diff --git a/app/backend/app.py b/app/backend/app.py
index 07d4d64027..927b961f7c 100644
--- a/app/backend/app.py
+++ b/app/backend/app.py
@@ -8,6 +8,7 @@
 from pathlib import Path
 from typing import Any, AsyncGenerator, Dict, Union, cast
 
+import aiohttp
 from azure.cognitiveservices.speech import (
     ResultReason,
     SpeechConfig,
@@ -133,6 +134,16 @@ async def content_file(path: str, auth_claims: Dict[str, Any]):
     if AZURE_ENFORCE_ACCESS_CONTROL is set to true, logged in users can only access files they have access to
     This is also slow and memory hungry.
     """
+    # if the path looks like issue-NNN.html, fetch it from github.com/Azure-samples/azure-search-openai-demo/issues instead
+    if path.startswith("issue-") and path.endswith(".html"):
+        issue_id = path.split("-")[1].split(".")[0]
+        url = f"https://github.com/Azure-Samples/azure-search-openai-demo/issues/{issue_id}"
+        async with aiohttp.ClientSession() as session:
+            async with session.get(url) as response:
+                if response.status != 200:
+                    abort(404)
+                return await response.text()
+
     # Remove page number from path, filename-1.txt -> filename.txt
     # This shouldn't typically be necessary as browsers don't send hash fragments to servers
     if path.find("#page=") > 0:
diff --git a/app/backend/approaches/approach.py b/app/backend/approaches/approach.py
index 44a1d6380a..c0c9a3fea9 100644
--- a/app/backend/approaches/approach.py
+++ b/app/backend/approaches/approach.py
@@ -28,6 +28,14 @@
 from core.authentication import AuthenticationHelper
 
 
+@dataclass
+class AzureAISearch:
+    aisearch_query: str
+
+@dataclass
+class GitHubIssueSearch:
+    github_query: str
+
 @dataclass
 class Document:
     id: Optional[str]
@@ -204,6 +212,34 @@ async def search(
 
         return qualified_documents
 
+
+    async def search_github_issues(self, github_issue_search: GitHubIssueSearch) -> list[Document]:
+        async with aiohttp.ClientSession() as session:
+            async with session.get(f'https://api.github.com/search/issues?q={github_issue_search.github_query}+repo:Azure-samples/azure-search-openai-demo+type:issue&per_page=10') as response:
+                if response.status == 200:
+                    issues = (await response.json()).get("items", [])
+                    documents = []
+                    # strip out image markdown from the body
+                    for issue in issues:
+                        body = issue["body"].replace("![", "").replace("](https://", "").replace(")", "")
+                        # turn html_url like https://github.com/Azure-Samples/azure-search-openai-demo/issues/2358 into issue-2358.html
+                        sourcefile = f"issue-{issue.get('number')}.html"
+                        documents.append(Document(
+                            id=issue.get("id"),
+                            content=f"# {issue.get('title')}\n\n{body}",
+                            sourcepage=sourcefile,
+                            sourcefile=sourcefile,
+                            embedding=[],
+                            image_embedding=[],
+                            category=None,
+                            oids=[],
+                            groups=[],
+                            captions=[],
+                            ))
+                    return documents
+                else:
+                    return []
+
     def get_sources_content(
         self, results: List[Document], use_semantic_captions: bool, use_image_citation: bool
     ) -> list[str]:
diff --git a/app/backend/approaches/chatapproach.py b/app/backend/approaches/chatapproach.py
index 1dc1b48171..5323f76e8b 100644
--- a/app/backend/approaches/chatapproach.py
+++ b/app/backend/approaches/chatapproach.py
@@ -1,11 +1,11 @@
 import json
 import re
 from abc import ABC, abstractmethod
-from typing import Any, AsyncGenerator, Optional
+from typing import Any, AsyncGenerator, List, Optional, Union
 
 from openai.types.chat import ChatCompletion, ChatCompletionMessageParam
 
-from approaches.approach import Approach
+from approaches.approach import Approach, AzureAISearch, GitHubIssueSearch
 
 
 class ChatApproach(Approach, ABC):
@@ -16,23 +16,27 @@ class ChatApproach(Approach, ABC):
     async def run_until_final_call(self, messages, overrides, auth_claims, should_stream) -> tuple:
         pass
 
-    def get_search_query(self, chat_completion: ChatCompletion, user_query: str):
+    def get_search_query(self, chat_completion: ChatCompletion, user_query: str) -> List[Union[AzureAISearch, GitHubIssueSearch]]:
         response_message = chat_completion.choices[0].message
+        search_queries = []
 
         if response_message.tool_calls:
             for tool in response_message.tool_calls:
                 if tool.type != "function":
                     continue
                 function = tool.function
-                if function.name == "search_sources":
+                if function.name == "azure_ai_search_docs":
                     arg = json.loads(function.arguments)
                     search_query = arg.get("search_query", self.NO_RESPONSE)
                     if search_query != self.NO_RESPONSE:
-                        return search_query
-        elif query_text := response_message.content:
-            if query_text.strip() != self.NO_RESPONSE:
-                return query_text
-        return user_query
+                        search_queries.append(AzureAISearch(aisearch_query=search_query))
+                elif function.name == "github_search_issues":
+                    arg = json.loads(function.arguments)
+                    search_query = arg.get("search_query", self.NO_RESPONSE)
+                    if search_query != self.NO_RESPONSE:
+                        search_queries.append(GitHubIssueSearch(github_query=search_query))
+
+        return search_queries
 
     def extract_followup_questions(self, content: Optional[str]):
         if content is None:
diff --git a/app/backend/approaches/chatreadretrieveread.py b/app/backend/approaches/chatreadretrieveread.py
index 7777b9a741..2caf8cebb0 100644
--- a/app/backend/approaches/chatreadretrieveread.py
+++ b/app/backend/approaches/chatreadretrieveread.py
@@ -12,7 +12,7 @@
 from openai_messages_token_helper import build_messages, get_token_limit
 
 from approaches.approach import ThoughtStep
-from approaches.chatapproach import ChatApproach
+from approaches.chatapproach import ChatApproach, GitHubIssueSearch
 from approaches.promptmanager import PromptManager
 from core.authentication import AuthenticationHelper
 
@@ -124,33 +124,40 @@ async def run_until_final_call(
             max_tokens=query_response_token_limit,  # Setting too low risks malformed JSON, setting too high may affect performance
             n=1,
             tools=tools,
+            tool_choice="auto",
             seed=seed,
         )
 
-        query_text = self.get_search_query(chat_completion, original_user_query)
-
-        # STEP 2: Retrieve relevant documents from the search index with the GPT optimized query
-
-        # If retrieval mode includes vectors, compute an embedding for the query
-        vectors: list[VectorQuery] = []
-        if use_vector_search:
-            vectors.append(await self.compute_text_embedding(query_text))
-
-        results = await self.search(
-            top,
-            query_text,
-            filter,
-            vectors,
-            use_text_search,
-            use_vector_search,
-            use_semantic_ranker,
-            use_semantic_captions,
-            minimum_search_score,
-            minimum_reranker_score,
-        )
+        search_queries = self.get_search_query(chat_completion, original_user_query)
+        results = []
+
+        for query in search_queries:
+            if isinstance(query, GitHubIssueSearch):
+                # Handle GitHub issue search
+                results.extend(await self.search_github_issues(query))
+            else:
+                # Handle regular AI search query
+
+                vectors: list[VectorQuery] = []
+                if use_vector_search:
+                    vectors.append(await self.compute_text_embedding(query.aisearch_query))
+
+                results.extend(await self.search(
+                    top,
+                    query.aisearch_query,
+                    filter,
+                    vectors,
+                    use_text_search,
+                    use_vector_search,
+                    use_semantic_ranker,
+                    use_semantic_captions,
+                    minimum_search_score,
+                    minimum_reranker_score,
+                ))
 
         # STEP 3: Generate a contextual and content specific answer using the search results and chat history
         text_sources = self.get_sources_content(results, use_semantic_captions, use_image_citation=False)
+
         rendered_answer_prompt = self.prompt_manager.render_prompt(
             self.answer_prompt,
             self.get_system_prompt_variables(overrides.get("prompt_template"))
@@ -186,7 +193,7 @@ async def run_until_final_call(
                 ),
                 ThoughtStep(
                     "Search using generated search query",
-                    query_text,
+                    search_queries,
                     {
                         "use_semantic_captions": use_semantic_captions,
                         "use_semantic_ranker": use_semantic_ranker,
@@ -222,4 +229,5 @@ async def run_until_final_call(
             stream=should_stream,
             seed=seed,
         )
+
         return (extra_info, chat_coroutine)
diff --git a/app/backend/approaches/prompts/chat_query_rewrite.prompty b/app/backend/approaches/prompts/chat_query_rewrite.prompty
index 7738a85cd7..7809bd200e 100644
--- a/app/backend/approaches/prompts/chat_query_rewrite.prompty
+++ b/app/backend/approaches/prompts/chat_query_rewrite.prompty
@@ -14,26 +14,14 @@ sample:
           content: "The Northwind Health Plus plan includes coverage for emergency services, mental health and substance abuse coverage, and out-of-network services, which are not included in the Northwind Standard plan. [Benefit_Options.pdf#page=3]"
 ---
 system:
-Below is a history of the conversation so far, and a new question asked by the user that needs to be answered by searching in a knowledge base.
-You have access to Azure AI Search index with 100's of documents.
-Generate a search query based on the conversation and the new question.
-Do not include cited source filenames and document names e.g. info.txt or doc.pdf in the search query terms.
-Do not include any text inside [] or <<>> in the search query terms.
-Do not include any special characters like '+'.
-If the question is not in English, translate the question to English before generating the search query.
-If you cannot generate a search query, return just the number 0.
-
-user:
-(EXAMPLE) How did crypto do last year?
+Below is a history of the conversation so far, and a new question asked by the user about the azure-search-openai-demo open source project.
+You have access to an Azure AI Search index with the documentation or to the GitHub issue tracker for the project documentation.
 
-assistant:
-Summarize Cryptocurrency Market Dynamics from last year
-
-user:
-(EXAMPLE) What are my health plans?
+Based on the conversation and the new question, suggest the optimal search query for the AI Search index or GitHub issue tracker.
+If the question is not in English, translate the question to English before generating the search query.
 
-assistant:
-Show available health plans
+If you cannot generate a search query for either AI Search or GitHub, return just the number 0.
+If you think that it would help to search both, then recommend both functions be called.
 
 {% for message in past_messages %}
 {{ message["role"] }}:
@@ -41,4 +29,4 @@ Show available health plans
 {% endfor %}
 
