Navicat Blog

Supercharging Your Queries with Navicat and ChatGPT Feb 9, 2023 by Robert Gravelle

It's official; the age of Artificial Intelligence (AI) has arrived! Until our new overlords decide to use us to power their machines, let's take the time to fully enjoy all the benefits they provide and the myriad of ways that they make our lives easier. Case in point, the AI-driven chatbot, ChatGPT, by OpenAI, has been lauded for its ability to produce tremendously spot-on answers to questions across a broad range of topics. And, although ChatGPT may not be making our jobs obsolete just yet, it has proven to be amazingly adept at working with data sets, much like a DBMS. In today's blog, we'll explore how ChatGPT could be utilized to supplement a professional database development and administration tool like Navicat.

Creating the Data Set

ChatGPT is able to model a formal dataset from a list of delimited values. All you need to do is tell it what to do using regular, conversational language. ChatGPT is also able to answer follow-up questions, admit its mistakes, challenge incorrect premises, and reject inappropriate requests. We can see an example on the OUseful.Info blog that created a table named "racerresults". Here are the instructions given to ChatGPT, along with a sampling of the input data:

Treat the following as a tab separated dataset. Using just the first, third and fourth columns, treat the data as if it were a relational SQL database table called "racerresults" with columns "Race", "Driver" and "Team", and the "Race" column as a primary key column. Display a SQL statement that could create the corresponding table and populate it with the data.

Bahrain	20 Mar 2022	Charles Leclerc	FERRARI	57	1:37:33.584
Saudi Arabia	27 Mar 2022	Max Verstappen	RED BULL RACING RBPT	50	1:24:19.293
Australia	10 Apr 2022	Charles Leclerc	FERRARI	58	1:27:46.548
Emilia Romagna	24 Apr 2022	Max Verstappen	RED BULL RACING RBPT	63	1:32:07.986
Miami	08 May 2022	Max Verstappen	RED BULL RACING RBPT	57	1:34:24.258
Spain	22 May 2022	Max Verstappen	RED BULL RACING RBPT	66	1:37:20.475
Monaco	29 May 2022	Sergio Perez	RED BULL RACING RBPT	64	1:56:30.265

From the above instructions and data, ChatGPT generated the following CREATE TABLE and INSERT statements:

raceresults_create_and_insert_statements (122K)

With the data in place, we're ready to run queries against it.

Querying a Data Set with ChatGPT

In terms of query formulation, ChatGPT shares some similarities with Navicat, in that both allow you to construct queries with little knowledge of SQL. To do that, Navicat features the Query Builder tool. Here it is in macOS:

queryBuilder (136K)

As for ChatGPT, it takes a question phrased in regular, conversational language, and produces the required SQL statement(s). For instance, given the following list of historical figures:

historical_figures (56K)

We can simply as ChatGPT how it would query for the oldest historical figure. Here is the resulting SQL statement and explanation offered by ChatGPT:

oldest_historical_figure_query (166K)

Fun with Data

ChatGPT can do a lot more than generate queries; it can also think creatively to assign emojis to each historical figure:

historical_figures_with_emojis (131K)

Final Thoughts on Supercharging Your Queries with Navicat and ChatGPT

While AI bots like ChatGPT are a long way from replacing traditional database tools, they do offer another tool to database practitioners who are looking for new and innovative ways of approaching data-related tasks. At the time of this writing, ChatGPT was at capacity and unable to accept new users, but once things die down a bit, I would urge you to give ChatGPT a try.

Navicat Blogs
Feed Entries
Blog Archives