first attempt

app.py

@@ -0,0 +1,32 @@
+import streamlit as st
+from haystack.nodes.connector import Crawler
+from haystack.nodes import TransformersSummarizer
+import validators
+import json
+
+output_dir = "crawled_files"
+crawler = Crawler(output_dir=output_dir)
+
+summarizer = TransformersSummarizer(model_name_or_path="google/pegasus-xsum")
+
+documents = []
+
+def crawl_url_and_write_content(url):
+    docs = crawler.crawl(urls=['https://www.deepset.ai/blog/haystack-node-for-information-extraction'], crawler_depth=0, overwrite_existing_files=True)
+    for doc in docs:
+        jsonObject = json.load(doc.open())
+        documents[0] = jsonObject
+# Streamlit App
+
+st.title('Summarizer Demo with Haystack Summarizer')
+    
+url_text = st.text_input("Please Enter a url here",value="https://www.rba.gov.au/media-releases/2022/mr-22-12.html")
+
+if validators.url(url_text):
+    crawl_url_and_write_content(url_text)
+
+summarize = st.button('Summarize')
+
+if summarize:
+    summary = summarizer.predict(documents=documents)
+    st.write(summary)

requirements.txt

@@ -0,0 +1,2 @@
+farm-haystack==1.4.0
+validators==0.18.2