 user:
-Generate search query for: {{ user_query }}
+{{ user_query }}
diff --git a/app/backend/approaches/prompts/chat_query_rewrite_tools.json b/app/backend/approaches/prompts/chat_query_rewrite_tools.json
index cf1743483c..6c3216479c 100644
--- a/app/backend/approaches/prompts/chat_query_rewrite_tools.json
+++ b/app/backend/approaches/prompts/chat_query_rewrite_tools.json
@@ -1,17 +1,40 @@
-[{
+[
+  {
     "type": "function",
     "function": {
-        "name": "search_sources",
-        "description": "Retrieve sources from the Azure AI Search index",
-        "parameters": {
-            "type": "object",
-            "properties": {
-                "search_query": {
-                    "type": "string",
-                    "description": "Query string to retrieve documents from azure search eg: 'Health care plan'"
-                }
-            },
-            "required": ["search_query"]
-        }
+      "name": "azure_ai_search_docs",
+      "description": "Retrieve sources from the Azure AI Search index. Use this function for questions like 'does the repo support user-based access control?'",
+      "parameters": {
+        "type": "object",
+        "properties": {
+          "search_query": {
+            "type": "string",
+            "description": "Query string to retrieve documents from azure search eg: 'data access control'. Do not include cited source filenames and document names e.g. info.txt or doc.pdf in the search query terms. Do not include any text inside [] or <<>> in the search query terms. Do not include any special characters like '+'."
+          }
+        },
+        "required": ["search_query"],
+        "additionalProperties": false
+      },
+      "strict": true
     }
-}]
+  },
+  {
+    "type": "function",
+    "function": {
+      "name": "github_search_issues",
+      "description": "Retrieve issues from the azure-search-openai-demo issue tracker. Use this function for questions like 'what are the top errors with deployment?'",
+      "parameters": {
+        "type": "object",
+        "properties": {
+          "search_query": {
+            "type": "string",
+            "description": "Query string to retrieve issues from github eg: 'Deployment failure' - should only contain the search terms, does not need 'issue' or 'issues' in the search query."
+          }
+        },
+        "required": ["search_query"],
+        "additionalProperties": false
+      },
+      "strict": true
+    }
+  }
+]
diff --git a/app/frontend/src/components/AnalysisPanel/ThoughtProcess.tsx b/app/frontend/src/components/AnalysisPanel/ThoughtProcess.tsx
index 4c3a62617e..b604f490c8 100644
--- a/app/frontend/src/components/AnalysisPanel/ThoughtProcess.tsx
+++ b/app/frontend/src/components/AnalysisPanel/ThoughtProcess.tsx
@@ -28,7 +28,7 @@ export const ThoughtProcess = ({ thoughts }: Props) => {
                                     </span>
                                 ))}
                         </Stack>
-                        {Array.isArray(t.description) ? (
+                        {Array.isArray(t.description) || typeof t.description === "object" ? (
                             <SyntaxHighlighter language="json" wrapLongLines className={styles.tCodeBlock} style={a11yLight}>
                                 {JSON.stringify(t.description, null, 2)}
                             </SyntaxHighlighter>
diff --git a/app/frontend/src/locales/en/translation.json b/app/frontend/src/locales/en/translation.json
index 07f657da8b..66fecde2be 100644
--- a/app/frontend/src/locales/en/translation.json
+++ b/app/frontend/src/locales/en/translation.json
@@ -37,9 +37,9 @@
     "chatEmptyStateTitle": "Chat with your data",
     "chatEmptyStateSubtitle": "Ask anything or try an example",
     "defaultExamples": {
-        "1": "What is included in my Northwind Health Plus plan that is not in standard?",
-        "2": "What happens in a performance review?",
-        "3": "What does a Product Manager do?",
+        "1": "summarize issues with manageacls.py?",
+        "2": "how to enable user-based access control?",
+        "3": "Summarize the available documentation and reported user issues around manageacls.py",
         "placeholder": "Type a new question (e.g. does my plan cover annual eye exams?)"
     },
     "askTitle": "Ask your data",
diff --git a/app/frontend/src/pages/chat/Chat.tsx b/app/frontend/src/pages/chat/Chat.tsx
index e3c0cfd77f..838e704fec 100644
--- a/app/frontend/src/pages/chat/Chat.tsx
+++ b/app/frontend/src/pages/chat/Chat.tsx
@@ -45,7 +45,7 @@ const Chat = () => {
     const [seed, setSeed] = useState<number | null>(null);
     const [minimumRerankerScore, setMinimumRerankerScore] = useState<number>(0);
     const [minimumSearchScore, setMinimumSearchScore] = useState<number>(0);
-    const [retrieveCount, setRetrieveCount] = useState<number>(3);
+    const [retrieveCount, setRetrieveCount] = useState<number>(5);
     const [retrievalMode, setRetrievalMode] = useState<RetrievalMode>(RetrievalMode.Hybrid);
     const [useSemanticRanker, setUseSemanticRanker] = useState<boolean>(true);
     const [shouldStream, setShouldStream] = useState<boolean>(true);
diff --git a/convertdocs.py b/convertdocs.py
new file mode 100644
index 0000000000..b427e0fbe0
--- /dev/null
+++ b/convertdocs.py
@@ -0,0 +1,25 @@
+import os
+
+import pypandoc
+
+
+def convert_md_to_html(directory):
+    # Ensure the output directory exists
+    html_output_dir = os.path.join(directory, 'html')
+    os.makedirs(html_output_dir, exist_ok=True)
+
+    # Iterate over all files in the directory
+    for filename in os.listdir(directory):
+        if filename.endswith('.md'):
+            filepath = os.path.join(directory, filename)
+            base_filename = os.path.splitext(filename)[0]
+
+            # Convert to HTML
+            html_output_path = os.path.join(html_output_dir, f'{base_filename}.html')
+            pypandoc.convert_file(filepath, 'html', outputfile=html_output_path)
+            print(f'Converted {filename} to {html_output_path}')
+
+if __name__ == '__main__':
+    # Specify the directory containing the Markdown files
+    directory = '.'
+    convert_md_to_html(directory)
\ No newline at end of file
diff --git a/data/README.html b/data/README.html
new file mode 100644
index 0000000000..f62f7513ca
--- /dev/null
+++ b/data/README.html
@@ -0,0 +1,35 @@
+<h1 id="additional-documentation">Additional documentation</h1>
+<p>Consult the main <a href="../README.md">README</a> for general
+information about the project. These are advanced topics that are not
+necessary for a basic deployment.</p>
+<ul>
+<li>Deploying:
+<ul>
+<li><a href="docs/deploy_troubleshooting.md">Troubleshooting
+deployment</a>
+<ul>
+<li><a href="appservice.md">Debugging the app on App Service</a></li>
+</ul></li>
+<li><a href="azd.md">Deploying with azd: deep dive and CI/CD</a></li>
+<li><a href="deploy_existing.md">Deploying with existing Azure
+resources</a></li>
+<li><a href="deploy_lowcost.md">Deploying from a free account</a></li>
+<li><a href="deploy_features.md">Enabling optional features</a>
+<ul>
+<li><a href="docs/deploy_features.md">All features</a></li>
+<li><a href="login_and_acl.md">Login and access control</a></li>
+<li><a href="gpt4v.md">GPT-4 Turbo with Vision</a></li>
+<li><a href="deploy_private.md">Private endpoints</a></li>
+</ul></li>
+<li><a href="sharing_environments.md">Sharing deployment
+environments</a></li>
+</ul></li>
+<li><a href="localdev.md">Local development</a></li>
+<li><a href="customization.md">Customizing the app</a></li>
+<li><a href="docs/evaluation.md">Evaluation</a></li>
+<li><a href="data_ingestion.md">Data ingestion</a></li>
+<li><a href="monitoring.md">Monitoring with Application
+Insights</a></li>
+<li><a href="productionizing.md">Productionizing</a></li>
+<li><a href="other_samples.md">Alternative RAG chat samples</a></li>
+</ul>
diff --git a/data/appservice.html b/data/appservice.html
new file mode 100644
index 0000000000..baaf22fcda
--- /dev/null
+++ b/data/appservice.html
@@ -0,0 +1,696 @@
+<h1 id="rag-chat-debugging-the-app-on-app-service">RAG chat: Debugging
+the app on App Service</h1>
+<p>When you run <code>azd up</code> or <code>azd deploy</code>, it
+deploys your application to App Service, and displays the deployed
+endpoint in the console.</p>
+<p>If you encounter an error with that deployed app, you can debug the
+deployment using the tips below.</p>
+<ul>
+<li><a href="#debugging-failed-azure-app-service-deployments">Debugging
+failed Azure App Service deployments</a></li>
+<li><a href="#checking-the-deployment-logs-for-errors">Checking the
+deployment logs for errors</a></li>
+<li><a href="#checking-the-app-logs-for-errors">Checking the app logs
+for errors</a></li>
+<li><a href="#checking-azure-monitor-for-errors">Checking Azure Monitor
+for errors</a></li>
+<li><a href="#configuring-log-levels">Configuring log levels</a></li>
+</ul>
+<h2 id="debugging-failed-azure-app-service-deployments">Debugging failed
+Azure App Service deployments</h2>
+<p>If you see a 500 error upon visiting your app after deployment,
+something went wrong during either the deployment or the server start
+script.</p>
+<p>We recommend always waiting 10 minutes, to give the server time to
+properly startup.</p>
+<p>If you still see a 500 error after 10 minutes:</p>
+<ol type="1">
+<li><a href="#checking-the-deployment-logs-for-errors">Check the
+deployment logs</a></li>
+<li><a href="#checking-the-app-logs-for-errors">Look for errors in the
+app logs</a></li>
+<li><a href="#checking-azure-monitor-for-errors">Look for errors in
+Azure Monitor</a></li>
+</ol>
+<h2 id="checking-the-deployment-logs-for-errors">Checking the deployment
+logs for errors</h2>
+<p>In the Azure portal, navigate to your App Service.</p>
+<p>Select <em>Deployment Center</em> from the side navigation menu, then
+select <em>Logs</em>. You should see a timestamped list of recent
+deploys:</p>
+<figure>
+<img src="images/screenshot_appservice_deploymentcenter.png"
+alt="Screenshot of deployment logs tab from Azure App Service" />
+<figcaption aria-hidden="true">Screenshot of deployment logs tab from
+Azure App Service</figcaption>
+</figure>
+<p>Check whether the status of the most recent deploy is “Success
+(Active)” or “Failed”. If it’s success, the deployment logs might still
+reveal issues, and if it’s failed, the logs should certainly reveal the
+issue.</p>
+<p>Click the commit ID to open the logs for the most recent deploy.
+First scroll down to see if any errors or warnings are reported at the
+end. This is what you’ll hopefully see if all went well:</p>
+<figure>
+<img src="images/screenshot_appservice_deployment_log.png"
+alt="Screenshot of successful deployment logs from Azure App Service" />
+<figcaption aria-hidden="true">Screenshot of successful deployment logs
+from Azure App Service</figcaption>
+</figure>
+<p>Now scroll back up to find the timestamp with the label “Running oryx
+build”. <a href="https://github.com/microsoft/Oryx">Oryx</a> is the open
+source tool that builds apps for App Service, Functions, and other
+platforms, across all the supported MS languages. Click the <em>Show
+logs</em> link next to that label. That will pop open detailed logs at
+the bottom. Scroll down.</p>
+<details>
+<summary>
+Expand to see the logs for a successful Oryx build for the application.
+</summary>
+<pre class="plaintext"><code>Command: oryx build /tmp/zipdeploy/extracted -o /home/site/wwwroot --platform python --platform-version 3.11 -p virtualenv_name=antenv --log-file /tmp/build-debug.log  -i /tmp/8dc28dad0e10acb --compress-destination-dir | tee /tmp/oryx-build.log
+Operation performed by Microsoft Oryx, https://github.com/Microsoft/Oryx
+You can report issues at https://github.com/Microsoft/Oryx/issues
+
+Oryx Version: 0.2.20230508.1, Commit: 7fe2bf39b357dd68572b438a85ca50b5ecfb4592, ReleaseTagName: 20230508.1
+
+Build Operation ID: 7440a33100749a32
+Repository Commit : b09bff8b-da36-4d70-9e2f-c7b9131d85bc
+OS Type           : bullseye
+Image Type        : githubactions
+
+Detecting platforms...
+Detected following platforms:
+  python: 3.11.7
+Version &#39;3.11.7&#39; of platform &#39;python&#39; is not installed. Generating script to install it...
+
+Using intermediate directory &#39;/tmp/8dc28dad0e10acb&#39;.
+
+Copying files to the intermediate directory...
+Done in 27 sec(s).
+
+Source directory     : /tmp/8dc28dad0e10acb
+Destination directory: /home/site/wwwroot
+
+
+Downloading and extracting &#39;python&#39; version &#39;3.11.7&#39; to &#39;/tmp/oryx/platforms/python/3.11.7&#39;...
+Detected image debian flavor: bullseye.
+Downloaded in 5 sec(s).
+Verifying checksum...
+Extracting contents...
+performing sha512 checksum for: python...
+Done in 48 sec(s).
+
+image detector file exists, platform is python..
+OS detector file exists, OS is bullseye..
+Python Version: /tmp/oryx/platforms/python/3.11.7/bin/python3.11
+Creating directory for command manifest file if it does not exist
+Removing existing manifest file
+Python Virtual Environment: antenv
+Creating virtual environment...
+Activating virtual environment...
+Running pip install...
+[19:21:31+0000] Collecting aiofiles==23.2.1 (from -r requirements.txt (line 7))
+[19:21:31+0000]   Obtaining dependency information for aiofiles==23.2.1 from https://files.pythonhosted.org/packages/c5/19/5af6804c4cc0fed83f47bff6e413a98a36618e7d40185cd36e69737f3b0e/aiofiles-23.2.1-py3-none-any.whl.metadata
+[19:21:31+0000]   Downloading aiofiles-23.2.1-py3-none-any.whl.metadata (9.7 kB)
+[19:21:35+0000] Collecting aiohttp==3.9.3 (from -r requirements.txt (line 9))
+[19:21:35+0000]   Obtaining dependency information for aiohttp==3.9.3 from https://files.pythonhosted.org/packages/84/bb/74c9f32e1a76fab04b54ed6cd4b0dc4a07bd9dc6f3bb37f630149a9c3068/aiohttp-3.9.3-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata
+[19:21:35+0000]   Downloading aiohttp-3.9.3-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (7.4 kB)
+[19:21:35+0000] Collecting aiosignal==1.3.1 (from -r requirements.txt (line 11))
+[19:21:35+0000]   Downloading aiosignal-1.3.1-py3-none-any.whl (7.6 kB)
+[19:21:36+0000] Collecting annotated-types==0.6.0 (from -r requirements.txt (line 13))
+[19:21:36+0000]   Obtaining dependency information for annotated-types==0.6.0 from https://files.pythonhosted.org/packages/28/78/d31230046e58c207284c6b2c4e8d96e6d3cb4e52354721b944d3e1ee4aa5/annotated_types-0.6.0-py3-none-any.whl.metadata
+[19:21:36+0000]   Downloading annotated_types-0.6.0-py3-none-any.whl.metadata (12 kB)
+[19:21:36+0000] Collecting anyio==4.2.0 (from -r requirements.txt (line 15))
+[19:21:36+0000]   Obtaining dependency information for anyio==4.2.0 from https://files.pythonhosted.org/packages/bf/cd/d6d9bb1dadf73e7af02d18225cbd2c93f8552e13130484f1c8dcfece292b/anyio-4.2.0-py3-none-any.whl.metadata
+[19:21:36+0000]   Downloading anyio-4.2.0-py3-none-any.whl.metadata (4.6 kB)
+[19:21:36+0000] Collecting asgiref==3.7.2 (from -r requirements.txt (line 19))
+[19:21:36+0000]   Obtaining dependency information for asgiref==3.7.2 from https://files.pythonhosted.org/packages/9b/80/b9051a4a07ad231558fcd8ffc89232711b4e618c15cb7a392a17384bbeef/asgiref-3.7.2-py3-none-any.whl.metadata
+[19:21:36+0000]   Downloading asgiref-3.7.2-py3-none-any.whl.metadata (9.2 kB)
+[19:21:36+0000] Collecting attrs==23.2.0 (from -r requirements.txt (line 21))
+[19:21:36+0000]   Obtaining dependency information for attrs==23.2.0 from https://files.pythonhosted.org/packages/e0/44/827b2a91a5816512fcaf3cc4ebc465ccd5d598c45cefa6703fcf4a79018f/attrs-23.2.0-py3-none-any.whl.metadata
+[19:21:36+0000]   Downloading attrs-23.2.0-py3-none-any.whl.metadata (9.5 kB)
+[19:21:36+0000] Collecting azure-common==1.1.28 (from -r requirements.txt (line 23))
+[19:21:36+0000]   Downloading azure_common-1.1.28-py2.py3-none-any.whl (14 kB)
+[19:21:36+0000] Collecting azure-core==1.29.7 (from -r requirements.txt (line 27))
+[19:21:36+0000]   Obtaining dependency information for azure-core==1.29.7 from https://files.pythonhosted.org/packages/ff/29/dbc7182bc207530c7b5858d59f429158465f878845d64a038afc1aa61e35/azure_core-1.29.7-py3-none-any.whl.metadata
+[19:21:36+0000]   Downloading azure_core-1.29.7-py3-none-any.whl.metadata (36 kB)
+[19:21:36+0000] Collecting azure-core-tracing-opentelemetry==1.0.0b11 (from -r requirements.txt (line 37))
+[19:21:36+0000]   Obtaining dependency information for azure-core-tracing-opentelemetry==1.0.0b11 from https://files.pythonhosted.org/packages/e6/6e/3ef6dfba8e0faa4692caa6d103c721ccba6ac37a24744848a3a10bb3fe89/azure_core_tracing_opentelemetry-1.0.0b11-py3-none-any.whl.metadata
+[19:21:36+0000]   Downloading azure_core_tracing_opentelemetry-1.0.0b11-py3-none-any.whl.metadata (8.5 kB)
+[19:21:37+0000] Collecting azure-identity==1.15.0 (from -r requirements.txt (line 39))
+[19:21:37+0000]   Obtaining dependency information for azure-identity==1.15.0 from https://files.pythonhosted.org/packages/30/10/5dbf755b368d10a28d55b06ac1f12512a13e88874a23db82defdea9a8cd9/azure_identity-1.15.0-py3-none-any.whl.metadata
+[19:21:37+0000]   Downloading azure_identity-1.15.0-py3-none-any.whl.metadata (75 kB)
+[19:21:37+0000]      ━━━━━━━━━━━━━━━━��━━━━━━━━━━━━━━━━━━━━━━━ 75.4/75.4 kB 6.2 MB/s eta 0:00:00
+[19:21:37+0000] Collecting azure-keyvault-secrets==4.7.0 (from -r requirements.txt (line 41))
+[19:21:37+0000]   Downloading azure_keyvault_secrets-4.7.0-py3-none-any.whl (348 kB)
+[19:21:37+0000]      ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 348.6/348.6 kB 19.6 MB/s eta 0:00:00
+[19:21:37+0000] Collecting azure-monitor-opentelemetry==1.2.0 (from -r requirements.txt (line 43))
+[19:21:37+0000]   Obtaining dependency information for azure-monitor-opentelemetry==1.2.0 from https://files.pythonhosted.org/packages/66/72/5a6bac11b8f3bd60825f19c144c4c770c46951165f8ee5fc10ab3eaadf59/azure_monitor_opentelemetry-1.2.0-py3-none-any.whl.metadata
+[19:21:37+0000]   Downloading azure_monitor_opentelemetry-1.2.0-py3-none-any.whl.metadata (19 kB)
+[19:21:37+0000] Collecting azure-monitor-opentelemetry-exporter==1.0.0b21 (from -r requirements.txt (line 45))
+[19:21:37+0000]   Obtaining dependency information for azure-monitor-opentelemetry-exporter==1.0.0b21 from https://files.pythonhosted.org/packages/4a/0d/18cb0da98b49c9a6724f6cae46a7e59b8325cda476bde13b64404a428ae8/azure_monitor_opentelemetry_exporter-1.0.0b21-py2.py3-none-any.whl.metadata
+[19:21:37+0000]   Downloading azure_monitor_opentelemetry_exporter-1.0.0b21-py2.py3-none-any.whl.metadata (31 kB)
+[19:21:37+0000] Collecting azure-search-documents==11.6.0b1 (from -r requirements.txt (line 47))
+[19:21:37+0000]   Obtaining dependency information for azure-search-documents==11.6.0b1 from https://files.pythonhosted.org/packages/7c/f6/b138d9a252f80db69c052c65410bc972dca375e29c71c472e27d0bae327d/azure_search_documents-11.6.0b1-py3-none-any.whl.metadata
+[19:21:37+0000]   Downloading azure_search_documents-11.6.0b1-py3-none-any.whl.metadata (23 kB)
+[19:21:37+0000] Collecting azure-storage-blob==12.19.0 (from -r requirements.txt (line 49))
+[19:21:37+0000]   Obtaining dependency information for azure-storage-blob==12.19.0 from https://files.pythonhosted.org/packages/f6/82/24b0d7cf67ea63af86f11092756b8fe2adc1d55323241dc4107f5f5748e2/azure_storage_blob-12.19.0-py3-none-any.whl.metadata
+[19:21:37+0000]   Downloading azure_storage_blob-12.19.0-py3-none-any.whl.metadata (26 kB)
+[19:21:37+0000] Collecting blinker==1.7.0 (from -r requirements.txt (line 51))
+[19:21:37+0000]   Obtaining dependency information for blinker==1.7.0 from https://files.pythonhosted.org/packages/fa/2a/7f3714cbc6356a0efec525ce7a0613d581072ed6eb53eb7b9754f33db807/blinker-1.7.0-py3-none-any.whl.metadata
+[19:21:37+0000]   Downloading blinker-1.7.0-py3-none-any.whl.metadata (1.9 kB)
+[19:21:37+0000] Collecting certifi==2023.11.17 (from -r requirements.txt (line 55))
+[19:21:37+0000]   Obtaining dependency information for certifi==2023.11.17 from https://files.pythonhosted.org/packages/64/62/428ef076be88fa93716b576e4a01f919d25968913e817077a386fcbe4f42/certifi-2023.11.17-py3-none-any.whl.metadata
+[19:21:37+0000]   Downloading certifi-2023.11.17-py3-none-any.whl.metadata (2.2 kB)
+[19:21:39+0000] Collecting cffi==1.16.0 (from -r requirements.txt (line 61))
+[19:21:39+0000]   Obtaining dependency information for cffi==1.16.0 from https://files.pythonhosted.org/packages/9b/89/a31c81e36bbb793581d8bba4406a8aac4ba84b2559301c44eef81f4cf5df/cffi-1.16.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata
+[19:21:39+0000]   Downloading cffi-1.16.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (1.5 kB)
+[19:21:40+0000] Collecting charset-normalizer==3.3.2 (from -r requirements.txt (line 63))
+[19:21:40+0000]   Obtaining dependency information for charset-normalizer==3.3.2 from https://files.pythonhosted.org/packages/40/26/f35951c45070edc957ba40a5b1db3cf60a9dbb1b350c2d5bef03e01e61de/charset_normalizer-3.3.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata
+[19:21:40+0000]   Downloading charset_normalizer-3.3.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (33 kB)
+[19:21:40+0000] Collecting click==8.1.7 (from -r requirements.txt (line 65))
+[19:21:40+0000]   Obtaining dependency information for click==8.1.7 from https://files.pythonhosted.org/packages/00/2e/d53fa4befbf2cfa713304affc7ca780ce4fc1fd8710527771b58311a3229/click-8.1.7-py3-none-any.whl.metadata
+[19:21:40+0000]   Downloading click-8.1.7-py3-none-any.whl.metadata (3.0 kB)
+[19:21:43+0000] Collecting cryptography==42.0.1 (from -r requirements.txt (line 70))
+[19:21:43+0000]   Obtaining dependency information for cryptography==42.0.1 from https://files.pythonhosted.org/packages/f8/46/2776ca9b602f79633fdf69824b5e18c94f2e0c5f09a94fc69e5b0887c14d/cryptography-42.0.1-cp39-abi3-manylinux_2_28_x86_64.whl.metadata
+[19:21:43+0000]   Downloading cryptography-42.0.1-cp39-abi3-manylinux_2_28_x86_64.whl.metadata (5.3 kB)
+[19:21:43+0000] Collecting deprecated==1.2.14 (from -r requirements.txt (line 78))
+[19:21:43+0000]   Obtaining dependency information for deprecated==1.2.14 from https://files.pythonhosted.org/packages/20/8d/778b7d51b981a96554f29136cd59ca7880bf58094338085bcf2a979a0e6a/Deprecated-1.2.14-py2.py3-none-any.whl.metadata
+[19:21:43+0000]   Downloading Deprecated-1.2.14-py2.py3-none-any.whl.metadata (5.4 kB)
+[19:21:43+0000] Collecting distro==1.9.0 (from -r requirements.txt (line 80))
+[19:21:43+0000]   Obtaining dependency information for distro==1.9.0 from https://files.pythonhosted.org/packages/12/b3/231ffd4ab1fc9d679809f356cebee130ac7daa00d6d6f3206dd4fd137e9e/distro-1.9.0-py3-none-any.whl.metadata
+[19:21:44+0000]   Downloading distro-1.9.0-py3-none-any.whl.metadata (6.8 kB)
+[19:21:44+0000] Collecting ecdsa==0.18.0 (from -r requirements.txt (line 82))
+[19:21:44+0000]   Downloading ecdsa-0.18.0-py2.py3-none-any.whl (142 kB)
+[19:21:44+0000]      ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 142.9/142.9 kB 3.7 MB/s eta 0:00:00
+[19:21:44+0000] Collecting fixedint==0.1.6 (from -r requirements.txt (line 84))
+[19:21:44+0000]   Downloading fixedint-0.1.6-py3-none-any.whl (12 kB)
+[19:21:45+0000] Collecting flask==3.0.1 (from -r requirements.txt (line 86))
+[19:21:45+0000]   Obtaining dependency information for flask==3.0.1 from https://files.pythonhosted.org/packages/bd/0e/63738e88e981ae57c23bad6c499898314a1110a4141f77d7bd929b552fb4/flask-3.0.1-py3-none-any.whl.metadata
+[19:21:45+0000]   Downloading flask-3.0.1-py3-none-any.whl.metadata (3.6 kB)
+[19:21:47+0000] Collecting frozenlist==1.4.1 (from -r requirements.txt (line 88))
+[19:21:47+0000]   Obtaining dependency information for frozenlist==1.4.1 from https://files.pythonhosted.org/packages/b3/c9/0bc5ee7e1f5cc7358ab67da0b7dfe60fbd05c254cea5c6108e7d1ae28c63/frozenlist-1.4.1-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata
+[19:21:47+0000]   Downloading frozenlist-1.4.1-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (12 kB)
+[19:21:47+0000] Collecting h11==0.14.0 (from -r requirements.txt (line 92))
+[19:21:47+0000]   Downloading h11-0.14.0-py3-none-any.whl (58 kB)
+[19:21:47+0000]      ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 58.3/58.3 kB 2.5 MB/s eta 0:00:00
+[19:21:48+0000] Collecting h2==4.1.0 (from -r requirements.txt (line 98))
+[19:21:48+0000]   Downloading h2-4.1.0-py3-none-any.whl (57 kB)
+[19:21:48+0000]      ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 57.5/57.5 kB 627.9 kB/s eta 0:00:00
+[19:21:48+0000] Collecting hpack==4.0.0 (from -r requirements.txt (line 100))
+[19:21:48+0000]   Downloading hpack-4.0.0-py3-none-any.whl (32 kB)
+[19:21:49+0000] Collecting httpcore==1.0.2 (from -r requirements.txt (line 102))
+[19:21:49+0000]   Obtaining dependency information for httpcore==1.0.2 from https://files.pythonhosted.org/packages/56/ba/78b0a99c4da0ff8b0f59defa2f13ca4668189b134bd9840b6202a93d9a0f/httpcore-1.0.2-py3-none-any.whl.metadata
+[19:21:49+0000]   Downloading httpcore-1.0.2-py3-none-any.whl.metadata (20 kB)
+[19:21:50+0000] Collecting httpx==0.26.0 (from -r requirements.txt (line 104))
+[19:21:50+0000]   Obtaining dependency information for httpx==0.26.0 from https://files.pythonhosted.org/packages/39/9b/4937d841aee9c2c8102d9a4eeb800c7dad25386caabb4a1bf5010df81a57/httpx-0.26.0-py3-none-any.whl.metadata
+[19:21:50+0000]   Downloading httpx-0.26.0-py3-none-any.whl.metadata (7.6 kB)
+[19:21:50+0000] Collecting hypercorn==0.16.0 (from -r requirements.txt (line 106))
+[19:21:50+0000]   Obtaining dependency information for hypercorn==0.16.0 from https://files.pythonhosted.org/packages/17/9e/700d764316399c20fbe8e98c6fff903b5d3f950043cc2fcbd0831a42c953/hypercorn-0.16.0-py3-none-any.whl.metadata
+[19:21:50+0000]   Downloading hypercorn-0.16.0-py3-none-any.whl.metadata (5.4 kB)
+[19:21:50+0000] Collecting hyperframe==6.0.1 (from -r requirements.txt (line 108))
+[19:21:50+0000]   Downloading hyperframe-6.0.1-py3-none-any.whl (12 kB)
+[19:21:51+0000] Collecting idna==3.6 (from -r requirements.txt (line 110))
+[19:21:51+0000]   Obtaining dependency information for idna==3.6 from https://files.pythonhosted.org/packages/c2/e7/a82b05cf63a603df6e68d59ae6a68bf5064484a0718ea5033660af4b54a9/idna-3.6-py3-none-any.whl.metadata
+[19:21:51+0000]   Downloading idna-3.6-py3-none-any.whl.metadata (9.9 kB)
+[19:21:51+0000] Collecting importlib-metadata==6.11.0 (from -r requirements.txt (line 116))
+[19:21:51+0000]   Obtaining dependency information for importlib-metadata==6.11.0 from https://files.pythonhosted.org/packages/59/9b/ecce94952ab5ea74c31dcf9ccf78ccd484eebebef06019bf8cb579ab4519/importlib_metadata-6.11.0-py3-none-any.whl.metadata
+[19:21:51+0000]   Downloading importlib_metadata-6.11.0-py3-none-any.whl.metadata (4.9 kB)
+[19:21:52+0000] Collecting isodate==0.6.1 (from -r requirements.txt (line 118))
+[19:21:52+0000]   Downloading isodate-0.6.1-py2.py3-none-any.whl (41 kB)
+[19:21:52+0000]      ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 41.7/41.7 kB 1.8 MB/s eta 0:00:00
+[19:21:52+0000] Collecting itsdangerous==2.1.2 (from -r requirements.txt (line 124))
+[19:21:52+0000]   Downloading itsdangerous-2.1.2-py3-none-any.whl (15 kB)
+[19:21:52+0000] Collecting jinja2==3.1.3 (from -r requirements.txt (line 128))
+[19:21:52+0000]   Obtaining dependency information for jinja2==3.1.3 from https://files.pythonhosted.org/packages/30/6d/6de6be2d02603ab56e72997708809e8a5b0fbfee080735109b40a3564843/Jinja2-3.1.3-py3-none-any.whl.metadata
+[19:21:52+0000]   Downloading Jinja2-3.1.3-py3-none-any.whl.metadata (3.3 kB)
+[19:21:53+0000] Collecting markupsafe==2.1.4 (from -r requirements.txt (line 132))
+[19:21:53+0000]   Obtaining dependency information for markupsafe==2.1.4 from https://files.pythonhosted.org/packages/d3/0a/c6dfffacc5a9a17c97019cb7cbec67e5abfb65c59a58ecba270fa224f88d/MarkupSafe-2.1.4-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata
+[19:21:53+0000]   Downloading MarkupSafe-2.1.4-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (3.0 kB)
+[19:21:53+0000] Collecting msal==1.26.0 (from -r requirements.txt (line 137))
+[19:21:53+0000]   Obtaining dependency information for msal==1.26.0 from https://files.pythonhosted.org/packages/b7/61/2756b963e84db6946e4b93a8e288595106286fc11c7129fcb869267ead67/msal-1.26.0-py2.py3-none-any.whl.metadata
+[19:21:53+0000]   Downloading msal-1.26.0-py2.py3-none-any.whl.metadata (11 kB)
+[19:21:54+0000] Collecting msal-extensions==1.1.0 (from -r requirements.txt (line 142))
+[19:21:54+0000]   Obtaining dependency information for msal-extensions==1.1.0 from https://files.pythonhosted.org/packages/78/8d/ecd0eb93196f25c722ba1b923fd54d190366feccfa5b159d48dacf2b1fee/msal_extensions-1.1.0-py3-none-any.whl.metadata
+[19:21:54+0000]   Downloading msal_extensions-1.1.0-py3-none-any.whl.metadata (7.7 kB)
+[19:21:54+0000] Collecting msrest==0.7.1 (from -r requirements.txt (line 144))
+[19:21:54+0000]   Downloading msrest-0.7.1-py3-none-any.whl (85 kB)
+[19:21:54+0000]      ━━━━━━━━━━━━━━━━━━━��━━━━━━━━━━━━━━━━━━━━ 85.4/85.4 kB 6.2 MB/s eta 0:00:00
+[19:21:59+0000] Collecting multidict==6.0.4 (from -r requirements.txt (line 146))
+[19:21:59+0000]   Downloading multidict-6.0.4-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (117 kB)
+[19:21:59+0000]      ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 117.4/117.4 kB 2.2 MB/s eta 0:00:00
+[19:22:05+0000] Collecting numpy==1.26.3 (from -r requirements.txt (line 150))
+[19:22:05+0000]   Obtaining dependency information for numpy==1.26.3 from https://files.pythonhosted.org/packages/5a/62/007b63f916aca1d27f5fede933fda3315d931ff9b2c28b9c2cf388cd8edb/numpy-1.26.3-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata
+[19:22:05+0000]   Downloading numpy-1.26.3-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (61 kB)
+[19:22:05+0000]      ━━━━━━━━━━━━━━━━━━━━━━━━━���━━━━━━━━━━━━━━ 61.2/61.2 kB 5.6 MB/s eta 0:00:00
+[19:22:05+0000] Collecting oauthlib==3.2.2 (from -r requirements.txt (line 155))
+[19:22:05+0000]   Downloading oauthlib-3.2.2-py3-none-any.whl (151 kB)
+[19:22:05+0000]      ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 151.7/151.7 kB 7.9 MB/s eta 0:00:00
+[19:22:06+0000] Collecting openai[datalib]==1.10.0 (from -r requirements.txt (line 157))
+[19:22:06+0000]   Obtaining dependency information for openai[datalib]==1.10.0 from https://files.pythonhosted.org/packages/46/85/8681046cd9cc13a36ac76e4a1b047338c90dbeab2e9b14fb36de7f314c93/openai-1.10.0-py3-none-any.whl.metadata
+[19:22:06+0000]   Downloading openai-1.10.0-py3-none-any.whl.metadata (18 kB)
+[19:22:06+0000] Collecting opentelemetry-api==1.22.0 (from -r requirements.txt (line 159))
+[19:22:06+0000]   Obtaining dependency information for opentelemetry-api==1.22.0 from https://files.pythonhosted.org/packages/fc/2e/a8509051aa446783e24ee03d74bd268c07d5d25a8d48686cfcf3429d5d32/opentelemetry_api-1.22.0-py3-none-any.whl.metadata
+[19:22:06+0000]   Downloading opentelemetry_api-1.22.0-py3-none-any.whl.metadata (1.4 kB)
+[19:22:07+0000] Collecting opentelemetry-instrumentation==0.43b0 (from -r requirements.txt (line 177))
+[19:22:07+0000]   Obtaining dependency information for opentelemetry-instrumentation==0.43b0 from https://files.pythonhosted.org/packages/91/f0/4a9f7cbcc697273d847040a9e4f98ceb07b642e1fe5fed56a0fb6b567665/opentelemetry_instrumentation-0.43b0-py3-none-any.whl.metadata
+[19:22:07+0000]   Downloading opentelemetry_instrumentation-0.43b0-py3-none-any.whl.metadata (5.9 kB)
+[19:22:08+0000] Collecting opentelemetry-instrumentation-aiohttp-client==0.43b0 (from -r requirements.txt (line 191))
+[19:22:08+0000]   Obtaining dependency information for opentelemetry-instrumentation-aiohttp-client==0.43b0 from https://files.pythonhosted.org/packages/23/75/ce33cd15bc706b1e170e5ce65235a8418e3332ad543419b902a9d24f079f/opentelemetry_instrumentation_aiohttp_client-0.43b0-py3-none-any.whl.metadata
+[19:22:08+0000]   Downloading opentelemetry_instrumentation_aiohttp_client-0.43b0-py3-none-any.whl.metadata (2.2 kB)
+[19:22:08+0000] Collecting opentelemetry-instrumentation-asgi==0.43b0 (from -r requirements.txt (line 193))
+[19:22:08+0000]   Obtaining dependency information for opentelemetry-instrumentation-asgi==0.43b0 from https://files.pythonhosted.org/packages/71/cd/a0456c8e4441d9ef5b412a3ffdf97629a81adeb331f8bb645df4f9153dd8/opentelemetry_instrumentation_asgi-0.43b0-py3-none-any.whl.metadata
+[19:22:08+0000]   Downloading opentelemetry_instrumentation_asgi-0.43b0-py3-none-any.whl.metadata (2.1 kB)
+[19:22:08+0000] Collecting opentelemetry-instrumentation-dbapi==0.43b0 (from -r requirements.txt (line 197))
+[19:22:08+0000]   Obtaining dependency information for opentelemetry-instrumentation-dbapi==0.43b0 from https://files.pythonhosted.org/packages/6d/96/f46bffb40e71f0abd82ad24ecfa7a8e29b6abca631f7d049d80afee83ff9/opentelemetry_instrumentation_dbapi-0.43b0-py3-none-any.whl.metadata
+[19:22:08+0000]   Downloading opentelemetry_instrumentation_dbapi-0.43b0-py3-none-any.whl.metadata (1.9 kB)
+[19:22:09+0000] Collecting opentelemetry-instrumentation-django==0.43b0 (from -r requirements.txt (line 199))
+[19:22:09+0000]   Obtaining dependency information for opentelemetry-instrumentation-django==0.43b0 from https://files.pythonhosted.org/packages/11/66/a6b5aadb04b5daf002fcbe97bb6bc83416c53b81a608de0e9ad886c59643/opentelemetry_instrumentation_django-0.43b0-py3-none-any.whl.metadata
+[19:22:09+0000]   Downloading opentelemetry_instrumentation_django-0.43b0-py3-none-any.whl.metadata (2.3 kB)
+[19:22:09+0000] Collecting opentelemetry-instrumentation-fastapi==0.43b0 (from -r requirements.txt (line 201))
+[19:22:09+0000]   Obtaining dependency information for opentelemetry-instrumentation-fastapi==0.43b0 from https://files.pythonhosted.org/packages/1d/51/429d04b8694fec2f87184ced4beeab1dd6db194a9444b0a6fca1675338b2/opentelemetry_instrumentation_fastapi-0.43b0-py3-none-any.whl.metadata
+[19:22:09+0000]   Downloading opentelemetry_instrumentation_fastapi-0.43b0-py3-none-any.whl.metadata (2.3 kB)
+[19:22:10+0000] Collecting opentelemetry-instrumentation-flask==0.43b0 (from -r requirements.txt (line 203))
+[19:22:10+0000]   Obtaining dependency information for opentelemetry-instrumentation-flask==0.43b0 from https://files.pythonhosted.org/packages/21/eb/4b0d6f98d2767c7117ebe497bcc58f00e70cc6b4ce97b99bd3eccf3d6644/opentelemetry_instrumentation_flask-0.43b0-py3-none-any.whl.metadata
+[19:22:10+0000]   Downloading opentelemetry_instrumentation_flask-0.43b0-py3-none-any.whl.metadata (2.4 kB)
+[19:22:10+0000] Collecting opentelemetry-instrumentation-httpx==0.43b0 (from -r requirements.txt (line 205))
+[19:22:10+0000]   Obtaining dependency information for opentelemetry-instrumentation-httpx==0.43b0 from https://files.pythonhosted.org/packages/7e/ed/a8d3951650145d7d7997c83e35c59c02c8bf632c24ff2e07ab065ad7dd48/opentelemetry_instrumentation_httpx-0.43b0-py3-none-any.whl.metadata
+[19:22:10+0000]   Downloading opentelemetry_instrumentation_httpx-0.43b0-py3-none-any.whl.metadata (7.1 kB)
+[19:22:11+0000] Collecting opentelemetry-instrumentation-psycopg2==0.43b0 (from -r requirements.txt (line 207))
+[19:22:11+0000]   Obtaining dependency information for opentelemetry-instrumentation-psycopg2==0.43b0 from https://files.pythonhosted.org/packages/0a/4e/f2085da8254b0f019a5dd57f737395c39274a23c25bf3dfe4030a4169325/opentelemetry_instrumentation_psycopg2-0.43b0-py3-none-any.whl.metadata
+[19:22:11+0000]   Downloading opentelemetry_instrumentation_psycopg2-0.43b0-py3-none-any.whl.metadata (2.1 kB)
+[19:22:11+0000] Collecting opentelemetry-instrumentation-requests==0.43b0 (from -r requirements.txt (line 209))
