#313- Add in-memory template caching to speed up PDF processing

src/file_manipulator.py

@@ -7,33 +7,19 @@
 class FileManipulator:
     def __init__(self):
         self.filler = Filler()
-        self.llm = LLM()
-
-    def create_template(self, pdf_path: str):
-        """
-        By using commonforms, we create an editable .pdf template and we store it.
-        """
-        template_path = pdf_path[:-4] + "_template.pdf"
-        prepare_form(pdf_path, template_path)
-        return template_path
 
     def fill_form(self, user_input: str, fields: list, pdf_form_path: str):
-        """
-        It receives the raw data, runs the PDF filling logic,
-        and returns the path to the newly created file.
-        """
         print("[1] Received request from frontend.")
         print(f"[2] PDF template path: {pdf_form_path}")
 
         if not os.path.exists(pdf_form_path):
             print(f"Error: PDF template not found at {pdf_form_path}")
-            return None  # Or raise an exception
+            return None
 
         print("[3] Starting extraction and PDF filling process...")
         try:
-            self.llm._target_fields = fields
-            self.llm._transcript_text = user_input
-            output_name = self.filler.fill_form(pdf_form=pdf_form_path, llm=self.llm)
+            llm = LLM(transcript_text=user_input, target_fields=fields, json={})
+            output_name = self.filler.fill_form(pdf_form=pdf_form_path, llm=llm)
 
             print("\n----------------------------------")
             print("✅ Process Complete.")
@@ -43,5 +29,12 @@ def fill_form(self, user_input: str, fields: list, pdf_form_path: str):
 
         except Exception as e:
             print(f"An error occurred during PDF generation: {e}")
-            # Re-raise the exception so the frontend can handle it
             raise e
+
+    def create_template(self, pdf_path: str):
+        """
+        By using commonforms, we create an editable .pdf template and we store it.
+        """
+        template_path = pdf_path[:-4] + "_template.pdf"
+        prepare_form(pdf_path, template_path)
+        return template_path

src/filler.py

@@ -5,7 +5,7 @@
 
 class Filler:
     def __init__(self):
-        pass
+        self._template_cache = {}
 
     def fill_form(self, pdf_form: str, llm: LLM):
         """
@@ -24,9 +24,17 @@ def fill_form(self, pdf_form: str, llm: LLM):
         textbox_answers = t2j.get_data()  # This is a dictionary
 
         answers_list = list(textbox_answers.values())
-
-        # Read PDF
-        pdf = PdfReader(pdf_form)
+        # --- NEW: CACHING LOGIC ---
+        if pdf_form not in self._template_cache:
+            print(f"[LOG] Template Cache Miss, parsing new PDF for {pdf_form}...")
+            # Read the file from the hard drive once and store it in RAM
+            with open(pdf_form, "rb") as f:
+                self._template_cache[pdf_form] = f.read()
+        else:
+            print(f"[LOG] Template Cache Hit! Reusing memory for {pdf_form}...")
+
+        # Load PDF instantly from RAM instead of hitting the slow hard drive
+        pdf = PdfReader(fdata=self._template_cache[pdf_form])
 
         # Loop through pages
         for page in pdf.pages:

src/llm.py

@@ -46,7 +46,7 @@ def build_prompt(self, current_field):
 
     def main_loop(self):
         # self.type_check_all()
-        for field in self._target_fields.keys():
+        for field in self._target_fields:
             prompt = self.build_prompt(field)
             # print(prompt)
             # ollama_url = "http://localhost:11434/api/generate"

src/main.py

@@ -3,6 +3,7 @@
 from commonforms import prepare_form 
 from pypdf import PdfReader
 from controller import Controller
+from typing import Union
 
 def input_fields(num_fields: int):
     fields = []
@@ -68,7 +69,7 @@ def run_pdf_fill_process(user_input: str, definitions: list, pdf_form_path: Unio
 if __name__ == "__main__":
     file = "./src/inputs/file.pdf"
     user_input = "Hi. The employee's name is John Doe. His job title is managing director. His department supervisor is Jane Doe. His phone number is 123456. His email is jdoe@ucsc.edu. The signature is <Mamañema>, and the date is 01/02/2005"
-    fields = ["Employee's name", "Employee's job title", "Employee's department supervisor", "Employee's phone number", "Employee's email", "Signature", "Date"]
+    descriptive_fields = ["Employee's name", "Employee's job title", "Employee's department supervisor", "Employee's phone number", "Employee's email", "Signature", "Date"]
     prepared_pdf = "temp_outfile.pdf"
     prepare_form(file, prepared_pdf)
 
@@ -78,6 +79,13 @@ def run_pdf_fill_process(user_input: str, definitions: list, pdf_form_path: Unio
         num_fields = len(fields)
     else:
         num_fields = 0
-        
+
     controller = Controller()
-    controller.fill_form(user_input, fields, file)
+
+    # --- TEST RUN 1 (Should be a Miss) ---
+    print("\n🚀 STARTING RUN 1...")
+    controller.fill_form(user_input, descriptive_fields, file)
+
+    # --- TEST RUN 2 (Should be a Hit!) ---
+    print("\n🚀 STARTING RUN 2...")
+    controller.fill_form(user_input, descriptive_fields, file)