Multipage shared code and download changes (#7)

* add multipage loader and app_user file

* fix pdf generation

* change download button name

* fix iframe styles

* fix imports

* fix dockerfile

Dockerfile

@@ -1,12 +1,10 @@
-
 # Copyright (c) Microsoft Corporation. All rights reserved.
-# Dockerfile
 # https://eng.ms/docs/more/containers-secure-supply-chain/approved-images
 FROM  mcr.microsoft.com/oryx/python:3.10
 
 RUN curl -fsSL https://packages.microsoft.com/keys/microsoft.asc | gpg --dearmor -o /usr/share/keyrings/microsoft-prod.gpg
 RUN apt-get update -y
-ENV IS_DOCKER True
+RUN apt-get install wkhtmltopdf -y
 
 # Install dependencies
 WORKDIR ./

app/Home.py

@@ -1,36 +1,11 @@
+# Copyright (c) 2024 Microsoft Corporation. All rights reserved.
 import streamlit as st
+from components.app_loader import load_multipage_app
 import util.mermaid as mermaid
-from streamlit_javascript import st_javascript
-import util.session_variables
-
-def get_user(sv):
-    if sv.mode.value != 'cloud':
-        return
-    css='''
-            [data-testid="stSidebarNavItems"] {
-                max-height: 100vh
-            }
-        '''
-    st.markdown(f'<style>{css}</style>', unsafe_allow_html=True)
-    js_code = """await fetch("/.auth/me")
-    .then(function(response) {return response.json();})
-    """
-    return_value = st_javascript(js_code)
-
-    username = None
-    if return_value == 0:
-        pass # this is the result before the actual value is returned 
-    elif isinstance(return_value, list) and len(return_value) > 0:
-        username = return_value[0]["user_id"]
-        sv.username.value = username
-        st.sidebar.write(f"Logged in as {username}")
-    else:
-        st.warning(f"Could not directly read username from azure active directory: {return_value}.")     
 
 def main():
     st.set_page_config(layout="wide", initial_sidebar_state="expanded", page_title='Intelligence Toolkit | Home')
-    sv = util.session_variables.SessionVariables('home')
-    get_user(sv)
+    load_multipage_app()
 
     transparency_faq = open('./app/TransparencyFAQ.md', 'r').read()
     st.markdown(transparency_faq + '\n\n' + f"""\

app/components/app_loader.py

@@ -0,0 +1,13 @@
+# Copyright (c) 2024 Microsoft Corporation. All rights reserved.
+from javascript.styles import add_styles
+import components.app_user as au
+import streamlit as st
+
+def load_multipage_app():
+    #Load user if logged in
+    user = au.app_user()
+    user.view_get_info()
+
+    #load css
+    add_styles()
+

app/components/app_user.py

@@ -0,0 +1,41 @@
+# Copyright (c) 2024 Microsoft Corporation. All rights reserved.
+import streamlit as st
+from javascript.scripts import get_auth_user
+from util.session_variables import SessionVariables
+
+class app_user:
+
+    sv = None
+
+    def __init__(self, sv = None):
+        if sv is not None:
+            self.sv = sv
+        else:
+            self.sv = SessionVariables('home')
+        self.login()
+
+    def _get_info(self):
+        return self.sv.username.value 
+
+    def _set_user(self, username):
+        self.sv.username.value = username
+
+    def view_get_info(self):
+        if self.sv.username.value:
+            st.sidebar.write(f"Logged in as {self.sv.username.value}")
+
+    def _view_error_info(self, return_value):
+        st.warning(f"Could not directly read username from azure active directory: {return_value}.")     
+
+    def login(self):
+        if self.sv.mode.value != 'cloud':
+            return
+        return_value = get_auth_user()
+        username = None
+        if return_value == 0:
+            pass # this is the result before the actual value is returned 
+        elif isinstance(return_value, list) and len(return_value) > 0:
+            username = return_value[0]["user_id"]
+            self._set_user(username)
+        else:
+            self._view_error_info(return_value)

