Update index.html to enhance accessibility and structure with improved headings and IDs for the Table of Contents and sections.

index.html

@@ -77,7 +77,7 @@
             queries into SQL, executing them, and intelligently processing the results, including handling complex
             scenarios
             like table joins.</p>
-        <h2>Table of Contents</h2>
+        <h2 id="table-of-contents">Table of Contents</h2>
         <ul>
             <li><a href="#intelligent-text-to-sql-agent-with-smolagents">Intelligent Text-to-SQL Agent with
                     <code>smolagents</code></a></li>
@@ -97,7 +97,8 @@
             <li><a href="#contributing">Contributing</a></li>
             <li><a href="#license">License</a></li>
         </ul>
-        <h2 class="code-line" data-line-start=29 data-line-end=30><a id="Why_an_AI_Agent_for_TexttoSQL_29"></a>Why an AI
+        <h2 id="why-an-ai-agent-for-text-to-sql" class="code-line" data-line-start=29 data-line-end=30><a
+                id="Why_an_AI_Agent_for_TexttoSQL_29"></a>Why an AI
             Agent for Text-to-SQL?</h2>
         <p class="has-line-data" data-line-start="31" data-line-end="32">Traditional Text-to-SQL pipelines often suffer
             from
@@ -122,7 +123,7 @@
             <li class="has-line-data" data-line-start="40" data-line-end="42"><strong>Enhanced Robustness:</strong>
                 Providing a more reliable and intelligent way to interact with databases from natural language.</li>
         </ul>
-        <h2 class="code-line" data-line-start=42 data-line-end=43><a id="Features_42"></a>Features</h2>
+        <h2 id="features" class="code-line" data-line-start=42 data-line-end=43><a id="Features_42"></a>Features</h2>
         <ul>
             <li class="has-line-data" data-line-start="44" data-line-end="45"><strong>Natural Language to SQL:</strong>
                 Translates user questions into executable SQL queries.</li>
@@ -139,7 +140,8 @@
                 Integrates
                 with various Large Language Models (LLMs) via <code>smolagents</code>.</li>
         </ul>
-        <h2 class="code-line" data-line-start=51 data-line-end=52><a id="Installation_51"></a>Installation</h2>
+        <h2 id="installation" class="code-line" data-line-start=51 data-line-end=52><a
+                id="Installation_51"></a>Installation</h2>
         <p class="has-line-data" data-line-start="53" data-line-end="54">To get started, clone this repository and
             install
             the required dependencies:</p>
@@ -147,33 +149,31 @@
 <span class="hljs-built_in">cd</span> text-to-sql-agent
 pip install smolagents python-dotenv sqlalchemy --upgrade -q
 </code></pre>
-        <pre><code class="has-line-data" data-line-start="62" data-line-end="148">
-**Note:** To interact with Large Language Models via inference providers (e.g., Hugging Face Inference API), you'll need a valid authentication token set as an environment variable, typically `HF_TOKEN`.
 
-## Project Structure
-
-The core logic of this project is encapsulated in `text_to_sql.py`.
-
-```text
-.
+        <p class="has-line-data" data-line-start="4" data-line-end="5"><strong>Note:</strong> To interact with Large
+            Language Models via inference providers (e.g., Hugging Face Inference API), you’ll need a valid
+            authentication token set as an environment variable, typically <code>HF_TOKEN</code>.</p>
+        <h2 class="code-line" data-line-start=6 data-line-end=7><a id="Project_Structure_6"></a>Project Structure</h2>
+        <p class="has-line-data" data-line-start="8" data-line-end="9">The core logic of this project is encapsulated in
+            <code>text_to_sql.py</code>.
+        </p>
+        <pre><code class="has-line-data" data-line-start="11" data-line-end="15" class="language-text">.
 ├── README.md
 └── text_to_sql.py
-```
-
-## Usage
-
-This section walks through the `text_to_sql.py` script, explaining each part of building and using the agent.
-
-### Setup and Dependencies
-
-First, load your environment variables, including your LLM token.
-
-```python
-# text_to_sql.py
-from dotenv import load_dotenv
+</code></pre>
+        <h2 class="code-line" data-line-start=16 data-line-end=17><a id="Usage_16"></a>Usage</h2>
+        <p class="has-line-data" data-line-start="18" data-line-end="19">This section walks through the
+            <code>text_to_sql.py</code> script, explaining each part of building and using the agent.
+        </p>
+        <h3 class="code-line" data-line-start=20 data-line-end=21><a id="Setup_and_Dependencies_20"></a>Setup and
+            Dependencies</h3>
+        <p class="has-line-data" data-line-start="22" data-line-end="23">First, load your environment variables,
+            including your LLM token.</p>
+        <pre><code class="has-line-data" data-line-start="25" data-line-end="45" class="language-python"><span class="hljs-comment"># text_to_sql.py</span>
+<span class="hljs-keyword">from</span> dotenv <span class="hljs-keyword">import</span> load_dotenv
 load_dotenv()
 