+[19:22:11+0000]   Obtaining dependency information for opentelemetry-instrumentation-requests==0.43b0 from https://files.pythonhosted.org/packages/3b/a9/98618c6383cad51313f448412cadd0bed43634f0287eaf67a3e71a536f9c/opentelemetry_instrumentation_requests-0.43b0-py3-none-any.whl.metadata
+[19:22:11+0000]   Downloading opentelemetry_instrumentation_requests-0.43b0-py3-none-any.whl.metadata (2.7 kB)
+[19:22:12+0000] Collecting opentelemetry-instrumentation-urllib==0.43b0 (from -r requirements.txt (line 213))
+[19:22:12+0000]   Obtaining dependency information for opentelemetry-instrumentation-urllib==0.43b0 from https://files.pythonhosted.org/packages/29/8a/c184945b2628ed44b9357e0df84dfc0974efd4e1360b3d89d2180ebfb3c0/opentelemetry_instrumentation_urllib-0.43b0-py3-none-any.whl.metadata
+[19:22:12+0000]   Downloading opentelemetry_instrumentation_urllib-0.43b0-py3-none-any.whl.metadata (3.4 kB)
+[19:22:12+0000] Collecting opentelemetry-instrumentation-urllib3==0.43b0 (from -r requirements.txt (line 215))
+[19:22:12+0000]   Obtaining dependency information for opentelemetry-instrumentation-urllib3==0.43b0 from https://files.pythonhosted.org/packages/a0/54/3e6fc502e06d6c4cba23f314426951225f950b1af3c2e6decb780cd64ff1/opentelemetry_instrumentation_urllib3-0.43b0-py3-none-any.whl.metadata
+[19:22:12+0000]   Downloading opentelemetry_instrumentation_urllib3-0.43b0-py3-none-any.whl.metadata (3.6 kB)
+[19:22:13+0000] Collecting opentelemetry-instrumentation-wsgi==0.43b0 (from -r requirements.txt (line 217))
+[19:22:13+0000]   Obtaining dependency information for opentelemetry-instrumentation-wsgi==0.43b0 from https://files.pythonhosted.org/packages/4a/37/6315abd394778d76b9bf206980436a8539cc13ddcd0bced709f4d9c3d1e8/opentelemetry_instrumentation_wsgi-0.43b0-py3-none-any.whl.metadata
+[19:22:13+0000]   Downloading opentelemetry_instrumentation_wsgi-0.43b0-py3-none-any.whl.metadata (2.1 kB)
+[19:22:13+0000] Collecting opentelemetry-resource-detector-azure==0.1.3 (from -r requirements.txt (line 221))
+[19:22:13+0000]   Obtaining dependency information for opentelemetry-resource-detector-azure==0.1.3 from https://files.pythonhosted.org/packages/99/c4/6790b15d360d0a14c5fb3a754d713470758da8a3635d90502aabb52febe2/opentelemetry_resource_detector_azure-0.1.3-py3-none-any.whl.metadata
+[19:22:13+0000]   Downloading opentelemetry_resource_detector_azure-0.1.3-py3-none-any.whl.metadata (4.6 kB)
+[19:22:14+0000] Collecting opentelemetry-sdk==1.22.0 (from -r requirements.txt (line 223))
+[19:22:14+0000]   Obtaining dependency information for opentelemetry-sdk==1.22.0 from https://files.pythonhosted.org/packages/ff/94/588f49e0dd9a62ec46102736d2378330032a55e19c79ff7e4febea7ebed1/opentelemetry_sdk-1.22.0-py3-none-any.whl.metadata
+[19:22:14+0000]   Downloading opentelemetry_sdk-1.22.0-py3-none-any.whl.metadata (1.5 kB)
+[19:22:14+0000] Collecting opentelemetry-semantic-conventions==0.43b0 (from -r requirements.txt (line 227))
+[19:22:14+0000]   Obtaining dependency information for opentelemetry-semantic-conventions==0.43b0 from https://files.pythonhosted.org/packages/e0/26/69be0f1a56a362c68fa0c7632d841b1b8f29d809bc6b1b897387c9f46973/opentelemetry_semantic_conventions-0.43b0-py3-none-any.whl.metadata
+[19:22:14+0000]   Downloading opentelemetry_semantic_conventions-0.43b0-py3-none-any.whl.metadata (2.3 kB)
+[19:22:15+0000] Collecting opentelemetry-util-http==0.43b0 (from -r requirements.txt (line 241))
+[19:22:15+0000]   Obtaining dependency information for opentelemetry-util-http==0.43b0 from https://files.pythonhosted.org/packages/74/91/a87a59baeeb917a93f2cc86fa670cf533328d18a2d09b0cef4f65e8b83e9/opentelemetry_util_http-0.43b0-py3-none-any.whl.metadata
+[19:22:15+0000]   Downloading opentelemetry_util_http-0.43b0-py3-none-any.whl.metadata (2.5 kB)
+[19:22:15+0000] Collecting packaging==23.2 (from -r requirements.txt (line 252))
+[19:22:15+0000]   Obtaining dependency information for packaging==23.2 from https://files.pythonhosted.org/packages/ec/1a/610693ac4ee14fcdf2d9bf3c493370e4f2ef7ae2e19217d7a237ff42367d/packaging-23.2-py3-none-any.whl.metadata
+[19:22:15+0000]   Downloading packaging-23.2-py3-none-any.whl.metadata (3.2 kB)
+[19:22:20+0000] Collecting pandas==2.2.0 (from -r requirements.txt (line 256))
+[19:22:20+0000]   Obtaining dependency information for pandas==2.2.0 from https://files.pythonhosted.org/packages/5b/7e/9fd11ba8e86a8add8f2ff4e11c7111f65ec6fd1b547222160bb969e2bf5e/pandas-2.2.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata
+[19:22:20+0000]   Downloading pandas-2.2.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (19 kB)
+[19:22:21+0000] Collecting pandas-stubs==2.1.4.231227 (from -r requirements.txt (line 258))
+[19:22:21+0000]   Obtaining dependency information for pandas-stubs==2.1.4.231227 from https://files.pythonhosted.org/packages/c0/6d/c5c23926fcc7526a5df32a8f3b3540948be8dd4c25f4a097f9091d40535c/pandas_stubs-2.1.4.231227-py3-none-any.whl.metadata
+[19:22:21+0000]   Downloading pandas_stubs-2.1.4.231227-py3-none-any.whl.metadata (9.6 kB)
+[19:22:26+0000] Collecting pillow==10.2.0 (from -r requirements.txt (line 260))
+[19:22:26+0000]   Obtaining dependency information for pillow==10.2.0 from https://files.pythonhosted.org/packages/66/9c/2e1877630eb298bbfd23f90deeec0a3f682a4163d5ca9f178937de57346c/pillow-10.2.0-cp311-cp311-manylinux_2_28_x86_64.whl.metadata
+[19:22:26+0000]   Downloading pillow-10.2.0-cp311-cp311-manylinux_2_28_x86_64.whl.metadata (9.7 kB)
+[19:22:27+0000] Collecting portalocker==2.8.2 (from -r requirements.txt (line 262))
+[19:22:27+0000]   Obtaining dependency information for portalocker==2.8.2 from https://files.pythonhosted.org/packages/17/9e/87671efcca80ba6203811540ed1f9c0462c1609d2281d7b7f53cef05da3d/portalocker-2.8.2-py3-none-any.whl.metadata
+[19:22:27+0000]   Downloading portalocker-2.8.2-py3-none-any.whl.metadata (8.5 kB)
+[19:22:28+0000] Collecting priority==2.0.0 (from -r requirements.txt (line 264))
+[19:22:28+0000]   Downloading priority-2.0.0-py3-none-any.whl (8.9 kB)
+[19:22:28+0000] Collecting pyasn1==0.5.1 (from -r requirements.txt (line 266))
+[19:22:28+0000]   Obtaining dependency information for pyasn1==0.5.1 from https://files.pythonhosted.org/packages/d1/75/4686d2872bf2fc0b37917cbc8bbf0dd3a5cdb0990799be1b9cbf1e1eb733/pyasn1-0.5.1-py2.py3-none-any.whl.metadata
+[19:22:29+0000]   Downloading pyasn1-0.5.1-py2.py3-none-any.whl.metadata (8.6 kB)
+[19:22:29+0000] Collecting pycparser==2.21 (from -r requirements.txt (line 270))
+[19:22:29+0000]   Downloading pycparser-2.21-py2.py3-none-any.whl (118 kB)
+[19:22:29+0000]      ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 118.7/118.7 kB 3.6 MB/s eta 0:00:00
+[19:22:31+0000] Collecting pydantic==2.6.0 (from -r requirements.txt (line 272))
+[19:22:31+0000]   Obtaining dependency information for pydantic==2.6.0 from https://files.pythonhosted.org/packages/e4/37/3ffe6e7daa1ea1b4bf5228807a92ccbae538cf57c0c50b93564c310c11a8/pydantic-2.6.0-py3-none-any.whl.metadata
+[19:22:31+0000]   Downloading pydantic-2.6.0-py3-none-any.whl.metadata (81 kB)
+[19:22:31+0000]      ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 81.8/81.8 kB 3.6 MB/s eta 0:00:00
+[19:22:39+0000] Collecting pydantic-core==2.16.1 (from -r requirements.txt (line 274))
+[19:22:39+0000]   Obtaining dependency information for pydantic-core==2.16.1 from https://files.pythonhosted.org/packages/98/19/955b83b6e33b7ac27914860069a918fe49b29c13bc149dc7bb7c60954812/pydantic_core-2.16.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata
+[19:22:39+0000]   Downloading pydantic_core-2.16.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (6.5 kB)
+[19:22:39+0000] Collecting pyjwt[crypto]==2.8.0 (from -r requirements.txt (line 276))
+[19:22:39+0000]   Obtaining dependency information for pyjwt[crypto]==2.8.0 from https://files.pythonhosted.org/packages/2b/4f/e04a8067c7c96c364cef7ef73906504e2f40d690811c021e1a1901473a19/PyJWT-2.8.0-py3-none-any.whl.metadata
+[19:22:39+0000]   Downloading PyJWT-2.8.0-py3-none-any.whl.metadata (4.2 kB)
+[19:22:39+0000] Collecting python-dateutil==2.8.2 (from -r requirements.txt (line 278))
+[19:22:39+0000]   Downloading python_dateutil-2.8.2-py2.py3-none-any.whl (247 kB)
+[19:22:39+0000]      ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 247.7/247.7 kB 19.2 MB/s eta 0:00:00
+[19:22:40+0000] Collecting python-jose[cryptography]==3.3.0 (from -r requirements.txt (line 280))
+[19:22:40+0000]   Downloading python_jose-3.3.0-py2.py3-none-any.whl (33 kB)
+[19:22:40+0000] Collecting pytz==2023.4 (from -r requirements.txt (line 282))
+[19:22:40+0000]   Obtaining dependency information for pytz==2023.4 from https://files.pythonhosted.org/packages/3b/dd/9b84302ba85ac6d3d3042d3e8698374838bde1c386b4adb1223d7a0efd4e/pytz-2023.4-py2.py3-none-any.whl.metadata
+[19:22:40+0000]   Downloading pytz-2023.4-py2.py3-none-any.whl.metadata (22 kB)
+[19:22:41+0000] Collecting quart==0.19.4 (from -r requirements.txt (line 284))
+[19:22:41+0000]   Obtaining dependency information for quart==0.19.4 from https://files.pythonhosted.org/packages/9a/2c/681b4fcecefd98627a90dd5aecdc6b57ba18c9ce07e173d86a0b1274f20b/quart-0.19.4-py3-none-any.whl.metadata
+[19:22:41+0000]   Downloading quart-0.19.4-py3-none-any.whl.metadata (5.7 kB)
+[19:22:42+0000] Collecting quart-cors==0.7.0 (from -r requirements.txt (line 288))
+[19:22:42+0000]   Obtaining dependency information for quart-cors==0.7.0 from https://files.pythonhosted.org/packages/60/fc/1ffe9042df05d48f5eaac4116708fee3f7bb18b696380cc4e3797c8fd510/quart_cors-0.7.0-py3-none-any.whl.metadata
+[19:22:42+0000]   Downloading quart_cors-0.7.0-py3-none-any.whl.metadata (9.4 kB)
+[19:22:50+0000] Collecting regex==2023.12.25 (from -r requirements.txt (line 290))
+[19:22:50+0000]   Obtaining dependency information for regex==2023.12.25 from https://files.pythonhosted.org/packages/8d/6b/2f6478814954c07c04ba60b78d688d3d7bab10d786e0b6c1db607e4f6673/regex-2023.12.25-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata
+[19:22:50+0000]   Downloading regex-2023.12.25-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (40 kB)
+[19:22:50+0000]      ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 40.9/40.9 kB 504.8 kB/s eta 0:00:00
+[19:22:50+0000] Collecting requests==2.31.0 (from -r requirements.txt (line 292))
+[19:22:50+0000]   Obtaining dependency information for requests==2.31.0 from https://files.pythonhosted.org/packages/70/8e/0e2d847013cb52cd35b38c009bb167a1a26b2ce6cd6965bf26b47bc0bf44/requests-2.31.0-py3-none-any.whl.metadata
+[19:22:50+0000]   Downloading requests-2.31.0-py3-none-any.whl.metadata (4.6 kB)
+[19:22:51+0000] Collecting requests-oauthlib==1.3.1 (from -r requirements.txt (line 299))
+[19:22:51+0000]   Downloading requests_oauthlib-1.3.1-py2.py3-none-any.whl (23 kB)
+[19:22:51+0000] Collecting rsa==4.9 (from -r requirements.txt (line 301))
+[19:22:51+0000]   Downloading rsa-4.9-py3-none-any.whl (34 kB)
+[19:22:51+0000] Collecting six==1.16.0 (from -r requirements.txt (line 303))
+[19:22:52+0000]   Downloading six-1.16.0-py2.py3-none-any.whl (11 kB)
+[19:22:53+0000] Collecting sniffio==1.3.0 (from -r requirements.txt (line 309))
+[19:22:53+0000]   Downloading sniffio-1.3.0-py3-none-any.whl (10 kB)
+[19:22:53+0000] Collecting tenacity==8.2.3 (from -r requirements.txt (line 314))
+[19:22:53+0000]   Obtaining dependency information for tenacity==8.2.3 from https://files.pythonhosted.org/packages/f4/f1/990741d5bb2487d529d20a433210ffa136a367751e454214013b441c4575/tenacity-8.2.3-py3-none-any.whl.metadata
+[19:22:53+0000]   Downloading tenacity-8.2.3-py3-none-any.whl.metadata (1.0 kB)
+[19:22:54+0000] Collecting tiktoken==0.5.2 (from -r requirements.txt (line 316))
+[19:22:54+0000]   Obtaining dependency information for tiktoken==0.5.2 from https://files.pythonhosted.org/packages/fb/a9/237dc2db35e6ec0fb7dd63e3d10ebe0377559203bd2a87e12a4adbfc8585/tiktoken-0.5.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata
+[19:22:54+0000]   Downloading tiktoken-0.5.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (6.6 kB)
+[19:22:55+0000] Collecting tqdm==4.66.1 (from -r requirements.txt (line 318))
+[19:22:55+0000]   Obtaining dependency information for tqdm==4.66.1 from https://files.pythonhosted.org/packages/00/e5/f12a80907d0884e6dff9c16d0c0114d81b8cd07dc3ae54c5e962cc83037e/tqdm-4.66.1-py3-none-any.whl.metadata
+[19:22:55+0000]   Downloading tqdm-4.66.1-py3-none-any.whl.metadata (57 kB)
+[19:22:55+0000]      ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 57.6/57.6 kB 2.4 MB/s eta 0:00:00
+[19:22:55+0000] Collecting types-pillow==10.2.0.20240206 (from -r requirements.txt (line 320))
+[19:22:55+0000]   Obtaining dependency information for types-pillow==10.2.0.20240206 from https://files.pythonhosted.org/packages/54/a1/9c24f95c637f5ed77f0a1de9077a06af018acc0d3ffe9bb0843abc13619c/types_Pillow-10.2.0.20240206-py3-none-any.whl.metadata
+[19:22:56+0000]   Downloading types_Pillow-10.2.0.20240206-py3-none-any.whl.metadata (1.6 kB)
+[19:22:56+0000] Collecting types-pytz==2023.4.0.20240130 (from -r requirements.txt (line 322))
+[19:22:56+0000]   Obtaining dependency information for types-pytz==2023.4.0.20240130 from https://files.pythonhosted.org/packages/83/cd/018e825d60d86c1798c7acccfcb3d7c31227793445e4b87423498e8c486d/types_pytz-2023.4.0.20240130-py3-none-any.whl.metadata
+[19:22:56+0000]   Downloading types_pytz-2023.4.0.20240130-py3-none-any.whl.metadata (1.5 kB)
+[19:22:57+0000] Collecting typing-extensions==4.9.0 (from -r requirements.txt (line 324))
+[19:22:57+0000]   Obtaining dependency information for typing-extensions==4.9.0 from https://files.pythonhosted.org/packages/b7/f4/6a90020cd2d93349b442bfcb657d0dc91eee65491600b2cb1d388bc98e6b/typing_extensions-4.9.0-py3-none-any.whl.metadata
+[19:22:57+0000]   Downloading typing_extensions-4.9.0-py3-none-any.whl.metadata (3.0 kB)
+[19:22:57+0000] Collecting tzdata==2023.4 (from -r requirements.txt (line 333))
+[19:22:57+0000]   Obtaining dependency information for tzdata==2023.4 from https://files.pythonhosted.org/packages/a3/fb/52b62131e21b24ee297e4e95ed41eba29647dad0e0051a92bb66b43c70ff/tzdata-2023.4-py2.py3-none-any.whl.metadata
+[19:22:57+0000]   Downloading tzdata-2023.4-py2.py3-none-any.whl.metadata (1.4 kB)
+[19:22:57+0000] Collecting urllib3==2.1.0 (from -r requirements.txt (line 335))
+[19:22:57+0000]   Obtaining dependency information for urllib3==2.1.0 from https://files.pythonhosted.org/packages/96/94/c31f58c7a7f470d5665935262ebd7455c7e4c7782eb525658d3dbf4b9403/urllib3-2.1.0-py3-none-any.whl.metadata
+[19:22:57+0000]   Downloading urllib3-2.1.0-py3-none-any.whl.metadata (6.4 kB)
+[19:22:58+0000] Collecting uvicorn==0.27.0.post1 (from -r requirements.txt (line 337))
+[19:22:58+0000]   Obtaining dependency information for uvicorn==0.27.0.post1 from https://files.pythonhosted.org/packages/c7/f3/29caa83f5795b20ed3aca357c648f3ae995ff6ff08e38b22387017abbdc5/uvicorn-0.27.0.post1-py3-none-any.whl.metadata
+[19:22:58+0000]   Downloading uvicorn-0.27.0.post1-py3-none-any.whl.metadata (6.4 kB)
+[19:22:59+0000] Collecting werkzeug==3.0.1 (from -r requirements.txt (line 339))
+[19:22:59+0000]   Obtaining dependency information for werkzeug==3.0.1 from https://files.pythonhosted.org/packages/c3/fc/254c3e9b5feb89ff5b9076a23218dafbc99c96ac5941e900b71206e6313b/werkzeug-3.0.1-py3-none-any.whl.metadata
+[19:22:59+0000]   Downloading werkzeug-3.0.1-py3-none-any.whl.metadata (4.1 kB)
+[19:23:00+0000] Collecting wrapt==1.16.0 (from -r requirements.txt (line 343))
+[19:23:00+0000]   Obtaining dependency information for wrapt==1.16.0 from https://files.pythonhosted.org/packages/6e/52/2da48b35193e39ac53cfb141467d9f259851522d0e8c87153f0ba4205fb1/wrapt-1.16.0-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata
+[19:23:00+0000]   Downloading wrapt-1.16.0-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (6.6 kB)
+[19:23:00+0000] Collecting wsproto==1.2.0 (from -r requirements.txt (line 350))
+[19:23:00+0000]   Downloading wsproto-1.2.0-py3-none-any.whl (24 kB)
+[19:23:03+0000] Collecting yarl==1.9.4 (from -r requirements.txt (line 352))
+[19:23:03+0000]   Obtaining dependency information for yarl==1.9.4 from https://files.pythonhosted.org/packages/9f/ea/94ad7d8299df89844e666e4aa8a0e9b88e02416cd6a7dd97969e9eae5212/yarl-1.9.4-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata
+[19:23:03+0000]   Downloading yarl-1.9.4-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (31 kB)
+[19:23:03+0000] Collecting zipp==3.17.0 (from -r requirements.txt (line 354))
+[19:23:03+0000]   Obtaining dependency information for zipp==3.17.0 from https://files.pythonhosted.org/packages/d9/66/48866fc6b158c81cc2bfecc04c480f105c6040e8b077bc54c634b4a67926/zipp-3.17.0-py3-none-any.whl.metadata
+[19:23:03+0000]   Downloading zipp-3.17.0-py3-none-any.whl.metadata (3.7 kB)
+[19:23:30+0000] Requirement already satisfied: setuptools&gt;=16.0 in ./antenv/lib/python3.11/site-packages (from opentelemetry-instrumentation==0.43b0-&gt;-r requirements.txt (line 177)) (65.5.0)
+[19:23:51+0000] Downloading aiofiles-23.2.1-py3-none-any.whl (15 kB)
+[19:23:51+0000] Downloading aiohttp-3.9.3-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (1.3 MB)
+[19:23:51+0000]    ━━━━━━━━━━━━━━━━━��━━━━━━━━━━━━━━━━━━━━━━ 1.3/1.3 MB 32.0 MB/s eta 0:00:00
+[19:23:51+0000] Downloading annotated_types-0.6.0-py3-none-any.whl (12 kB)
+[19:23:51+0000] Downloading anyio-4.2.0-py3-none-any.whl (85 kB)
+[19:23:51+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 85.5/85.5 kB 6.9 MB/s eta 0:00:00
+[19:23:51+0000] Downloading asgiref-3.7.2-py3-none-any.whl (24 kB)
+[19:23:51+0000] Downloading attrs-23.2.0-py3-none-any.whl (60 kB)
+[19:23:51+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 60.8/60.8 kB 5.9 MB/s eta 0:00:00
+[19:23:51+0000] Downloading azure_core-1.29.7-py3-none-any.whl (192 kB)
+[19:23:51+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 192.9/192.9 kB 6.3 MB/s eta 0:00:00
+[19:23:51+0000] Downloading azure_core_tracing_opentelemetry-1.0.0b11-py3-none-any.whl (10 kB)
+[19:23:51+0000] Downloading azure_identity-1.15.0-py3-none-any.whl (164 kB)
+[19:23:52+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 164.7/164.7 kB 5.7 MB/s eta 0:00:00
+[19:23:52+0000] Downloading azure_monitor_opentelemetry-1.2.0-py3-none-any.whl (20 kB)
+[19:23:52+0000] Downloading azure_monitor_opentelemetry_exporter-1.0.0b21-py2.py3-none-any.whl (78 kB)
+[19:23:52+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 78.5/78.5 kB 787.5 kB/s eta 0:00:00
+[19:23:52+0000] Downloading azure_search_documents-11.6.0b1-py3-none-any.whl (315 kB)
+[19:23:52+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 315.3/315.3 kB 5.6 MB/s eta 0:00:00
+[19:23:52+0000] Downloading azure_storage_blob-12.19.0-py3-none-any.whl (394 kB)
+[19:23:52+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 394.2/394.2 kB 19.4 MB/s eta 0:00:00
+[19:23:52+0000] Downloading blinker-1.7.0-py3-none-any.whl (13 kB)
+[19:23:52+0000] Downloading certifi-2023.11.17-py3-none-any.whl (162 kB)
+[19:23:52+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 162.5/162.5 kB 6.3 MB/s eta 0:00:00
+[19:23:52+0000] Downloading cffi-1.16.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (464 kB)
+[19:23:52+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 464.8/464.8 kB 3.1 MB/s eta 0:00:00
+[19:23:52+0000] Downloading charset_normalizer-3.3.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (140 kB)
+[19:23:52+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 140.3/140.3 kB 5.7 MB/s eta 0:00:00
+[19:23:52+0000] Downloading click-8.1.7-py3-none-any.whl (97 kB)
+[19:23:53+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 97.9/97.9 kB 831.8 kB/s eta 0:00:00
+[19:23:53+0000] Downloading cryptography-42.0.1-cp39-abi3-manylinux_2_28_x86_64.whl (4.6 MB)
+[19:23:53+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 4.6/4.6 MB 15.8 MB/s eta 0:00:00
+[19:23:53+0000] Downloading Deprecated-1.2.14-py2.py3-none-any.whl (9.6 kB)
+[19:23:53+0000] Downloading distro-1.9.0-py3-none-any.whl (20 kB)
+[19:23:53+0000] Downloading flask-3.0.1-py3-none-any.whl (101 kB)
+[19:23:53+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 101.2/101.2 kB 1.1 MB/s eta 0:00:00
+[19:23:53+0000] Downloading frozenlist-1.4.1-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl (272 kB)
+[19:23:53+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 272.3/272.3 kB 12.9 MB/s eta 0:00:00
+[19:23:53+0000] Downloading httpcore-1.0.2-py3-none-any.whl (76 kB)
+[19:23:53+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 76.9/76.9 kB 7.0 MB/s eta 0:00:00
+[19:23:53+0000] Downloading httpx-0.26.0-py3-none-any.whl (75 kB)
+[19:23:54+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 75.9/75.9 kB 20.7 MB/s eta 0:00:00
+[19:23:54+0000] Downloading hypercorn-0.16.0-py3-none-any.whl (59 kB)
+[19:23:54+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 59.9/59.9 kB 17.8 MB/s eta 0:00:00
+[19:23:54+0000] Downloading idna-3.6-py3-none-any.whl (61 kB)
+[19:23:54+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 61.6/61.6 kB 3.2 MB/s eta 0:00:00
+[19:23:54+0000] Downloading importlib_metadata-6.11.0-py3-none-any.whl (23 kB)
+[19:23:54+0000] Downloading Jinja2-3.1.3-py3-none-any.whl (133 kB)
+[19:23:54+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 133.2/133.2 kB 9.8 MB/s eta 0:00:00
+[19:23:54+0000] Downloading MarkupSafe-2.1.4-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (28 kB)
+[19:23:54+0000] Downloading msal-1.26.0-py2.py3-none-any.whl (99 kB)
+[19:23:54+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 99.0/99.0 kB 10.0 MB/s eta 0:00:00
+[19:23:54+0000] Downloading msal_extensions-1.1.0-py3-none-any.whl (19 kB)
+[19:23:54+0000] Downloading numpy-1.26.3-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (18.3 MB)
+[19:23:55+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 18.3/18.3 MB 9.5 MB/s eta 0:00:00
+[19:23:55+0000] Downloading opentelemetry_api-1.22.0-py3-none-any.whl (57 kB)
+[19:23:55+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 57.9/57.9 kB 3.1 MB/s eta 0:00:00
+[19:23:55+0000] Downloading opentelemetry_instrumentation-0.43b0-py3-none-any.whl (28 kB)
+[19:23:55+0000] Downloading opentelemetry_instrumentation_aiohttp_client-0.43b0-py3-none-any.whl (11 kB)
+[19:23:55+0000] Downloading opentelemetry_instrumentation_asgi-0.43b0-py3-none-any.whl (14 kB)
+[19:23:55+0000] Downloading opentelemetry_instrumentation_dbapi-0.43b0-py3-none-any.whl (10 kB)
+[19:23:55+0000] Downloading opentelemetry_instrumentation_django-0.43b0-py3-none-any.whl (18 kB)
+[19:23:55+0000] Downloading opentelemetry_instrumentation_fastapi-0.43b0-py3-none-any.whl (11 kB)
+[19:23:55+0000] Downloading opentelemetry_instrumentation_flask-0.43b0-py3-none-any.whl (14 kB)
+[19:23:55+0000] Downloading opentelemetry_instrumentation_httpx-0.43b0-py3-none-any.whl (12 kB)
+[19:23:55+0000] Downloading opentelemetry_instrumentation_psycopg2-0.43b0-py3-none-any.whl (10 kB)
+[19:23:55+0000] Downloading opentelemetry_instrumentation_requests-0.43b0-py3-none-any.whl (12 kB)
+[19:23:55+0000] Downloading opentelemetry_instrumentation_urllib-0.43b0-py3-none-any.whl (11 kB)
+[19:23:55+0000] Downloading opentelemetry_instrumentation_urllib3-0.43b0-py3-none-any.whl (11 kB)
+[19:23:56+0000] Downloading opentelemetry_instrumentation_wsgi-0.43b0-py3-none-any.whl (13 kB)
+[19:23:56+0000] Downloading opentelemetry_resource_detector_azure-0.1.3-py3-none-any.whl (10 kB)
+[19:23:56+0000] Downloading opentelemetry_sdk-1.22.0-py3-none-any.whl (105 kB)
+[19:23:56+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 105.6/105.6 kB 1.6 MB/s eta 0:00:00
+[19:23:56+0000] Downloading opentelemetry_semantic_conventions-0.43b0-py3-none-any.whl (36 kB)
+[19:23:56+0000] Downloading opentelemetry_util_http-0.43b0-py3-none-any.whl (6.9 kB)
+[19:23:56+0000] Downloading packaging-23.2-py3-none-any.whl (53 kB)
+[19:23:56+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 53.0/53.0 kB 1.9 MB/s eta 0:00:00
+[19:23:56+0000] Downloading pandas-2.2.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (13.0 MB)
+[19:23:58+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 13.0/13.0 MB 7.7 MB/s eta 0:00:00
+[19:23:58+0000] Downloading pandas_stubs-2.1.4.231227-py3-none-any.whl (153 kB)
+[19:23:58+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 153.6/153.6 kB 8.1 MB/s eta 0:00:00
+[19:23:58+0000] Downloading pillow-10.2.0-cp311-cp311-manylinux_2_28_x86_64.whl (4.5 MB)
+[19:23:58+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 4.5/4.5 MB 16.5 MB/s eta 0:00:00
+[19:23:58+0000] Downloading portalocker-2.8.2-py3-none-any.whl (17 kB)
+[19:23:58+0000] Downloading pyasn1-0.5.1-py2.py3-none-any.whl (84 kB)
+[19:23:58+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 84.9/84.9 kB 6.8 MB/s eta 0:00:00
+[19:23:58+0000] Downloading pydantic-2.6.0-py3-none-any.whl (394 kB)
+[19:23:58+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 394.2/394.2 kB 9.3 MB/s eta 0:00:00
+[19:23:59+0000] Downloading pydantic_core-2.16.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (2.2 MB)
+[19:23:59+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 2.2/2.2 MB 5.6 MB/s eta 0:00:00
+[19:23:59+0000] Downloading pytz-2023.4-py2.py3-none-any.whl (506 kB)
+[19:23:59+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 506.5/506.5 kB 7.1 MB/s eta 0:00:00
+[19:23:59+0000] Downloading quart-0.19.4-py3-none-any.whl (77 kB)
+[19:23:59+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 77.8/77.8 kB 6.5 MB/s eta 0:00:00
+[19:23:59+0000] Downloading quart_cors-0.7.0-py3-none-any.whl (8.0 kB)
+[19:23:59+0000] Downloading regex-2023.12.25-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (785 kB)
+[19:23:59+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 785.1/785.1 kB 13.7 MB/s eta 0:00:00
+[19:23:59+0000] Downloading requests-2.31.0-py3-none-any.whl (62 kB)
+[19:23:59+0000]    ━━━━━━━━━━━━━━━━━━━━━━━���━━━━━━━━━━━━━━━━ 62.6/62.6 kB 1.2 MB/s eta 0:00:00
+[19:23:59+0000] Downloading tenacity-8.2.3-py3-none-any.whl (24 kB)
+[19:23:59+0000] Downloading tiktoken-0.5.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (2.0 MB)
+[19:24:00+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 2.0/2.0 MB 15.3 MB/s eta 0:00:00
+[19:24:00+0000] Downloading tqdm-4.66.1-py3-none-any.whl (78 kB)
+[19:24:00+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 78.3/78.3 kB 6.9 MB/s eta 0:00:00
+[19:24:00+0000] Downloading types_Pillow-10.2.0.20240206-py3-none-any.whl (52 kB)
+[19:24:00+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 52.9/52.9 kB 1.1 MB/s eta 0:00:00
+[19:24:00+0000] Downloading types_pytz-2023.4.0.20240130-py3-none-any.whl (5.1 kB)
+[19:24:00+0000] Downloading typing_extensions-4.9.0-py3-none-any.whl (32 kB)
+[19:24:00+0000] Downloading tzdata-2023.4-py2.py3-none-any.whl (346 kB)
+[19:24:00+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 346.6/346.6 kB 11.7 MB/s eta 0:00:00
+[19:24:00+0000] Downloading urllib3-2.1.0-py3-none-any.whl (104 kB)
+[19:24:00+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 104.6/104.6 kB 1.2 MB/s eta 0:00:00
+[19:24:00+0000] Downloading uvicorn-0.27.0.post1-py3-none-any.whl (60 kB)
+[19:24:00+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 60.7/60.7 kB 5.7 MB/s eta 0:00:00
+[19:24:00+0000] Downloading werkzeug-3.0.1-py3-none-any.whl (226 kB)
+[19:24:00+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 226.7/226.7 kB 12.6 MB/s eta 0:00:00
+[19:24:00+0000] Downloading wrapt-1.16.0-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl (80 kB)
+[19:24:00+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 80.7/80.7 kB 8.0 MB/s eta 0:00:00
+[19:24:00+0000] Downloading yarl-1.9.4-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (328 kB)
+[19:24:00+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 328.1/328.1 kB 20.1 MB/s eta 0:00:00
+[19:24:00+0000] Downloading zipp-3.17.0-py3-none-any.whl (7.4 kB)
+[19:24:00+0000] Downloading openai-1.10.0-py3-none-any.whl (225 kB)
+[19:24:00+0000]    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 225.1/225.1 kB 11.7 MB/s eta 0:00:00
+[19:24:01+0000] Downloading PyJWT-2.8.0-py3-none-any.whl (22 kB)
+[19:24:07+0000] Installing collected packages: pytz, fixedint, azure-common, zipp, wrapt, urllib3, tzdata, typing-extensions, types-pytz, types-pillow, tqdm, tenacity, sniffio, six, regex, pyjwt, pycparser, pyasn1, priority, portalocker, pillow, packaging, opentelemetry-util-http, opentelemetry-semantic-conventions, oauthlib, numpy, multidict, markupsafe, itsdangerous, idna, hyperframe, hpack, h11, frozenlist, distro, click, charset-normalizer, certifi, blinker, attrs, asgiref, annotated-types, aiofiles, yarl, wsproto, werkzeug, uvicorn, rsa, requests, python-dateutil, pydantic-core, pandas-stubs, jinja2, isodate, importlib-metadata, httpcore, h2, ecdsa, deprecated, cffi, anyio, aiosignal, tiktoken, requests-oauthlib, python-jose, pydantic, pandas, opentelemetry-api, hypercorn, httpx, flask, cryptography, azure-core, aiohttp, quart, opentelemetry-sdk, opentelemetry-instrumentation, openai, msrest, azure-storage-blob, azure-search-documents, azure-keyvault-secrets, azure-core-tracing-opentelemetry, quart-cors, opentelemetry-resource-detector-azure, opentelemetry-instrumentation-wsgi, opentelemetry-instrumentation-urllib3, opentelemetry-instrumentation-urllib, opentelemetry-instrumentation-requests, opentelemetry-instrumentation-httpx, opentelemetry-instrumentation-dbapi, opentelemetry-instrumentation-asgi, opentelemetry-instrumentation-aiohttp-client, msal, azure-monitor-opentelemetry-exporter, opentelemetry-instrumentation-psycopg2, opentelemetry-instrumentation-flask, opentelemetry-instrumentation-fastapi, opentelemetry-instrumentation-django, msal-extensions, azure-monitor-opentelemetry, azure-identity
+[19:25:31+0000] Successfully installed aiofiles-23.2.1 aiohttp-3.9.3 aiosignal-1.3.1 annotated-types-0.6.0 anyio-4.2.0 asgiref-3.7.2 attrs-23.2.0 azure-common-1.1.28 azure-core-1.29.7 azure-core-tracing-opentelemetry-1.0.0b11 azure-identity-1.15.0 azure-keyvault-secrets-4.7.0 azure-monitor-opentelemetry-1.2.0 azure-monitor-opentelemetry-exporter-1.0.0b21 azure-search-documents-11.6.0b1 azure-storage-blob-12.19.0 blinker-1.7.0 certifi-2023.11.17 cffi-1.16.0 charset-normalizer-3.3.2 click-8.1.7 cryptography-42.0.1 deprecated-1.2.14 distro-1.9.0 ecdsa-0.18.0 fixedint-0.1.6 flask-3.0.1 frozenlist-1.4.1 h11-0.14.0 h2-4.1.0 hpack-4.0.0 httpcore-1.0.2 httpx-0.26.0 hypercorn-0.16.0 hyperframe-6.0.1 idna-3.6 importlib-metadata-6.11.0 isodate-0.6.1 itsdangerous-2.1.2 jinja2-3.1.3 markupsafe-2.1.4 msal-1.26.0 msal-extensions-1.1.0 msrest-0.7.1 multidict-6.0.4 numpy-1.26.3 oauthlib-3.2.2 openai-1.10.0 opentelemetry-api-1.22.0 opentelemetry-instrumentation-0.43b0 opentelemetry-instrumentation-aiohttp-client-0.43b0 opentelemetry-instrumentation-asgi-0.43b0 opentelemetry-instrumentation-dbapi-0.43b0 opentelemetry-instrumentation-django-0.43b0 opentelemetry-instrumentation-fastapi-0.43b0 opentelemetry-instrumentation-flask-0.43b0 opentelemetry-instrumentation-httpx-0.43b0 opentelemetry-instrumentation-psycopg2-0.43b0 opentelemetry-instrumentation-requests-0.43b0 opentelemetry-instrumentation-urllib-0.43b0 opentelemetry-instrumentation-urllib3-0.43b0 opentelemetry-instrumentation-wsgi-0.43b0 opentelemetry-resource-detector-azure-0.1.3 opentelemetry-sdk-1.22.0 opentelemetry-semantic-conventions-0.43b0 opentelemetry-util-http-0.43b0 packaging-23.2 pandas-2.2.0 pandas-stubs-2.1.4.231227 pillow-10.2.0 portalocker-2.8.2 priority-2.0.0 pyasn1-0.5.1 pycparser-2.21 pydantic-2.6.0 pydantic-core-2.16.1 pyjwt-2.8.0 python-dateutil-2.8.2 python-jose-3.3.0 pytz-2023.4 quart-0.19.4 quart-cors-0.7.0 regex-2023.12.25 requests-2.31.0 requests-oauthlib-1.3.1 rsa-4.9 six-1.16.0 sniffio-1.3.0 tenacity-8.2.3 tiktoken-0.5.2 tqdm-4.66.1 types-pillow-10.2.0.20240206 types-pytz-2023.4.0.20240130 typing-extensions-4.9.0 tzdata-2023.4 urllib3-2.1.0 uvicorn-0.27.0.post1 werkzeug-3.0.1 wrapt-1.16.0 wsproto-1.2.0 yarl-1.9.4 zipp-3.17.0
+
+[notice] A new release of pip is available: 23.2.1 -&gt; 24.0
+[notice] To update, run: pip install --upgrade pip
+Not a vso image, so not writing build commands
+Preparing output...
+
+Copying files to destination directory &#39;/tmp/_preCompressedDestinationDir&#39;...
+Done in 48 sec(s).
+Compressing content of directory &#39;/tmp/_preCompressedDestinationDir&#39;...
+Copied the compressed output to &#39;/home/site/wwwroot&#39;
+
+Removing existing manifest file
+Creating a manifest file...
+Manifest file created.
+Copying .ostype to manifest output directory.
+
+Done in 522 sec(s).</code></pre>
+</details>
+<p>Look for these important steps in the Oryx build:</p>
+<ul>
+<li><em>Detected following platforms: python: 3.11.7</em> That should
+match your runtime in the App Service configuration.</li>
+<li><em>Running pip install…</em> That should install all the
+requirements in your requirements.txt - if it didn’t find your
+requirements.txt, then you won’t see the packages installed.</li>
+</ul>
+<p>If you see all those steps in the Oryx build, then that’s a good sign
+that the build went well, and you can move on to checking the App
+Service logs.</p>
+<h2 id="checking-the-app-logs-for-errors">Checking the app logs for
+errors</h2>
+<p>Select <em>Advanced Tools</em> from the side nav:</p>
+<figure>
+<img src="images/screenshot_appservice_tools.png"
+alt="Advanced Tools" />
+<figcaption aria-hidden="true">Advanced Tools</figcaption>
+</figure>
+<p>Select <em>Go</em> to open the Kudu website.</p>
+<p>When the Kudu website loads, find the <em>Current Docker Logs</em>
+link and select <em>Download as zip</em> next to it:</p>
+<figure>
+<img src="images/screenshot_appservice_dockerlogs.png"
+alt="Screenshot of section with Download logs links" />
+<figcaption aria-hidden="true">Screenshot of section with Download logs
+links</figcaption>
+</figure>
+<p>In the downloaded zip file, find the filename that starts with the
+most recent date and ends with “_default_docker.log”:</p>
+<figure>
+<img src="images/screenshot_appservice_downloadedlogs.png"
+alt="Screenshot of downloaded logs" />
+<figcaption aria-hidden="true">Screenshot of downloaded
+logs</figcaption>
+</figure>
+<p>Open that file to see the full logs, with the most recent logs at the
+bottom.</p>
+<details>
+<summary>
+Here are the full logs for the app successfully starting:
+</summary>
+<pre class="plaintext"><code>
+2024-02-08T19:30:27.900249002Z    _____
+2024-02-08T19:30:27.900282702Z   /  _  \ __________ _________   ____
+2024-02-08T19:30:27.900288002Z  /  /_\  \\___   /  |  \_  __ \_/ __ \
+2024-02-08T19:30:27.900291902Z /    |    \/    /|  |  /|  | \/\  ___/
+2024-02-08T19:30:27.900295502Z \____|__  /_____ \____/ |__|    \___  &gt;
+2024-02-08T19:30:27.900299602Z         \/      \/                  \/
+2024-02-08T19:30:27.900303402Z A P P   S E R V I C E   O N   L I N U X
+2024-02-08T19:30:27.900307003Z
+2024-02-08T19:30:27.900310303Z Documentation: http://aka.ms/webapp-linux
+2024-02-08T19:30:27.900313903Z Python 3.11.4
+2024-02-08T19:30:27.900317303Z Note: Any data outside &#39;/home&#39; is not persisted
+2024-02-08T19:30:32.956710361Z Starting OpenBSD Secure Shell server: sshd.
+2024-02-08T19:30:33.441385332Z Site&#39;s appCommandLine: python3 -m gunicorn main:app
+2024-02-08T19:30:33.703536564Z Launching oryx with: create-script -appPath /home/site/wwwroot -output /opt/startup/startup.sh -virtualEnvName antenv -defaultApp /opt/defaultsite -userStartupCommand &#39;python3 -m gunicorn main:app&#39;
+2024-02-08T19:30:33.703598264Z Found build manifest file at &#39;/home/site/wwwroot/oryx-manifest.toml&#39;. Deserializing it...
+2024-02-08T19:30:33.703605164Z Build Operation ID: 7440a33100749a32
+2024-02-08T19:30:33.703609765Z Oryx Version: 0.2.20230707.1, Commit: 0bd28e69919b5e8beba451e8677e3345f0be8361, ReleaseTagName: 20230707.1