app/javascript/scripts.py

@@ -0,0 +1,9 @@
+# Copyright (c) 2024 Microsoft Corporation. All rights reserved.
+from streamlit_javascript import st_javascript
+import streamlit as st
+
+def get_auth_user():
+    js_code = """await fetch("/.auth/me")
+    .then(function(response) {return response.json();})
+    """
+    return st_javascript(js_code)

app/javascript/styles.py

@@ -0,0 +1,25 @@
+# Copyright (c) 2024 Microsoft Corporation. All rights reserved.
+import streamlit as st
+
+style_sidebar = '''
+    [data-testid="stSidebarNavItems"] {
+        max-height: 100vh
+    }
+'''
+
+style_pdf = '''body {
+    font-family: 'helvetica';
+}
+'''
+
+style_iframes = '''
+    iframe {
+        display: none;
+    }
+'''
+
+def add_styles():
+    st.markdown(f'''<style>
+        {style_sidebar}
+        {style_iframes}
+    </style>''', unsafe_allow_html=True)

app/pages/Attribute_Patterns.py

@@ -1,6 +1,11 @@
+# Copyright (c) 2024 Microsoft Corporation. All rights reserved.
 import workflows.attribute_patterns.workflow
+from components.app_loader import load_multipage_app
+import streamlit as st
 
 def main():
+    st.set_page_config(layout="wide", initial_sidebar_state="collapsed", page_title='Intelligence Toolkit | Attribute Patterns')
+    load_multipage_app()
     workflows.attribute_patterns.workflow.create()
 
 if __name__ == '__main__':

app/pages/Data_Synthesis.py

@@ -1,6 +1,11 @@
+# Copyright (c) 2024 Microsoft Corporation. All rights reserved.
 import workflows.data_synthesis.workflow
+from components.app_loader import load_multipage_app
+import streamlit as st
 
 def main():
+    st.set_page_config(layout="wide", initial_sidebar_state="collapsed", page_title='Intelligence Toolkit | Data Synthesis')
+    load_multipage_app()
     workflows.data_synthesis.workflow.create()
 
 if __name__ == '__main__':

app/pages/Group_Narratives.py

@@ -1,6 +1,11 @@
+# Copyright (c) 2024 Microsoft Corporation. All rights reserved.
 import workflows.group_narratives.workflow
+import streamlit as st
+from components.app_loader import load_multipage_app
 
 def main():
+    st.set_page_config(layout="wide", initial_sidebar_state="collapsed", page_title='Intelligence Toolkit | Group Narratives')
+    load_multipage_app()
     workflows.group_narratives.workflow.create()
 
 if __name__ == '__main__':

app/pages/Question_Answering.py

@@ -1,6 +1,12 @@
+# Copyright (c) 2024 Microsoft Corporation. All rights reserved.
 import workflows.question_answering.workflow
+from components.app_loader import load_multipage_app
+import streamlit as st
+
 
 def main():
+    st.set_page_config(layout="wide", initial_sidebar_state="collapsed", page_title='Intelligence Toolkit | Question Answering')
+    load_multipage_app()
     workflows.question_answering.workflow.create()
 
 if __name__ == '__main__':

app/pages/Record_Matching.py

@@ -1,6 +1,11 @@
+# Copyright (c) 2024 Microsoft Corporation. All rights reserved.
 import workflows.record_matching.workflow
+from components.app_loader import load_multipage_app
+import streamlit as st
 
 def main():
+    st.set_page_config(layout="wide", initial_sidebar_state="collapsed", page_title='Intelligence Toolkit | Record Matching')
+    load_multipage_app()
     workflows.record_matching.workflow.create()
 
 if __name__ == '__main__':

app/pages/Risk_Networks.py

@@ -1,6 +1,11 @@
+# Copyright (c) 2024 Microsoft Corporation. All rights reserved.
 import workflows.risk_networks.workflow
+from components.app_loader import load_multipage_app
+import streamlit as st
 
 def main():
+    st.set_page_config(layout="wide", initial_sidebar_state="collapsed", page_title='Intelligence Toolkit | Risk Networks')
+    load_multipage_app()
     workflows.risk_networks.workflow.create()
 
 if __name__ == '__main__':

app/util/download_pdf.py

@@ -3,20 +3,17 @@
 import pdfkit
 from util.wkhtmltopdf import config_pdfkit, pdfkit_options
 import streamlit as st
+from javascript.styles import style_pdf
 