-from sqlalchemy import (
+<span class="hljs-keyword">from</span> sqlalchemy <span class="hljs-keyword">import</span> (
     create_engine,
     MetaData,
     Table,
@@ -185,71 +185,66 @@ from sqlalchemy import (
     inspect,
     text,
 )
-from smolagents import tool, CodeAgent, InferenceClientModel
-
-# ... (rest of the code)
-```
+<span class="hljs-keyword">from</span> smolagents <span class="hljs-keyword">import</span> tool, CodeAgent, InferenceClientModel
 
-### Database Initialization
-
-We set up an in-memory SQLite database using SQLAlchemy, defining `receipts` and `waiters` tables and populating them with sample data.
-
-````python
-# text_to_sql.py
-engine = create_engine(&quot;sqlite:///:memory:&quot;)
+<span class="hljs-comment"># ... (rest of the code)</span>
+</code></pre>
+        <h3 class="code-line" data-line-start=46 data-line-end=47><a id="Database_Initialization_46"></a>Database
+            Initialization</h3>
+        <p class="has-line-data" data-line-start="48" data-line-end="49">We set up an in-memory SQLite database using
+            SQLAlchemy, defining <code>receipts</code> and <code>waiters</code> tables and populating them with sample
+            data.</p>
+        <pre><code class="has-line-data" data-line-start="51" data-line-end="88" class="language-python"><span class="hljs-comment"># text_to_sql.py</span>
+engine = create_engine(<span class="hljs-string">"sqlite:///:memory:"</span>)
 metadata_obj = MetaData()
 
-def insert_rows_into_table(rows, table, engine=engine):
-    for row in rows:
+<span class="hljs-function"><span class="hljs-keyword">def</span> <span class="hljs-title">insert_rows_into_table</span><span class="hljs-params">(rows, table, engine=engine)</span>:</span>
+    <span class="hljs-keyword">for</span> row <span class="hljs-keyword">in</span> rows:
         stmt = insert(table).values(**row)
-        with engine.begin() as connection:
+        <span class="hljs-keyword">with</span> engine.begin() <span class="hljs-keyword">as</span> connection:
             connection.execute(stmt)
 
-# Define the 'receipts' table
+<span class="hljs-comment"># Define the 'receipts' table</span>
 receipts = Table(
-    &quot;receipts&quot;,
+    <span class="hljs-string">"receipts"</span>,
     metadata_obj,
-    Column(&quot;receipt_id&quot;, Integer, primary_key=True),
-    Column(&quot;customer_name&quot;, String(255)), # Adjusted from String(16) for longer names
-    Column(&quot;price&quot;, Float),
-    Column(&quot;tip&quot;, Float),
+    Column(<span class="hljs-string">"receipt_id"</span>, Integer, primary_key=<span class="hljs-keyword">True</span>),
+    Column(<span class="hljs-string">"customer_name"</span>, String(<span class="hljs-number">255</span>)), <span class="hljs-comment"># Adjusted from String(16) for longer names</span>
+    Column(<span class="hljs-string">"price"</span>, Float),
+    Column(<span class="hljs-string">"tip"</span>, Float),
 )
 metadata_obj.create_all(engine)
 