+2024-02-08T19:30:33.712124127Z Output is compressed. Extracting it...
+2024-02-08T19:30:33.712151827Z Extracting &#39;/home/site/wwwroot/output.tar.gz&#39; to directory &#39;/tmp/8dc28dad0e10acb&#39;...
+2024-02-08T19:31:08.047051747Z App path is set to &#39;/tmp/8dc28dad0e10acb&#39;
+2024-02-08T19:31:08.073259604Z Writing output script to &#39;/opt/startup/startup.sh&#39;
+2024-02-08T19:31:08.431803481Z Using packages from virtual environment antenv located at /tmp/8dc28dad0e10acb/antenv.
+2024-02-08T19:31:08.431842281Z Updated PYTHONPATH to &#39;/opt/startup/app_logs:/tmp/8dc28dad0e10acb/antenv/lib/python3.11/site-packages&#39;
+2024-02-08T19:31:11.043306496Z [2024-02-08 19:31:11 +0000] [75] [INFO] Starting gunicorn 20.1.0
+2024-02-08T19:31:11.060556234Z [2024-02-08 19:31:11 +0000] [75] [INFO] Listening at: http://0.0.0.0:8000 (75)
+2024-02-08T19:31:11.060586534Z [2024-02-08 19:31:11 +0000] [75] [INFO] Using worker: uvicorn.workers.UvicornWorker
+2024-02-08T19:31:11.069707155Z [2024-02-08 19:31:11 +0000] [76] [INFO] Booting worker with pid: 76
+2024-02-08T19:31:11.188073718Z [2024-02-08 19:31:11 +0000] [77] [INFO] Booting worker with pid: 77
+2024-02-08T19:31:11.415802023Z [2024-02-08 19:31:11 +0000] [78] [INFO] Booting worker with pid: 78
+2024-02-08T19:32:20.509338341Z [2024-02-08 19:32:20 +0000] [76] [INFO] Started server process [76]
+2024-02-08T19:32:20.521167526Z [2024-02-08 19:32:20 +0000] [77] [INFO] Started server process [77]
+2024-02-08T19:32:20.521189626Z [2024-02-08 19:32:20 +0000] [77] [INFO] Waiting for application startup.
+2024-02-08T19:32:20.521207626Z [2024-02-08 19:32:20 +0000] [78] [INFO] Started server process [78]
+2024-02-08T19:32:20.521212726Z [2024-02-08 19:32:20 +0000] [78] [INFO] Waiting for application startup.
+2024-02-08T19:32:20.521217126Z [2024-02-08 19:32:20 +0000] [76] [INFO] Waiting for application startup.
+2024-02-08T19:32:20.726894213Z [2024-02-08 19:32:20 +0000] [76] [INFO] Application startup complete.
+2024-02-08T19:32:20.726936214Z [2024-02-08 19:32:20 +0000] [78] [INFO] Application startup complete.
+2024-02-08T19:32:20.726942614Z [2024-02-08 19:32:20 +0000] [77] [INFO] Application startup complete.</code></pre>
+</details>
+<p>A few notable logs:</p>
+<ul>
+<li><code>2024-02-08T19:30:33.441385332Z Site's appCommandLine: python3 -m gunicorn main:app</code>
+This log indicates that App Service was correctly configured with a
+custom startup command to run the app.</li>
+<li><code>[2024-02-08 19:31:11 +0000] [75] [INFO] Starting gunicorn 20.1.0</code>
+That’s the start of the gunicorn server serving the app.</li>
+<li><code>2024-02-08T19:32:20.726942614Z [2024-02-08 19:32:20 +0000] [77] [INFO] Application startup complete.</code>
+At this point, the app has started successfully.</li>
+</ul>
+<p>If you do not see any errors in those logs, then the app should be
+running successfully. If you do see errors, then try looking in Azure
+Monitor.</p>
+<h2 id="checking-azure-monitor-for-errors">Checking Azure Monitor for
+errors</h2>
+<p>By default, deployed apps use Application Insights to trace and log
+errors. (If you explicitly opted out of Application Insights, then you
+won’t have this feature.)</p>
+<p>In the Azure Portal, navigate to the Application Insights for your
+app.</p>
+<p>To see any exceptions and server errors, navigate to the
+<em>Investigate -&gt; Failures</em> blade and browse through the
+exceptions.</p>
+<figure>
+<img src="images/screenshot_appinsights_failures.png"
+alt="Screenshot of Application Insights Failures tab" />
+<figcaption aria-hidden="true">Screenshot of Application Insights
+Failures tab</figcaption>
+</figure>
+<h2 id="configuring-log-levels">Configuring log levels</h2>
+<p>By default, the deployed app only logs messages from packages with a
+level of <code>WARNING</code> or higher, but logs all messages from the
+app with a level of <code>INFO</code> or higher.</p>
+<p>These lines of code in <code>app/backend/app.py</code> configure the
+logging level:</p>
+<div class="sourceCode" id="cb3"><pre
+class="sourceCode python"><code class="sourceCode python"><span id="cb3-1"><a href="#cb3-1" aria-hidden="true" tabindex="-1"></a><span class="co"># Set root level to </span><span class="al">WARNING</span><span class="co"> to avoid seeing overly verbose logs from SDKS</span></span>
+<span id="cb3-2"><a href="#cb3-2" aria-hidden="true" tabindex="-1"></a>logging.basicConfig(level<span class="op">=</span>logging.WARNING)</span>
+<span id="cb3-3"><a href="#cb3-3" aria-hidden="true" tabindex="-1"></a><span class="co"># Set the app logger level to INFO by default</span></span>
+<span id="cb3-4"><a href="#cb3-4" aria-hidden="true" tabindex="-1"></a>default_level <span class="op">=</span> <span class="st">&quot;INFO&quot;</span></span>
+<span id="cb3-5"><a href="#cb3-5" aria-hidden="true" tabindex="-1"></a>app.logger.setLevel(os.getenv(<span class="st">&quot;APP_LOG_LEVEL&quot;</span>, default_level))</span></code></pre></div>
+<p>To change the default level, either change <code>default_level</code>
+or set the <code>APP_LOG_LEVEL</code> environment variable to one of the
+<a
+href="https://docs.python.org/3/library/logging.html#logging-levels">allowed
+log levels</a>: <code>DEBUG</code>, <code>INFO</code>,
+<code>WARNING</code>, <code>ERROR</code>, <code>CRITICAL</code>.</p>
+<p>If you need to log in a route handler, use the the global variable
+<code>current_app</code>’s logger:</p>
+<div class="sourceCode" id="cb4"><pre
+class="sourceCode python"><code class="sourceCode python"><span id="cb4-1"><a href="#cb4-1" aria-hidden="true" tabindex="-1"></a><span class="cf">async</span> <span class="kw">def</span> chat():</span>
+<span id="cb4-2"><a href="#cb4-2" aria-hidden="true" tabindex="-1"></a>    current_app.logger.info(<span class="st">&quot;Received /chat request&quot;</span>)</span></code></pre></div>
+<p>Otherwise, use the <code>logging</code> module’s root logger:</p>
+<div class="sourceCode" id="cb5"><pre
+class="sourceCode python"><code class="sourceCode python"><span id="cb5-1"><a href="#cb5-1" aria-hidden="true" tabindex="-1"></a>logging.info(<span class="st">&quot;System message: </span><span class="sc">%s</span><span class="st">&quot;</span>, system_message)</span></code></pre></div>
+<p>If you’re having troubles finding the logs in App Service, read the
+section above on <a href="#checking-the-app-logs-for-errors">checking
+app logs</a> or watch <a
+href="https://www.youtube.com/watch?v=f0-aYuvws54">this video about
+viewing App Service logs</a>.</p>
diff --git a/data/azd.html b/data/azd.html
new file mode 100644
index 0000000000..9a98ed915e
--- /dev/null
+++ b/data/azd.html
@@ -0,0 +1,77 @@
+<h1 id="rag-chat-deploying-with-the-azure-developer-cli">RAG chat:
+Deploying with the Azure Developer CLI</h1>
+<p>This guide includes advanced topics that are not necessary for a
+basic deployment. If you are new to the project, please consult the main
+<a href="../README.md#deploying">README</a> for steps on deploying the
+project.</p>
+<p><a href="https://www.youtube.com/watch?v=mDFZdmn7nhk">📺 Watch:
+Deployment of your chat app</a></p>
+<ul>
+<li><a href="#how-does-azd-up-work">How does <code>azd up</code>
+work?</a></li>
+<li><a href="#configuring-continuous-deployment">Configuring continuous
+deployment</a>
+<ul>
+<li><a href="#github-actions">GitHub actions</a></li>
+<li><a href="#azure-devops">Azure DevOps</a></li>
+</ul></li>
+</ul>
+<h2 id="how-does-azd-up-work">How does <code>azd up</code> work?</h2>
+<p>The <code>azd up</code> command comes from the <a
+href="https://learn.microsoft.com/azure/developer/azure-developer-cli/overview">Azure
+Developer CLI</a>, and takes care of both provisioning the Azure
+resources and deploying code to the selected Azure hosts.</p>
+<p>The <code>azd up</code> command uses the <code>azure.yaml</code> file
+combined with the infrastructure-as-code <code>.bicep</code> files in
+the <code>infra/</code> folder. The <code>azure.yaml</code> file for
+this project declares several “hooks” for the prepackage step and
+postprovision steps. The <code>up</code> command first runs the
+<code>prepackage</code> hook which installs Node dependencies and builds
+the React.JS-based JavaScript files. It then packages all the code (both
+frontend and backend) into a zip file which it will deploy later.</p>
+<p>Next, it provisions the resources based on <code>main.bicep</code>
+and <code>main.parameters.json</code>. At that point, since there is no
+default value for the OpenAI resource location, it asks you to pick a
+location from a short list of available regions. Then it will send
+requests to Azure to provision all the required resources. With
+everything provisioned, it runs the <code>postprovision</code> hook to
+process the local data and add it to an Azure AI Search index.</p>
+<p>Finally, it looks at <code>azure.yaml</code> to determine the Azure
+host and uploads the zip to Azure App Service. The <code>azd up</code>
+command is now complete, but it may take another 5-10 minutes for the
+App Service app to be fully available and working, especially for the
+initial deploy.</p>
+<p>Related commands are <code>azd provision</code> for just provisioning
+(if infra files change) and <code>azd deploy</code> for just deploying
+updated app code.</p>
+<h2 id="configuring-continuous-deployment">Configuring continuous
+deployment</h2>
+<p>This repository includes both a GitHub Actions workflow and an Azure
+DevOps pipeline for continuous deployment with every push to
+<code>main</code>. The GitHub Actions workflow is the default, but you
+can switch to Azure DevOps if you prefer.</p>
+<p>More details are available in <a
+href="https://learn.microsoft.com/azure/developer/azure-developer-cli/configure-devops-pipeline?tabs=GitHub">Learn.com:
+Configure a pipeline and push updates</a></p>
+<h3 id="github-actions">GitHub actions</h3>
+<p>After you have deployed the app once with <code>azd up</code>, you
+can enable continuous deployment with GitHub Actions.</p>
+<p>Run this command to set up a Service Principal account for CI
+deployment and to store your <code>azd</code> environment variables in
+GitHub Actions secrets:</p>
+<pre class="shell"><code>azd pipeline config</code></pre>
+<p>You can trigger the “Deploy” workflow manually from your GitHub
+actions, or wait for the next push to main.</p>
+<p>If you change your <code>azd</code> environment variables at any time
+(via <code>azd env set</code> or as a result of provisioning), re-run
+that command in order to update the GitHub Actions secrets.</p>
+<h3 id="azure-devops">Azure DevOps</h3>
+<p>After you have deployed the app once with <code>azd up</code>, you
+can enable continuous deployment with Azure DevOps.</p>
+<p>Run this command to set up a Service Principal account for CI
+deployment and to store your <code>azd</code> environment variables in
+GitHub Actions secrets:</p>
+<pre class="shell"><code>azd pipeline config --provider azdo</code></pre>
+<p>If you change your <code>azd</code> environment variables at any time
+(via <code>azd env set</code> or as a result of provisioning), re-run
+that command in order to update the GitHub Actions secrets.</p>
diff --git a/data/azure_app_service.html b/data/azure_app_service.html
new file mode 100644
index 0000000000..706465d4be
--- /dev/null
+++ b/data/azure_app_service.html
@@ -0,0 +1,45 @@
+<h1 id="rag-chat-deploying-on-azure-app-service">RAG chat: Deploying on
+Azure App Service</h1>
+<p>Due to <a href="https://github.com/Azure/azure-dev/issues/2736">a
+limitation</a> of the Azure Developer CLI (<code>azd</code>), there can
+be only one host option in the <a href="../azure.yaml">azure.yaml</a>
+file. By default, <code>host: containerapp</code> is used and
+<code>host: appservice</code> is commented out.</p>
+<p>To deploy to Azure App Service, please follow the following
+steps:</p>
+<ol type="1">
+<li><p>Comment out <code>host: containerapp</code> and uncomment
+<code>host: appservice</code> in the <a
+href="../azure.yaml">azure.yaml</a> file.</p></li>
+<li><p>Login to your Azure account:</p>
+<div class="sourceCode" id="cb1"><pre
+class="sourceCode bash"><code class="sourceCode bash"><span id="cb1-1"><a href="#cb1-1" aria-hidden="true" tabindex="-1"></a><span class="ex">azd</span> auth login</span></code></pre></div></li>
+<li><p>Create a new <code>azd</code> environment to store the deployment
+parameters:</p>
+<div class="sourceCode" id="cb2"><pre
+class="sourceCode bash"><code class="sourceCode bash"><span id="cb2-1"><a href="#cb2-1" aria-hidden="true" tabindex="-1"></a><span class="ex">azd</span> env new</span></code></pre></div>
+<p>Enter a name that will be used for the resource group. This will
+create a new folder in the <code>.azure</code> folder, and set it as the
+active environment for any calls to <code>azd</code> going
+forward.</p></li>
+<li><p>Set the deployment target to <code>appservice</code>:</p>
+<div class="sourceCode" id="cb3"><pre
+class="sourceCode bash"><code class="sourceCode bash"><span id="cb3-1"><a href="#cb3-1" aria-hidden="true" tabindex="-1"></a><span class="ex">azd</span> env set DEPLOYMENT_TARGET appservice</span></code></pre></div></li>
+<li><p>(Optional) This is the point where you can customize the
+deployment by setting other <code>azd</code> environment variables, in
+order to <a href="deploy_existing.md">use existing resources</a>, <a
+href="deploy_features.md">enable optional features (such as auth or
+vision)</a>, or <a href="deploy_lowcost.md">deploy to free
+tiers</a>.</p></li>
+<li><p>Provision the resources and deploy the code:</p>
+<div class="sourceCode" id="cb4"><pre
+class="sourceCode bash"><code class="sourceCode bash"><span id="cb4-1"><a href="#cb4-1" aria-hidden="true" tabindex="-1"></a><span class="ex">azd</span> up</span></code></pre></div>
+<p>This will provision Azure resources and deploy this sample to those
+resources, including building the search index based on the files found
+in the <code>./data</code> folder.</p>
+<p><strong>Important</strong>: Beware that the resources created by this
+command will incur immediate costs, primarily from the AI Search
+resource. These resources may accrue costs even if you interrupt the
+command before it is fully executed. You can run <code>azd down</code>
+or delete the resources manually to avoid unnecessary spending.</p></li>
+</ol>
diff --git a/data/azure_container_apps.html b/data/azure_container_apps.html
new file mode 100644
index 0000000000..f3294ec2a2
--- /dev/null
+++ b/data/azure_container_apps.html
@@ -0,0 +1,62 @@
+<h1 id="rag-chat-deploying-on-azure-container-apps">RAG chat: Deploying
+on Azure Container Apps</h1>
+<p>Due to <a href="https://github.com/Azure/azure-dev/issues/2736">a
+limitation</a> of the Azure Developer CLI (<code>azd</code>), there can
+be only one host option in the <a href="../azure.yaml">azure.yaml</a>
+file. By default, <code>host: containerapp</code> is used and
+<code>host: appservice</code> is commented out.</p>
+<p>However, if you have an older version of the repo, you may need to
+follow these steps to deploy to Container Apps instead, or you can stick
+with Azure App Service.</p>
+<p>To deploy to Azure Container Apps, please follow the following
+steps:</p>
+<ol type="1">
+<li><p>Comment out <code>host: appservice</code> and uncomment
+<code>host: containerapp</code> in the <a
+href="../azure.yaml">azure.yaml</a> file.</p></li>
+<li><p>Login to your Azure account:</p>
+<div class="sourceCode" id="cb1"><pre
+class="sourceCode bash"><code class="sourceCode bash"><span id="cb1-1"><a href="#cb1-1" aria-hidden="true" tabindex="-1"></a><span class="ex">azd</span> auth login</span></code></pre></div></li>
+<li><p>Create a new <code>azd</code> environment to store the deployment
+parameters:</p>
+<div class="sourceCode" id="cb2"><pre
+class="sourceCode bash"><code class="sourceCode bash"><span id="cb2-1"><a href="#cb2-1" aria-hidden="true" tabindex="-1"></a><span class="ex">azd</span> env new</span></code></pre></div>
+<p>Enter a name that will be used for the resource group. This will
+create a new folder in the <code>.azure</code> folder, and set it as the
+active environment for any calls to <code>azd</code> going
+forward.</p></li>
+<li><p>Set the deployment target to <code>containerapps</code>:</p>
+<div class="sourceCode" id="cb3"><pre
+class="sourceCode bash"><code class="sourceCode bash"><span id="cb3-1"><a href="#cb3-1" aria-hidden="true" tabindex="-1"></a><span class="ex">azd</span> env set DEPLOYMENT_TARGET containerapps</span></code></pre></div></li>
+<li><p>(Optional) This is the point where you can customize the
+deployment by setting other `azd1 environment variables, in order to <a
+href="docs/deploy_existing.md">use existing resources</a>, <a
+href="docs/deploy_features.md">enable optional features (such as auth or
+vision)</a>, or <a href="docs/deploy_lowcost.md">deploy to free
+tiers</a>.</p></li>
+<li><p>Provision the resources and deploy the code:</p>
+<div class="sourceCode" id="cb4"><pre
+class="sourceCode bash"><code class="sourceCode bash"><span id="cb4-1"><a href="#cb4-1" aria-hidden="true" tabindex="-1"></a><span class="ex">azd</span> up</span></code></pre></div>
+<p>This will provision Azure resources and deploy this sample to those
+resources, including building the search index based on the files found
+in the <code>./data</code> folder.</p>
+<p><strong>Important</strong>: Beware that the resources created by this
+command will incur immediate costs, primarily from the AI Search
+resource. These resources may accrue costs even if you interrupt the
+command before it is fully executed. You can run <code>azd down</code>
+or delete the resources manually to avoid unnecessary spending.</p></li>
+</ol>
+<h2 id="customizing-workload-profile">Customizing Workload Profile</h2>
+<p>The default workload profile is Consumption. If you want to use a
+dedicated workload profile like D4, please run:</p>
+<div class="sourceCode" id="cb5"><pre
+class="sourceCode bash"><code class="sourceCode bash"><span id="cb5-1"><a href="#cb5-1" aria-hidden="true" tabindex="-1"></a><span class="ex">azd</span> env AZURE_CONTAINER_APPS_WORKLOAD_PROFILE D4</span></code></pre></div>
+<p>For a full list of workload profiles, please check <a
+href="https://learn.microsoft.com/azure/container-apps/workload-profiles-overview#profile-types">here</a>.
+Please note dedicated workload profiles have a different billing model
+than Consumption plan. Please check <a
+href="https://learn.microsoft.com/azure/container-apps/billing">here</a>
+for details.</p>
+<h2 id="private-endpoints">Private endpoints</h2>
+<p>Private endpoints is still in private preview for Azure Container
+Apps and not supported for now.</p>
diff --git a/data/customization.html b/data/customization.html
new file mode 100644
index 0000000000..0bf2389b6b
--- /dev/null
+++ b/data/customization.html
@@ -0,0 +1,309 @@
+<h1 id="rag-chat-customizing-the-chat-app">RAG chat: Customizing the
+chat app</h1>
+<p><a href="https://www.youtube.com/watch?v=D3slfMqydHc">📺 Watch: (RAG
+Deep Dive series) Customizing the app</a></p>
+<p>This guide provides more details for customizing the RAG chat
+app.</p>
+<ul>
+<li><a href="#using-your-own-data">Using your own data</a></li>
+<li><a href="#customizing-the-ui">Customizing the UI</a></li>
+<li><a href="#customizing-the-backend">Customizing the backend</a>
+<ul>
+<li><a href="#chatask-tabs">Chat/Ask tabs</a>
+<ul>
+<li><a href="#chat-approach">Chat approach</a>
+<ul>
+<li><a href="#chat-with-vision">Chat with vision</a></li>
+</ul></li>
+<li><a href="#ask-tab">Ask tab</a>
+<ul>
+<li><a href="#ask-with-vision">Ask with vision</a></li>
+</ul></li>
+</ul></li>
+</ul></li>
+<li><a href="#improving-answer-quality">Improving answer quality</a>
+<ul>
+<li><a href="#identify-the-problem-point">Identify the problem
+point</a></li>
+<li><a href="#improving-openai-chatcompletion-results">Improving OpenAI
+ChatCompletion results</a></li>
+<li><a href="#improving-azure-ai-search-results">Improving Azure AI
+Search results</a></li>
+<li><a href="#evaluating-answer-quality">Evaluating answer
+quality</a></li>
+</ul></li>
+</ul>
+<h2 id="using-your-own-data">Using your own data</h2>
+<p>The Chat App is designed to work with any PDF documents. The sample
+data is provided to help you get started quickly, but you can easily
+replace it with your own data. You’ll want to first remove all the
+existing data, then add your own. See the <a
+href="data_ingestion.md">data ingestion guide</a> for more details.</p>
+<h2 id="customizing-the-ui">Customizing the UI</h2>
+<p>The frontend is built using <a href="https://reactjs.org/">React</a>
+and <a href="https://react.fluentui.dev/">Fluent UI components</a>. The
+frontend components are stored in the <code>app/frontend/src</code>
+folder. To modify the page title, header text, example questions, and
+other UI elements, you can customize the
+<code>app/frontend/src/locales/{en/es/fr/jp/it}/translation.json</code>
+file for different languages(English is the default). The primary
+strings and labels used throughout the application are defined within
+these files.</p>
+<h2 id="customizing-the-backend">Customizing the backend</h2>
+<p>The backend is built using <a
+href="https://quart.palletsprojects.com/">Quart</a>, a Python framework
+for asynchronous web applications. The backend code is stored in the
+<code>app/backend</code> folder. The frontend and backend communicate
+using the <a href="https://aka.ms/chatprotocol">AI Chat HTTP
+Protocol</a>.</p>
+<h3 id="chatask-tabs">Chat/Ask tabs</h3>
+<p>Typically, the primary backend code you’ll want to customize is the
+<code>app/backend/approaches</code> folder, which contains the classes
+powering the Chat and Ask tabs. Each class uses a different RAG
+(Retrieval Augmented Generation) approach, which include system messages
+that should be changed to match your data</p>
+<h4 id="chat-approach">Chat approach</h4>
+<p>The chat tab uses the approach programmed in <a
+href="https://github.com/Azure-Samples/azure-search-openai-demo/blob/main/app/backend/approaches/chatreadretrieveread.py">chatreadretrieveread.py</a>.</p>
+<ol type="1">
+<li>It calls the OpenAI ChatCompletion API to turn the user question
+into a good search query, using the prompt and tools from <a
+href="https://github.com/Azure-Samples/azure-search-openai-demo/blob/main/app/backend/approaches/prompts/chat_query_rewrite.prompty">chat_query_rewrite.prompty</a>.</li>
+<li>It queries Azure AI Search for search results for that query
+(optionally using the vector embeddings for that query).</li>
+<li>It then calls the OpenAI ChatCompletion API to answer the question
+based on the sources, using the prompt from <a
+href="https://github.com/Azure-Samples/azure-search-openai-demo/blob/main/app/backend/approaches/prompts/chat_answer_question.prompty">chat_answer_question.prompty</a>.
+That call includes the past message history as well (or as many messages
+fit inside the model’s token limit).</li>
+</ol>
+<p>The prompts are currently tailored to the sample data since they
+start with “Assistant helps the company employees with their healthcare
+plan questions, and questions about the employee handbook.” Modify the
+<a
+href="https://github.com/Azure-Samples/azure-search-openai-demo/blob/main/app/backend/approaches/prompts/chat_query_rewrite.prompty">chat_query_rewrite.prompty</a>
+and <a
+href="https://github.com/Azure-Samples/azure-search-openai-demo/blob/main/app/backend/approaches/prompts/chat_answer_question.prompty">chat_answer_question.prompty</a>
+prompts to match your data.</p>
+<h5 id="chat-with-vision">Chat with vision</h5>
+<p>If you followed the instructions in <a
+href="gpt4v.md">docs/gpt4v.md</a> to enable a GPT Vision model and then
+select “Use GPT vision model”, then the chat tab will use the
+<code>chatreadretrievereadvision.py</code> approach instead. This
+approach is similar to the <code>chatreadretrieveread.py</code>
+approach, with a few differences:</p>
+<ol type="1">
+<li>Step 1 is the same as before, except it uses the GPT-4 Vision model
+instead of the default GPT-3.5 model.</li>
+<li>For this step, it also calculates a vector embedding for the user
+question using <a
+href="https://learn.microsoft.com/azure/ai-services/computer-vision/how-to/image-retrieval#call-the-vectorize-text-api">the
+Computer Vision vectorize text API</a>, and passes that to the Azure AI
+Search to compare against the <code>imageEmbeddings</code> fields in the
+indexed documents. For each matching document, it downloads the image
+blob and converts it to a base 64 encoding.</li>
+<li>When it combines the search results and user question, it includes
+the base 64 encoded images, and sends along both the text and images to
+the GPT4 Vision model (similar to this <a
+href="https://platform.openai.com/docs/guides/vision/quick-start">documentation
+example</a>). The model generates a response that includes citations to
+the images, and the UI renders the base64 encoded images when a citation
+is clicked.</li>
+</ol>
+<p>The prompt for step 2 is currently tailored to the sample data since
+it starts with “You are an intelligent assistant helping analyze the
+Annual Financial Report of Contoso Ltd.”. Modify the <a
+href="https://github.com/Azure-Samples/azure-search-openai-demo/blob/main/app/backend/approaches/prompts/chat_answer_question_vision.prompty">chat_answer_question_vision.prompty</a>
+prompt to match your data.</p>
+<h4 id="ask-tab">Ask tab</h4>
+<p>The ask tab uses the approach programmed in <a
+href="https://github.com/Azure-Samples/azure-search-openai-demo/blob/main/app/backend/approaches/retrievethenread.py">retrievethenread.py</a>.</p>
+<ol type="1">
+<li>It queries Azure AI Search for search results for the user question
+(optionally using the vector embeddings for that question).</li>
+<li>It then combines the search results and user question, and calls the
+OpenAI ChatCompletion API to answer the question based on the sources,
+using the prompt from <a
+href="https://github.com/Azure-Samples/azure-search-openai-demo/blob/main/app/backend/approaches/prompts/ask_answer_question.prompty">ask_answer_question.prompty</a>.</li>
+</ol>
+<p>The prompt for step 2 is currently tailored to the sample data since
+it starts with “You are an intelligent assistant helping Contoso Inc
+employees with their healthcare plan questions and employee handbook
+questions.” Modify <a
+href="https://github.com/Azure-Samples/azure-search-openai-demo/blob/main/app/backend/approaches/prompts/ask_answer_question.prompty">ask_answer_question.prompty</a>
+to match your data.</p>
+<h4 id="ask-with-vision">Ask with vision</h4>
+<p>If you followed the instructions in <a
+href="gpt4v.md">docs/gpt4v.md</a> to enable the GPT-4 Vision model and
+then select “Use GPT vision model”, then the ask tab will use the
+<code>retrievethenreadvision.py</code> approach instead. This approach
+is similar to the <code>retrievethenread.py</code> approach, with a few
+differences:</p>
+<ol type="1">
+<li>For this step, it also calculates a vector embedding for the user
+question using <a
+href="https://learn.microsoft.com/azure/ai-services/computer-vision/how-to/image-retrieval#call-the-vectorize-text-api">the
+Computer Vision vectorize text API</a>, and passes that to the Azure AI
+Search to compare against the <code>imageEmbeddings</code> fields in the
+indexed documents. For each matching document, it downloads the image
+blob and converts it to a base 64 encoding.</li>
+<li>When it combines the search results and user question, it includes
+the base 64 encoded images, and sends along both the text and images to
+the GPT4 Vision model (similar to this <a
+href="https://platform.openai.com/docs/guides/vision/quick-start">documentation
+example</a>). The model generates a response that includes citations to
+the images, and the UI renders the base64 encoded images when a citation
+is clicked.</li>
+</ol>
+<p>The prompt for step 2 is currently tailored to the sample data since
+it starts with “You are an intelligent assistant helping analyze the
+Annual Financial Report of Contoso Ltd”. Modify the <a
+href="https://github.com/Azure-Samples/azure-search-openai-demo/blob/main/app/backend/approaches/prompts/ask_answer_question_vision.prompty">ask_answer_question_vision.prompty</a>
+prompt to match your data.</p>
+<h4 id="making-settings-overrides-permanent">Making settings overrides
+permanent</h4>
+<p>The UI provides a “Developer Settings” menu for customizing the
+approaches, like disabling semantic ranker or using vector search. Those
+settings are passed in the “context” field of the request to the
+backend, and are not saved permanently. However, if you find a setting
+that you do want to make permanent, there are two approaches:</p>
+<ol type="1">
+<li><p>Change the defaults in the frontend. You’ll find the defaults in
+<code>Chat.tsx</code> and <code>Ask.tsx</code>. For example, this line
+of code sets the default retrieval mode to Hybrid:</p>
+<div class="sourceCode" id="cb1"><pre
+class="sourceCode typescript"><code class="sourceCode typescript"><span id="cb1-1"><a href="#cb1-1" aria-hidden="true" tabindex="-1"></a><span class="kw">const</span> [retrievalMode<span class="op">,</span> setRetrievalMode] <span class="op">=</span> <span class="fu">useState</span><span class="op">&lt;</span>RetrievalMode<span class="op">&gt;</span>(RetrievalMode<span class="op">.</span><span class="at">Hybrid</span>)<span class="op">;</span></span></code></pre></div>
+<p>You can change the default to Text by changing the code to:</p>
+<div class="sourceCode" id="cb2"><pre
+class="sourceCode typescript"><code class="sourceCode typescript"><span id="cb2-1"><a href="#cb2-1" aria-hidden="true" tabindex="-1"></a><span class="kw">const</span> [retrievalMode<span class="op">,</span> setRetrievalMode] <span class="op">=</span> <span class="fu">useState</span><span class="op">&lt;</span>RetrievalMode<span class="op">&gt;</span>(RetrievalMode<span class="op">.</span><span class="at">Text</span>)<span class="op">;</span></span></code></pre></div></li>
+<li><p>Change the overrides in the backend. Each of the approaches has a
+<code>run</code> method that takes a <code>context</code> parameter, and
+the first line of code extracts the overrides from that
+<code>context</code>. That’s where you can override any of the settings.
+For example, to change the retrieval mode to text:</p>
+<div class="sourceCode" id="cb3"><pre
+class="sourceCode python"><code class="sourceCode python"><span id="cb3-1"><a href="#cb3-1" aria-hidden="true" tabindex="-1"></a>overrides <span class="op">=</span> context.get(<span class="st">&quot;overrides&quot;</span>, {})</span>
+<span id="cb3-2"><a href="#cb3-2" aria-hidden="true" tabindex="-1"></a>overrides[<span class="st">&quot;retrieval_mode&quot;</span>] <span class="op">=</span> <span class="st">&quot;text&quot;</span></span></code></pre></div>
+<p>By changing the setting on the backend, you can safely remove the
+Developer Settings UI from the frontend, if you don’t wish to expose
+that to your users.</p></li>
+</ol>
+<h2 id="improving-answer-quality">Improving answer quality</h2>
+<p>Once you are running the chat app on your own data and with your own
+tailored system prompt, the next step is to test the app with questions
+and note the quality of the answers. If you notice any answers that
+aren’t as good as you’d like, here’s a process for improving them.</p>
+<h3 id="identify-the-problem-point">Identify the problem point</h3>
+<p>The first step is to identify where the problem is occurring. For
+example, if using the Chat tab, the problem could be:</p>
+<ol type="1">
+<li>OpenAI ChatCompletion API is not generating a good search query
+based on the user question</li>
+<li>Azure AI Search is not returning good search results for the
+query</li>
+<li>OpenAI ChatCompletion API is not generating a good answer based on
+the search results and user question</li>
+</ol>
+<p>You can look at the “Thought process” tab in the chat app to see each
+of those steps, and determine which one is the problem.</p>
+<h3 id="improving-openai-chatcompletion-results">Improving OpenAI
+ChatCompletion results</h3>
+<p>If the problem is with the ChatCompletion API calls (steps 1 or 3
+above), you can try changing the relevant prompt.</p>
+<p>Once you’ve changed the prompt, make sure you ask the same question
+multiple times to see if the overall quality has improved, and <a
+href="#evaluating-answer-quality">run an evaluation</a> when you’re
+satisfied with the changes. The ChatCompletion API can yield different
+results every time, even for a temperature of 0.0, but especially for a
+higher temperature than that (like our default of 0.7 for step 3).</p>
+<p>You can also try changing the ChatCompletion parameters, like
+temperature, to see if that improves results for your domain.</p>
+<h3 id="improving-azure-ai-search-results">Improving Azure AI Search
+results</h3>
+<p>If the problem is with Azure AI Search (step 2 above), the first step
+is to check what search parameters you’re using. Generally, the best
+results are found with hybrid search (text + vectors) plus the
+additional semantic re-ranking step, and that’s what we’ve enabled by
+default. There may be some domains where that combination isn’t optimal,
+however. Check out this blog post which <a
+href="https://techcommunity.microsoft.com/blog/azure-ai-services-blog/azure-ai-search-outperforming-vector-search-with-hybrid-retrieval-and-ranking-ca/3929167">evaluates
+AI search strategies</a> for a better understanding of the differences,
+or watch this <a href="https://www.youtube.com/watch?v=ugJy9QkgLYg">RAG
+Deep Dive video on AI Search</a>.</p>
+<h4 id="configuring-parameters-in-the-app">Configuring parameters in the
+app</h4>
+<p>You can change many of the search parameters in the “Developer
+settings” in the frontend and see if results improve for your queries.
+The most relevant options:</p>
+<figure>
+<img src="images/screenshot_searchoptions.png"
+alt="Screenshot of search options in developer settings" />
+<figcaption aria-hidden="true">Screenshot of search options in developer
+settings</figcaption>
+</figure>
+<h4 id="configuring-parameters-in-the-azure-portal">Configuring
+parameters in the Azure Portal</h4>
+<p>You may find it easier to experiment with search options with the
+index explorer in the Azure Portal. Open up the Azure AI Search
+resource, select the Indexes tab, and select the index there.</p>
+<p>Then use the JSON view of the search explorer, and make sure you
+specify the same options you’re using in the app. For example, this
+query represents a search with semantic ranker configured:</p>
+<div class="sourceCode" id="cb4"><pre
+class="sourceCode json"><code class="sourceCode json"><span id="cb4-1"><a href="#cb4-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>
+<span id="cb4-2"><a href="#cb4-2" aria-hidden="true" tabindex="-1"></a>  <span class="dt">&quot;search&quot;</span><span class="fu">:</span> <span class="st">&quot;eye exams&quot;</span><span class="fu">,</span></span>
+<span id="cb4-3"><a href="#cb4-3" aria-hidden="true" tabindex="-1"></a>  <span class="dt">&quot;queryType&quot;</span><span class="fu">:</span> <span class="st">&quot;semantic&quot;</span><span class="fu">,</span></span>
+<span id="cb4-4"><a href="#cb4-4" aria-hidden="true" tabindex="-1"></a>  <span class="dt">&quot;semanticConfiguration&quot;</span><span class="fu">:</span> <span class="st">&quot;default&quot;</span><span class="fu">,</span></span>
