error handling, UI changes, and Logs

ai_functions.py

@@ -1,10 +1,13 @@
-from settings import gpt_model, RPFAAP2, RPFAAP1, TDRP, TDRP_COORDS
 import openai
 import json
 import logging
-from helpers import remove_na, filter_tables, merge_strings
+import sys
 import os
+from settings import gpt_model, RPFAAP2, RPFAAP1, TDRP, TDRP_COORDS
+from helpers import remove_na, filter_tables, merge_strings, Logger
+
 logging.basicConfig(filename='app.log', level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
+sys.stdout = Logger("output.log")
 
 def chat_gpt_image(content, context):
     openai.api_key = os.environ['GPT_API_KEY']
@@ -20,6 +23,7 @@ def chat_gpt_image(content, context):
 
     final_content = (" ".join(sequence))
     logging.info(final_content)
+    print(final_content)
     completion = openai.ChatCompletion.create(
     model=gpt_model,
     user="1",
@@ -29,6 +33,7 @@ def chat_gpt_image(content, context):
     ]
     )
     logging.info(completion.choices[0].message.content)
+    print(completion.choices[0].message.content)
     return(completion.choices[0].message.content)
 
 def chat_gpt_document(content, document_type, context):
@@ -43,17 +48,17 @@ def chat_gpt_document(content, document_type, context):
     content_name = content[3]
 
     if document_type == "RPFAA Building P1":
-        document = "RPFAAP1.json"
+        document = "./templates/RPFAAP1.json"
         desired_format = RPFAAP1
         tables = [3]
         input_coords = TDRP_COORDS
     elif document_type == "RPFAA Building P2":
-        document = "RPFAAP2.json"
+        document = "./templates/RPFAAP2.json"
         desired_format = RPFAAP2
         tables = []
         input_coords = TDRP_COORDS
     elif document_type == "TDRP":
-        document = "TDRP.json"
+        document = "./templates/TDRP.json"
         desired_format = TDRP
         tables = [0]
         input_coords = TDRP_COORDS
@@ -68,6 +73,7 @@ def chat_gpt_document(content, document_type, context):
 
     content_1 = (" ".join(sequence_1))
     logging.info(content_1)
+    print(content_1)
 
     completion_1 = openai.ChatCompletion.create(
     model=gpt_model,
@@ -78,6 +84,7 @@ def chat_gpt_document(content, document_type, context):
     ]
     )
     logging.info(completion_1.choices[0].message.content)
+    print(completion_1.choices[0].message.content)
     input_string = remove_na(completion_1.choices[0].message.content)
     input_string = merge_strings(input_string,input_coords,document_content)
 

app.py

@@ -2,16 +2,18 @@ import openai
 import gradio as gr
 import json
 import time
-from tqdm import tqdm
-from azure.core.exceptions import HttpResponseError
 import logging
 import requests
-import google 
+import sys
+from tqdm import tqdm
+from azure.core.exceptions import HttpResponseError
 from ocr_functions import detect_document, detect_image
 from ai_functions import chat_gpt_document, chat_gpt_image
-from helpers import save_json
+from helpers import save_json, read_logs, clear_logs, Logger
+from css import css
 
 logging.basicConfig(filename='app.log', level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
+sys.stdout = Logger("output.log")
 
 def retry_unprocessed_documents():
     global global_document_type
@@ -19,12 +21,15 @@ def retry_unprocessed_documents():
     global unprocessed_documents
     if unprocessed_documents:
         output = batch_document(unprocessed_documents, global_document_type, global_context, "None")
-        unprocessed_documents = []
         return output
     else:
-        unprocessed_documents = []
-        return save_json("No Unprocessed Documents", "No Unprocessed Documents")
+        return save_json("No Unprocessed Documents", "No Unprocessed Documents"), "All Documents Processed"
     
+def clear_unprocessed_documents():
+    global unprocessed_documents
+    unprocessed_documents = []
+    return "All Documents Processed"
+
 def combine_json_files(json_files, progress=gr.Progress()):
     combined_data = []
     progress(0, desc="Starting")
@@ -33,19 +38,21 @@ def combine_json_files(json_files, progress=gr.Progress()):
             data = json.load(json_file)
             combined_data.extend(data)
     logging.info("Combined JSON File: ", combined_data)
+    print("Combined JSON File: ", combined_data)
     return save_json(combined_data, "Combined Json")
 
 unprocessed_documents = []
 global_document_type = None
 global_context = None
 def batch_document(content, document_type, context, progress = gr.Progress()):
-    logging.info(content)
     combined_data = []
     global global_document_type
     global global_context
     global_document_type = document_type
     global_context = context
 