-css = '''body {
-    font-family: 'helvetica';
-}
-'''
 #itk-label
 text_label = 'Report generated using Intelligence Toolkit (https://aka.ms/itk)'
 
 def add_download_pdf(name, text, button_text='Download PDF', is_markdown=True, disabled=False):
     if not name.endswith('.pdf'):
         name += '.pdf'
-    text = f'<style>{css}</style>{text}<hr>{text_label}'
     # Convert text to HTML if it's in Markdown format
     text = markdown2.markdown(text) if is_markdown else text
+    text = f'<style>{style_pdf}</style> \n\n {text} <hr>{text_label}'
 
     # Generate PDF from HTML string
     config_pdf = config_pdfkit()

app/util/session_variable.py

@@ -1,3 +1,4 @@
+# Copyright (c) 2024 Microsoft Corporation. All rights reserved.
 from typing import Any
 import streamlit as st
 import traceback

app/workflows/attribute_patterns/classes.py

@@ -1,6 +1,4 @@
-import streamlit as st
-import pandas as pd
-import networkx as nx
+# Copyright (c) 2024 Microsoft Corporation. All rights reserved.
 import numpy as np
 from collections import defaultdict
 

app/workflows/attribute_patterns/workflow.py

@@ -1,3 +1,4 @@
+# Copyright (c) 2024 Microsoft Corporation. All rights reserved.
 import streamlit as st
 import pandas as pd
 import numpy as np
@@ -21,7 +22,6 @@
 
 def create():
     workflow = 'attribute_patterns'
-    st.set_page_config(layout="wide", initial_sidebar_state="collapsed", page_title='Intelligence Toolkit | Attribute Patterns')
     sv = vars.SessionVariables('attribute_patterns')
     intro_tab, uploader_tab, detect_tab, explain_tab = st.tabs(['Attribute patterns workflow:', 'Create graph model', 'Detect patterns', 'Generate AI pattern reports'])
     df = None
@@ -178,4 +178,4 @@ def create():
                 is_download_disabled = report_data == ''
                 name = 'attribute_pattern_report'
                 add_download_pdf(f'{name}.pdf', report_data, 'Download PDF report', disabled=is_download_disabled)
-                st.download_button('Download markdown report', data=report_data, file_name=f'{name}.md', mime='text/markdown', disabled=is_download_disabled)
+                st.download_button('Download MD report', data=report_data, file_name=f'{name}.md', mime='text/markdown', disabled=is_download_disabled)

app/workflows/data_synthesis/workflow.py

@@ -1,4 +1,4 @@
-
+# Copyright (c) 2024 Microsoft Corporation. All rights reserved.
 import streamlit as st
 import pandas as pd
 import plotly.io as pio
@@ -21,7 +21,6 @@
 
 def create():
     workflow = 'data_synthesis'
-    st.set_page_config(layout="wide", initial_sidebar_state="collapsed", page_title='Intelligence Toolkit | Data Synthesis')
     sv = vars.SessionVariables('data_synthesis')
 
     intro_tab, prepare_tab, generate_tab, queries_tab = st.tabs(['Data synthesis workflow:', 'Upload deidentified sensitive data', 'Generate anonymous synthetic data', 'Query and visualize data'])

app/workflows/group_narratives/workflow.py

@@ -1,3 +1,4 @@
+# Copyright (c) 2024 Microsoft Corporation. All rights reserved.
 import streamlit as st
 import pandas as pd
 