+<span id="cb4-5"><a href="#cb4-5" aria-hidden="true" tabindex="-1"></a>  <span class="dt">&quot;queryLanguage&quot;</span><span class="fu">:</span> <span class="st">&quot;en-us&quot;</span><span class="fu">,</span></span>
+<span id="cb4-6"><a href="#cb4-6" aria-hidden="true" tabindex="-1"></a>  <span class="dt">&quot;speller&quot;</span><span class="fu">:</span> <span class="st">&quot;lexicon&quot;</span><span class="fu">,</span></span>
+<span id="cb4-7"><a href="#cb4-7" aria-hidden="true" tabindex="-1"></a>  <span class="dt">&quot;top&quot;</span><span class="fu">:</span> <span class="dv">3</span></span>
+<span id="cb4-8"><a href="#cb4-8" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>
+<p>You can also use the <code>highlight</code> parameter to see what
+text is being matched in the <code>content</code> field in the search
+results.</p>
+<div class="sourceCode" id="cb5"><pre
+class="sourceCode json"><code class="sourceCode json"><span id="cb5-1"><a href="#cb5-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>
+<span id="cb5-2"><a href="#cb5-2" aria-hidden="true" tabindex="-1"></a>    <span class="dt">&quot;search&quot;</span><span class="fu">:</span> <span class="st">&quot;eye exams&quot;</span><span class="fu">,</span></span>
+<span id="cb5-3"><a href="#cb5-3" aria-hidden="true" tabindex="-1"></a>    <span class="dt">&quot;highlight&quot;</span><span class="fu">:</span> <span class="st">&quot;content&quot;</span></span>
+<span id="cb5-4"><a href="#cb5-4" aria-hidden="true" tabindex="-1"></a>    <span class="er">...</span></span>
+<span id="cb5-5"><a href="#cb5-5" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>
+<figure>
+<img src="images/screenshot_searchindex.png"
+alt="Screenshot of search explorer with highlighted results" />
+<figcaption aria-hidden="true">Screenshot of search explorer with
+highlighted results</figcaption>
+</figure>
+<p>The search explorer works well for testing text, but is harder to use
+with vectors, since you’d also need to compute the vector embedding and
+send it in. It is probably easier to use the app frontend for testing
+vectors/hybrid search.</p>
+<h4 id="other-approaches-to-improve-search-results">Other approaches to
+improve search results</h4>
+<p>Here are additional ways for improving the search results:</p>
+<ul>
+<li>Adding additional metadata to the “content” field, like the document
+title, so that it can be matched in the search results. Modify <a
+href="https://github.com/Azure-Samples/azure-search-openai-demo/blob/main/app/backend/prepdocslib/searchmanager.py">searchmanager.py</a>
+to include more text in the <code>content</code> field.</li>
+<li>Making additional fields searchable by the full text search step.
+For example, the “sourcepage” field is not currently searchable, but you
+could make that into a <code>SearchableField</code> with
+<code>searchable=True</code> in <a
+href="https://github.com/Azure-Samples/azure-search-openai-demo/blob/main/app/backend/prepdocslib/searchmanager.py">searchmanager.py</a>.
+A change like that requires <a
+href="https://learn.microsoft.com/azure/search/search-howto-reindex#change-an-index-schema">re-building
+the index</a>.</li>
+<li>Using function calling to search by particular fields, like
+searching by the filename. See this blog post on <a
+href="https://blog.pamelafox.org/2024/03/rag-techniques-using-function-calling.html">function
+calling for structured retrieval</a>.</li>
+<li>Using a different splitting strategy for the documents, or modifying
+the existing ones, to improve the chunks that are indexed. You can find
+the currently available splitters in <a
+href="https://github.com/Azure-Samples/azure-search-openai-demo/blob/main/app/backend/prepdocslib/textsplitter.py">textsplitter.py</a>.</li>
+</ul>
+<h3 id="evaluating-answer-quality">Evaluating answer quality</h3>
+<p>Once you’ve made changes to the prompts or settings, you’ll want to
+rigorously evaluate the results to see if they’ve improved. Follow the
+<a href="./evaluation.md">evaluation guide</a> to learn how to run
+evaluations, review results, and compare answers across runs.</p>
diff --git a/data/data_ingestion.html b/data/data_ingestion.html
new file mode 100644
index 0000000000..b843344202
--- /dev/null
+++ b/data/data_ingestion.html
@@ -0,0 +1,230 @@
+<h1 id="rag-chat-data-ingestion">RAG chat: Data ingestion</h1>
+<p>The <a href="/">azure-search-openai-demo</a> project can set up a
+full RAG chat app on Azure AI Search and OpenAI so that you can chat on
+custom data, like internal enterprise data or domain-specific knowledge
+sets. For full instructions on setting up the project, consult the <a
+href="/README.md">main README</a>, and then return here for detailed
+instructions on the data ingestion component.</p>
+<p>The chat app provides two ways to ingest data: manual indexing and
+integrated vectorization. This document explains the differences between
+the two approaches and provides an overview of the manual indexing
+process.</p>
+<ul>
+<li><a href="#supported-document-formats">Supported document
+formats</a></li>
+<li><a href="#manual-indexing-process">Manual indexing process</a>
+<ul>
+<li><a href="#chunking">Chunking</a></li>
+<li><a
+href="#enhancing-search-functionality-with-data-categorization">Categorizing
+data for enhanced search</a></li>
+<li><a href="#indexing-additional-documents">Indexing additional
+documents</a></li>
+<li><a href="#removing-documents">Removing documents</a></li>
+</ul></li>
+<li><a href="#integrated-vectorization">Integrated Vectorization</a>
+<ul>
+<li><a href="#indexing-of-additional-documents">Indexing of additional
+documents</a></li>
+<li><a href="#removal-of-documents">Removal of documents</a></li>
+<li><a href="#scheduled-indexing">Scheduled indexing</a></li>
+</ul></li>
+<li><a href="#debugging-tips">Debugging tips</a></li>
+</ul>
+<h2 id="supported-document-formats">Supported document formats</h2>
+<p>In order to ingest a document format, we need a tool that can turn it
+into text. By default, the manual indexing uses Azure Document
+Intelligence (DI in the table below), but we also have local parsers for
+several formats. The local parsers are not as sophisticated as Azure
+Document Intelligence, but they can be used to decrease charges.</p>
+<table>
+<colgroup>
+<col style="width: 9%" />
+<col style="width: 54%" />
+<col style="width: 36%" />
+</colgroup>
+<thead>
+<tr>
+<th>Format</th>
+<th>Manual indexing</th>
+<th>Integrated Vectorization</th>
+</tr>
+</thead>
+<tbody>
+<tr>
+<td>PDF</td>
+<td>Yes (DI or local with PyPDF)</td>
+<td>Yes</td>
+</tr>
+<tr>
+<td>HTML</td>
+<td>Yes (DI or local with BeautifulSoup)</td>
+<td>Yes</td>
+</tr>
+<tr>
+<td>DOCX, PPTX, XLSX</td>
+<td>Yes (DI)</td>
+<td>Yes</td>
+</tr>
+<tr>
+<td>Images (JPG, PNG, BPM, TIFF, HEIFF)</td>
+<td>Yes (DI)</td>
+<td>Yes</td>
+</tr>
+<tr>
+<td>TXT</td>
+<td>Yes (Local)</td>
+<td>Yes</td>
+</tr>
+<tr>
+<td>JSON</td>
+<td>Yes (Local)</td>
+<td>Yes</td>
+</tr>
+<tr>
+<td>CSV</td>
+<td>Yes (Local)</td>
+<td>Yes</td>
+</tr>
+</tbody>
+</table>
+<p>The Blob indexer used by the Integrated Vectorization approach also
+supports a few <a
+href="https://learn.microsoft.com/azure/search/search-howto-indexing-azure-blob-storage#supported-document-formats">additional
+formats</a>.</p>
+<h2 id="manual-indexing-process">Manual indexing process</h2>
+<p>The <a href="../app/backend/prepdocs.py"><code>prepdocs.py</code></a>
+script is responsible for both uploading and indexing documents. The
+typical usage is to call it using <code>scripts/prepdocs.sh</code>
+(Mac/Linux) or <code>scripts/prepdocs.ps1</code> (Windows), as these
+scripts will set up a Python virtual environment and pass in the
+required parameters based on the current <code>azd</code> environment.
+You can pass additional arguments directly to the script, for example
+<code>scripts/prepdocs.ps1 --removeall</code>. Whenever
+<code>azd up</code> or <code>azd provision</code> is run, the script is
+called automatically.</p>
+<figure>
+<img src="images/diagram_prepdocs.png"
+alt="Diagram of the indexing process" />
+<figcaption aria-hidden="true">Diagram of the indexing
+process</figcaption>
+</figure>
+<p>The script uses the following steps to index documents:</p>
+<ol type="1">
+<li>If it doesn’t yet exist, create a new index in Azure AI Search.</li>
+<li>Upload the PDFs to Azure Blob Storage.</li>
+<li>Split the PDFs into chunks of text.</li>
+<li>Upload the chunks to Azure AI Search. If using vectors (the
+default), also compute the embeddings and upload those alongside the
+text.</li>
+</ol>
+<h3 id="chunking">Chunking</h3>
+<p>We’re often asked why we need to break up the PDFs into chunks when
+Azure AI Search supports searching large documents.</p>
+<p>Chunking allows us to limit the amount of information we send to
+OpenAI due to token limits. By breaking up the content, it allows us to
+easily find potential chunks of text that we can inject into OpenAI. The
+method of chunking we use leverages a sliding window of text such that
+sentences that end one chunk will start the next. This allows us to
+reduce the chance of losing the context of the text.</p>
+<p>If needed, you can modify the chunking algorithm in
+<code>app/backend/prepdocslib/textsplitter.py</code>.</p>
+<h3
+id="enhancing-search-functionality-with-data-categorization">Enhancing
+search functionality with data categorization</h3>
+<p>To enhance search functionality, categorize data during the ingestion
+process with the <code>--category</code> argument, for example
+<code>scripts/prepdocs.ps1 --category ExampleCategoryName</code>. This
+argument specifies the category to which the data belongs, enabling you
+to filter search results based on these categories.</p>
+<p>After running the script with the desired category, ensure these
+categories are added to the ‘Include Category’ dropdown list. This can
+be found in the developer settings in <a
+href="https://github.com/Azure-Samples/azure-search-openai-demo/blob/main/app/frontend/src/components/Settings/Settings.tsx"><code>Settings.tsx</code></a>.
+The default option for this dropdown is “All”. By including specific
+categories, you can refine your search results more effectively.</p>
+<h3 id="indexing-additional-documents">Indexing additional
+documents</h3>
+<p>To upload more PDFs, put them in the data/ folder and run
+<code>./scripts/prepdocs.sh</code> or
+<code>./scripts/prepdocs.ps1</code>.</p>
+<p>A <a
+href="https://github.com/Azure-Samples/azure-search-openai-demo/pull/835">recent
+change</a> added checks to see what’s been uploaded before. The prepdocs
+script now writes an .md5 file with an MD5 hash of each file that gets
+uploaded. Whenever the prepdocs script is re-run, that hash is checked
+against the current hash and the file is skipped if it hasn’t
+changed.</p>
+<h3 id="removing-documents">Removing documents</h3>
+<p>You may want to remove documents from the index. For example, if
+you’re using the sample data, you may want to remove the documents that
+are already in the index before adding your own.</p>
+<p>To remove all documents, use
+<code>./scripts/prepdocs.sh --removeall</code> or
+<code>./scripts/prepdocs.ps1 --removeall</code>.</p>
+<p>You can also remove individual documents by using the
+<code>--remove</code> flag. Open either <code>scripts/prepdocs.sh</code>
+or <code>scripts/prepdocs.ps1</code> and replace <code>/data/*</code>
+with <code>/data/YOUR-DOCUMENT-FILENAME-GOES-HERE.pdf</code>. Then run
+<code>scripts/prepdocs.sh --remove</code> or
+<code>scripts/prepdocs.ps1 --remove</code>.</p>
+<h2 id="integrated-vectorization">Integrated Vectorization</h2>
+<p>Azure AI Search includes an <a
+href="https://techcommunity.microsoft.com/blog/azure-ai-services-blog/announcing-the-public-preview-of-integrated-vectorization-in-azure-ai-search/3960809">integrated
+vectorization feature</a>, a cloud-based approach to data ingestion.
+Integrated vectorization takes care of document format cracking, data
+extraction, chunking, vectorization, and indexing, all with Azure
+technologies.</p>
+<p>See <a
+href="https://github.com/Azure/azure-search-vector-samples/blob/main/demo-python/code/integrated-vectorization/azure-search-integrated-vectorization-sample.ipynb">this
+notebook</a> to understand the process of setting up integrated
+vectorization. We have integrated that code into our
+<code>prepdocs</code> script, so you can use it without needing to
+understand the details.</p>
+<p>You must first explicitly <a
+href="./deploy_features.md#enabling-integrated-vectorization">enable
+integrated vectorization</a> in the <code>azd</code> environment to use
+this feature.</p>
+<p>This feature cannot be used on existing index. You need to create a
+new index or drop and recreate an existing index. In the newly created
+index schema, a new field ‘parent_id’ is added. This is used internally
+by the indexer to manage life cycle of chunks.</p>
+<p>This feature is not supported in the free SKU for Azure AI
+Search.</p>
+<h3 id="indexing-of-additional-documents">Indexing of additional
+documents</h3>
+<p>To add additional documents to the index, first upload them to your
+data source (Blob storage, by default). Then navigate to the Azure
+portal, find the index, and run it. The Azure AI Search indexer will
+identify the new documents and ingest them into the index.</p>
+<h3 id="removal-of-documents">Removal of documents</h3>
+<p>To remove documents from the index, remove them from your data source
+(Blob storage, by default). Then navigate to the Azure portal, find the
+index, and run it. The Azure AI Search indexer will take care of
+removing those documents from the index.</p>
+<h3 id="scheduled-indexing">Scheduled indexing</h3>
+<p>If you would like the indexer to run automatically, you can set it up
+to <a
+href="https://learn.microsoft.com/azure/search/search-howto-schedule-indexers">run
+on a schedule</a>.</p>
+<h2 id="debugging-tips">Debugging tips</h2>
+<p>If you are not sure if a file successfully uploaded, you can query
+the index from the Azure Portal or from the REST API. Open the index and
+paste the queries below into the search bar.</p>
+<p>To see all the filenames uploaded to the index:</p>
+<div class="sourceCode" id="cb1"><pre
+class="sourceCode json"><code class="sourceCode json"><span id="cb1-1"><a href="#cb1-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>
+<span id="cb1-2"><a href="#cb1-2" aria-hidden="true" tabindex="-1"></a>  <span class="dt">&quot;search&quot;</span><span class="fu">:</span> <span class="st">&quot;*&quot;</span><span class="fu">,</span></span>
+<span id="cb1-3"><a href="#cb1-3" aria-hidden="true" tabindex="-1"></a>  <span class="dt">&quot;count&quot;</span><span class="fu">:</span> <span class="kw">true</span><span class="fu">,</span></span>
+<span id="cb1-4"><a href="#cb1-4" aria-hidden="true" tabindex="-1"></a>  <span class="dt">&quot;top&quot;</span><span class="fu">:</span> <span class="dv">1</span><span class="fu">,</span></span>
+<span id="cb1-5"><a href="#cb1-5" aria-hidden="true" tabindex="-1"></a>  <span class="dt">&quot;facets&quot;</span><span class="fu">:</span> <span class="ot">[</span><span class="st">&quot;sourcefile&quot;</span><span class="ot">]</span></span>
+<span id="cb1-6"><a href="#cb1-6" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>
+<p>To search for specific filenames:</p>
+<div class="sourceCode" id="cb2"><pre
+class="sourceCode json"><code class="sourceCode json"><span id="cb2-1"><a href="#cb2-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>
+<span id="cb2-2"><a href="#cb2-2" aria-hidden="true" tabindex="-1"></a>  <span class="dt">&quot;search&quot;</span><span class="fu">:</span> <span class="st">&quot;*&quot;</span><span class="fu">,</span></span>
+<span id="cb2-3"><a href="#cb2-3" aria-hidden="true" tabindex="-1"></a>  <span class="dt">&quot;count&quot;</span><span class="fu">:</span> <span class="kw">true</span><span class="fu">,</span></span>
+<span id="cb2-4"><a href="#cb2-4" aria-hidden="true" tabindex="-1"></a>  <span class="dt">&quot;top&quot;</span><span class="fu">:</span> <span class="dv">1</span><span class="fu">,</span></span>
+<span id="cb2-5"><a href="#cb2-5" aria-hidden="true" tabindex="-1"></a>  <span class="dt">&quot;filter&quot;</span><span class="fu">:</span> <span class="st">&quot;sourcefile eq &#39;employee_handbook.pdf&#39;&quot;</span><span class="fu">,</span></span>
+<span id="cb2-6"><a href="#cb2-6" aria-hidden="true" tabindex="-1"></a>  <span class="dt">&quot;facets&quot;</span><span class="fu">:</span> <span class="ot">[</span><span class="st">&quot;sourcefile&quot;</span><span class="ot">]</span></span>
+<span id="cb2-7"><a href="#cb2-7" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>
diff --git a/data/deploy_existing.html b/data/deploy_existing.html
new file mode 100644
index 0000000000..2a8cef80a3
--- /dev/null
+++ b/data/deploy_existing.html
@@ -0,0 +1,209 @@
+<h1 id="rag-chat-deploying-with-existing-azure-resources">RAG chat:
+Deploying with existing Azure resources</h1>
+<p>If you already have existing Azure resources, or if you want to
+specify the exact name of new Azure Resource, you can do so by setting
+<code>azd</code> environment values. You should set these values before
+running <code>azd up</code>. Once you’ve set them, return to the <a
+href="../README.md#deploying">deployment steps</a>.</p>
+<ul>
+<li><a href="#resource-group">Resource group</a></li>
+<li><a href="#openai-resource">OpenAI resource</a></li>
+<li><a href="#azure-ai-search-resource">Azure AI Search
+resource</a></li>
+<li><a href="#azure-app-service-plan-and-app-service-resources">Azure
+App Service Plan and App Service resources</a></li>
+<li><a href="#azure-application-insights-and-related-resources">Azure
+Application Insights and related resources</a></li>
+<li><a href="#azure-computer-vision-resources">Azure Computer Vision
+resources</a></li>
+<li><a href="#azure-document-intelligence-resource">Azure Document
+Intelligence resource</a></li>
+<li><a href="#azure-speech-resource">Azure Speech resource</a></li>
+<li><a href="#other-azure-resources">Other Azure resources</a></li>
+</ul>
+<h2 id="resource-group">Resource group</h2>
+<ol type="1">
+<li>Run
+<code>azd env set AZURE_RESOURCE_GROUP {Name of existing resource group}</code></li>
+<li>Run
+<code>azd env set AZURE_LOCATION {Location of existing resource group}</code></li>
+</ol>
+<h2 id="openai-resource">OpenAI resource</h2>
+<h3 id="azure-openai">Azure OpenAI</h3>
+<ol type="1">
+<li>Run
+<code>azd env set AZURE_OPENAI_SERVICE {Name of existing OpenAI service}</code></li>
+<li>Run
+<code>azd env set AZURE_OPENAI_RESOURCE_GROUP {Name of existing resource group that OpenAI service is provisioned to}</code></li>
+<li>Run
+<code>azd env set AZURE_OPENAI_LOCATION {Location of existing OpenAI service}</code></li>
+<li>Run
+<code>azd env set AZURE_OPENAI_CHATGPT_DEPLOYMENT {Name of existing chat deployment}</code>.
+Only needed if your chat deployment is not the default ‘chat’.</li>
+<li>Run
+<code>azd env set AZURE_OPENAI_CHATGPT_MODEL {Model name of existing chat deployment}</code>.
+Only needed if your chat model is not the default ‘gpt-35-turbo’.</li>
+<li>Run
+<code>azd env set AZURE_OPENAI_CHATGPT_DEPLOYMENT_VERSION {Version string for existing chat deployment}</code>.
+Only needed if your chat deployment model version is not the default
+‘0125’. You definitely need to change this if you changed the
+model.</li>
+<li>Run
+<code>azd env set AZURE_OPENAI_CHATGPT_DEPLOYMENT_SKU {Name of SKU for existing chat deployment}</code>.
+Only needed if your chat deployment SKU is not the default ‘Standard’,
+like if it is ‘GlobalStandard’ instead.</li>
+<li>Run
+<code>azd env set AZURE_OPENAI_EMB_DEPLOYMENT {Name of existing embedding deployment}</code>.
+Only needed if your embeddings deployment is not the default
+‘embedding’.</li>
+<li>Run
+<code>azd env set AZURE_OPENAI_EMB_MODEL_NAME {Model name of existing embedding deployment}</code>.
+Only needed if your embeddings model is not the default
+‘text-embedding-ada-002’.</li>
+<li>Run
+<code>azd env set AZURE_OPENAI_EMB_DIMENSIONS {Dimensions for existing embedding deployment}</code>.
+Only needed if your embeddings model is not the default
+‘text-embedding-ada-002’.</li>
+<li>Run
+<code>azd env set AZURE_OPENAI_EMB_DEPLOYMENT_VERSION {Version string for existing embedding deployment}</code>.
+If your embeddings deployment is one of the ‘text-embedding-3’ models,
+set this to the number 1.</li>
+<li>This project does <em>not</em> use keys when authenticating to Azure
+OpenAI. However, if your Azure OpenAI service must have key access
+enabled for some reason (like for use by other projects), then run
+<code>azd env set AZURE_OPENAI_DISABLE_KEYS false</code>. The default
+value is <code>true</code> so you should only run the command if you
+need key access.</li>
+</ol>
+<p>When you run <code>azd up</code> after and are prompted to select a
+value for <code>openAiResourceGroupLocation</code>, make sure to select
+the same location as the existing OpenAI resource group.</p>
+<h3 id="openai.com-openai">Openai.com OpenAI</h3>
+<ol type="1">
+<li>Run <code>azd env set OPENAI_HOST openai</code></li>
+<li>Run
+<code>azd env set OPENAI_ORGANIZATION {Your OpenAI organization}</code></li>
+<li>Run
+<code>azd env set OPENAI_API_KEY {Your OpenAI API key}</code></li>
+<li>Run <code>azd up</code></li>
+</ol>
+<p>You can retrieve your OpenAI key by checking <a
+href="https://platform.openai.com/account/api-keys">your user page</a>
+and your organization by navigating to <a
+href="https://platform.openai.com/account/org-settings">your
+organization page</a>. Learn more about creating an OpenAI free trial at
+<a href="https://openai.com/pricing">this link</a>. Do <em>not</em>
+check your key into source control.</p>
+<p>When you run <code>azd up</code> after and are prompted to select a
+value for <code>openAiResourceGroupLocation</code>, you can select any
+location as it will not be used.</p>
+<h2 id="azure-ai-search-resource">Azure AI Search resource</h2>
+<ol type="1">
+<li>Run
+<code>azd env set AZURE_SEARCH_SERVICE {Name of existing Azure AI Search service}</code></li>
+<li>Run
+<code>azd env set AZURE_SEARCH_SERVICE_RESOURCE_GROUP {Name of existing resource group with ACS service}</code></li>
+<li>If that resource group is in a different location than the one
+you’ll pick for the <code>azd up</code> step, then run
+<code>azd env set AZURE_SEARCH_SERVICE_LOCATION {Location of existing service}</code></li>
+<li>If the search service’s SKU is not standard, then run
+<code>azd env set AZURE_SEARCH_SERVICE_SKU {Name of SKU}</code>. If you
+specify the free tier, then your app will no longer be able to use
+semantic ranker. Be advised that <a
+href="https://learn.microsoft.com/azure/search/search-sku-tier#tier-upgrade-or-downgrade">search
+SKUs cannot be changed</a>. (<a
+href="https://learn.microsoft.com/azure/templates/microsoft.search/searchservices?pivots=deployment-language-bicep#sku">See
+other possible SKU values</a>)</li>
+<li>If you have an existing index that is set up with all the expected
+fields, then run
+<code>azd env set AZURE_SEARCH_INDEX {Name of existing index}</code>.
+Otherwise, the <code>azd up</code> command will create a new index.</li>
+</ol>
+<p>You can also customize the search service (new or existing) for
+non-English searches:</p>
+<ol type="1">
+<li>To configure the language of the search query to a value other than
+“en-US”, run
+<code>azd env set AZURE_SEARCH_QUERY_LANGUAGE {Name of query language}</code>.
+(<a
+href="https://learn.microsoft.com/rest/api/searchservice/preview-api/search-documents#queryLanguage">See
+other possible values</a>)</li>
+<li>To turn off the spell checker, run
+<code>azd env set AZURE_SEARCH_QUERY_SPELLER none</code>. Consult <a
+href="https://learn.microsoft.com/rest/api/searchservice/preview-api/search-documents#queryLanguage">this
+table</a> to determine if spell checker is supported for your query
+language.</li>
+<li>To configure the name of the analyzer to use for a searchable text
+field to a value other than “en.microsoft”, run
+<code>azd env set AZURE_SEARCH_ANALYZER_NAME {Name of analyzer name}</code>.
+(<a
+href="https://learn.microsoft.com/dotnet/api/microsoft.azure.search.models.field.analyzer?view=azure-dotnet-legacy&amp;viewFallbackFrom=azure-dotnet">See
+other possible values</a>)</li>
+</ol>
+<h2 id="azure-app-service-plan-and-app-service-resources">Azure App
+Service Plan and App Service resources</h2>
+<ol type="1">
+<li>Run
+<code>azd env set AZURE_APP_SERVICE_PLAN {Name of existing Azure App Service Plan}</code></li>
+<li>Run
+<code>azd env set AZURE_APP_SERVICE {Name of existing Azure App Service}</code>.</li>
+<li>Run
+<code>azd env set AZURE_APP_SERVICE_SKU {SKU of Azure App Service, defaults to B1}</code>.</li>
+</ol>
+<h2 id="azure-application-insights-and-related-resources">Azure
+Application Insights and related resources</h2>
+<ol type="1">
+<li>Run
+<code>azd env set AZURE_APPLICATION_INSIGHTS {Name of existing Azure App Insights}</code>.</li>
+<li>Run
+<code>azd env set AZURE_APPLICATION_INSIGHTS_DASHBOARD {Name of existing Azure App Insights Dashboard}</code>.</li>
+<li>Run
+<code>azd env set AZURE_LOG_ANALYTICS {Name of existing Azure Log Analytics Workspace Name}</code>.</li>
+</ol>
+<h2 id="azure-computer-vision-resources">Azure Computer Vision
+resources</h2>
+<ol type="1">
+<li>Run
+<code>azd env set AZURE_COMPUTER_VISION_SERVICE {Name of existing Azure Computer Vision Service Name}</code></li>
+<li>Run
+<code>azd env set AZURE_COMPUTER_VISION_RESOURCE_GROUP {Name of existing Azure Computer Vision Resource Group Name}</code></li>
+<li>Run
+<code>azd env set AZURE_COMPUTER_VISION_LOCATION {Name of existing Azure Computer Vision Location}</code></li>
+<li>Run
+<code>azd env set AZURE_COMPUTER_VISION_SKU {SKU of Azure Computer Vision service, defaults to F0}</code></li>
+</ol>
+<h2 id="azure-document-intelligence-resource">Azure Document
+Intelligence resource</h2>
+<p>In order to support analysis of many document formats, this
+repository uses a preview version of Azure Document Intelligence
+(formerly Form Recognizer) that is only available in <a
+href="https://learn.microsoft.com/azure/ai-services/document-intelligence/concept-layout">limited
+regions</a>. If your existing resource is in one of those regions, then
+you can re-use it by setting the following environment variables:</p>
+<ol type="1">
+<li>Run
+<code>azd env set AZURE_DOCUMENTINTELLIGENCE_SERVICE {Name of existing Azure AI Document Intelligence service}</code></li>
+<li>Run
+<code>azd env set AZURE_DOCUMENTINTELLIGENCE_LOCATION {Location of existing service}</code></li>
+<li>Run
+<code>azd env set AZURE_DOCUMENTINTELLIGENCE_RESOURCE_GROUP {Name of resource group with existing service, defaults to main resource group}</code></li>
+<li>Run
+<code>azd env set AZURE_DOCUMENTINTELLIGENCE_SKU {SKU of existing service, defaults to S0}</code></li>
+</ol>
+<h2 id="azure-speech-resource">Azure Speech resource</h2>
+<ol type="1">
+<li>Run
+<code>azd env set AZURE_SPEECH_SERVICE {Name of existing Azure Speech service}</code></li>
+<li>Run
+<code>azd env set AZURE_SPEECH_SERVICE_RESOURCE_GROUP {Name of existing resource group with speech service}</code></li>
+<li>If that resource group is in a different location than the one
+you’ll pick for the <code>azd up</code> step, then run
+<code>azd env set AZURE_SPEECH_SERVICE_LOCATION {Location of existing service}</code></li>
+<li>If the speech service’s SKU is not “S0”, then run
+<code>azd env set AZURE_SPEECH_SERVICE_SKU {Name of SKU}</code>.</li>
+</ol>
+<h2 id="other-azure-resources">Other Azure resources</h2>
+<p>You can also use existing Azure AI Storage Accounts. See
+<code>./infra/main.parameters.json</code> for list of environment
+variables to pass to <code>azd env set</code> to configure those
+existing resources.</p>
diff --git a/data/deploy_features.html b/data/deploy_features.html
new file mode 100644
index 0000000000..2e5568a5b8
--- /dev/null
+++ b/data/deploy_features.html
@@ -0,0 +1,395 @@
+<h1 id="rag-chat-enabling-optional-features">RAG chat: Enabling optional
+features</h1>
+<p>This document covers optional features that can be enabled in the
+deployed Azure resources. You should typically enable these features
+before running <code>azd up</code>. Once you’ve set them, return to the
+<a href="../README.md#deploying">deployment steps</a>.</p>
+<ul>
+<li><a href="#using-gpt-4">Using GPT-4</a></li>
+<li><a href="#using-text-embedding-3-models">Using text-embedding-3
+models</a></li>
+<li><a href="#enabling-gpt-4-turbo-with-vision">Enabling GPT-4 Turbo
+with Vision</a></li>
+<li><a
+href="#enabling-media-description-with-azure-content-understanding">Enabling
+media description with Azure Content Understanding</a></li>
+<li><a href="#enabling-client-side-chat-history">Enabling client-side
+chat history</a></li>
+<li><a
+href="#enabling-persistent-chat-history-with-azure-cosmos-db">Enabling
+persistent chat history with Azure Cosmos DB</a></li>
+<li><a href="#enabling-language-picker">Enabling language
+picker</a></li>
+<li><a href="#enabling-speech-inputoutput">Enabling speech
+input/output</a></li>
+<li><a href="#enabling-integrated-vectorization">Enabling Integrated
+Vectorization</a></li>
+<li><a href="#enabling-authentication">Enabling authentication</a></li>
+<li><a href="#enabling-login-and-document-level-access-control">Enabling
+login and document level access control</a></li>
+<li><a href="#enabling-user-document-upload">Enabling user document
+upload</a></li>
+<li><a href="#enabling-cors-for-an-alternate-frontend">Enabling CORS for
+an alternate frontend</a></li>
+<li><a href="#adding-an-openai-load-balancer">Adding an OpenAI load
+balancer</a></li>
+<li><a href="#deploying-with-private-endpoints">Deploying with private
+endpoints</a></li>
+<li><a href="#using-local-parsers">Using local parsers</a></li>
+</ul>
+<h2 id="using-gpt-4">Using GPT-4</h2>
+<p>(Instructions for <strong>GPT-4</strong>, <strong>GPT-4o</strong>,
+and <strong>GPT-4o mini</strong> models are also included here.)</p>
+<p>We generally find that most developers are able to get high-quality
+answers using GPT-3.5. However, if you want to try GPT-4, GPT-4o, or
+GPT-4o mini, you can do so by following these steps:</p>
+<p>Execute the following commands inside your terminal:</p>
+<ol type="1">
+<li><p>To set the name of the deployment, run this command with a unique
+name in your Azure OpenAI account. You can use any deployment name, as
+long as it’s unique in your Azure OpenAI account.</p>
+<div class="sourceCode" id="cb1"><pre
+class="sourceCode bash"><code class="sourceCode bash"><span id="cb1-1"><a href="#cb1-1" aria-hidden="true" tabindex="-1"></a><span class="ex">azd</span> env set AZURE_OPENAI_CHATGPT_DEPLOYMENT <span class="op">&lt;</span>your-deployment-name<span class="op">&gt;</span></span></code></pre></div>
+<p>For example:</p>
+<div class="sourceCode" id="cb2"><pre
+class="sourceCode bash"><code class="sourceCode bash"><span id="cb2-1"><a href="#cb2-1" aria-hidden="true" tabindex="-1"></a><span class="ex">azd</span> env set AZURE_OPENAI_CHATGPT_DEPLOYMENT chat4</span></code></pre></div></li>
+<li><p>To set the GPT model name to a <strong>gpt-4</strong>,
+<strong>gpt-4o</strong>, or <strong>gpt-4o mini</strong> version from
+the <a
+href="https://learn.microsoft.com/azure/ai-services/openai/concepts/models">available
+models</a>, run this command with the appropriate GPT model name.</p>
+<p>For GPT-4:</p>
+<div class="sourceCode" id="cb3"><pre
+class="sourceCode bash"><code class="sourceCode bash"><span id="cb3-1"><a href="#cb3-1" aria-hidden="true" tabindex="-1"></a><span class="ex">azd</span> env set AZURE_OPENAI_CHATGPT_MODEL gpt-4</span></code></pre></div>
+<p>For GPT-4o:</p>
+<div class="sourceCode" id="cb4"><pre
+class="sourceCode bash"><code class="sourceCode bash"><span id="cb4-1"><a href="#cb4-1" aria-hidden="true" tabindex="-1"></a><span class="ex">azd</span> env set AZURE_OPENAI_CHATGPT_MODEL gpt-4o</span></code></pre></div>
+<p>For GPT-4o mini:</p>
+<div class="sourceCode" id="cb5"><pre
+class="sourceCode bash"><code class="sourceCode bash"><span id="cb5-1"><a href="#cb5-1" aria-hidden="true" tabindex="-1"></a><span class="ex">azd</span> env set AZURE_OPENAI_CHATGPT_MODEL gpt-4o-mini</span></code></pre></div></li>
+<li><p>To set the Azure OpenAI deployment SKU name, run this command
+with <a
+href="https://learn.microsoft.com/azure/ai-services/openai/how-to/deployment-types#deployment-types">the
+desired SKU name</a>.</p>
+<div class="sourceCode" id="cb6"><pre
+class="sourceCode bash"><code class="sourceCode bash"><span id="cb6-1"><a href="#cb6-1" aria-hidden="true" tabindex="-1"></a><span class="ex">azd</span> env set AZURE_OPENAI_CHATGPT_DEPLOYMENT_SKU GlobalStandard</span></code></pre></div></li>
+<li><p>To set the Azure OpenAI deployment capacity, run this command
+with the desired capacity.</p>
+<div class="sourceCode" id="cb7"><pre
+class="sourceCode bash"><code class="sourceCode bash"><span id="cb7-1"><a href="#cb7-1" aria-hidden="true" tabindex="-1"></a><span class="ex">azd</span> env set AZURE_OPENAI_CHATGPT_DEPLOYMENT_CAPACITY 10</span></code></pre></div></li>
+<li><p>To set the Azure OpenAI deployment version from the <a
+href="https://learn.microsoft.com/azure/ai-services/openai/concepts/models">available
+versions</a>, run this command with the appropriate version.</p>
+<p>For GPT-4:</p>
+<div class="sourceCode" id="cb8"><pre
+class="sourceCode bash"><code class="sourceCode bash"><span id="cb8-1"><a href="#cb8-1" aria-hidden="true" tabindex="-1"></a><span class="ex">azd</span> env set AZURE_OPENAI_CHATGPT_DEPLOYMENT_VERSION turbo-2024-04-09</span></code></pre></div>
+<p>For GPT-4o:</p>
+<div class="sourceCode" id="cb9"><pre
+class="sourceCode bash"><code class="sourceCode bash"><span id="cb9-1"><a href="#cb9-1" aria-hidden="true" tabindex="-1"></a><span class="ex">azd</span> env set AZURE_OPENAI_CHATGPT_DEPLOYMENT_VERSION 2024-05-13</span></code></pre></div>
+<p>For GPT-4o mini:</p>
+<div class="sourceCode" id="cb10"><pre
+class="sourceCode bash"><code class="sourceCode bash"><span id="cb10-1"><a href="#cb10-1" aria-hidden="true" tabindex="-1"></a><span class="ex">azd</span> env set AZURE_OPENAI_CHATGPT_DEPLOYMENT_VERSION 2024-07-18</span></code></pre></div></li>
+<li><p>To update the deployment with the new parameters, run this
+command.</p>
+<div class="sourceCode" id="cb11"><pre
+class="sourceCode bash"><code class="sourceCode bash"><span id="cb11-1"><a href="#cb11-1" aria-hidden="true" tabindex="-1"></a><span class="ex">azd</span> up</span></code></pre></div></li>
+</ol>
+<blockquote>
+<p>[!NOTE] To revert back to GPT 3.5, run the following commands:</p>
+<ul>
+<li><code>azd env set AZURE_OPENAI_CHATGPT_DEPLOYMENT chat</code> to set
+the name of your old GPT 3.5 deployment.</li>
+<li><code>azd env set AZURE_OPENAI_CHATGPT_MODEL gpt-35-turbo</code> to
+set the name of your old GPT 3.5 model.</li>
+<li><code>azd env set AZURE_OPENAI_CHATGPT_DEPLOYMENT_CAPACITY 30</code>
+to set the capacity of your old GPT 3.5 deployment.</li>
+<li><code>azd env set AZURE_OPENAI_CHATGPT_DEPLOYMENT_SKU Standard</code>
+to set the Sku name back to Standard.</li>
+<li><code>azd env set AZURE_OPENAI_CHATGPT_DEPLOYMENT_VERSION 0125</code>
+to set the version number of your old GPT 3.5.</li>
+<li><code>azd up</code> to update the provisioned resources.</li>
+</ul>
+<p>Note that this does not delete your GPT-4 deployment; it just makes
+your application create a new or reuse an old GPT 3.5 deployment. If you
