_info,
        split="train",
        version="0.1",
        include_answers=True,
    )
    val_imdb = create_imdb_from_json(
        val_data,
        documents_metadata=documents_metadata,
        documents_ocr_information=doc_ocr_info,
        split="train",
        version="0.1",
        include_answers=True,
    )
    np.save("Imdb/train_imdb.npy", train_imdb)
    np.save("Imdb/val_imdb.npy", val_imdb)

    document_paths = []
    num_jobs = 6
    block_size = int(len(document_ids) / num_jobs) + 1
    print(f"{block_size} * {num_jobs} = {block_size*num_jobs} ({len(document_ids)})")
    parallel_results = Parallel(n_jobs=num_jobs)(
        delayed(get_document_metadata_block)(documents_metadata, documents, documents_blocks[i])
        for i in range(num_jobs)
    )
    """