Users of the extension providing source code analysis support for SQL files.
Packaging, delivering and analyzing your source code
Before you can use the extension, you may need to use a third-party DDL extractor for your RDBMS and generate one or several *.sql files containing the required DDL. In other words, you need to "export" the database schema with a tool that generates DDL from the database or at least have DDL scripts that can be used to re-build the database. This export should contain exclusively *.sql files which can then be delivered via the CAST Delivery Manager Tool.
Please note that the analyzer is not able to follow the chronology of the DDL script. In other words, we do not guarantee that files will be analyzed in the same order as the creation, modification date or versioning.
What about superfluous clauses exported in your SQL files?
When performing the DDL extraction, CAST recommends that you DO NOT export storage definitions or any other superfluous clauses in your SQL files. All minor changes in the exported SQL files (e.g.: Engine for MariaDB or even the collation) will have an impact on the analysis results: i.e. objects will be marked as updated between successive analyses.
Therefore CAST's recommendation is to extract the SQL in the same manner, with the same tool, and using the same options that will extract unchanged objects.
How to extract DDL and DLM files using third-party tools
This is a non-exhaustive list. For general extraction you can also consider http://www.sql-workbench.net/dev-download.html.
CAST documentation: SQL Analyzer - postgreSQL schema export or extraction as operator
See also http://stackoverflow.com/questions/1884758/generate-ddl-programmatically-on-postgresql
CAST documentation: SQL Analyzer - MariaDB and MySQL DDL example export or extraction
http://sqlitebrowser.org/ (export database function)
src and uaxDirectory files
Alongside .SQL files, the extension will also accept also src and uaxDirectory files for analysis.
Packaging and delivery
Using the CAST Delivery Manager Tool:
- create a new Version
- create a new Package for your SQL Script (*.sql) files using the Files on your file system option:
- Enter a name for the Package and then define the root folder in which your .sql files are stored:
- Run the Package action: the CAST Delivery Manager Tool will not find any "projects" related to the .sql files - this is the expected behaviour. However, if your SQL script related source code is part of a larger application (for example a JEE application), then other projects may be found during the package action. The CAST Delivery Manager Tool, should, however, identify the .sql files:
- Deliver the Version
Using the CAST Management Studio:
- Accept and deploy the Version in the CAST Management Studio. No Analysis Units will be created automatically relating to the SQL Script files - this is the expected behaviour. However, if your SQL Script files are part of a larger application (for example a JEE project), then other Analysis Units may be created automatically:
- In the Current Version tab, add a new Analysis Unit specifically for your SQL Script files, selecting the Add new Universal Analysis Unit option:
- Edit the new Analysis Unit and configure in the Source Settings tab:
- a name for the Analysis Unit
- ensure you tick the SQL Analyzer option
- define the location of the deployed SQL script files (the CAST Management Studio will locate this automatically in the Deployment folder):
- Run a test analysis on the Analysis Unit before you generate a new snapshot.
- Add dependencies between client technologies (Java, C#, Mainframe etc...) to that Analysis Unit in order to get client to server links.