+want to delete it, you can go to your Azure OpenAI studio and do so.</p>
+</blockquote>
+<h2 id="using-text-embedding-3-models">Using text-embedding-3
+models</h2>
+<p>By default, the deployed Azure web app uses the
+<code>text-embedding-ada-002</code> embedding model. If you want to use
+one of the text-embedding-3 models, you can do so by following these
+steps:</p>
+<ol type="1">
+<li><p>Run one of the following commands to set the desired model:</p>
+<pre class="shell"><code>azd env set AZURE_OPENAI_EMB_MODEL_NAME text-embedding-3-small</code></pre>
+<pre class="shell"><code>azd env set AZURE_OPENAI_EMB_MODEL_NAME text-embedding-3-large</code></pre></li>
+<li><p>Specify the desired dimensions of the model: (from 256-3072,
+model dependent)</p>
+<pre class="shell"><code>azd env set AZURE_OPENAI_EMB_DIMENSIONS 256</code></pre></li>
+<li><p>Set the model version to “1” (the only version as of March
+2024):</p>
+<pre class="shell"><code>azd env set AZURE_OPENAI_EMB_DEPLOYMENT_VERSION 1</code></pre></li>
+<li><p>When prompted during <code>azd up</code>, make sure to select a
+region for the OpenAI resource group location that supports the
+text-embedding-3 models. There are <a
+href="https://learn.microsoft.com/azure/ai-services/openai/concepts/models#embeddings-models">limited
+regions available</a>.</p></li>
+</ol>
+<p>If you have already deployed:</p>
+<ul>
+<li>You’ll need to change the deployment name by running
+<code>azd env set AZURE_OPENAI_EMB_DEPLOYMENT &lt;new-deployment-name&gt;</code></li>
+<li>You’ll need to create a new index, and re-index all of the data
+using the new model. You can either delete the current index in the
+Azure Portal, or create an index with a different name by running
+<code>azd env set AZURE_SEARCH_INDEX new-index-name</code>. When you
+next run <code>azd up</code>, the new index will be created and the data
+will be re-indexed.</li>
+<li>If your OpenAI resource is not in one of the supported regions, you
+should delete <code>openAiResourceGroupLocation</code> from
+<code>.azure/YOUR-ENV-NAME/config.json</code>. When running
+<code>azd up</code>, you will be prompted to select a new region.</li>
+</ul>
+<blockquote>
+<p>![NOTE] The text-embedding-3 models are not currently supported by
+the integrated vectorization feature.</p>
+</blockquote>
+<h2 id="enabling-gpt-4-turbo-with-vision">Enabling GPT-4 Turbo with
+Vision</h2>
+<p>⚠️ This feature is not currently compatible with <a
+href="#enabling-integrated-vectorization">integrated
+vectorization</a>.</p>
+<p>This section covers the integration of GPT-4 Vision with Azure AI
+Search. Learn how to enhance your search capabilities with the power of
+image and text indexing, enabling advanced search functionalities over
+diverse document types. For a detailed guide on setup and usage, visit
+our <a href="gpt4v.md">Enabling GPT-4 Turbo with Vision</a> page.</p>
+<h2
+id="enabling-media-description-with-azure-content-understanding">Enabling
+media description with Azure Content Understanding</h2>
+<p>⚠️ This feature is not currently compatible with <a
+href="#enabling-integrated-vectorization">integrated vectorization</a>.
+It is compatible with <a href="./gpt4v.md">GPT vision integration</a>,
+but the features provide similar functionality.</p>
+<p>By default, if your documents contain image-like figures, the data
+ingestion process will ignore those figures, so users will not be able
+to ask questions about them.</p>
+<p>You can optionably enable the description of media content using
+Azure Content Understanding. When enabled, the data ingestion process
+will send figures to Azure Content Understanding and replace the figure
+with the description in the indexed document.</p>
+<p>To enable media description with Azure Content Understanding,
+run:</p>
+<pre class="shell"><code>azd env set USE_MEDIA_DESCRIBER_AZURE_CU true</code></pre>
+<p>If you have already run <code>azd up</code>, you will need to run
+<code>azd provision</code> to create the new Content Understanding
+service. If you have already indexed your documents and want to re-index
+them with the media descriptions, first <a
+href="./data_ingestion.md#removing-documents">remove the existing
+documents</a> and then <a
+href="./data_ingestion.md#indexing-additional-documents">re-ingest the
+data</a>.</p>
+<p>⚠️ This feature does not yet support DOCX, PPTX, or XLSX formats. If
+you have figures in those formats, they will be ignored. Convert them
+first to PDF or image formats to enable media description.</p>
+<h2 id="enabling-client-side-chat-history">Enabling client-side chat
+history</h2>
+<p><a href="https://www.youtube.com/watch?v=1YiTFnnLVIA">📺 Watch: (RAG
+Deep Dive series) Storing chat history</a></p>
+<p>This feature allows users to view the chat history of their
+conversation, stored in the browser using <a
+href="https://developer.mozilla.org/docs/Web/API/IndexedDB_API">IndexedDB</a>.
+That means the chat history will be available only on the device where
+the chat was initiated. To enable browser-stored chat history, run:</p>
+<pre class="shell"><code>azd env set USE_CHAT_HISTORY_BROWSER true</code></pre>
+<h2 id="enabling-persistent-chat-history-with-azure-cosmos-db">Enabling
+persistent chat history with Azure Cosmos DB</h2>
+<p><a href="https://www.youtube.com/watch?v=1YiTFnnLVIA">📺 Watch: (RAG
+Deep Dive series) Storing chat history</a></p>
+<p>This feature allows authenticated users to view the chat history of
+their conversations, stored in the server-side storage using <a
+href="https://learn.microsoft.com/azure/cosmos-db/">Azure Cosmos
+DB</a>.This option requires that authentication be enabled. The chat
+history will be persistent and accessible from any device where the user
+logs in with the same account. To enable server-stored chat history,
+run:</p>
+<pre class="shell"><code>azd env set USE_CHAT_HISTORY_COSMOS true</code></pre>
+<p>When both the browser-stored and Cosmos DB options are enabled,
+Cosmos DB will take precedence over browser-stored chat history.</p>
+<h2 id="enabling-language-picker">Enabling language picker</h2>
+<p>You can optionally enable the language picker to allow users to
+switch between different languages. Currently, it supports English,
+Spanish, French, and Japanese.</p>
+<p>To add support for additional languages, create new locale files and
+update <code>app/frontend/src/i18n/config.ts</code> accordingly. To
+enable language picker, run:</p>
+<pre class="shell"><code>azd env set ENABLE_LANGUAGE_PICKER true</code></pre>
+<h2 id="enabling-speech-inputoutput">Enabling speech input/output</h2>
+<p><a href="https://www.youtube.com/watch?v=BwiHUjlLY_U">📺 Watch a
+short video of speech input/output</a></p>
+<p>You can optionally enable speech input/output by setting the azd
+environment variables.</p>
+<h3 id="speech-input">Speech Input</h3>
+<p>The speech input feature uses the browser’s built-in <a
+href="https://developer.mozilla.org/docs/Web/API/SpeechRecognition">Speech
+Recognition API</a>. It may not work in all browser/OS combinations. To
+enable speech input, run:</p>
+<pre class="shell"><code>azd env set USE_SPEECH_INPUT_BROWSER true</code></pre>
+<h3 id="speech-output">Speech Output</h3>
+<p>The speech output feature uses <a
+href="https://learn.microsoft.com/azure/ai-services/speech-service/overview">Azure
+Speech Service</a> for speech-to-text. Additional costs will be incurred
+for using the Azure Speech Service. <a
+href="https://azure.microsoft.com/pricing/details/cognitive-services/speech-services/">See
+pricing</a>. To enable speech output, run:</p>
+<pre class="shell"><code>azd env set USE_SPEECH_OUTPUT_AZURE true</code></pre>
+<p>To set <a
+href="https://learn.microsoft.com/azure/ai-services/speech-service/language-support?tabs=tts">the
+voice</a> for the speech output, run:</p>
+<pre class="shell"><code>azd env set AZURE_SPEECH_SERVICE_VOICE en-US-AndrewMultilingualNeural</code></pre>
+<p>Alternatively you can use the browser’s built-in <a
+href="https://developer.mozilla.org/docs/Web/API/SpeechSynthesis">Speech
+Synthesis API</a>. It may not work in all browser/OS combinations. To
+enable speech output, run:</p>
+<pre class="shell"><code>azd env set USE_SPEECH_OUTPUT_BROWSER true</code></pre>
+<h2 id="enabling-integrated-vectorization">Enabling Integrated
+Vectorization</h2>
+<p>⚠️ This feature is not currently compatible with the <a
+href="./gpt4v.md">GPT vision integration</a>.</p>
+<p>Azure AI search recently introduced an <a
+href="https://techcommunity.microsoft.com/blog/azure-ai-services-blog/announcing-the-public-preview-of-integrated-vectorization-in-azure-ai-search/3960809">integrated
+vectorization feature in preview mode</a>. This feature is a cloud-based
+approach to data ingestion, which takes care of document format
+cracking, data extraction, chunking, vectorization, and indexing, all
+with Azure technologies.</p>
+<p>To enable integrated vectorization with this sample:</p>
+<ol type="1">
+<li><p>If you’ve previously deployed, delete the existing search index.
+🗑️</p></li>
+<li><p>To enable the use of integrated vectorization, run:</p>
+<pre class="shell"><code>azd env set USE_FEATURE_INT_VECTORIZATION true</code></pre></li>
+<li><p>If you’ve already deployed your app, then you can run just the
+<code>provision</code> step:</p>
+<pre class="shell"><code>azd provision</code></pre>
+<p>That will set up necessary RBAC roles and configure the integrated
+vectorization feature on your search service.</p>
+<p>If you haven’t deployed your app yet, then you should run the full
+<code>azd up</code> after configuring all optional features.</p></li>
+<li><p>You can view the resources such as the indexer and skillset in
+Azure Portal and monitor the status of the vectorization
+process.</p></li>
+</ol>
+<h2 id="enabling-authentication">Enabling authentication</h2>
+<p>By default, the deployed Azure web app will have no authentication or
+access restrictions enabled, meaning anyone with routable network access
+to the web app can chat with your indexed data. If you’d like to
+automatically setup authentication and user login as part of the
+<code>azd up</code> process, see <a href="./login_and_acl.md">this
+guide</a>.</p>
+<p>Alternatively, you can manually require authentication to your Azure
+Active Directory by following the <a
+href="https://learn.microsoft.com/azure/app-service/scenario-secure-app-authentication-app-service">Add
+app authentication</a> tutorial and set it up against the deployed web
+app.</p>
+<p>To then limit access to a specific set of users or groups, you can
+follow the steps from <a
+href="https://learn.microsoft.com/entra/identity-platform/howto-restrict-your-app-to-a-set-of-users">Restrict
+your Microsoft Entra app to a set of users</a> by changing “Assignment
+Required?” option under the Enterprise Application, and then assigning
+users/groups access. Users not granted explicit access will receive the
+error message -AADSTS50105: Your administrator has configured the
+application <app_name> to block users unless they are specifically
+granted (‘assigned’) access to the application.-</p>
+<h2 id="enabling-login-and-document-level-access-control">Enabling login
+and document level access control</h2>
+<p>By default, the deployed Azure web app allows users to chat with all
+your indexed data. You can enable an optional login system using Azure
+Active Directory to restrict access to indexed data based on the logged
+in user. Enable the optional login and document level access control
+system by following <a href="./login_and_acl.md">this guide</a>.</p>
+<h2 id="enabling-user-document-upload">Enabling user document
+upload</h2>
+<p>You can enable an optional user document upload system to allow users
+to upload their own documents and chat with them. This feature requires
+you to first <a href="./login_and_acl.md">enable login and document
+level access control</a>. Then you can enable the optional user document
+upload system by setting an azd environment variable:</p>
+<p><code>azd env set USE_USER_UPLOAD true</code></p>
+<p>Then you’ll need to run <code>azd up</code> to provision an Azure
+Data Lake Storage Gen2 account for storing the user-uploaded documents.
+When the user uploads a document, it will be stored in a directory in
+that account with the same name as the user’s Entra object id, and will
+have ACLs associated with that directory. When the ingester runs, it
+will also set the <code>oids</code> of the indexed chunks to the user’s
+Entra object id.</p>
+<p>If you are enabling this feature on an existing index, you should
+also update your index to have the new <code>storageUrl</code>
+field:</p>
+<pre class="shell"><code>python ./scripts/manageacl.py  -v --acl-action enable_acls</code></pre>
+<p>And then update existing search documents with the storage URL of the
+main Blob container:</p>
+<pre class="shell"><code>python ./scripts/manageacl.py  -v --acl-action update_storage_urls --url &lt;https://YOUR-MAIN-STORAGE-ACCOUNT.blob.core.windows.net/content/&gt;</code></pre>
+<p>Going forward, all uploaded documents will have their
+<code>storageUrl</code> set in the search index. This is necessary to
+disambiguate user-uploaded documents from admin-uploaded documents.</p>
+<h2 id="enabling-cors-for-an-alternate-frontend">Enabling CORS for an
+alternate frontend</h2>
+<p>By default, the deployed Azure web app will only allow requests from
+the same origin. To enable CORS for a frontend hosted on a different
+origin, run:</p>
+<ol type="1">
+<li>Run
+<code>azd env set ALLOWED_ORIGIN https://&lt;your-domain.com&gt;</code></li>
+<li>Run <code>azd up</code></li>
+</ol>
+<p>For the frontend code, change <code>BACKEND_URI</code> in
+<code>api.ts</code> to point at the deployed backend URL, so that all
+fetch requests will be sent to the deployed backend.</p>
+<p>For an alternate frontend that’s written in Web Components and
+deployed to Static Web Apps, check out <a
+href="https://github.com/Azure-Samples/azure-search-openai-javascript">azure-search-openai-javascript</a>
+and its guide on <a
+href="https://github.com/Azure-Samples/azure-search-openai-javascript#using-a-different-backend">using
+a different backend</a>. Both these repositories adhere to the same <a
+href="https://aka.ms/chatprotocol">HTTP protocol for AI chat
+apps</a>.</p>
+<h2 id="adding-an-openai-load-balancer">Adding an OpenAI load
+balancer</h2>
+<p>As discussed in more details in our <a
+href="./productionizing.md">productionizing guide</a>, you may want to
+consider implementing a load balancer between OpenAI instances if you
+are consistently going over the TPM limit. Fortunately, this repository
+is designed for easy integration with other repositories that create
+load balancers for OpenAI instances. For seamless integration
+instructions with this sample, please check:</p>
+<ul>
+<li><a
+href="https://learn.microsoft.com/azure/developer/python/get-started-app-chat-scaling-with-azure-api-management">Scale
+Azure OpenAI for Python with Azure API Management</a></li>
+<li><a
+href="https://learn.microsoft.com/azure/developer/python/get-started-app-chat-scaling-with-azure-container-apps">Scale
+Azure OpenAI for Python chat using RAG with Azure Container
+Apps</a></li>
+</ul>
+<h2 id="deploying-with-private-endpoints">Deploying with private
+endpoints</h2>
+<p>It is possible to deploy this app with public access disabled, using
+Azure private endpoints and private DNS Zones. For more details, read <a
+href="./deploy_private.md">the private deployment guide</a>. That
+requires a multi-stage provisioning, so you will need to do more than
+just <code>azd up</code> after setting the environment variables.</p>
+<h2 id="using-local-parsers">Using local parsers</h2>
+<p>If you want to decrease the charges by using local parsers instead of
+Azure Document Intelligence, you can set environment variables before
+running the <a href="./data_ingestion.md">data ingestion script</a>.
+Note that local parsers will generally be not as sophisticated.</p>
+<ol type="1">
+<li>Run <code>azd env set USE_LOCAL_PDF_PARSER true</code> to use the
+local PDF parser.</li>
+<li>Run <code>azd env set USE_LOCAL_HTML_PARSER true</code> to use the
+local HTML parser.</li>
+</ol>
+<p>The local parsers will be used the next time you run the data
+ingestion script. To use these parsers for the user document upload
+system, you’ll need to run <code>azd provision</code> to update the web
+app to use the local parsers.</p>
diff --git a/data/deploy_freetrial.html b/data/deploy_freetrial.html
new file mode 100644
index 0000000000..1638b5418a
--- /dev/null
+++ b/data/deploy_freetrial.html
@@ -0,0 +1,40 @@
+<h1 id="rag-chat-deploying-with-a-free-trial-account">RAG chat:
+Deploying with a free trial account</h1>
+<p>If you have just created an Azure free trial account and are using
+the free trial credits, there are several modifications you need to
+make, due to restrictions on the free trial account.</p>
+<p>Follow these instructions <em>before</em> you run
+<code>azd up</code>.</p>
+<h2 id="accomodate-for-low-openai-quotas">Accomodate for low OpenAI
+quotas</h2>
+<p>The free trial accounts currently get a max of 1K TPM
+(tokens-per-minute), whereas our Bicep templates try to allocate 30K
+TPM.</p>
+<p>To reduce the TPM allocation, run these commands:</p>
+<pre class="shell"><code>azd env set AZURE_OPENAI_CHATGPT_DEPLOYMENT_CAPACITY 1
+azd env set AZURE_OPENAI_EMB_DEPLOYMENT_CAPACITY 1</code></pre>
+<p>Alternatively, if you have an OpenAI.com account, you can use that
+instead:</p>
+<pre class="shell"><code>azd env set OPENAI_HOST openai
+azd env set OPENAI_ORGANIZATION {Your OpenAI organization}
+azd env set OPENAI_API_KEY {Your OpenAI API key}</code></pre>
+<h2 id="accomodate-for-azure-container-apps-restrictions">Accomodate for
+Azure Container Apps restrictions</h2>
+<p>By default, this project deploys to Azure Container Apps, using a
+remote build process that builds the Docker image in the cloud.
+Unfortunately, free trial accounts cannot use that remote build
+process.</p>
+<p>You have two options:</p>
+<ol type="1">
+<li><p>Comment out or delete <code>remoteBuild: true</code> in
+<code>azure.yaml</code>, and make sure you have Docker installed in your
+environment.</p></li>
+<li><p>Deploy using App Service instead:</p>
+<ul>
+<li><p>Comment out <code>host: containerapp</code> and uncomment
+<code>host: appservice</code> in the <a
+href="../azure.yaml">azure.yaml</a> file.</p></li>
+<li><p>Set the deployment target to <code>appservice</code>:</p>
+<pre class="shell"><code>azd env set DEPLOYMENT_TARGET appservice</code></pre></li>
+</ul></li>
+</ol>
diff --git a/data/deploy_lowcost.html b/data/deploy_lowcost.html
new file mode 100644
index 0000000000..866b9c1a23
--- /dev/null
+++ b/data/deploy_lowcost.html
@@ -0,0 +1,122 @@
+<h1 id="rag-chat-deploying-with-minimal-costs">RAG chat: Deploying with
+minimal costs</h1>
+<p>This AI RAG chat application is designed to be easily deployed using
+the Azure Developer CLI, which provisions the infrastructure according
+to the Bicep files in the <code>infra</code> folder. Those files
+describe each of the Azure resources needed, and configures their SKU
+(pricing tier) and other parameters. Many Azure services offer a free
+tier, but the infrastructure files in this project do <em>not</em>
+default to the free tier as there are often limitations in that
+tier.</p>
+<p>However, if your goal is to minimize costs while prototyping your
+application, follow the steps below <em>before</em> running
+<code>azd up</code>. Once you’ve gone through these steps, return to the
+<a href="../README.md#deploying">deployment steps</a>.</p>
+<p><a href="https://www.youtube.com/watch?v=nlIyos0RXHw">📺 Live stream:
+Deploying from a free account</a></p>
+<ol type="1">
+<li><p>Log in to your Azure account using the Azure Developer CLI:</p>
+<pre class="shell"><code>azd auth login</code></pre></li>
+<li><p>Create a new azd environment for the free resource group:</p>
+<pre class="shell"><code>azd env new</code></pre>
+<p>Enter a name that will be used for the resource group. This will
+create a new folder in the <code>.azure</code> folder, and set it as the
+active environment for any calls to <code>azd</code> going
+forward.</p></li>
+<li><p>Switch from Azure Container Apps to the free tier of Azure App
+Service:</p>
+<p>Azure Container Apps has a consumption-based pricing model that is
+very low cost, but it is not free, plus Azure Container Registry costs a
+small amount each month.</p>
+<p>To deploy to App Service instead:</p>
+<ul>
+<li><p>Comment out <code>host: containerapp</code> and uncomment
+<code>host: appservice</code> in the <a
+href="../azure.yaml">azure.yaml</a> file.</p></li>
+<li><p>Set the deployment target to <code>appservice</code>:</p>
+<pre class="shell"><code>azd env set DEPLOYMENT_TARGET appservice</code></pre></li>
+<li><p>Set the App Service SKU to the free tier:</p>
+<pre class="shell"><code>azd env set AZURE_APP_SERVICE_SKU F1</code></pre></li>
+</ul>
+<p>Limitation: You are only allowed a certain number of free App Service
+instances per region. If you have exceeded your limit in a region, you
+will get an error during the provisioning stage. If that happens, you
+can run <code>azd down</code>, then <code>azd env new</code> to create a
+new environment with a new region.</p></li>
+<li><p>Use the free tier of Azure AI Search:</p>
+<pre class="shell"><code>azd env set AZURE_SEARCH_SERVICE_SKU free</code></pre>
+<p>Limitations:</p>
+<ol type="1">
+<li>You are only allowed one free search service across all regions. If
+you have one already, either delete that service or follow instructions
+to reuse your <a
+href="../README.md#existing-azure-ai-search-resource">existing search
+service</a>.</li>
+<li>The free tier does not support semantic ranker, so the app UI will
+no longer display the option to use the semantic ranker. Note that will
+generally result in <a
+href="https://techcommunity.microsoft.com/blog/azure-ai-services-blog/azure-ai-search-outperforming-vector-search-with-hybrid-retrieval-and-ranking-ca/3929167">decreased
+search relevance</a>.</li>
+</ol></li>
+<li><p>Use the free tier of Azure Document Intelligence (used in
+analyzing files):</p>
+<pre class="shell"><code>azd env set AZURE_DOCUMENTINTELLIGENCE_SKU F0</code></pre>
+<p><strong>Limitation for PDF files:</strong></p>
+<p>The free tier will only scan the first two pages of each PDF. In our
+sample documents, those first two pages are just title pages, so you
+won’t be able to get answers from the documents. You can either use your
+own documents that are only 2-pages long, or you can use a local Python
+package for PDF parsing by setting:</p>
+<pre class="shell"><code>azd env set USE_LOCAL_PDF_PARSER true</code></pre>
+<p><strong>Limitation for HTML files:</strong></p>
+<p>The free tier will only scan the first two pages of each HTML file.
+So, you might not get very accurate answers from the files. You can
+either use your own files that are only 2-pages long, or you can use a
+local Python package for HTML parsing by setting:</p>
+<pre class="shell"><code>azd env set USE_LOCAL_HTML_PARSER true</code></pre></li>
+<li><p>Use the free tier of Azure Cosmos DB:</p>
+<pre class="shell"><code>azd env set AZURE_COSMOSDB_SKU free</code></pre>
+<p>Limitation: You can have only one free Cosmos DB account. To keep
+your account free of charge, ensure that you do not exceed the free tier
+limits. For more information, see the <a
+href="https://learn.microsoft.com/azure/cosmos-db/free-tier">Azure
+Cosmos DB lifetime free tier</a>.</p></li>
+<li><p>⚠️ This step is currently only possible if you’re deploying to
+App Service (<a
+href="https://github.com/Azure-Samples/azure-search-openai-demo/issues/2281">see
+issue 2281</a>):</p>
+<p>Turn off Azure Monitor (Application Insights):</p>
+<pre class="shell"><code>azd env set AZURE_USE_APPLICATION_INSIGHTS false</code></pre>
+<p>Application Insights is quite inexpensive already, so turning this
+off may not be worth the costs saved, but it is an option for those who
+want to minimize costs.</p></li>
+<li><p>Use OpenAI.com instead of Azure OpenAI: This should not be
+necessary, as the costs are same for both services, but you may need
+this step if your account does not have access to Azure OpenAI for some
+reason.</p>
+<pre class="shell"><code>azd env set OPENAI_HOST openai
+azd env set OPENAI_ORGANIZATION {Your OpenAI organization}
+azd env set OPENAI_API_KEY {Your OpenAI API key}</code></pre>
+<p>Both Azure OpenAI and openai.com OpenAI accounts will incur costs,
+based on tokens used, but the costs are fairly low for the amount of
+sample data (less than $10).</p></li>
+<li><p>Disable vector search:</p>
+<pre class="shell"><code>azd env set USE_VECTORS false</code></pre>
+<p>By default, the application computes vector embeddings for documents
+during the data ingestion phase, and then computes a vector embedding
+for user questions asked in the application. Those computations require
+an embedding model, which incurs costs per tokens used. The costs are
+fairly low, so the benefits of vector search would typically outweigh
+the costs, but it is possible to disable vector support. If you do so,
+the application will fall back to a keyword search, which is less
+accurate.</p></li>
+<li><p>Once you’ve made the desired customizations, follow the steps in
+the README <a href="../README.md#deploying-from-scratch">to run
+<code>azd up</code></a>. We recommend using “eastus” as the region, for
+availability reasons.</p></li>
+</ol>
+<h2 id="reducing-costs-locally">Reducing costs locally</h2>
+<p>To save costs for local development, you could use an
+OpenAI-compatible model. Follow steps in <a
+href="localdev.md#using-a-local-openai-compatible-api">local development
+guide</a>.</p>
diff --git a/data/deploy_private.html b/data/deploy_private.html
new file mode 100644
index 0000000000..a22c4d8fb4
--- /dev/null
+++ b/data/deploy_private.html
@@ -0,0 +1,93 @@
+<!--
+---
+name: RAG chat with private endpoints
+description: Configure access to a chat app so that it's only accessible from private endpoints.
+languages:
+- python
+- typescript
+- bicep
+- azdeveloper
+products:
+- azure-openai
+- azure-cognitive-search
+- azure-app-service
+- azure
+page_type: sample
+urlFragment: azure-search-openai-demo-private-access
+---
+-->
+<h1 id="rag-chat-deploying-with-private-access">RAG chat: Deploying with
+private access</h1>
+<p><a href="https://www.youtube.com/watch?v=08wtL1eB15g">📺 Watch: (RAG
+Deep Dive series) Private network deployment</a></p>
+<p>The <a href="/">azure-search-openai-demo</a> project can set up a
+full RAG chat app on Azure AI Search and OpenAI so that you can chat on
+custom data, like internal enterprise data or domain-specific knowledge
+sets. For full instructions on setting up the project, consult the <a
+href="/README.md">main README</a>, and then return here for detailed
+instructions on configuring private endpoints.</p>
+<p>⚠️ This feature is not yet compatible with Azure Container Apps, so
+you will need to <a href="./azure_app_service.md">deploy to Azure App
+Service</a> instead.</p>
+<p>If you want to disable public access when deploying the Chat App, you
+can do so by setting <code>azd</code> environment values.</p>
+<h2 id="before-you-begin">Before you begin</h2>
+<p>Deploying with public access disabled adds additional cost to your
+deployment. Please see pricing for the following products:</p>
+<ol type="1">
+<li><a
+href="https://azure.microsoft.com/pricing/details/private-link/">Private
+Endpoints</a>
+<ol type="1">
+<li>The exact number of private endpoints created depends on the <a
+href="./deploy_features.md">optional features</a> used.</li>
+</ol></li>
+<li><a href="https://azure.microsoft.com/pricing/details/dns/">Private
+DNS Zones</a></li>
+</ol>
+<h2 id="environment-variables-controlling-private-access">Environment
+variables controlling private access</h2>
+<ol type="1">
+<li><code>AZURE_PUBLIC_NETWORK_ACCESS</code>: Controls the value of
+public network access on supported Azure resources. Valid values are
+‘Enabled’ or ‘Disabled’.
+<ol type="1">
+<li>When public network access is ‘Enabled’, Azure resources are open to
+the internet.</li>
+<li>When public network access is ‘Disabled’, Azure resources are only
+accessible over a virtual network.</li>
+</ol></li>
+<li><code>AZURE_USE_PRIVATE_ENDPOINT</code>: Controls deployment of <a
+href="https://learn.microsoft.com/azure/private-link/private-endpoint-overview">private
+endpoints</a> which connect Azure resources to the virtual network.
+<ol type="1">
+<li>When set to ‘true’, ensures private endpoints are deployed for
+connectivity even when <code>AZURE_PUBLIC_NETWORK_ACCESS</code> is
+‘Disabled’.</li>
+<li>Note that private endpoints do not make the chat app accessible from
+the internet. Connections must be initiated from inside the virtual
+network.</li>
+</ol></li>
+</ol>
+<h2 id="recommended-deployment-strategy-for-private-access">Recommended
+deployment strategy for private access</h2>
+<ol type="1">
+<li>Deploy the app with private endpoints enabled and public access
+enabled.</li>
+</ol>
+<pre class="shell"><code>azd env set AZURE_USE_PRIVATE_ENDPOINT true
+azd env set AZURE_PUBLIC_NETWORK_ACCESS Enabled
+azd up</code></pre>
+<ol type="1">
+<li>Validate that you can connect to the chat app and it’s working as
+expected from the internet.</li>
+<li>Re-provision the app with public access disabled.</li>
+</ol>
+<pre class="shell"><code>azd env set AZURE_PUBLIC_NETWORK_ACCESS Disabled
+azd provision</code></pre>
+<ol type="1">
+<li>Log into your network using a tool like <a
+href="https://azure.microsoft.com/services/vpn-gateway/">Azure VPN
+Gateway</a> and validate that you can connect to the chat app from
+inside the network.</li>
+</ol>
diff --git a/data/deploy_troubleshooting.html b/data/deploy_troubleshooting.html
new file mode 100644
index 0000000000..38e534bd00
--- /dev/null
+++ b/data/deploy_troubleshooting.html
@@ -0,0 +1,35 @@
+<h1 id="rag-chat-troubleshooting-deployment">RAG chat: Troubleshooting
+deployment</h1>
+<p>If you are experiencing an error when deploying the RAG chat solution
+using the <a href="../README.md#deploying">deployment steps</a>, this
+guide will help you troubleshoot common issues.</p>
+<ol type="1">
+<li><p>You’re attempting to create resources in regions not enabled for
+Azure OpenAI (e.g. East US 2 instead of East US), or where the model
+you’re trying to use isn’t enabled. See <a
+href="https://aka.ms/oai/models">this matrix of model
+availability</a>.</p></li>
+<li><p>You’ve exceeded a quota, most often number of resources per
+region. See <a href="https://aka.ms/oai/quotas">this article on quotas
+and limits</a>.</p></li>
+<li><p>You’re getting “same resource name not allowed” conflicts. That’s
+likely because you’ve run the sample multiple times and deleted the
+resources you’ve been creating each time, but are forgetting to purge
+them. Azure keeps resources for 48 hours unless you purge from soft
+delete. See <a
+href="https://learn.microsoft.com/azure/cognitive-services/manage-resources?tabs=azure-portal#purge-a-deleted-resource">this
+article on purging resources</a>.</p></li>
+<li><p>You see <code>CERTIFICATE_VERIFY_FAILED</code> when the
+<code>prepdocs.py</code> script runs. That’s typically due to incorrect
+SSL certificates setup on your machine. Try the suggestions in this <a
+href="https://stackoverflow.com/questions/35569042/ssl-certificate-verify-failed-with-python3/43855394#43855394">StackOverflow
+answer</a>.</p></li>
+<li><p>After running <code>azd up</code> and visiting the website, you
+see a ‘404 Not Found’ in the browser. Wait 10 minutes and try again, as
+it might be still starting up. Then try running <code>azd deploy</code>
+and wait again. If you still encounter errors with the deployed app and
+are deploying to App Service, consult the <a
+href="/docs/appservice.md">guide on debugging App Service
+deployments</a>. Please file an issue if the logs don’t help you resolve
+the error.</p></li>
+</ol>
diff --git a/data/evaluation.html b/data/evaluation.html
new file mode 100644
index 0000000000..500bcda7ea
--- /dev/null
+++ b/data/evaluation.html
@@ -0,0 +1,131 @@
+<h1 id="evaluating-the-rag-answer-quality">Evaluating the RAG answer
+quality</h1>
+<p><a href="https://www.youtube.com/watch?v=lyCLu53fb3g">📺 Watch: (RAG
+Deep Dive series) Evaluating RAG answer quality</a></p>
+<p>Follow these steps to evaluate the quality of the answers generated
+by the RAG flow.</p>
+<ul>
+<li><a href="#deploy-an-evaluation-model">Deploy an evaluation
+model</a></li>
+<li><a href="#setup-the-evaluation-environment">Setup the evaluation
+environment</a></li>
+<li><a href="#generate-ground-truth-data">Generate ground truth
+data</a></li>
+<li><a href="#run-bulk-evaluation">Run bulk evaluation</a></li>
+<li><a href="#review-the-evaluation-results">Review the evaluation
+results</a></li>
+<li><a href="#run-bulk-evaluation-on-a-pr">Run bulk evaluation on a
+PR</a></li>
+</ul>
+<h2 id="deploy-an-evaluation-model">Deploy an evaluation model</h2>
+<ol type="1">
+<li><p>Run this command to tell <code>azd</code> to deploy a GPT-4 level
+model for evaluation:</p>
+<pre class="shell"><code>azd env set USE_EVAL true</code></pre></li>
+<li><p>Set the capacity to the highest possible value to ensure that the
+evaluation runs relatively quickly. Even with a high capacity, it can
+take a long time to generate ground truth data and run bulk
+evaluations.</p>
+<pre class="shell"><code>azd env set AZURE_OPENAI_EVAL_DEPLOYMENT_CAPACITY 100</code></pre>
+<p>By default, that will provision a <code>gpt-4o</code> model, version
+<code>2024-08-06</code>. To change those settings, set the azd
+environment variables <code>AZURE_OPENAI_EVAL_MODEL</code> and
+<code>AZURE_OPENAI_EVAL_MODEL_VERSION</code> to the desired
+values.</p></li>
+<li><p>Then, run the following command to provision the model:</p>
+<pre class="shell"><code>azd provision</code></pre></li>
+</ol>
+<h2 id="setup-the-evaluation-environment">Setup the evaluation
+environment</h2>
+<p>Make a new Python virtual environment and activate it. This is
+currently required due to incompatibilities between the dependencies of
+the evaluation script and the main project.</p>
+<div class="sourceCode" id="cb4"><pre
+class="sourceCode bash"><code class="sourceCode bash"><span id="cb4-1"><a href="#cb4-1" aria-hidden="true" tabindex="-1"></a><span class="ex">python</span> <span class="at">-m</span> venv .evalenv</span></code></pre></div>
+<div class="sourceCode" id="cb5"><pre
+class="sourceCode bash"><code class="sourceCode bash"><span id="cb5-1"><a href="#cb5-1" aria-hidden="true" tabindex="-1"></a><span class="bu">source</span> .evalenv/bin/activate</span></code></pre></div>
+<p>Install all the dependencies for the evaluation script by running the
+following command:</p>
+<div class="sourceCode" id="cb6"><pre
+class="sourceCode bash"><code class="sourceCode bash"><span id="cb6-1"><a href="#cb6-1" aria-hidden="true" tabindex="-1"></a><span class="ex">pip</span> install <span class="at">-r</span> evals/requirements.txt</span></code></pre></div>
+<h2 id="generate-ground-truth-data">Generate ground truth data</h2>
+<p>Modify the search terms and tasks in
+<code>evals/generate_config.json</code> to match your domain.</p>
+<p>Generate ground truth data by running the following command:</p>
+<div class="sourceCode" id="cb7"><pre
+class="sourceCode bash"><code class="sourceCode bash"><span id="cb7-1"><a href="#cb7-1" aria-hidden="true" tabindex="-1"></a><span class="ex">python</span> evals/generate_ground_truth.py <span class="at">--numquestions</span><span class="op">=</span>200 <span class="at">--numsearchdocs</span><span class="op">=</span>1000</span></code></pre></div>