+    unprocessed_docs_temp = []
+
     if progress == "None":
         for x in content:
             retries = 3
@@ -55,17 +62,23 @@ def batch_document(content, document_type, context, progress = gr.Progress()):
                 try: 
                     data = json.loads(chat_gpt_document(detect_document(x),document_type,context))
                     combined_data.append(data)
-                    i = 0
                     break
                 except (openai.error.APIConnectionError, openai.error.AuthenticationError, openai.error.RateLimitError, HttpResponseError, requests.exceptions.RequestException) as e:
                     logging.error(f'Retry {i+1} failed: {e}')
+                    print(f'Retry {i+1} failed: {e}')
                     if i < retries - 1:
                         logging.error(f'Retrying in {timeout} seconds...')
+                        print(f'Retrying in {timeout} seconds...')
                         time.sleep(timeout)
                         i += 1
                     else:
-                        unprocessed_documents.append(x)
+                        unprocessed_docs_temp.append(x)
                         break
+                except Exception as e: # catch any other exceptions
+                    logging.error(f'Unexpected error {e}')
+                    print(f'Unexpected error {e}')
+                    unprocessed_docs_temp.append(x)
+                    break
 
     else: 
         progress(0, desc="Starting")
@@ -77,23 +90,42 @@ def batch_document(content, document_type, context, progress = gr.Progress()):
                 try: 
                     data = json.loads(chat_gpt_document(detect_document(x),document_type,context))
                     combined_data.append(data)
-                    i = 0
                     break
                 except (openai.error.APIConnectionError, openai.error.AuthenticationError, openai.error.RateLimitError, HttpResponseError, requests.exceptions.RequestException) as e:
                     logging.error(f'Retry {i+1} failed: {e}')
+                    print(f'Retry {i+1} failed: {e}')
                     if i < retries - 1:
                         logging.error(f'Retrying in {timeout} seconds...')
+                        print(f'Retrying in {timeout} seconds...')
                         time.sleep(timeout)
                         i += 1
                     else:
                         unprocessed_documents.append(x)
                         break
+                except Exception as e: # catch any other exceptions
+                    logging.error(f'Unexpected error {e}')
+                    print(f'Unexpected error {e}')
+                    unprocessed_documents.append(x)
+                    break
     
     logging.info(combined_data)
-    logging.info(unprocessed_documents)
+    print(combined_data)
+
     if document_type == "":
         document_type = "error"
-    return save_json(combined_data, document_type)
+
+    if unprocessed_documents: 
+        unprocessed = "\n".join([doc.name.split('\\')[-1].split('/')[-1].split('.')[0] for doc in unprocessed_documents])
+        logging.info(unprocessed)
+        print(unprocessed)
+    elif unprocessed_docs_temp:
+        unprocessed_documents.extend(unprocessed_docs_temp)
+        unprocessed = "\n".join([doc.name.split('\\')[-1].split('/')[-1].split('.')[0] for doc in unprocessed_documents])
+        logging.info(unprocessed)
+        print(unprocessed)
+    else:
+        unprocessed = "All Documents Processed"
+    return save_json(combined_data, document_type), unprocessed
 
 def image(content, context):
     retries = 3
@@ -105,8 +137,10 @@ def image(content, context):
             break
         except (openai.error.APIConnectionError, openai.error.AuthenticationError, openai.error.RateLimitError, HttpResponseError, requests.exceptions.RequestException) as e:
             logging.error(f'Retry {i+1} failed: {e}')
+            print(f'Retry {i+1} failed: {e}')
             if i < retries - 1:
                 logging.error(f'Retrying in {timeout} seconds...')
+                print(f'Retrying in {timeout} seconds...')
                 time.sleep(timeout)
                 i += 1
             else:
@@ -129,9 +163,13 @@ def document(content, document_type, context):
                 i += 1
             else:
                 break
+        except Exception as e: # catch any other exceptions
+            logging.error(f'Unexpected error {e}')
+            print(f'Unexpected error {e}')
+            break
     return data
 
