Update app.py

app.py

@@ -4,132 +4,82 @@ import os
 import gradio as gr
 import tempfile
 
-# Set JAVA_HOME environment variable
-# os.environ['JAVA_HOME'] = '/usr/lib/jvm/java-11-openjdk-amd64'
-
 def grade_pdf(input_pdf, pdffigures2_jar):
-    """
-    Process a PDF file using pdffigures2 and return the extracted JSON output.
-
-    Args:
-        input_pdf (str): Path to the input PDF file.
-        output_dir (str): Path to the directory where JSON output will be saved.
-        pdffigures2_jar (str): Path to the pdffigures2 JAR file.
-
-    Returns:
-        dict: Parsed JSON data extracted by pdffigures2.
-
-    Raises:
-        FileNotFoundError: If the input PDF or JAR file does not exist.
-        Exception: If the command fails or the JSON output is not generated.
-    """
-    # Check if input file and JAR file exist
     if not os.path.exists(input_pdf):
         raise FileNotFoundError(f"Input PDF file not found: {input_pdf}")
     if not os.path.exists(pdffigures2_jar):
         raise FileNotFoundError(f"pdffigures2 JAR file not found: {pdffigures2_jar}")
 
-
-    # Command to execute pdffigures2
     command = [
         "java", "-jar", pdffigures2_jar,
         input_pdf,
-        "-d", '_'
+        "-d", "_"
     ]
-    print(input_pdf)
 
     try:
-        # Run the command
         subprocess.run(command, capture_output=True, text=True, check=True)
 
-        # Construct the output JSON file path
-        # print()
-        output_json_file = "_"+os.path.splitext(os.path.basename(input_pdf))[0] +".json"
-        print("output_json_file: ", output_json_file)
+        output_json_file = "_" + os.path.splitext(os.path.basename(input_pdf))[0] + ".json"
 
-        # Read and return the JSON data
         if os.path.exists(output_json_file):
             with open(output_json_file, "r") as f:
                 regions = json.load(f)
 
-            issueList = list()
+            issue_list = []
 
             for region in regions:
-                issues = dict()
-                if(region["captionBoundary"]["y1"]>region["regionBoundary"]["y1"]):
-                    region["captionLocation"]="Below"
+                issues = {}
+                if region["captionBoundary"]["y1"] > region["regionBoundary"]["y1"]:
+                    region["captionLocation"] = "Below"
                 else:
-                    region["captionLocation"]="Above"
+                    region["captionLocation"] = "Above"
 
-                if(region["captionLocation"]=="Below" and region["figType"] == "Table"):
-                    issues["page_number"]=region["page"]
-                    issues["caption_coordinate"]=region["captionBoundary"]
+                if region["captionLocation"] == "Below" and region["figType"] == "Table":
+                    issues["page_number"] = region["page"]+1
+                    issues["caption_coordinate"] = region["captionBoundary"]
                     issues["fig_type"] = region["figType"]
                     issues["caption_location"] = region["captionLocation"]
-                    issues["description"]="Location of the caption for tables must be above the table."
-                    issueList.append(issues)
-                    
-                if(region["captionLocation"]=="Above" and region["figType"] == "Figure"):
-                    issues["page_number"]=region["page"]
-                    issues["caption_coordinate"]=region["captionBoundary"]
+                    issues["description"] = "Location of the caption for tables must be above the table."
+                    issue_list.append(issues)
+
+                if region["captionLocation"] == "Above" and region["figType"] == "Figure":
+                    issues["page_number"] = region["page"]+1
+                    issues["caption_coordinate"] = region["captionBoundary"]
                     issues["fig_type"] = region["figType"]
-                    issues["description"]="Location of the caption for figures must be below the table."
-                    issueList.append(issues)
-                    
-            return issueList
+                    issues["caption_location"] = region["captionLocation"]
+                    issues["description"] = "Location of the caption for figures must be below the figure."
+                    issue_list.append(issues)
 
+            return issue_list
         else:
             raise Exception(f"JSON output file not found: {output_json_file}")
 
     except subprocess.CalledProcessError as e:
-        raise Exception(f"Error while running pdffigures2: {e.stderr}")
+        raise Exception(f"Error while running pdffigures2:\nSTDOUT:\n{e.stdout}\nSTDERR:\n{e.stderr}")
 
 def process_file(file):
-    """
-    Wrapper to process a file via grade_pdf and return JSON data.
-
-    Args:
-        file (str): Path to the uploaded PDF file.
-
-    Returns:
-        str: JSON string of extracted data.
-    """
-    # Define paths
-    pdffigures2_jar_path = "pdffigures2.jar"  # Path to pdffigures2 JAR file
-    
-    temp_input = tempfile.NamedTemporaryFile(delete=False, suffix='.pdf')
-    temp_input.write(file)
-    temp_input_path = temp_input.name
-    print("Path of the input file: ",temp_input_path)
+    pdffigures2_jar_path = "pdffigures2.jar"
+
+    with tempfile.NamedTemporaryFile(delete=False, suffix=".pdf", mode="wb") as temp_input:
+        temp_input.write(file.read())
+        temp_input_path = temp_input.name
 
     try:
-        # Process the PDF and get JSON data
-        # print(file.name)
         extracted_data = grade_pdf(temp_input_path, pdffigures2_jar_path)
-        # Clean up the temporary input file
         os.unlink(temp_input_path)
-        
-
-        # Return JSON as a formatted string
         return json.dumps(extracted_data, indent=4)
     except Exception as e:
-        return f"An error occurred: {e}"
+        return f"An error occurred:\n{str(e)}"
 
-# Gradio Interface
 def gradio_interface():
     interface = gr.Interface(
         fn=process_file,
-        
-        inputs=gr.File(label="Upload PDF File", file_types=[".pdf"], type="binary"),
-        
+        inputs=gr.File(label="Upload PDF File", file_types=[".pdf"]),
         outputs=gr.Textbox(label="Extracted JSON Data", lines=20),
-        
         title="PDF Grading Interface",
-        
-        description="Upload a PDF file, and this tool will extract figures and captions as JSON."
+        description="Upload a PDF to detect improper caption positions for tables and figures."
     )
     return interface
 
 if __name__ == "__main__":
-    # Launch the Gradio app
-    gradio_interface().launch()
+    gradio_interface().launch()