feat: include a pdf reader

app_langchain.py

@@ -2,10 +2,19 @@ from langchain_core.prompts import ChatPromptTemplate
 from langchain_core.runnables import RunnablePassthrough
 
 import streamlit as st
+import fitz
 
 from src.utils import load_config_values
 from src.dev_llm import FakeLLM
 
+def extract_text_from_pdf(pdf_file):
+    document = fitz.open(stream=pdf_file.read(), filetype="pdf")
+    text = ""
+    for page_num in range(len(document)):
+        page = document.load_page(page_num)
+        text += page.get_text()
+    return text
+
 # TODO: Change this to reflect prod model rather than dev models
 # Initalise fake values and a fake LLM to test out the full pipeline
 tmp_llm = FakeLLM()
@@ -23,20 +32,23 @@ system_message, context_message, model_id = load_config_values(
 prompt = ChatPromptTemplate.from_template(
     template=context_message,
 )
+uploaded_doc = st.file_uploader(
+    label="Upload PDF:\n",
+    type=".pdf"
+)
+if uploaded_doc is not None:
+    uploaded_text = extract_text_from_pdf(uploaded_doc)
 
 chain = (
     {
         "system_message": lambda x: system_message,
-        "pdf_text": lambda x: tmp_pdf_text,
+        "pdf_text": lambda x: uploaded_text,
         "data_to_extract": RunnablePassthrough()
     }
     |prompt
     |tmp_llm
 )
 
-st.file_uploader(
-    label="Upload PDF:\n",
-    type=".pdf"
-)
+user_input_1 = st.text_input(label="input")
 
-st.text(chain.invoke("{\"appointment_date\"}"))
+st.text(chain.invoke(user_input_1))