-with gr.Blocks(title="Axon OCR", css=".markdown {text-align: center;}") as app:
+with gr.Blocks(title="Axon OCR", css=css) as app:
     gr.Markdown("""# Axon OCR
     Attach Images or Files below and convert them to Text.""", elem_classes="markdown")
     with gr.Tab("Scan Image"):
@@ -140,7 +178,7 @@ with gr.Blocks(title="Axon OCR", css=".markdown {text-align: center;}") as app:
                 image_input = [gr.Image(type="pil"),
                     gr.Textbox(label="What kind of Image is this? (Optional)", placeholder="This is an image of an Official Reciept")]
             image_output = gr.Textbox(label="Result")
-        image_button = gr.Button("Scan")
+        image_button = gr.Button("Scan", variant="primary")
     with gr.Tab("Scan Document"):
         with gr.Row():
             with gr.Column():
@@ -148,33 +186,42 @@ with gr.Blocks(title="Axon OCR", css=".markdown {text-align: center;}") as app:
                     gr.Dropdown(["RPFAA Building P1", "RPFAA Building P2", "TDRP"], label="File Type", info="What type of document is this?"),
                     gr.Textbox(label="Any additional information? (Optional)", placeholder="This is document is an Official Reciept")]
             document_output = gr.Textbox(label="Result")
-        document_button = gr.Button("Scan")
+        document_button = gr.Button("Scan", variant="primary")
     with gr.Tab("Batch Scan"):
         with gr.Row():
             with gr.Column():
                 batch_document_input = [gr.File(file_types=["pdf","tiff","image","text"], file_count="multiple"),
                     gr.Dropdown(["RPFAA Building P1", "RPFAA Building P2", "TDRP"], label="File Type", info="What type of document is this?"),
                     gr.Textbox(label="Any additional information? (Optional)", placeholder="This is document is an Official Reciept")]
-            batch_document_output = gr.File(label="Result")
-        batch_document_button = gr.Button("Scan")
+            with gr.Column():
+                batch_document_output = gr.File(label="Result") 
+                with gr.Accordion("Unprocessed Documents", open=False):
+                    batch_unprocessed = gr.Textbox(info="Download the file before retrying Unprocessed Documents and clear unprocessed documents after every scan to avoid overlaps", show_label=False, elem_classes="unprocessed_textbox")
+                    clear_unprocessed_button = gr.Button("Clear Unprocessed Documents")
+        batch_document_button = gr.Button("Scan", variant="primary")
         with gr.Row():
             with gr.Column():
-                retry_button = gr.Button("Retry Unprocessed Documents", label="Retry")
+                retry_button = gr.Button("Retry Unprocessed Documents")
             with gr.Column():
-                stop_button = gr.Button("Stop Processing Document", label="Stop")
+                stop_button = gr.Button("Stop Processing Document", variant="stop")
     with gr.Tab("Combine JSON"):
         with gr.Row():
             with gr.Column():
                 json_files_input = gr.File(file_types=[".json"], file_count="multiple", label='Upload JSON files')
                 combined_json_output = gr.File(label="Result")
-                combine_button = gr.Button('Combine JSON files')
-
+                combine_button = gr.Button('Combine JSON files', variant="primary")
+    with gr.Accordion("Logs", open=False):
+        logs = gr.Textbox(max_lines=10, show_label=False, elem_classes="log_textbox")
+        app.load(read_logs, None, logs, every=1)
+        clear_button = gr.Button("Clear Logs")
+        clear_button.click(clear_logs)
+    clear_unprocessed_button.click(clear_unprocessed_documents, outputs=batch_unprocessed)
     image_button.click(image, inputs=image_input, outputs=image_output)
     document_button.click(document, inputs=document_input, outputs=document_output)
-    batch_document_event = batch_document_button.click(batch_document, inputs=batch_document_input, outputs=batch_document_output)
-    retry_button.click(retry_unprocessed_documents, outputs=batch_document_output)
+    batch_document_event = batch_document_button.click(batch_document, inputs=batch_document_input, outputs=[batch_document_output,batch_unprocessed])
+    retry_button.click(retry_unprocessed_documents, outputs=[batch_document_output,batch_unprocessed])
     stop_button.click(fn=None, inputs=None, outputs=None, cancels=[batch_document_event])
     combine_button.click(combine_json_files, inputs=json_files_input, outputs=combined_json_output)
 
 app.queue()
-app.launch(auth=("username", "password"))
+app.launch(auth=("username", "password"), favicon_path="assets/logo.png")

css.py

@@ -0,0 +1,11 @@
+css = """
+.log_textbox textarea {
+    height: 300px !important;
+}
+.markdown {
+    text-align: center;
+}
+.unprocessed_textbox textarea {
+    height: 100px !important;
+}
+"""