-# Sample data for 'receipts'
+<span class="hljs-comment"># Sample data for 'receipts'</span>
 rows = [
-    {&quot;receipt_id&quot;: 1, &quot;customer_name&quot;: &quot;Alan Payne&quot;, &quot;price&quot;: 12.06, &quot;tip&quot;: 1.20},
-    {&quot;receipt_id&quot;: 2, &quot;customer_name&quot;: &quot;Alex Mason&quot;, &quot;price&quot;: 23.86, &quot;tip&quot;: 0.24},
-    {&quot;receipt_id&quot;: 3, &quot;customer_name&quot;: &quot;Woodrow Wilson&quot;, &quot;price&quot;: 53.43, &quot;tip&quot;: 5.43},
-    {&quot;receipt_id&quot;: 4, &quot;customer_name&quot;: &quot;Margaret James&quot;, &quot;price&quot;: 21.11, &quot;tip&quot;: 1.00},
+    {<span class="hljs-string">"receipt_id"</span>: <span class="hljs-number">1</span>, <span class="hljs-string">"customer_name"</span>: <span class="hljs-string">"Alan Payne"</span>, <span class="hljs-string">"price"</span>: <span class="hljs-number">12.06</span>, <span class="hljs-string">"tip"</span>: <span class="hljs-number">1.20</span>},
+    {<span class="hljs-string">"receipt_id"</span>: <span class="hljs-number">2</span>, <span class="hljs-string">"customer_name"</span>: <span class="hljs-string">"Alex Mason"</span>, <span class="hljs-string">"price"</span>: <span class="hljs-number">23.86</span>, <span class="hljs-string">"tip"</span>: <span class="hljs-number">0.24</span>},
+    {<span class="hljs-string">"receipt_id"</span>: <span class="hljs-number">3</span>, <span class="hljs-string">"customer_name"</span>: <span class="hljs-string">"Woodrow Wilson"</span>, <span class="hljs-string">"price"</span>: <span class="hljs-number">53.43</span>, <span class="hljs-string">"tip"</span>: <span class="hljs-number">5.43</span>},
+    {<span class="hljs-string">"receipt_id"</span>: <span class="hljs-number">4</span>, <span class="hljs-string">"customer_name"</span>: <span class="hljs-string">"Margaret James"</span>, <span class="hljs-string">"price"</span>: <span class="hljs-number">21.11</span>, <span class="hljs-string">"tip"</span>: <span class="hljs-number">1.00</span>},
 ]
 insert_rows_into_table(rows, receipts)
 
