MATLAB Database Toolbox™ Users Guide (The MathWorks, Inc.) (Z-Library)

# MATLAB Database Toolbox™ User's Guide — Reading Guide ## 【One-Line Pitch】 A comprehensive reference for connecting MATLAB to relational and NoSQL databases—covering setup, querying, importing, and analyzing data—essential for engineers and data scientists who want to move data between MATLAB and enterprise databases without leaving their analysis environment. ## 【Book Arc】 - **Opening (~0%–10%)**: Covers getting started with Database Toolbox, including driver installation verification, configuring JDBC and ODBC data sources for Microsoft SQL Server (Windows and Linux), and establishing initial database connections through both the Database Explorer app and command-line functions. - **Early (~10%–23%)**: Moves into practical data import workflows—using the Database Explorer app to browse tables and build queries, importing data with functions like `sqlread` and `fetch`, handling custom data types (including Java objects from Oracle ref cursors), and executing SQL scripts. Introduces SQLite workflows for lightweight local databases. - **Early (~23%–32%)**: Expands to native interfaces for MySQL and PostgreSQL, demonstrating CRUD operations (insert, delete, update) through MATLAB functions. Introduces NoSQL connectivity with Apache Cassandra (using `upsert`, `partitionRead`) and graph database workflows with Neo4j, including searching nodes/relationships and converting graph data to MATLAB `digraph` objects for visualization. - **Middle (~39%–52%)**: Focuses on programmatic data source configuration using `SQLConnectionOptions` objects—setting JDBC driver locations, server names, ports, and authentication types, then testing and saving configurations. Covers direct JDBC connections using driver class names and URLs, plus working with `databaseDatastore` objects for streaming large query results in chunks. - **Late (~52%–end)**: The excerpts suggest continued coverage of advanced data import/export patterns, memory management for large datasets, and the complete function reference with syntax, input arguments, and compatibility notes (including deprecated functions like `testConnection`). ## 【Key Takeaways】 - **Driver installation is the first hurdle** (Early): Before any MATLAB-to-database workflow, verify JDBC/ODBC drivers are installed and TCP/IP is enabled on the database server—most connection failures trace back to these prerequisites. - **Two connection paths exist: app and command line** (Early): The Database Explorer app provides visual table browsing and query building for exploratory work, while functions like `odbc`, `mysql`, and `postgresql` enable scripted, reproducible connections for production workflows. - **MATLAB tables map cleanly to database tables** (Early): Use `sqlwrite` to push MATLAB tables into databases and `fetch`/`sqlread` to pull data back—column names and data types transfer predictably, making round-trip workflows straightforward. - **NoSQL support extends beyond relational databases** (Early): The toolbox includes native interfaces for Apache Cassandra (columnar NoSQL) and Neo4j (graph databases), with functions like `upsert`, `partitionRead`, `searchGraph`, and `neo4jStruct2Digraph` for specialized data models. - **Programmatic data source configuration scales** (Middle): The `SQLConnectionOptions` object lets you define, test, and save JDBC data sources entirely in code—essential for sharing reproducible setups across teams or deploying to new environments. - **Direct JDBC connections offer maximum flexibility** (Middle): When a preconfigured data source isn't available, connect directly using driver class names and JDBC URLs (e.g., `org.postgresql.Driver` with `jdbc:postgresql://host:port/dbname`), giving you control over every connection parameter. - **Large datasets require streaming patterns** (Middle): The `databaseDatastore` object reads query results incrementally with configurable `ReadSize`, preventing memory exhaustion when working with tables containing millions of rows. ## 【Reading Tips】 - **Skim the driver installation tutorials** (~0%–10%) if you're experienced with database connections—they're vendor-specific and repetitive. Return to them only when troubleshooting connection issues. - **Deep-read the SQLite and MySQL/PostgreSQL workflow chapters** (~13%–23%) if you work with relational databases—they show complete end-to-end patterns from connection to data manipulation that transfer across vendors. - **Treat the NoSQL chapters** (~23%–32%) as reference material—skim to understand what's possible with Cassandra and Neo4j, then consult specific function documentation when you need those capabilities. - **Pay attention to the function reference section** (Late) for syntax details, input argument types, and compatibility notes—the excerpts show deprecated functions (like `testConnection`) being phased out, so check for modern alternatives. - **Focus on the connection object properties** shown in examples (Middle)—they reveal what a successful connection looks like (empty `Message` property, populated catalog/schema lists) and help you debug failures faster. ## 【Coverage Limits】 This guide synthesizes excerpts covering roughly the first half of the book (through ~52%). Later chapters on advanced data import/export, memory management details, and the complete function reference are only partially represented in the source material. ##