+<p>The options are:</p>
+<ul>
+<li><code>numquestions</code>: The number of questions to generate. We
+suggest at least 200.</li>
+<li><code>numsearchdocs</code>: The number of documents (chunks) to
+retrieve from your search index. You can leave off the option to fetch
+all documents, but that will significantly increase time it takes to
+generate ground truth data. You may want to at least start with a
+subset.</li>
+<li><code>kgfile</code>: An existing RAGAS knowledge base JSON file,
+which is usually <code>ground_truth_kg.json</code>. You may want to
+specify this if you already created a knowledge base and just want to
+tweak the question generation steps.</li>
+<li><code>groundtruthfile</code>: The file to write the generated ground
+truth answwers. By default, this is
+<code>evals/ground_truth.jsonl</code>.</li>
+</ul>
+<p>🕰️ This may take a long time, possibly several hours, depending on
+the size of the search index.</p>
+<p>Review the generated data in <code>evals/ground_truth.jsonl</code>
+after running that script, removing any question/answer pairs that don’t
+seem like realistic user input.</p>
+<h2 id="run-bulk-evaluation">Run bulk evaluation</h2>
+<p>Review the configuration in <code>evals/eval_config.json</code> to
+ensure that everything is correctly setup. You may want to adjust the
+metrics used. See <a
+href="https://github.com/Azure-Samples/ai-rag-chat-evaluator">the
+ai-rag-chat-evaluator README</a> for more information on the available
+metrics.</p>
+<p>By default, the evaluation script will evaluate every question in the
+ground truth data. Run the evaluation script by running the following
+command:</p>
+<div class="sourceCode" id="cb8"><pre
+class="sourceCode bash"><code class="sourceCode bash"><span id="cb8-1"><a href="#cb8-1" aria-hidden="true" tabindex="-1"></a><span class="ex">python</span> evals/evaluate.py</span></code></pre></div>
+<p>The options are:</p>
+<ul>
+<li><code>numquestions</code>: The number of questions to evaluate. By
+default, this is all questions in the ground truth data.</li>
+<li><code>resultsdir</code>: The directory to write the evaluation
+results. By default, this is a timestamped folder in
+<code>evals/results</code>. This option can also be specified in
+<code>eval_config.json</code>.</li>
+<li><code>targeturl</code>: The URL of the running application to
+evaluate. By default, this is <code>http://localhost:50505</code>. This
+option can also be specified in <code>eval_config.json</code>.</li>
+</ul>
+<p>🕰️ This may take a long time, possibly several hours, depending on
+the number of ground truth questions, and the TPM capacity of the
+evaluation model, and the number of GPT metrics requested.</p>
+<h2 id="review-the-evaluation-results">Review the evaluation
+results</h2>
+<p>The evaluation script will output a summary of the evaluation
+results, inside the <code>evals/results</code> directory.</p>
+<p>You can see a summary of results across all evaluation runs by
+running the following command:</p>
+<div class="sourceCode" id="cb9"><pre
+class="sourceCode bash"><code class="sourceCode bash"><span id="cb9-1"><a href="#cb9-1" aria-hidden="true" tabindex="-1"></a><span class="ex">python</span> <span class="at">-m</span> evaltools summary evals/results</span></code></pre></div>
+<p>Compare answers to the ground truth by running the following
+command:</p>
+<div class="sourceCode" id="cb10"><pre
+class="sourceCode bash"><code class="sourceCode bash"><span id="cb10-1"><a href="#cb10-1" aria-hidden="true" tabindex="-1"></a><span class="ex">python</span> <span class="at">-m</span> evaltools diff evals/results/baseline/</span></code></pre></div>
+<p>Compare answers across two runs by running the following command:</p>
+<div class="sourceCode" id="cb11"><pre
+class="sourceCode bash"><code class="sourceCode bash"><span id="cb11-1"><a href="#cb11-1" aria-hidden="true" tabindex="-1"></a><span class="ex">python</span> <span class="at">-m</span> evaltools diff evals/results/baseline/ evals/results/SECONDRUNHERE</span></code></pre></div>
+<h2 id="run-bulk-evaluation-on-a-pr">Run bulk evaluation on a PR</h2>
+<p>This repository includes a GitHub Action workflow
+<code>evaluate.yaml</code> that can be used to run the evaluation on the
+changes in a PR.</p>
+<p>In order for the workflow to run successfully, you must first set up
+<a href="./azd.md#github-actions">continuous integration</a> for the
+repository.</p>
+<p>To run the evaluation on the changes in a PR, a repository member can
+post a <code>/evaluate</code> comment to the PR. This will trigger the
+evaluation workflow to run the evaluation on the PR changes and will
+post the results to the PR.</p>
diff --git a/data/gpt4v.html b/data/gpt4v.html
new file mode 100644
index 0000000000..cc52c482e6
--- /dev/null
+++ b/data/gpt4v.html
@@ -0,0 +1,88 @@
+<h1 id="rag-chat-using-gpt-vision-model-with-rag-approach">RAG chat:
+Using GPT vision model with RAG approach</h1>
+<p><a href="https://www.youtube.com/watch?v=5FfIy7G2WW0">📺 Watch: (RAG
+Deep Dive series) Multimedia data ingestion</a></p>
+<p>This repository includes an optional feature that uses the GPT vision
+model to generate responses based on retrieved content. This feature is
+useful for answering questions based on the visual content of documents,
+such as photos and charts.</p>
+<h2 id="how-it-works">How it works</h2>
+<p>When this feature is enabled, the following changes are made to the
+application:</p>
+<ul>
+<li><strong>Search index</strong>: We added a new field to the Azure AI
+Search index to store the embedding returned by the multimodal Azure AI
+Vision API (while keeping the existing field that stores the OpenAI text
+embeddings).</li>
+<li><strong>Data ingestion</strong>: In addition to our usual PDF
+ingestion flow, we also convert each PDF document page to an image,
+store that image with the filename rendered on top, and add the
+embedding to the index.</li>
+<li><strong>Question answering</strong>: We search the index using both
+the text and multimodal embeddings. We send both the text and the image
+to gpt-4o, and ask it to answer the question based on both kinds of
+sources.</li>
+<li><strong>Citations</strong>: The frontend displays both image sources
+and text sources, to help users understand how the answer was
+generated.</li>
+</ul>
+<p>For more details on how this feature works, read <a
+href="https://techcommunity.microsoft.com/blog/azuredevcommunityblog/integrating-vision-into-rag-applications/4239460">this
+blog post</a> or watch <a
+href="https://www.youtube.com/live/C3Zq3z4UQm4?si=SSPowBBJoTBKZ9WW&amp;t=89">this
+video</a>.</p>
+<h2 id="using-the-feature">Using the feature</h2>
+<h3 id="prerequisites">Prerequisites</h3>
+<ul>
+<li>Create a <a
+href="https://ms.portal.azure.com/#create/Microsoft.CognitiveServicesComputerVision">AI
+Vision account in Azure Portal first</a>, so that you can agree to the
+Responsible AI terms for that resource. You can delete that account
+after agreeing.</li>
+<li>The ability to deploy a gpt-4o model in the <a
+href="https://learn.microsoft.com/azure/ai-services/openai/concepts/models#standard-deployment-model-availability">supported
+regions</a>. If you’re not sure, try to create a gpt-4o deployment from
+your Azure OpenAI deployments page.</li>
+<li>Ensure that you can deploy the Azure OpenAI resource group in <a
+href="https://learn.microsoft.com/azure/cognitive-services/openai/concepts/models#model-summary-table-and-region-availability">a
+region where all required components are available</a>:
+<ul>
+<li>Azure OpenAI models
+<ul>
+<li>gpt-35-turbo</li>
+<li>text-embedding-ada-002</li>
+<li>gpt-4o</li>
+</ul></li>
+<li><a
+href="https://learn.microsoft.com/azure/ai-services/computer-vision/">Azure
+AI Vision</a></li>
+</ul></li>
+</ul>
+<h3 id="deployment">Deployment</h3>
+<ol type="1">
+<li><p><strong>Enable GPT vision approach:</strong></p>
+<p>First, make sure you do <em>not</em> have integrated vectorization
+enabled, since that is currently incompatible:</p>
+<pre class="shell"><code>azd env set USE_FEATURE_INT_VECTORIZATION false</code></pre>
+<p>Then set the environment variable for enabling vision support:</p>
+<pre class="shell"><code>azd env set USE_GPT4V true</code></pre>
+<p>When set, that flag will provision a Azure AI Vision resource and
+gpt-4o model, upload image versions of PDFs to Blob storage, upload
+embeddings of images in a new <code>imageEmbedding</code> field, and
+enable the vision approach in the UI.</p></li>
+<li><p><strong>Clean old deployments (optional):</strong> Run
+<code>azd down --purge</code> for a fresh setup.</p></li>
+<li><p><strong>Start the application:</strong> Execute
+<code>azd up</code> to build, provision, deploy, and initiate document
+preparation.</p></li>
+<li><p><strong>Try out the feature:</strong></p>
+<ul>
+<li>Access the developer options in the web app and select “Use GPT
+vision model”.</li>
+<li>New sample questions will show up in the UI that are based on the
+sample financial document.</li>
+<li>Try out a question and see the answer generated by the GPT vision
+model.</li>
+<li>Check the ‘Thought process’ and ‘Supporting content’ tabs.</li>
+</ul></li>
+</ol>
diff --git a/data/localdev.html b/data/localdev.html
new file mode 100644
index 0000000000..43e3f0b1c8
--- /dev/null
+++ b/data/localdev.html
@@ -0,0 +1,124 @@
+<h1 id="rag-chat-local-development-of-chat-app">RAG chat: Local
+development of chat app</h1>
+<p>After deploying the app to Azure, you may want to continue
+development locally. This guide explains how to run the app locally,
+including hot reloading and debugging.</p>
+<ul>
+<li><a href="#running-development-server-from-the-command-line">Running
+development server from the command line</a></li>
+<li><a href="#hot-reloading-frontend-and-backend-files">Hot reloading
+frontend and backend files</a></li>
+<li><a href="#using-vs-code-run-and-debug">Using VS Code “Run and
+Debug”</a></li>
+<li><a href="#using-a-local-openai-compatible-api">Using a local
+OpenAI-compatible API</a>
+<ul>
+<li><a href="#using-ollama-server">Using Ollama server</a></li>
+<li><a href="#using-llamafile-server">Using llamafile server</a></li>
+</ul></li>
+</ul>
+<h2 id="running-development-server-from-the-command-line">Running
+development server from the command line</h2>
+<p>You can only run locally <strong>after</strong> having successfully
+run the <code>azd up</code> command. If you haven’t yet, follow the
+steps in <a href="../README.md#azure-deployment">Azure deployment</a>
+above.</p>
+<ol type="1">
+<li>Run <code>azd auth login</code></li>
+<li>Start the server:</li>
+</ol>
+<p>Windows:</p>
+<pre class="shell"><code>./app/start.ps1</code></pre>
+<p>Linux/Mac:</p>
+<pre class="shell"><code>./app/start.sh</code></pre>
+<p>VS Code: Run the “VS Code Task: Start App” task.</p>
+<h2 id="hot-reloading-frontend-and-backend-files">Hot reloading frontend
+and backend files</h2>
+<p>When you run <code>./start.ps1</code> or <code>./start.sh</code>, the
+backend files will be watched and reloaded automatically. However, the
+frontend files will not be watched and reloaded automatically.</p>
+<p>To enable hot reloading of frontend files, open a new terminal and
+navigate to the frontend directory:</p>
+<pre class="shell"><code>cd app/frontend</code></pre>
+<p>Then run:</p>
+<pre class="shell"><code>npm run dev</code></pre>
+<p>You should see:</p>
+<pre class="shell"><code>&gt; frontend@0.0.0 dev
+&gt; vite
+
+
+  VITE v4.5.1  ready in 957 ms
+
+  ➜  Local:   http://localhost:5173/
+  ➜  Network: use --host to expose
+  ➜  press h to show help</code></pre>
+<p>Navigate to the URL shown in the terminal (in this case,
+<code>http://localhost:5173/</code>). This local server will watch and
+reload frontend files. All backend requests will be routed to the Python
+server according to <code>vite.config.ts</code>.</p>
+<p>Then, whenever you make changes to frontend files, the changes will
+be automatically reloaded, without any browser refresh needed.</p>
+<h2 id="using-vs-code-run-and-debug">Using VS Code “Run and Debug”</h2>
+<p>This project includes configurations defined in
+<code>.vscode/launch.json</code> that allow you to run and debug the app
+directly from VS Code:</p>
+<ul>
+<li>“Backend (Python)”: Starts the Python backend server, defaulting to
+port 50505.</li>
+<li>“Frontend”: Starts the frontend server using Vite, typically at port
+5173.</li>
+<li>“Frontend &amp; Backend”: A compound configuration that starts both
+the frontend and backend servers.</li>
+</ul>
+<p>When you run these configurations, you can set breakpoints in your
+code and debug as you would in a normal VS Code debugging session.</p>
+<h2 id="using-a-local-openai-compatible-api">Using a local
+OpenAI-compatible API</h2>
+<p>You may want to save costs by developing against a local LLM server,
+such as <a
+href="https://github.com/Mozilla-Ocho/llamafile/">llamafile</a>. Note
+that a local LLM will generally be slower and not as sophisticated.</p>
+<p>Once the local LLM server is running and serving an OpenAI-compatible
+endpoint, set these environment variables:</p>
+<pre class="shell"><code>azd env set USE_VECTORS false
+azd env set OPENAI_HOST local
+azd env set OPENAI_BASE_URL &lt;your local endpoint&gt;
+azd env set AZURE_OPENAI_CHATGPT_MODEL local-model-name</code></pre>
+<p>Then restart the local development server. You should now be able to
+use the “Ask” tab.</p>
+<p>⚠️ Limitations:</p>
+<ul>
+<li>The “Chat” tab will only work if the local language model supports
+function calling.</li>
+<li>Your search mode must be text only (no vectors), since the search
+index is only populated with OpenAI-generated embeddings, and the local
+OpenAI host can’t generate those.</li>
+<li>The conversation history will be truncated using the GPT tokenizers,
+which may not be the same as the local model’s tokenizer, so if you have
+a long conversation, you may end up with token limit errors.</li>
+</ul>
+<blockquote>
+<p>[!NOTE] You must set <code>OPENAI_HOST</code> back to a non-local
+value (“azure”, “azure_custom”, or “openai”) before running
+<code>azd up</code> or <code>azd provision</code>, since the deployed
+backend can’t access your local server.</p>
+</blockquote>
+<h3 id="using-ollama-server">Using Ollama server</h3>
+<p>For example, to point at a local Ollama server running the
+<code>llama3.1:8b</code> model:</p>
+<pre class="shell"><code>azd env set OPENAI_HOST local
+azd env set OPENAI_BASE_URL http://localhost:11434/v1
+azd env set AZURE_OPENAI_CHATGPT_MODEL llama3.1:8b
+azd env set USE_VECTORS false</code></pre>
+<p>If you’re running the app inside a VS Code Dev Container, use this
+local URL instead:</p>
+<pre class="shell"><code>azd env set OPENAI_BASE_URL http://host.docker.internal:11434/v1</code></pre>
+<h3 id="using-llamafile-server">Using llamafile server</h3>
+<p>To point at a local llamafile server running on its default port:</p>
+<pre class="shell"><code>azd env set OPENAI_HOST local
+azd env set OPENAI_BASE_URL http://localhost:8080/v1
+azd env set USE_VECTORS false</code></pre>
+<p>Llamafile does <em>not</em> require a model name to be specified.</p>
+<p>If you’re running the app inside a VS Code Dev Container, use this
+local URL instead:</p>
+<pre class="shell"><code>azd env set OPENAI_BASE_URL http://host.docker.internal:8080/v1</code></pre>
diff --git a/data/login_and_acl.html b/data/login_and_acl.html
new file mode 100644
index 0000000000..35cda45c12
--- /dev/null
+++ b/data/login_and_acl.html
@@ -0,0 +1,689 @@
+<!--
+---
+name: RAG chat with document security
+description: This guide demonstrates how to add an optional login and document level access control system to a RAG chat app for your domain data. This system can be used to restrict access to indexed data to specific users.
+languages:
+- python
+- typescript
+- bicep
+- azdeveloper
+products:
+- azure-openai
+- azure-cognitive-search
+- azure-app-service
+- azure
+page_type: sample
+urlFragment: azure-search-openai-demo-document-security
+---
+-->
+<h1
+id="rag-chat-setting-up-optional-login-and-document-level-access-control">RAG
+chat: Setting up optional login and document level access control</h1>
+<p><a href="https://www.youtube.com/watch?v=GwEiYJgM8Vw">📺 Watch: (RAG
+Deep Dive series) Login and access control</a></p>
+<p>The <a href="/">azure-search-openai-demo</a> project can set up a
+full RAG chat app on Azure AI Search and OpenAI so that you can chat on
+custom data, like internal enterprise data or domain-specific knowledge
+sets. For full instructions on setting up the project, consult the <a
+href="/README.md">main README</a>, and then return here for detailed
+instructions on configuring login and access control.</p>
+<h2 id="table-of-contents">Table of Contents</h2>
+<ul>
+<li><a href="#requirements">Requirements</a></li>
+<li><a href="#setting-up-microsoft-entra-applications">Setting up
+Microsoft Entra applications</a>
+<ul>
+<li><a href="#automatic-setup">Automatic Setup</a></li>
+<li><a href="#manual-setup">Manual Setup</a>
+<ul>
+<li><a href="#server-app">Server App</a></li>
+<li><a href="#client-app">Client App</a></li>
+<li><a href="#configure-server-app-known-client-applications">Configure
+Server App Known Client Applications</a></li>
+<li><a href="#testing">Testing</a></li>
+<li><a href="#programmatic-access-with-authentication">Programmatic
+Access With Authentication</a></li>
+</ul></li>
+<li><a href="#troubleshooting">Troubleshooting</a></li>
+</ul></li>
+<li><a href="#adding-data-with-document-level-access-control">Adding
+data with document level access control</a>
+<ul>
+<li><a href="#using-the-add-documents-api">Using the Add Documents
+API</a></li>
+<li><a href="#azure-data-lake-storage-gen2-setup">Azure Data Lake
+Storage Gen2 and prepdocs</a></li>
+</ul></li>
+<li><a href="#environment-variables-reference">Environment variables
+reference</a>
+<ul>
+<li><a href="#authentication-behavior-by-environment">Authentication
+behavior by environment</a></li>
+</ul></li>
+</ul>
+<p>This guide demonstrates how to add an optional login and document
+level access control system to the sample. This system can be used to
+restrict access to indexed data to specific users based on what <a
+href="https://learn.microsoft.com/entra/fundamentals/how-to-manage-groups">Microsoft
+Entra groups</a> they are a part of, or their <a
+href="https://learn.microsoft.com/partner-center/find-ids-and-domain-names#find-the-user-object-id">user
+object id</a>.</p>
+<figure>
+<img src="/docs/images/applogincomponents.png"
+alt="AppLoginArchitecture" />
+<figcaption aria-hidden="true">AppLoginArchitecture</figcaption>
+</figure>
+<h2 id="requirements">Requirements</h2>
+<p><strong>IMPORTANT:</strong> In order to add optional login and
+document level access control, you’ll need the following in addition to
+the normal sample requirements</p>
+<ul>
+<li><strong>Azure account permissions</strong>: Your Azure account must
+have <a
+href="https://learn.microsoft.com/entra/identity/role-based-access-control/permissions-reference#cloud-application-administrator">permission
+to manage applications in Microsoft Entra</a>.</li>
+</ul>
+<h2 id="setting-up-microsoft-entra-applications">Setting up Microsoft
+Entra applications</h2>
+<p>Two Microsoft Entra applications must be registered in order to make
+the optional login and document level access control system work
+correctly. One app is for the client UI. The client UI is implemented as
+a <a
+href="https://learn.microsoft.com/entra/identity-platform/scenario-spa-app-registration">single
+page application</a>. The other app is for the API server. The API
+server uses a <a
+href="https://learn.microsoft.com/entra/identity-platform/msal-client-applications">confidential
+client</a> to call the <a
+href="https://learn.microsoft.com/graph/use-the-api">Microsoft Graph
+API</a>.</p>
+<h3 id="automatic-setup">Automatic Setup</h3>
+<p>The easiest way to setup the two apps is to use the <code>azd</code>
+CLI. We’ve written scripts that will automatically create the two apps
+and configure them for use with the sample. To trigger the automatic
+setup, run the following commands:</p>
+<ol type="1">
+<li>Run <code>azd env set AZURE_USE_AUTHENTICATION true</code> to enable
+the login UI and use App Service authentication by default.</li>
+<li>Ensure access control is enabled on your search index. If your index
+doesn’t exist yet, run prepdocs with
+<code>AZURE_USE_AUTHENTICATION</code> set to <code>true</code>. If your
+index already exists, run
+<code>python ./scripts/manageacl.py --acl-action enable_acls</code>.</li>
+<li>(Optional) To require access control when using the app, run
+<code>azd env set AZURE_ENFORCE_ACCESS_CONTROL true</code>.
+Authentication is always required to search on documents with access
+control assigned, regardless of if unauthenticated access is enabled or
+not.</li>
+<li>(Optional) To allow authenticated users to search on documents that
+have no access controls assigned, even when access control is required,
+run
+<code>azd env set AZURE_ENABLE_GLOBAL_DOCUMENT_ACCESS true</code>.</li>
+<li>(Optional) To allow unauthenticated users to use the app, even when
+access control is enforced, run
+<code>azd env set AZURE_ENABLE_UNAUTHENTICATED_ACCESS true</code>.
+<code>AZURE_ENABLE_GLOBAL_DOCUMENT_ACCESS</code> should also be set to
+true if you want unauthenticated users to be able to search on documents
+with no access control.</li>
+<li>Run
+<code>azd env set AZURE_AUTH_TENANT_ID &lt;YOUR-TENANT-ID&gt;</code> to
+set the tenant ID associated with authentication.</li>
+<li>If your auth tenant ID is different from your currently logged in
+tenant ID, run
+<code>azd auth login --tenant-id &lt;YOUR-TENANT-ID&gt;</code> to login
+to the authentication tenant simultaneously.</li>
+<li>Run <code>azd up</code> to deploy the app.</li>
+</ol>
+<h3 id="manual-setup">Manual Setup</h3>
+<p>The following instructions explain how to setup the two apps using
+the Azure Portal.</p>
+<h4 id="server-app">Server App</h4>
+<ul>
+<li><p>Sign in to the <a href="https://portal.azure.com/">Azure
+portal</a>.</p></li>
+<li><p>Select the Microsoft Entra ID service.</p></li>
+<li><p>In the left hand menu, select <strong>Application
+Registrations</strong>.</p></li>
+<li><p>Select <strong>New Registration</strong>.</p>
+<ul>
+<li>In the <strong>Name</strong> section, enter a meaningful application
+name. This name will be displayed to users of the app, for example
+<code>Azure Search OpenAI Chat API</code>.</li>
+<li>Under <strong>Supported account types</strong>, select
+<strong>Accounts in this organizational directory only</strong>.</li>
+</ul></li>
+<li><p>Select <strong>Register</strong> to create the
+application</p></li>
+<li><p>In the app’s registration screen, find the <strong>Application
+(client) ID</strong>.</p>
+<ul>
+<li>Run the following <code>azd</code> command to save this ID:
+<code>azd env set AZURE_SERVER_APP_ID &lt;Application (client) ID&gt;</code>.</li>
+</ul></li>
+<li><p>Microsoft Entra supports three types of credentials to
+authenticate an app using the <a
+href="https://learn.microsoft.com/entra/identity-platform/v2-oauth2-client-creds-grant-flow">client
+credentials</a>: passwords (app secrets), certificates, and federated
+identity credentials. For a higher level of security, either <a
+href="https://learn.microsoft.com/entra/identity-platform/howto-create-self-signed-certificate">certificates</a>
+or federated identity credentials are recommended. This sample currently
+uses an app secret for ease of provisioning.</p></li>
+<li><p>Select <strong>Certificates &amp; secrets</strong> in the left
+hand menu.</p></li>
+<li><p>In the <strong>Client secrets</strong> section, select
+<strong>New client secret</strong>.</p>
+<ul>
+<li>Type a description, for example
+<code>Azure Search OpenAI Chat Key</code>.</li>
+<li>Select one of the available key durations.</li>
+<li>The generated key value will be displayed after you select
+<strong>Add</strong>.</li>
+<li>Copy the generated key value and run the following <code>azd</code>
+command to save this ID:
+<code>azd env set AZURE_SERVER_APP_SECRET &lt;generated key value&gt;</code>.</li>
+</ul></li>
+<li><p>Select <strong>API Permissions</strong> in the left hand menu. By
+default, the <a
+href="https://learn.microsoft.com/graph/permissions-reference#user-permissions">delegated
+<code>User.Read</code></a> permission should be present. This permission
+is required to read the signed-in user’s profile to get the security
+information used for document level access control. If this permission
+is not present, it needs to be added to the application.</p>
+<ul>
+<li>Select <strong>Add a permission</strong>, and then <strong>Microsoft
+Graph</strong>.</li>
+<li>Select <strong>Delegated permissions</strong>.</li>
+<li>Search for and and select <code>User.Read</code>.</li>
+<li>Select <strong>Add permissions</strong>.</li>
+</ul></li>
+<li><p>Select <strong>Expose an API</strong> in the left hand menu. The
+server app works by using the <a
+href="https://learn.microsoft.com/entra/identity-platform/v2-oauth2-on-behalf-of-flow#protocol-diagram">On
+Behalf Of Flow</a>, which requires the server app to expose at least 1
+API.</p>
+<ul>
+<li>The application must define a URI to expose APIs. Select
+<strong>Add</strong> next to <strong>Application ID URI</strong>.
+<ul>
+<li>By default, the Application ID URI is set to
+<code>api://&lt;application client id&gt;</code>. Accept the default by
+selecting <strong>Save</strong>.</li>
+</ul></li>
+<li>Under <strong>Scopes defined by this API</strong>, select
+<strong>Add a scope</strong>.</li>
+<li>Fill in the values as indicated:
+<ul>
+<li>For <strong>Scope name</strong>, use
+<strong>access_as_user</strong>.</li>
+<li>For <strong>Who can consent?</strong>, select <strong>Admins and
+users</strong>.</li>
+<li>For <strong>Admin consent display name</strong>, type <strong>Access
+Azure Search OpenAI Chat API</strong>.</li>
+<li>For <strong>Admin consent description</strong>, type <strong>Allows
+the app to access Azure Search OpenAI Chat API as the signed-in
+user.</strong>.</li>
+<li>For <strong>User consent display name</strong>, type <strong>Access
+Azure Search OpenAI Chat API</strong>.</li>
+<li>For <strong>User consent description</strong>, type <strong>Allow
+the app to access Azure Search OpenAI Chat API on your
+behalf</strong>.</li>
+<li>Leave <strong>State</strong> set to <strong>Enabled</strong>.</li>
+<li>Select <strong>Add scope</strong> at the bottom to save the
+scope.</li>
+</ul></li>
+</ul></li>
+<li><p>(Optional) Enable group claims. Include which Microsoft Entra
+groups the user is part of as part of the login in the <a
+href="https://learn.microsoft.com/entra/identity-platform/optional-claims">optional
+claims</a>. The groups are used for <a
+href="https://learn.microsoft.com/azure/search/search-security-trimming-for-azure-search">optional
+security filtering</a> in the search results.</p>
+<ul>
+<li>In the left hand menu, select <strong>Token
+configuration</strong></li>
+<li>Under <strong>Optional claims</strong>, select <strong>Add groups
+claim</strong></li>
+<li>Select which <a
+href="https://learn.microsoft.com/entra/identity/hybrid/connect/how-to-connect-fed-group-claims">group
+types</a> to include in the claim. Note that a <a
+href="https://learn.microsoft.com/entra/identity-platform/access-token-claims-reference#groups-overage-claim">overage
+claim</a> will be emitted if the user is part of too many groups. In
+this case, the API server will use the <a
+href="https://learn.microsoft.com/graph/api/user-list-memberof?view=graph-rest-*0&amp;tabs=http">Microsoft
+Graph</a> to list the groups the user is part of instead of relying on
+the groups in the claim.</li>
+<li>Select <strong>Add</strong> to save your changes</li>
+</ul></li>
+</ul>
+<h4 id="client-app">Client App</h4>
+<ul>
+<li>Sign in to the <a href="https://portal.azure.com/">Azure
+portal</a>.</li>
+<li>Select the Microsoft Entra ID service.</li>
+<li>In the left hand menu, select <strong>Application
+Registrations</strong>.</li>
+<li>Select <strong>New Registration</strong>.
+<ul>
+<li>In the <strong>Name</strong> section, enter a meaningful application
+name. This name will be displayed to users of the app, for example
+<code>Azure Search OpenAI Chat Web App</code>.</li>
+<li>Under <strong>Supported account types</strong>, select
+<strong>Accounts in this organizational directory only</strong>.</li>
+<li>Under <code>Redirect URI (optional)</code> section, select
+<code>Single-page application (SPA)</code> in the combo-box and enter
+the following redirect URI:
+<ul>
+<li>If you are running the sample locally, add the endpoints
+<code>http://localhost:50505/redirect</code> and
+<code>http://localhost:5173/redirect</code></li>
+<li>If you are running the sample on Azure, add the endpoints provided
+by <code>azd up</code>:
+<code>https://&lt;your-endpoint&gt;.azurewebsites.net/redirect</code>.</li>
+<li>If you are running the sample from Github Codespaces, add the
+Codespaces endpoint:
+<code>https://&lt;your-codespace&gt;-50505.app.github.dev/redirect</code></li>
+</ul></li>
+</ul></li>
+<li>Select <strong>Register</strong> to create the application</li>
+<li>In the app’s registration screen, find the <strong>Application
+(client) ID</strong>.
+<ul>
+<li>Run the following <code>azd</code> command to save this ID:
+<code>azd env set AZURE_CLIENT_APP_ID &lt;Application (client) ID&gt;</code>.</li>
+</ul></li>
+<li>In the left hand menu, select <strong>Authentication</strong>.
+<ul>
+<li>Under Web, add a redirect URI with the endpoint provided by
+<code>azd up</code>:
+<code>https://&lt;your-endpoint&gt;.azurewebsites.net/.auth/login/aad/callback</code>.</li>
+<li>Under <strong>Implicit grant and hybrid flows</strong>, select
+<strong>ID Tokens (used for implicit and hybrid flows)</strong></li>
+<li>Select <strong>Save</strong></li>
+</ul></li>
+<li>In the left hand menu, select <strong>API permissions</strong>. You
+will add permission to access the <strong>access_as_user</strong> API on
+the server app. This permission is required for the <a
+href="https://learn.microsoft.com/entra/identity-platform/v2-oauth2-on-behalf-of-flow#protocol-diagram">On
+Behalf Of Flow</a> to work.
+<ul>
+<li>Select <strong>Add a permission</strong>, and then <strong>My
+APIs</strong>.</li>
+<li>In the list of applications, select your server application
+<strong>Azure Search OpenAI Chat API</strong></li>
+<li>Ensure <strong>Delegated permissions</strong> is selected.</li>
+<li>In the <strong>Select permissions</strong> section, select the
+<strong>access_as_user</strong> permission</li>
+<li>Select <strong>Add permissions</strong>.</li>
+</ul></li>
+<li>Stay in the <strong>API permissions</strong> section and select
+<strong>Add a permission</strong>.
+<ul>
+<li>Select <strong>Microsoft Graph</strong>.</li>
+<li>Select <strong>Delegated permissions</strong>.</li>
+<li>Search for and select <code>User.Read</code>.</li>
+<li>Select <strong>Add permissions</strong>.</li>
+</ul></li>
+</ul>
+<h4 id="configure-server-app-known-client-applications">Configure Server
+App Known Client Applications</h4>
+<p>Consent from the user must be obtained for use of the client and
+server app. The client app can prompt the user for consent through a
+dialog when they log in. The server app has no ability to show a dialog
+for consent. Client apps can be <a
+href="https://learn.microsoft.com/entra/identity-platform/v2-oauth2-on-behalf-of-flow#gaining-consent-for-the-middle-tier-application">added
+to the list of known clients</a> to access the server app, so a consent
+dialog is shown for the server app.</p>
+<ul>
+<li>Navigate to the server app registration</li>
+<li>In the left hand menu, select <strong>Manifest</strong></li>
+<li>Replace <code>"knownClientApplications": []</code> with
+<code>"knownClientApplications": ["&lt;client application id&gt;"]</code></li>
+<li>Select <strong>Save</strong></li>
+</ul>
+<h4 id="testing">Testing</h4>
+<p>If you are running setup for the first time, ensure you have run
+<code>azd env set AZURE_ADLS_GEN2_STORAGE_ACCOUNT &lt;YOUR-STORAGE_ACCOUNT&gt;</code>
+before running <code>azd up</code>. If you do not set this environment
+variable, your index will not be initialized with access control support
+when <code>prepdocs</code> is run for the first time. To manually enable
+access control in your index, use the <a
+href="#azure-data-lake-storage-gen2-setup">manual setup script</a>.</p>
+<p>Ensure you run <code>azd env set AZURE_USE_AUTHENTICATION</code> to
+enable the login UI once you have setup the two Microsoft Entra apps
+before you deploy or run the application. The login UI will not appear
+unless all <a href="#environment-variables-reference">required
+environment variables</a> have been setup.</p>
+<p>In both the chat and ask a question modes, under <strong>Developer
+settings</strong> optional <strong>Use oid security filter</strong> and
+<strong>Use groups security filter</strong> checkboxes will appear. The
+oid (User ID) filter maps to the <code>oids</code> field in the search
+index and the groups (Group ID) filter maps to the <code>groups</code>
+field in the search index. If <code>AZURE_ENFORCE_ACCESS_CONTROL</code>
+has been set, then both the <strong>Use oid security filter</strong> and
+<strong>Use groups security filter</strong> options are always enabled
+and cannot be disabled.</p>
+<h4 id="programmatic-access-with-authentication">Programmatic Access
+with Authentication</h4>
+<p>If you want to use the chat endpoint without the UI and still use
+authentication, you must disable <a
+href="https://learn.microsoft.com/azure/app-service/overview-authentication-authorization">App
+Service built-in authentication</a> and use only the app’s MSAL-based
+authentication flow. Ensure the
+<code>AZURE_DISABLE_APP_SERVICES_AUTHENTICATION</code> environment
+variable is set before deploying.</p>
+<p>Get an access token that can be used for calling the chat API using
+the following code:</p>
+<div class="sourceCode" id="cb1"><pre
+class="sourceCode python"><code class="sourceCode python"><span id="cb1-1"><a href="#cb1-1" aria-hidden="true" tabindex="-1"></a><span class="im">from</span> azure.identity <span class="im">import</span> DefaultAzureCredential</span>
+<span id="cb1-2"><a href="#cb1-2" aria-hidden="true" tabindex="-1"></a><span class="im">import</span> os</span>
+<span id="cb1-3"><a href="#cb1-3" aria-hidden="true" tabindex="-1"></a></span>
+<span id="cb1-4"><a href="#cb1-4" aria-hidden="true" tabindex="-1"></a>token <span class="op">=</span> DefaultAzureCredential().get_token(<span class="ss">f&quot;api://</span><span class="sc">{</span>os<span class="sc">.</span>environ[<span class="st">&#39;AZURE_SERVER_APP_ID&#39;</span>]<span class="sc">}</span><span class="ss">/access_as_user&quot;</span>, tenant_id<span class="op">=</span>os.getenv(<span class="st">&#39;AZURE_AUTH_TENANT_ID&#39;</span>, os.getenv(<span class="st">&#39;AZURE_TENANT_ID&#39;</span>)))</span>
+<span id="cb1-5"><a href="#cb1-5" aria-hidden="true" tabindex="-1"></a></span>
+<span id="cb1-6"><a href="#cb1-6" aria-hidden="true" tabindex="-1"></a><span class="bu">print</span>(token.token)</span></code></pre></div>
+<h3 id="troubleshooting">Troubleshooting</h3>
+<ul>
+<li>If your primary tenant restricts the ability to create Entra
+applications, you’ll need to use a separate tenant to create the Entra