helpers.py

@@ -1,7 +1,26 @@
 from settings import char_remove
 import re
 import json
+import sys
 import logging
+
+class Logger:
+    def __init__(self, filename):
+        self.terminal = sys.stdout
+        self.log = open(filename, "w")
+
+    def write(self, message):
+        self.terminal.write(message)
+        self.log.write(message)
+        
+    def flush(self):
+        self.terminal.flush()
+        self.log.flush()
+        
+    def isatty(self):
+        return False
+    
+sys.stdout = Logger("output.log")
 logging.basicConfig(filename='app.log', level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
 
 def remove_na(string):
@@ -29,8 +48,12 @@ def filter_tables(input_string, table_numbers):
 
     for table_number in table_numbers:
         # Picking the specific table
-        table_str = tables[table_number]
-
+        try: 
+            table_str = tables[table_number]
+        except (IndexError, UnboundLocalError) as e:
+            logging.error(f"Error: {e}, Please check document configuration or document type")
+            print(f"Error: {e}, Please check document configuration or document type")
+            raise e
         # Extracting cell coordinates and contents
         cells = re.findall(r"Cell\[(\d+)\]\[(\d+)\] has content '(.*?)'", table_str)
 
@@ -78,10 +101,14 @@ def merge_strings(input_string, input_coords, extract_coords):
     # Filter out empty lines and strip leading/trailing whitespaces
     lines2 = [line.strip() for line in lines2 if line.strip()]
 
-    logging.info(lines2)
     # Creating dictionaries to store the key-value pairs
-    dict1 = {line.split(": ")[0]: line.split(": ")[1] for line in lines1}
-    dict2 = {line.split(": ")[0]: line.split(": ")[1] for line in lines2}
+    try: 
+        dict1 = {line.split(": ")[0]: line.split(": ")[1] for line in lines1}
+        dict2 = {line.split(": ")[0]: line.split(": ")[1] for line in lines2}
+    except (IndexError, UnboundLocalError) as e:
+        logging.error(f"Error: {e}, Please check document configuration or document type")
+        print(f"Error: {e}, Please check document configuration or document type")
+        raise e
 
     # Updating the values in dict1 with the ones from dict2 if they share the same key
     for key in dict1.keys():
@@ -103,4 +130,14 @@ def merge_strings(input_string, input_coords, extract_coords):
     # Constructing the updated string1
     input_string = '\n'.join([f"{key}: {value}" for key, value in dict1.items()])
 
-    return input_string
+    return input_string
+
+def read_logs():
+    sys.stdout.flush()
+    with open("output.log","r",encoding="utf-8") as f:
+        lines = f.readlines()
+    return ''.join(lines[-100:])
+    
+def clear_logs():
+    with open("output.log","w",encoding="utf-8") as f:
+        f.write("")

ocr_functions.py

@@ -1,11 +1,14 @@
 from azure.core.credentials import AzureKeyCredential
 from azure.ai.formrecognizer import DocumentAnalysisClient
 from io import BytesIO
-from helpers import format_polygon
+from helpers import format_polygon, Logger
 import logging
 import os
+import sys 
 
 logging.basicConfig(filename='app.log', level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
+sys.stdout = Logger("output.log")
+
 endpoint = os.environ['AZURE_API_ENDPOINT']
 key  = os.environ['AZURE_API_KEY']
 
@@ -28,6 +31,7 @@ def detect_document(content):
                 kv_pair.value.content
             )
     logging.info(pair_content)
+    print(pair_content)
     document_content = "----Lines found in document----\n"
     for page in result.pages:
         for line_idx, line in enumerate(page.lines):
@@ -37,6 +41,7 @@ def detect_document(content):
                     format_polygon(line.polygon),
                 )
     logging.info(document_content)
+    print(document_content)
     table_content = "----Tables found in document----\n"
     for table_idx, table in enumerate(result.tables):
         table_content += "Table # {} has {} rows and {} columns\n".format(
@@ -49,6 +54,7 @@ def detect_document(content):
                 cell.content,
             )
     logging.info(table_content)
+    print(table_content)
     name = content.name.split('\\')[-1]
     name = name.split("/")[-1]
     name = name.split('.')[0]
@@ -67,4 +73,5 @@ def detect_image(content):
 
     result = poller.result()
     logging.info(result.content)
+    print(result.content)
     return(result.content)

RPFAAP1.json → templates/RPFAAP1.json

@@ -1,7 +1,7 @@
 {"File Name": "%s",
 "General Information": {
     "ARP No.": "",
-    "Owner": "",
+    "OWNER": "",
     "Address": "",
     "Tel No.": "",
     "Administrator/Beneficial User": "",