@@ -10,7 +11,6 @@
 
 def create():
     workflow = 'group_narratives'
-    st.set_page_config(layout="wide", initial_sidebar_state="collapsed", page_title='Intelligence Toolkit | Group Narratives')
     sv = vars.SessionVariables('group_narratives')
 
     intro_tab, prepare_tab, summarize_tab, generate_tab = st.tabs(['Group narratives workflow:', 'Upload data to narrate', 'Prepare data summary', 'Generate AI group reports',])
@@ -205,4 +205,4 @@ def create():
                 is_download_disabled = report_data == ''
                 reports_name = 'narrative_report'
                 add_download_pdf(f'{reports_name}.pdf', report_data, button_text='Download PDF report', disabled=is_download_disabled)
-                st.download_button('Download markdown report', data=report_data, file_name=f'{reports_name}.md', mime='text/markdown', disabled=is_download_disabled)
+                st.download_button('Download MD report', data=report_data, file_name=f'{reports_name}.md', mime='text/markdown', disabled=is_download_disabled)

app/workflows/question_answering/classes.py

@@ -1,3 +1,4 @@
+# Copyright (c) 2024 Microsoft Corporation. All rights reserved.
 import numpy as np
 import tiktoken
 import scipy.spatial.distance

app/workflows/question_answering/workflow.py

@@ -1,3 +1,4 @@
+# Copyright (c) 2024 Microsoft Corporation. All rights reserved.
 import numpy as np
 import streamlit as st
 from collections import Counter
@@ -17,7 +18,6 @@
 embedder = util.Embedder.create_embedder(config.cache_dir)
 
 def create():
-    st.set_page_config(layout="wide", initial_sidebar_state="collapsed", page_title='Intelligence Toolkit | Question Answering')
     sv = vars.SessionVariables('question_answering')
 
     intro_tab, uploader_tab, mining_tab, report_tab = st.tabs(['Question answering workflow:', 'Upload data', 'Mine & match questions', 'Generate AI answer reports'])
@@ -224,4 +224,4 @@ def create():
                 full_text = sv.answering_lazy_answer_text.value + '\n\n## Supporting FAQ\n\n' + re.sub(r' Q[\d]+: ', ' ', '\n\n'.join(sv.answering_matches.value.split('\n\n')[2:]), re.MULTILINE).replace('###### ', '### ')
 
                 add_download_pdf(f'{name}.pdf', full_text, 'Download PDF report', disabled=is_download_disabled)
-                st.download_button('Download markdown report', data=full_text, file_name=f'{name}.md', mime='text/markdown', disabled=sv.answering_lazy_answer_text.value == '', key='lazy_download_button')      
+                st.download_button('Download MD report', data=full_text, file_name=f'{name}.md', mime='text/markdown', disabled=sv.answering_lazy_answer_text.value == '', key='lazy_download_button')      

app/workflows/record_matching/workflow.py

@@ -1,3 +1,4 @@
+# Copyright (c) 2024 Microsoft Corporation. All rights reserved.
 import streamlit as st
 import polars as pl
 import pandas as pd
@@ -18,7 +19,6 @@
 embedder = util.Embedder.create_embedder(config.cache_dir)
 
 def create():
-    st.set_page_config(layout="wide", initial_sidebar_state="collapsed", page_title='Intelligence Toolkit | Record Matching')
     sv = vars.SessionVariables('record_matching')
 
     if not os.path.exists(config.outputs_dir):

app/workflows/risk_networks/workflow.py

@@ -1,3 +1,4 @@
+# Copyright (c) 2024 Microsoft Corporation. All rights reserved.
 import streamlit as st
 import pandas as pd
 import networkx as nx
@@ -28,7 +29,6 @@
 embedder = util.Embedder.create_embedder(config.cache_dir)
 
 def create():
-    st.set_page_config(layout="wide", initial_sidebar_state="collapsed", page_title='Intelligence Toolkit | Risk Networks')
     sv = vars.SessionVariables('risk_networks')
 
     if not os.path.exists(config.outputs_dir):