-# Print table schema (for LLM context)
+<span class="hljs-comment"># Print table schema (for LLM context)</span>
 inspector = inspect(engine)
-columns_info = [(col[&quot;name&quot;], col[&quot;type&quot;]) for col in inspector.get_columns(&quot;receipts&quot;)]
-table_description = &quot;Columns:\n&quot; + &quot;\n&quot;.join([f&quot;  - {name}: {col_type}&quot; for name, col_type in columns_info])
+columns_info = [(col[<span class="hljs-string">"name"</span>], col[<span class="hljs-string">"type"</span>]) <span class="hljs-keyword">for</span> col <span class="hljs-keyword">in</span> inspector.get_columns(<span class="hljs-string">"receipts"</span>)]
+table_description = <span class="hljs-string">"Columns:\n"</span> + <span class="hljs-string">"\n"</span>.join([f<span class="hljs-string">"  - {name}: {col_type}"</span> <span class="hljs-keyword">for</span> name, col_type <span class="hljs-keyword">in</span> columns_info])
 print(table_description)```
-
-**Output:**
 </code></pre>
-        <p class="has-line-data" data-line-start="149" data-line-end="150">Columns:</p>
-        <ul>
-            <li class="has-line-data" data-line-start="151" data-line-end="152">receipt_id: INTEGER</li>
-            <li class="has-line-data" data-line-start="152" data-line-end="153">customer_name: VARCHAR(255)</li>
-            <li class="has-line-data" data-line-start="153" data-line-end="154">price: FLOAT</li>
-            <li class="has-line-data" data-line-start="154" data-line-end="156">tip: FLOAT</li>
-        </ul>
-        <pre><code class="has-line-data" data-line-start="157" data-line-end="189">
-### Creating the SQL Tool
-
-The `sql_engine` function acts as the agent's interface to the database. Its detailed docstring provides the LLM with crucial information about its functionality and the database schema.
-
-```python
-# text_to_sql.py
-@tool
-def sql_engine(query: str) -&gt; str:
-    &quot;&quot;&quot;
+        <b>Output:</b>
+        <p class="has-line-data" data-line-start="0" data-line-end="1">Columns:</p>
+        <p class="has-line-data" data-line-start="2" data-line-end="6">receipt_id: INTEGER<br>
+            customer_name: VARCHAR(255)<br>
+            price: FLOAT<br>
+            tip: FLOAT</p>
+        <h3 class="code-line" data-line-start=7 data-line-end=8><a id="Creating_the_SQL_Tool_7"></a>Creating the SQL
+            Tool</h3>
+        <p class="has-line-data" data-line-start="9" data-line-end="10">The <code>sql_engine</code> function acts as the
+            agent’s interface to the database. Its detailed docstring provides the LLM with crucial information about
+            its functionality and the database schema.</p>
+        <pre><code class="has-line-data" data-line-start="12" data-line-end="37" class="language-python"><span class="hljs-comment"># text_to_sql.py</span>
+<span class="hljs-decorator">@tool</span>
+<span class="hljs-function"><span class="hljs-keyword">def</span> <span class="hljs-title">sql_engine</span><span class="hljs-params">(query: str)</span> -&gt; <span class="hljs-title">str</span>:</span>
+    <span class="hljs-string">"""
     Enables execution of SQL queries against the database.
     Outputs the query results as a formatted string.
 
@@ -263,16 +258,16 @@ def sql_engine(query: str) -&gt; str:
 
