Spaces:
Paused
Paused
Delete json_handling.py
Browse files- json_handling.py +0 -75
json_handling.py
DELETED
|
@@ -1,75 +0,0 @@
|
|
| 1 |
-
# import torch
|
| 2 |
-
# from model_loader import model, processor, device
|
| 3 |
-
# from processor_utils import load_input
|
| 4 |
-
# from prompt import get_prompt
|
| 5 |
-
# from inference import process_document
|
| 6 |
-
# import json
|
| 7 |
-
|
| 8 |
-
# # def process_whole_doc(file_path):
|
| 9 |
-
# # images = load_input(file_path)
|
| 10 |
-
# # complete_json = {}
|
| 11 |
-
# # for image in images:
|
| 12 |
-
# # json = process_documents(image)
|
| 13 |
-
# # def process_whole_doc(file_path):
|
| 14 |
-
# # images = load_input(file_path)
|
| 15 |
-
# # complete_json = {}
|
| 16 |
-
|
| 17 |
-
# # for image in images:
|
| 18 |
-
# # partial_json = process_document(image)
|
| 19 |
-
|
| 20 |
-
# # for part_name, part_value in partial_json.items():
|
| 21 |
-
|
| 22 |
-
# # # Ensure structure exists
|
| 23 |
-
# # if part_name not in complete_json:
|
| 24 |
-
# # complete_json[part_name] = [{}]
|
| 25 |
-
|
| 26 |
-
# # if not complete_json[part_name]:
|
| 27 |
-
# # complete_json[part_name].append({})
|
| 28 |
-
|
| 29 |
-
# # # Merge data
|
| 30 |
-
# # if isinstance(part_value, list) and part_value:
|
| 31 |
-
# # complete_json[part_name][0].update(part_value[0])
|
| 32 |
-
|
| 33 |
-
# # return complete_json
|
| 34 |
-
|
| 35 |
-
# def process_whole_doc(file_path):
|
| 36 |
-
# images = load_input(file_path)
|
| 37 |
-
|
| 38 |
-
# complete_json = "["
|
| 39 |
-
# first = True
|
| 40 |
-
|
| 41 |
-
# for image in images:
|
| 42 |
-
# partial_json = process_document(image) # returns string
|
| 43 |
-
|
| 44 |
-
# if not first:
|
| 45 |
-
# complete_json += ","
|
| 46 |
-
# complete_json += partial_json
|
| 47 |
-
# first = False
|
| 48 |
-
|
| 49 |
-
# complete_json += "]"
|
| 50 |
-
|
| 51 |
-
# return complete_json
|
| 52 |
-
# # complete_json = ""
|
| 53 |
-
|
| 54 |
-
|
| 55 |
-
# # # PART_NAME = "PART-1 - BILL OF ENTRY SUMMARY" # you can change per doc type
|
| 56 |
-
# # for i, image in enumerate(images, start=1):
|
| 57 |
-
# # partial_json = process_document(image)
|
| 58 |
-
# # # print(partial_json)
|
| 59 |
-
# # complete_json = complete_json+partial_json
|
| 60 |
-
# # # # 🔹 Extract only "result"
|
| 61 |
-
# # # result_data = partial_json.get("result", {})
|
| 62 |
-
# # # page_key = f"Page {i}"
|
| 63 |
-
# # # Ensure structure exists
|
| 64 |
-
# # # # if PART_NAME not in complete_json:
|
| 65 |
-
# # # complete_json[] = [{}]
|
| 66 |
-
|
| 67 |
-
# # # if not complete_json[PART_NAME]:
|
| 68 |
-
# # # complete_json[PART_NAME].append({})
|
| 69 |
-
|
| 70 |
-
# # # ✅ Merge directly
|
| 71 |
-
# # # complete_json[page_key]=result_data
|
| 72 |
-
|
| 73 |
-
# # return complete_json
|
| 74 |
-
|
| 75 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|