+applications. You can create a new tenant by following <a
+href="https://learn.microsoft.com/entra/identity-platform/quickstart-create-new-tenant">these
+instructions</a>. Then run
+<code>azd env set AZURE_AUTH_TENANT_ID &lt;YOUR-AUTH-TENANT-ID&gt;</code>
+before running <code>azd up</code>.</li>
+<li>If any Entra apps need to be recreated, you can avoid redeploying
+the app by <a
+href="https://learn.microsoft.com/azure/app-service/configure-common?tabs=portal#configure-app-settings">changing
+the app settings in the portal</a>. Any of the <a
+href="#environment-variables-reference">required environment
+variables</a> can be changed. Once the environment variables have been
+changed, restart the web app.</li>
+<li>It’s possible a consent dialog will not appear when you log into the
+app for the first time. If this consent dialog doesn’t appear, you will
+be unable to use the security filters because the API server app does
+not have permission to read your authorization information. A consent
+dialog can be forced to appear by adding
+<code>"prompt": "consent"</code> to the <code>loginRequest</code>
+property in <a
+href="/app/backend/core/authentication.py"><code>authentication.py</code></a></li>
+<li>It’s possible that your tenant admin has placed a restriction on
+consent to apps with <a
+href="https://learn.microsoft.com/entra/identity-platform/publisher-verification-overview">unverified
+publishers</a>. In this case, only admins may consent to the client and
+server apps, and normal user accounts are unable to use the login system
+until the admin consents on behalf of the entire organization.</li>
+<li>It’s possible that your tenant admin requires <a
+href="https://learn.microsoft.com/entra/identity/enterprise-apps/manage-consent-requests">admin
+approval of all new apps</a>. Regardless of whether you select the
+delegated or admin permissions, the app will not work without tenant
+admin consent. See this guide for <a
+href="https://learn.microsoft.com/entra/identity/enterprise-apps/grant-admin-consent?pivots=portal">granting
+consent to an app</a>.</li>
+</ul>
+<h2 id="adding-data-with-document-level-access-control">Adding data with
+document level access control</h2>
+<p>The sample supports 2 main strategies for adding data with document
+level access control.</p>
+<ul>
+<li><a href="#using-the-add-documents-api">Using the Add Documents
+API</a>. Sample scripts are provided which use the Azure AI Search
+Service Add Documents API to directly manage access control information
+on <em>existing documents</em> in the index.</li>
+<li><a href="#azure-data-lake-storage-gen2-setup">Using prepdocs and
+Azure Data Lake Storage Gen 2</a>. Sample scripts are provided which set
+up an <a
+href="https://learn.microsoft.com/azure/storage/blobs/data-lake-storage-introduction">Azure
+Data Lake Storage Gen 2</a> account, set the <a
+href="https://learn.microsoft.com/azure/storage/blobs/data-lake-storage-access-control">access
+control information</a> on files and folders stored there, and ingest
+those documents into the search index with their access control
+information.</li>
+</ul>
+<h3 id="using-the-add-documents-api">Using the Add Documents API</h3>
+<p>Manually enable document level access control on a search index and
+manually set access control values using the <a
+href="/scripts/manageacl.py">manageacl.py</a> script.</p>
+<p>Prior to running the script:</p>
+<ul>
+<li>Run <code>azd up</code> or use <code>azd env set</code> to manually
+set the <code>AZURE_SEARCH_SERVICE</code> and
+<code>AZURE_SEARCH_INDEX</code> azd environment variables</li>
+<li>Activate the Python virtual environment for your shell session</li>
+</ul>
+<p>The script supports the following commands. All commands support
+<code>-v</code> for verbose logging.</p>
+<ul>
+<li><p><code>python ./scripts/manageacl.py --acl-action enable_acls</code>:
+Creates the required <code>oids</code> (User ID) and <code>groups</code>
+(Group IDs) <a
+href="https://learn.microsoft.com/azure/search/search-security-trimming-for-azure-search">security
+filter</a> fields for document level access control on your index, as
+well as the <code>storageUrl</code> field for storing the Blob storage
+URL. Does nothing if these fields already exist.</p>
+<p>Example usage:</p>
+<pre class="shell"><code>python ./scripts/manageacl.py -v --acl-action enable_acls</code></pre></li>
+<li><p><code>python ./scripts/manageacl.py --acl-type [oids or groups]--acl-action view --url [https://url.pdf]</code>:
+Prints access control values associated with either User IDs or Group
+IDs for the document at the specified URL.</p>
+<p>Example to view all Group IDs:</p>
+<pre class="shell"><code>python ./scripts/manageacl.py -v --acl-type groups --acl-action view --url https://st12345.blob.core.windows.net/content/Benefit_Options.pdf</code></pre></li>
+<li><p><code>python ./scripts/manageacl.py --acl-type [oids or groups]--acl-action add --acl [ID of group or user] --url [https://url.pdf]</code>:
+Adds an access control value associated with either User IDs or Group
+IDs for the document at the specified URL.</p>
+<p>Example to add a Group ID:</p>
+<pre class="shell"><code>python ./scripts/manageacl.py -v --acl-type groups --acl-action add --acl xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx --url https://st12345.blob.core.windows.net/content/Benefit_Options.pdf</code></pre></li>
+<li><p><code>python ./scripts/manageacl.py --acl-type [oids or groups]--acl-action remove_all --url [https://url.pdf]</code>:
+Removes all access control values associated with either User IDs or
+Group IDs for a specific document.</p>
+<p>Example to remove all Group IDs:</p>
+<pre class="shell"><code>python ./scripts/manageacl.py -v --acl-type groups --acl-action remove_all --url https://st12345.blob.core.windows.net/content/Benefit_Options.pdf</code></pre></li>
+<li><p><code>python ./scripts/manageacl.py --url [https://url.pdf] --acl-type [oids or groups]--acl-action remove --acl [ID of group or user]</code>:
+Removes an access control value associated with either User IDs or Group
+IDs for a specific document.</p>
+<p>Example to remove a specific User ID:</p>
+<pre class="shell"><code>python ./scripts/manageacl.py -v --acl-type oids --acl-action remove --acl xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx --url https://st12345.blob.core.windows.net/content/Benefit_Options.pdf</code></pre></li>
+</ul>
+<h3 id="azure-data-lake-storage-gen2-setup">Azure Data Lake Storage Gen2
+Setup</h3>
+<p><a
+href="https://learn.microsoft.com/azure/storage/blobs/data-lake-storage-introduction">Azure
+Data Lake Storage Gen2</a> implements an <a
+href="https://learn.microsoft.com/azure/storage/blobs/data-lake-storage-access-control">access
+control model</a> that can be used for document level access control.
+The <a href="/scripts/adlsgen2setup.py">adlsgen2setup.py</a> script
+uploads the sample data included in the <a href="./data">data</a> folder
+to a Data Lake Storage Gen2 storage account. The <a
+href="https://learn.microsoft.com/azure/storage/blobs/data-lake-storage-access-control-model#role-based-access-control-azure-rbac">Storage
+Blob Data Owner</a> role is required to use the script.</p>
+<p>In order to use this script, an existing Data Lake Storage Gen2
+storage account is required. Run
+<code>azd env set AZURE_ADLS_GEN2_STORAGE_ACCOUNT &lt;your-storage-account&gt;</code>
+prior to running the script.</p>
+<p>Then run the script inside your Python environment:</p>
+<pre class="shell"><code>python /scripts/adlsgen2setup.py &#39;./data/*&#39; --data-access-control &#39;./scripts/sampleacls.json&#39; -v</code></pre>
+<p>The script performs the following steps:</p>
+<ul>
+<li>Creates example <a
+href="https://learn.microsoft.com/entra/fundamentals/how-to-manage-groups">groups</a>
+listed in the <a href="/scripts/sampleacls.json">sampleacls.json</a>
+file.</li>
+<li>Creates a filesystem / container <code>gptkbcontainer</code> in the
+storage account.</li>
+<li>Creates the directories listed in the <a
+href="/scripts/sampleacls.json">sampleacls.json</a> file.</li>
+<li>Uploads the sample PDFs referenced in the <a
+href="/scripts/sampleacls.json">sampleacls.json</a> file into the
+appropriate directories.</li>
+<li><a
+href="https://learn.microsoft.com/azure/storage/blobs/data-lake-storage-acl-cli">Recursively
+sets Access Control Lists (ACLs)</a> using the information from the <a
+href="/scripts/sampleacls.json">sampleacls.json</a> file.</li>
+</ul>
+<p>In order to use the sample access control, you need to join these
+groups in your Microsoft Entra tenant.</p>
+<p>Note that this optional script may not work in Codespaces if your
+administrator has applied a <a
+href="https://learn.microsoft.com/entra/identity/conditional-access/overview">Conditional
+Access policy</a> to your tenant.</p>
+<h4 id="azure-data-lake-storage-gen2-prep-docs">Azure Data Lake Storage
+Gen2 Prep Docs</h4>
+<p>Once a Data Lake Storage Gen2 storage account has been setup with
+sample data and access control lists, <a
+href="/app/backend/prepdocs.py">prepdocs.py</a> can be used to
+automatically process PDFs in the storage account and store them with
+their <a
+href="https://learn.microsoft.com/azure/storage/blobs/data-lake-storage-access-control">access
+control lists in the search index</a>.</p>
+<p>To run this script with a Data Lake Storage Gen2 account, first set
+the following environment variables:</p>
+<ul>
+<li><code>AZURE_ADLS_GEN2_STORAGE_ACCOUNT</code>: Name of existing <a
+href="https://learn.microsoft.com/azure/storage/blobs/data-lake-storage-introduction">Data
+Lake Storage Gen2 storage account</a>.</li>
+<li>(Optional) <code>AZURE_ADLS_GEN2_FILESYSTEM</code>: Name of existing
+Data Lake Storage Gen2 filesystem / container in the storage account. If
+empty, <code>gptkbcontainer</code> is used.</li>
+<li>(Optional) <code>AZURE_ADLS_GEN2_FILESYSTEM_PATH</code>: Specific
+path in the Data Lake Storage Gen2 filesystem / container to process.
+Only PDFs contained in this path will be processed.</li>
+</ul>
+<p>Once the environment variables are set, run the script using the
+following command: <code>/scripts/prepdocs.ps1</code> or
+<code>/scripts/prepdocs.sh</code>.</p>
+<h2 id="environment-variables-reference">Environment variables
+reference</h2>
+<p>The following environment variables are used to setup the optional
+login and document level access control:</p>
+<ul>
+<li><code>AZURE_USE_AUTHENTICATION</code>: Enables Entra ID login and
+document level access control. Set to true before running
+<code>azd up</code>.</li>
+<li><code>AZURE_ENFORCE_ACCESS_CONTROL</code>: Enforces Entra ID based
+login and document level access control on documents with access control
+assigned. Set to true before running <code>azd up</code>. If
+<code>AZURE_ENFORCE_ACCESS_CONTROL</code> is enabled and
+<code>AZURE_ENABLE_UNAUTHENTICATED_ACCESS</code> is not enabled, then
+authentication is required to use the app.</li>
+<li><code>AZURE_ENABLE_GLOBAL_DOCUMENT_ACCESS</code>: Allows users to
+search on documents that have no access controls assigned</li>
+<li><code>AZURE_ENABLE_UNAUTHENTICATED_ACCESS</code>: Allows
+unauthenticated users to access the chat app, even when
+<code>AZURE_ENFORCE_ACCESS_CONTROL</code> is enabled.
+<code>AZURE_ENABLE_GLOBAL_DOCUMENT_ACCESS</code> should be set to true
+to allow unauthenticated users to search on documents that have no
+access control assigned. Unauthenticated users cannot search on
+documents with access control assigned.</li>
+<li><code>AZURE_DISABLE_APP_SERVICES_AUTHENTICATION</code>: Disables <a
+href="https://learn.microsoft.com/azure/app-service/overview-authentication-authorization">use
+of built-in authentication for App Services</a>. An authentication flow
+based on the MSAL SDKs is used instead. Useful when you want to provide
+programmatic access to the chat endpoints with authentication.</li>
+<li><code>AZURE_SERVER_APP_ID</code>: (Required) Application ID of the
+Microsoft Entra app for the API server.</li>
+<li><code>AZURE_SERVER_APP_SECRET</code>: <a
+href="https://learn.microsoft.com/entra/identity-platform/v2-oauth2-client-creds-grant-flow">Client
+secret</a> used by the API server to authenticate using the Microsoft
+Entra server app.</li>
+<li><code>AZURE_CLIENT_APP_ID</code>: Application ID of the Microsoft
+Entra app for the client UI.</li>
+<li><code>AZURE_AUTH_TENANT_ID</code>: <a
+href="https://learn.microsoft.com/entra/fundamentals/how-to-find-tenant">Tenant
+ID</a> associated with the Microsoft Entra tenant used for login and
+document level access control. Defaults to <code>AZURE_TENANT_ID</code>
+if not defined.</li>
+<li><code>AZURE_ADLS_GEN2_STORAGE_ACCOUNT</code>: (Optional) Name of
+existing <a
+href="https://learn.microsoft.com/azure/storage/blobs/data-lake-storage-introduction">Data
+Lake Storage Gen2 storage account</a> for storing sample data with <a
+href="https://learn.microsoft.com/azure/storage/blobs/data-lake-storage-access-control">access
+control lists</a>. Only used with the optional Data Lake Storage Gen2 <a
+href="#azure-data-lake-storage-gen2-setup">setup</a> and <a
+href="#azure-data-lake-storage-gen2-prep-docs">prep docs</a>
+scripts.</li>
+<li><code>AZURE_ADLS_GEN2_FILESYSTEM</code>: (Optional) Name of existing
+<a
+href="https://learn.microsoft.com/azure/storage/blobs/data-lake-storage-introduction">Data
+Lake Storage Gen2 filesystem</a> for storing sample data with <a
+href="https://learn.microsoft.com/azure/storage/blobs/data-lake-storage-access-control">access
+control lists</a>. Only used with the optional Data Lake Storage Gen2 <a
+href="#azure-data-lake-storage-gen2-setup">setup</a> and <a
+href="#azure-data-lake-storage-gen2-prep-docs">prep docs</a>
+scripts.</li>
+<li><code>AZURE_ADLS_GEN2_FILESYSTEM_PATH</code>: (Optional) Name of
+existing path in a <a
+href="https://learn.microsoft.com/azure/storage/blobs/data-lake-storage-introduction">Data
+Lake Storage Gen2 filesystem</a> for storing sample data with <a
+href="https://learn.microsoft.com/azure/storage/blobs/data-lake-storage-access-control">access
+control lists</a>. Only used with the optional Data Lake Storage Gen2 <a
+href="#azure-data-lake-storage-gen2-prep-docs">prep docs</a>
+script.</li>
+</ul>
+<h3 id="authentication-behavior-by-environment">Authentication behavior
+by environment</h3>
+<p>This application uses an in-memory token cache. User sessions are
+only available in memory while the application is running. When the
+application server is restarted, all users will need to log-in
+again.</p>
+<p>The following table describes the impact of the
+<code>AZURE_USE_AUTHENTICATION</code> and
+<code>AZURE_ENFORCE_ACCESS_CONTROL</code> variables depending on the
+environment you are deploying the application in:</p>
+<table>
+<colgroup>
+<col style="width: 25%" />
+<col style="width: 25%" />
+<col style="width: 25%" />
+<col style="width: 25%" />
+</colgroup>
+<thead>
+<tr>
+<th>AZURE_USE_AUTHENTICATION</th>
+<th>AZURE_ENFORCE_ACCESS_CONTROL</th>
+<th>Environment</th>
+<th>Default Behavior</th>
+</tr>
+</thead>
+<tbody>
+<tr>
+<td>True</td>
+<td>False</td>
+<td>App Services</td>
+<td>Use integrated auth <br /> Login page blocks access to app <br />
+User can opt-into access control in developer settings <br /> Allows
+unrestricted access to sources</td>
+</tr>
+<tr>
+<td>True</td>
+<td>True</td>
+<td>App Services</td>
+<td>Use integrated auth <br /> Login page blocks access to app <br />
+User must use access control</td>
+</tr>
+<tr>
+<td>True</td>
+<td>False</td>
+<td>Local or Codespaces</td>
+<td>Do not use integrated auth <br /> Can use app without login <br />
+User can opt-into access control in developer settings <br /> Allows
+unrestricted access to sources</td>
+</tr>
+<tr>
+<td>True</td>
+<td>True</td>
+<td>Local or Codespaces</td>
+<td>Do not use integrated auth <br /> Cannot use app without login
+<br /> Behavior is chat box is greyed out with default “Please login
+message” <br /> User must use login button to make chat box usable
+<br /> User must use access control when logged in</td>
+</tr>
+<tr>
+<td>False</td>
+<td>False</td>
+<td>All</td>
+<td>No login or access control</td>
+</tr>
+<tr>
+<td>False</td>
+<td>True</td>
+<td>All</td>
+<td>Invalid setting</td>
+</tr>
+</tbody>
+</table>
diff --git a/data/monitoring.html b/data/monitoring.html
new file mode 100644
index 0000000000..ab9016f90b
--- /dev/null
+++ b/data/monitoring.html
@@ -0,0 +1,73 @@
+<h1 id="rag-chat-monitoring-with-application-insights">RAG chat:
+Monitoring with Application Insights</h1>
+<p>By default, deployed apps use Application Insights for the tracing of
+each request, along with the logging of errors.</p>
+<ul>
+<li><a href="#performance">Performance</a></li>
+<li><a href="#failures">Failures</a></li>
+<li><a href="#dashboard">Dashboard</a></li>
+<li><a href="#customizing-the-traces">Customizing the traces</a></li>
+</ul>
+<h2 id="performance">Performance</h2>
+<p>To see the performance data, go to the Application Insights resource
+in your resource group, click on the “Investigate -&gt; Performance”
+blade and navigate to any HTTP request to see the timing data. To
+inspect the performance of chat requests, use the “Drill into Samples”
+button to see end-to-end traces of all the API calls made for any chat
+request:</p>
+<figure>
+<img src="images/transaction-tracing.png" alt="Tracing screenshot" />
+<figcaption aria-hidden="true">Tracing screenshot</figcaption>
+</figure>
+<h2 id="failures">Failures</h2>
+<p>To see any exceptions and server errors, navigate to the “Investigate
+-&gt; Failures” blade and use the filtering tools to locate a specific
+exception. You can see Python stack traces on the right-hand side.</p>
+<h2 id="dashboard">Dashboard</h2>
+<p>You can see chart summaries on a dashboard by running the following
+command:</p>
+<pre class="shell"><code>azd monitor</code></pre>
+<p>You can modify the contents of that dashboard by updating
+<code>infra/backend-dashboard.bicep</code>, which is a Bicep file that
+defines the dashboard contents and layout.</p>
+<h2 id="customizing-the-traces">Customizing the traces</h2>
+<p>The tracing is done using these OpenTelemetry Python packages:</p>
+<ul>
+<li><a
+href="https://pypi.org/project/azure-monitor-opentelemetry/">azure-monitor-opentelemetry</a></li>
+<li><a
+href="https://pypi.org/project/opentelemetry-instrumentation-asgi/">opentelemetry-instrumentation-asgi</a></li>
+<li><a
+href="https://pypi.org/project/opentelemetry-instrumentation-httpx/">opentelemetry-instrumentation-httpx</a></li>
+<li><a
+href="https://pypi.org/project/opentelemetry-instrumentation-aiohttp-client/">opentelemetry-instrumentation-aiohttp-client</a></li>
+<li><a
+href="https://pypi.org/project/opentelemetry-instrumentation-openai/">opentelemetry-instrumentation-openai</a></li>
+</ul>
+<p>Those packages are configured in the <code>app.py</code> file:</p>
+<div class="sourceCode" id="cb2"><pre
+class="sourceCode python"><code class="sourceCode python"><span id="cb2-1"><a href="#cb2-1" aria-hidden="true" tabindex="-1"></a><span class="cf">if</span> os.getenv(<span class="st">&quot;APPLICATIONINSIGHTS_CONNECTION_STRING&quot;</span>):</span>
+<span id="cb2-2"><a href="#cb2-2" aria-hidden="true" tabindex="-1"></a>    configure_azure_monitor()</span>
+<span id="cb2-3"><a href="#cb2-3" aria-hidden="true" tabindex="-1"></a>    <span class="co"># This tracks HTTP requests made by aiohttp:</span></span>
+<span id="cb2-4"><a href="#cb2-4" aria-hidden="true" tabindex="-1"></a>    AioHttpClientInstrumentor().instrument()</span>
+<span id="cb2-5"><a href="#cb2-5" aria-hidden="true" tabindex="-1"></a>    <span class="co"># This tracks HTTP requests made by httpx:</span></span>
+<span id="cb2-6"><a href="#cb2-6" aria-hidden="true" tabindex="-1"></a>    HTTPXClientInstrumentor().instrument()</span>
+<span id="cb2-7"><a href="#cb2-7" aria-hidden="true" tabindex="-1"></a>    <span class="co"># This tracks OpenAI SDK requests:</span></span>
+<span id="cb2-8"><a href="#cb2-8" aria-hidden="true" tabindex="-1"></a>    OpenAIInstrumentor().instrument()</span>
+<span id="cb2-9"><a href="#cb2-9" aria-hidden="true" tabindex="-1"></a>    <span class="co"># This middleware tracks app route requests:</span></span>
+<span id="cb2-10"><a href="#cb2-10" aria-hidden="true" tabindex="-1"></a>    app.asgi_app <span class="op">=</span> OpenTelemetryMiddleware(app.asgi_app)</span></code></pre></div>
+<p>You can pass in parameters to <code>configure_azure_monitor()</code>
+to customize the tracing, like to add custom span processors. You can
+also set <a
+href="https://opentelemetry.io/docs/reference/specification/sdk-environment-variables/">OpenTelemetry
+environment variables</a> to customize the tracing, like to set the
+sampling rate. See the <a
+href="https://pypi.org/project/azure-monitor-opentelemetry/">azure-monitor-opentelemetry</a>
+documentation for more details.</p>
+<p>By default, <a
+href="https://pypi.org/project/opentelemetry-instrumentation-openai/">opentelemetry-instrumentation-openai</a>
+traces all requests made to the OpenAI API, including the messages and
+responses. To disable that for privacy reasons, set the
+<code>TRACELOOP_TRACE_CONTENT=false</code> environment variable.</p>
+<p>To set environment variables, update <code>appEnvVariables</code> in
+<code>infra/main.bicep</code> and re-run <code>azd up</code>.</p>
diff --git a/data/other_samples.html b/data/other_samples.html
new file mode 100644
index 0000000000..715f29e12c
--- /dev/null
+++ b/data/other_samples.html
@@ -0,0 +1,139 @@
+<h1 id="rag-chat-alternative-rag-chat-samples">RAG chat: Alternative RAG
+chat samples</h1>
+<p>There are an increasingly large number of ways to build RAG chat
+apps.</p>
+<h2 id="other-language-samples">Other language samples</h2>
+<p>Inspired by this repo, there are similar apps for other
+languages:</p>
+<ul>
+<li><a
+href="https://aka.ms/azai/js/code"><strong>JavaScript</strong></a></li>
+<li><a
+href="https://aka.ms/azai/net/code"><strong>.NET</strong></a></li>
+<li><a
+href="https://aka.ms/azai/java/code"><strong>Java</strong></a></li>
+</ul>
+<p>They do not all support the same features as this repo, but they
+provide a good starting point for building a RAG chat app in your
+preferred language.</p>
+<h2 id="other-python-samples">Other Python samples</h2>
+<p>Another popular repository for this use case is here: <a
+href="https://github.com/Microsoft/sample-app-aoai-chatGPT/">https://github.com/Microsoft/sample-app-aoai-chatGPT/</a></p>
+<p>That repository is designed for use by customers using Azure OpenAI
+studio and Azure Portal for setup. It also includes <code>azd</code>
+support for folks who want to deploy it completely from scratch.</p>
+<p>The primary differences:</p>
+<ul>
+<li>This repository includes multiple RAG (retrieval-augmented
+generation) approaches that chain the results of multiple API calls (to
+Azure OpenAI and ACS) together in different ways. The other repository
+uses only the built-in data sources option for the ChatCompletions API,
+which uses a RAG approach on the specified ACS index. That should work
+for most uses, but if you needed more flexibility, this sample may be a
+better option.</li>
+<li>This repository is also a bit more experimental in other ways, since
+it’s not tied to the Azure OpenAI Studio like the other repository.</li>
+</ul>
+<p>Feature comparison:</p>
+<table>
+<colgroup>
+<col style="width: 33%" />
+<col style="width: 33%" />
+<col style="width: 33%" />
+</colgroup>
+<thead>
+<tr>
+<th>Feature</th>
+<th>azure-search-openai-demo</th>
+<th>sample-app-aoai-chatGPT</th>
+</tr>
+</thead>
+<tbody>
+<tr>
+<td>RAG approach</td>
+<td>Multiple approaches</td>
+<td>Only via ChatCompletion API data_sources</td>
+</tr>
+<tr>
+<td>Vector support</td>
+<td>✅ Yes</td>
+<td>✅ Yes</td>
+</tr>
+<tr>
+<td>Data ingestion</td>
+<td>✅ Yes (<a href="data_ingestion.md#supported-document-formats">Many
+formats</a>)</td>
+<td>✅ Yes (<a
+href="https://learn.microsoft.com/azure/ai-services/openai/concepts/use-your-data?tabs=ai-search#data-formats-and-file-types">Many
+formats</a>)</td>
+</tr>
+<tr>
+<td>Persistent chat history</td>
+<td>✅ Yes</td>
+<td>✅ Yes</td>
+</tr>
+<tr>
+<td>User feedback</td>
+<td>❌ No</td>
+<td>✅ Yes</td>
+</tr>
+<tr>
+<td>GPT-4-vision</td>
+<td>✅ Yes</td>
+<td>❌ No</td>
+</tr>
+<tr>
+<td>Auth + ACL</td>
+<td>✅ Yes</td>
+<td>✅ Yes</td>
+</tr>
+<tr>
+<td>User upload</td>
+<td>✅ Yes</td>
+<td>❌ No</td>
+</tr>
+<tr>
+<td>Speech I/O</td>
+<td>✅ Yes</td>
+<td>❌ No</td>
+</tr>
+</tbody>
+</table>
+<p>Technology comparison:</p>
+<table>
+<colgroup>
+<col style="width: 33%" />
+<col style="width: 33%" />
+<col style="width: 33%" />
+</colgroup>
+<thead>
+<tr>
+<th>Tech</th>
+<th>azure-search-openai-demo</th>
+<th>sample-app-aoai-chatGPT</th>
+</tr>
+</thead>
+<tbody>
+<tr>
+<td>Frontend</td>
+<td>React</td>
+<td>React</td>
+</tr>
+<tr>
+<td>Backend</td>
+<td>Python (Quart)</td>
+<td>Python (Quart)</td>
+</tr>
+<tr>
+<td>Vector DB</td>
+<td>Azure AI Search</td>
+<td>Azure AI Search, CosmosDB Mongo vCore, ElasticSearch, Pinecone,
+AzureML</td>
+</tr>
+<tr>
+<td>Deployment</td>
+<td>Azure Developer CLI (azd)</td>
+<td>Azure Portal, az, azd</td>
+</tr>
+</tbody>
+</table>
diff --git a/data/productionizing.html b/data/productionizing.html
new file mode 100644
index 0000000000..3e1bf5f62e
--- /dev/null
+++ b/data/productionizing.html
@@ -0,0 +1,157 @@
+<h1 id="rag-chat-productionizing-the-app">RAG chat: Productionizing the
+app</h1>
+<p>This sample is designed to be a starting point for your own
+production application, but you should do a thorough review of the
+security and performance before deploying to production. Here are some
+things to consider:</p>
+<ul>
+<li><a href="#azure-resource-configuration">Azure resource
+configuration</a></li>
+<li><a href="#additional-security-measures">Additional security
+measures</a></li>
+<li><a href="#load-testing">Load testing</a></li>
+<li><a href="#evaluation">Evaluation</a></li>
+</ul>
+<h2 id="azure-resource-configuration">Azure resource configuration</h2>
+<h3 id="openai-capacity">OpenAI Capacity</h3>
+<p>The default TPM (tokens per minute) is set to 30K. That is equivalent
+to approximately 30 conversations per minute (assuming 1K per user
+message/response). You can increase the capacity by changing the
+<code>chatGptDeploymentCapacity</code> and
+<code>embeddingDeploymentCapacity</code> parameters in
+<code>infra/main.bicep</code> to your account’s maximum capacity. You
+can also view the Quotas tab in <a href="https://oai.azure.com/">Azure
+OpenAI studio</a> to understand how much capacity you have.</p>
+<p>If the maximum TPM isn’t enough for your expected load, you have a
+few options:</p>
+<ul>
+<li><p>Use a backoff mechanism to retry the request. This is helpful if
+you’re running into a short-term quota due to bursts of activity but
+aren’t over long-term quota. The <a
+href="https://tenacity.readthedocs.io/en/latest/">tenacity</a> library
+is a good option for this, and this <a
+href="https://github.com/Azure-Samples/azure-search-openai-demo/pull/500">pull
+request</a> shows how to apply it to this app.</p></li>
+<li><p>If you are consistently going over the TPM, then consider
+implementing a load balancer between OpenAI instances. Most developers
+implement that using Azure API Management or container-based load
+balancers. A native Python approach that integrates with the OpenAI
+Python API Library is also possible. For integration instructions with
+this sample, please check:</p>
+<ul>
+<li><a
+href="https://learn.microsoft.com/azure/developer/python/get-started-app-chat-scaling-with-azure-api-management">Scale
+Azure OpenAI for Python with Azure API Management</a></li>
+<li><a
+href="https://learn.microsoft.com/azure/developer/python/get-started-app-chat-scaling-with-azure-container-apps">Scale
+Azure OpenAI for Python chat using RAG with Azure Container
+Apps</a></li>
+<li><a
+href="https://github.com/Azure-Samples/azure-search-openai-demo/pull/1626">Pull
+request: Scale Azure OpenAI for Python with the Python
+openai-priority-loadbalancer</a></li>
+</ul></li>
+</ul>
+<h3 id="azure-storage">Azure Storage</h3>
+<p>The default storage account uses the <code>Standard_LRS</code> SKU.
+To improve your resiliency, we recommend using <code>Standard_ZRS</code>
+for production deployments, which you can specify using the
+<code>sku</code> property under the <code>storage</code> module in
+<code>infra/main.bicep</code>.</p>
+<h3 id="azure-ai-search">Azure AI Search</h3>
+<p>The default search service uses the “Basic” SKU with the free
+semantic ranker option, which gives you 1000 free queries a month. After
+1000 queries, you will get an error message about exceeding the semantic
+ranker free capacity.</p>
+<ul>
+<li><p>Assuming your app will experience more than 1000 questions per
+month, you should upgrade the semantic ranker SKU from “free” to
+“standard” SKU:</p>
+<pre class="shell"><code>azd env set AZURE_SEARCH_SEMANTIC_RANKER standard</code></pre>
+<p>Or disable semantic search entirely:</p>
+<pre class="shell"><code>azd env set AZURE_SEARCH_SEMANTIC_RANKER disabled</code></pre></li>
+<li><p>The search service can handle fairly large indexes, but it does
+have per-SKU limits on storage sizes, maximum vector dimensions, etc.
+You may want to upgrade the SKU to either a Standard or Storage
+Optimized SKU, depending on your expected load. However, you <a
+href="https://learn.microsoft.com/azure/search/search-sku-tier#tier-upgrade-or-downgrade">cannot
+change the SKU</a> of an existing search service, so you will need to
+re-index the data or manually copy it over. You can change the SKU by
+setting the <code>AZURE_SEARCH_SERVICE_SKU</code> azd environment
+variable to <a
+href="https://learn.microsoft.com/azure/templates/microsoft.search/searchservices?pivots=deployment-language-bicep#sku">an
+allowed SKU</a>.</p>
+<pre class="shell"><code>azd env set AZURE_SEARCH_SERVICE_SKU standard</code></pre>
+<p>See the <a
+href="https://learn.microsoft.com/azure/search/search-limits-quotas-capacity">Azure
+AI Search service limits documentation</a> for more details.</p></li>
+<li><p>If you see errors about search service capacity being exceeded,
+you may find it helpful to increase the number of replicas by changing
+<code>replicaCount</code> in
+<code>infra/core/search/search-services.bicep</code> or manually scaling
+it from the Azure Portal.</p></li>
+</ul>
+<h3 id="azure-app-service">Azure App Service</h3>
+<p>The default app service plan uses the <code>Basic</code> SKU with 1
+CPU core and 1.75 GB RAM. We recommend using a Premium level SKU,
+starting with 1 CPU core. You can use auto-scaling rules or scheduled
+scaling rules, and scale up the maximum/minimum based on load.</p>
+<h2 id="additional-security-measures">Additional security measures</h2>
+<ul>
+<li><strong>Authentication</strong>: By default, the deployed app is
+publicly accessible. We recommend restricting access to authenticated
+users. See <a
+href="./deploy_features.md#enabling-authentication">Enabling
+authentication</a> to learn how to enable authentication.</li>
+<li><strong>Networking</strong>: We recommend <a
+href="./deploy_private.md">deploying inside a Virtual Network</a>. If
+the app is only for internal enterprise use, use a private DNS zone.
+Also consider using Azure API Management (APIM) for firewalls and other
+forms of protection. For more details, read <a
+href="https://techcommunity.microsoft.com/blog/azurearchitectureblog/azure-openai-landing-zone-reference-architecture/3882102">Azure
+OpenAI Landing Zone reference architecture</a>.</li>
+</ul>
+<h2 id="load-testing">Load testing</h2>
+<p>We recommend running a loadtest for your expected number of users.
+You can use the <a href="https://docs.locust.io/">locust tool</a> with
+the <code>locustfile.py</code> in this sample or set up a loadtest with
+Azure Load Testing.</p>
+<p>To use locust, first install the dev requirements that includes
+locust:</p>
+<pre class="shell"><code>python -m pip install -r requirements-dev.txt</code></pre>
+<p>Or manually install locust:</p>
+<pre class="shell"><code>python -m pip install locust</code></pre>
+<p>Then run the locust command, specifying the name of the User class to
+use from <code>locustfile.py</code>. We’ve provided a
+<code>ChatUser</code> class that simulates a user asking questions and
+receiving answers, as well as a <code>ChatVisionUser</code> to simulate
+a user asking questions with the <a href="/docs/gpt4v.md">GPT-4 vision
+mode enabled</a>.</p>
+<pre class="shell"><code>locust ChatUser</code></pre>
+<p>Open the locust UI at <a
+href="http://localhost:8089/">http://localhost:8089/</a>, the URI
+displayed in the terminal.</p>
+<p>Start a new test with the URI of your website,
+e.g. <code>https://my-chat-app.azurewebsites.net</code>. Do <em>not</em>
+end the URI with a slash. You can start by pointing at your localhost if
+you’re concerned more about load on OpenAI/AI Search than the host
+platform.</p>
+<p>For the number of users and spawn rate, we recommend starting with 20
+users and 1 users/second. From there, you can keep increasing the number
+of users to simulate your expected load.</p>
+<p>Here’s an example loadtest for 50 users and a spawn rate of 1 per
+second:</p>
+<figure>
+<img src="images/screenshot_locust.png"
+alt="Screenshot of Locust charts showing 5 requests per second" />
+<figcaption aria-hidden="true">Screenshot of Locust charts showing 5
+requests per second</figcaption>
+</figure>
+<p>After each test, check the local or App Service logs to see if there
+are any errors.</p>
+<h2 id="evaluation">Evaluation</h2>
+<p>Before you make your chat app available to users, you’ll want to
+rigorously evaluate the answer quality. You can use tools in <a
+href="https://github.com/Azure-Samples/ai-rag-chat-evaluator">the AI RAG
+Chat evaluator</a> repository to run evaluations, review results, and
+compare answers across runs.</p>
diff --git a/data/sharing_environments.html b/data/sharing_environments.html
new file mode 100644
index 0000000000..d3a08fd66d
--- /dev/null
+++ b/data/sharing_environments.html
@@ -0,0 +1,27 @@
+<h1 id="rag-chat-sharing-deployment-environments">RAG chat: Sharing
+deployment environments</h1>
+<p>If you’ve deployed the RAG chat solution already following the steps
+in the <a href="../README.md#deploying">deployment guide</a>, you may
+want to share the environment with a colleague. Either you or they can
+follow these steps:</p>
+<ol type="1">
+<li>Install the <a
+href="https://learn.microsoft.com/cli/azure/install-azure-cli">Azure
+CLI</a></li>
+<li>Run <code>azd init -t azure-search-openai-demo</code> or clone this
+repository.</li>
+<li>Run <code>azd env refresh -e {environment name}</code> They will
+need the azd environment name, subscription ID, and location to run this
+command. You can find those values in your
+<code>.azure/{env name}/.env</code> file. This will populate their azd
+environment’s <code>.env</code> file with all the settings needed to run
+the app locally.</li>
+<li>Set the environment variable <code>AZURE_PRINCIPAL_ID</code> either
+in that <code>.env</code> file or in the active shell to their Azure ID,
+which they can get with <code>az ad signed-in-user show</code>.</li>
+<li>Run <code>./scripts/roles.ps1</code> or
+<code>.scripts/roles.sh</code> to assign all of the necessary roles to
+the user. If they do not have the necessary permission to create roles
+in the subscription, then you may need to run this script for them. Once
+the script runs, they should be able to run the app locally.</li>
+</ol>