     Args:
         query: The precise SQL query string to be executed.
-               Example: &quot;SELECT customer_name FROM receipts WHERE price &gt; 10.0;&quot;
-    &quot;&quot;&quot;
-    output = &quot;&quot;
-    with engine.connect() as con:
+               Example: "SELECT customer_name FROM receipts WHERE price &gt; 10.0;"
+    """</span>
+    output = <span class="hljs-string">""</span>
+    <span class="hljs-keyword">with</span> engine.connect() <span class="hljs-keyword">as</span> con:
         rows = con.execute(text(query))
-        for row in rows:
-            output += &quot;\n&quot; + str(row)
-    return output
-</code></pre>
-        <h3 class="code-line" data-line-start=190 data-line-end=191><a
+        <span class="hljs-keyword">for</span> row <span class="hljs-keyword">in</span> rows:
+            output += <span class="hljs-string">"\n"</span> + str(row)
+    <span class="hljs-keyword">return</span> output</code></pre>
+
+        <h3 id="instantiating-the-agent-single-table" class="code-line" data-line-start=190 data-line-end=191><a
                 id="Instantiating_the_Agent_Single_Table_190"></a>Instantiating the Agent (Single Table)</h3>
         <p class="has-line-data" data-line-start="192" data-line-end="193">We create a <code>CodeAgent</code> and
             provide it
@@ -283,7 +278,7 @@ agent = CodeAgent(
     model=InferenceClientModel(model_id=<span class="hljs-string">"meta-llama/Llama-3.1-8B-Instruct"</span>),
 )
 </code></pre>
-        <h3 class="code-line" data-line-start=202 data-line-end=203><a
+        <h3 id="querying-the-agent-single-table" class="code-line" data-line-start=202 data-line-end=203><a
                 id="Querying_the_Agent_Single_Table_202"></a>Querying
             the Agent: Single Table</h3>
         <p class="has-line-data" data-line-start="204" data-line-end="205">Now, we can ask the agent a question and
@@ -301,7 +296,8 @@ agent.run(<span class="hljs-string">"Can you give me the name of the client who
             and execute <code>SELECT MAX(price), customer_name FROM receipts ORDER BY price DESC LIMIT 1</code>, parse
             the
             result <code>(53.43, 'Woodrow Wilson')</code>, and identify ‘Woodrow Wilson’.</p>
-        <h3 class="code-line" data-line-start=214 data-line-end=215><a id="Extending_for_Table_Joins_214"></a>Extending
+        <h3 id="extending-for-table-joins" class="code-line" data-line-start=214 data-line-end=215><a
+                id="Extending_for_Table_Joins_214"></a>Extending
             for
             Table Joins</h3>
         <p class="has-line-data" data-line-start="216" data-line-end="217">To handle more complex queries, we add a
@@ -356,7 +352,7 @@ Table 'waiters':
     - receipt_id: INTEGER
     - waiter_name: VARCHAR(16)
 </code></pre>
-        <h3 class="code-line" data-line-start=272 data-line-end=273><a
+        <h3 id="querying-the-agent-multi-table" class="code-line" data-line-start=272 data-line-end=273><a
                 id="Querying_the_Agent_MultiTable_272"></a>Querying
             the Agent: Multi-Table</h3>
         <p class="has-line-data" data-line-start="274" data-line-end="275">We switch to a more powerful LLM
@@ -377,7 +373,8 @@ agent.run(<span class="hljs-string">"Which waiter received the highest total amo
             then process the results in Python to sum tips per waiter and identify “Michael Watts” as having the highest
             total tips.
         </p>
-        <h2 class="code-line" data-line-start=289 data-line-end=290><a id="How_it_Works_289"></a>How it Works</h2>
+        <h2 id="how-it-works" class="code-line" data-line-start=289 data-line-end=290><a id="How_it_Works_289"></a>How
+            it Works</h2>
         <p class="has-line-data" data-line-start="291" data-line-end="292">The <code>smolagents</code>
             <code>CodeAgent</code> operates on the <strong>ReAct (Reasoning + Acting)</strong> framework:
         </p>
@@ -408,7 +405,8 @@ agent.run(<span class="hljs-string">"Which waiter received the highest total amo
             solve
             complex problems and recover from errors, making it more robust than traditional direct translation methods.
         </p>
-        <h2 class="code-line" data-line-start=300 data-line-end=301><a id="Key_Concepts_Demonstrated_300"></a>Key
+        <h2 id="key-concepts-demonstrated" class="code-line" data-line-start=300 data-line-end=301><a
+                id="Key_Concepts_Demonstrated_300"></a>Key
             Concepts
             Demonstrated</h2>
         <ul>
@@ -430,11 +428,12 @@ agent.run(<span class="hljs-string">"Which waiter received the highest total amo
             <li class="has-line-data" data-line-start="307" data-line-end="309"><strong>ReAct Paradigm:</strong> The
                 iterative cycle of reasoning, acting, and observation that enables self-correction.</li>
         </ul>
-        <h2 class="code-line" data-line-start=309 data-line-end=310><a id="Contributing_309"></a>Contributing</h2>
+        <h2 id="contibuting" class="code-line" data-line-start=309 data-line-end=310><a
+                id="Contributing_309"></a>Contributing</h2>
         <p class="has-line-data" data-line-start="311" data-line-end="312">Feel free to open issues or submit pull
             requests
             if you have suggestions or improvements!</p>
-        <h2 class="code-line" data-line-start=313 data-line-end=314><a id="License_313"></a>License</h2>
+        <h2 id="license" class="code-line" data-line-start=313 data-line-end=314><a id="License_313"></a>License</h2>
         <p class="has-line-data" data-line-start="315" data-line-end="316">This project is open-sourced under the MIT
             License. See the <code>LICENSE</code> file for more details.</p>
     </div>