SchemaMan is a schema manager and query wrapper for multiple database backends, with Version and Change Management.


License
MIT
Install
pip install schemaman==0.1.45

Documentation

SchemaMan

SchemaMan is a schema manager for multiple databases, so that schema confguration can be stored in meta data files, and applied equally to any database.

SchemaMan can be used as a generic wrapper for working with databases, or migrating data between databases, who have SchemaMan schemas defined.

The focus of SchemaMan is on smaller low-volume data sets, such as found in System or Network Configuration management systems.

Version Management

Allows versioning data.

Change Managament

Approvals can be required to commit data from a Working Version into the datasource.

Version and Change Management Definitions

  • VMCM - Version Management and Change Management system
  • Working Change Set - A list of record fields that a Data Set User is making edits to. These are not applied to any live datasource, and are not integration until all the edits are finished, a Pending Change Set is created, for review.
  • Pending Change Set - A list of record fields that a Data Set User has specified as edits in a Working Change Set, and promoted to a Pending Change Set for review. The approval process starts here.
  • Approved Change Set - A list of record fields that a Data Set User promoted to a Pending Change Set, and then all required approvers marked "Approved" moving the state from Pending to Approved. The next step is Deployed Change Set.
  • Deployed Change Set - A list of record fields that has been approved. A Data Set will need to trigger a Deployment of the changes, which is when the changes are actually applied into the Data Set's data sources. Before this point, no data changes were made to the Data Set's data sources. This write must be done by the Data Set itself, as ChangeIn has no knowledge of how this would happen or authorization to perform this work. Call backs should be implemented for pre- and post- deployment scripts, so that they stages can be processed in ChangeIn, tracking the progress of the work.
  • Committed Change Set - A list of record fields that has been deployed and written into a Data Set's data sources. The committed data is kept in ChangeIn to provide auditing functionality, and also for retrieval for use in Roll Backs.
  • Collected Change Set - A list of record fields that was either rejected or otherwise brought out of the approval process, but is still collected together for editing purposes.
  • Approval Group - All record field changes have an Approval Group consisting of 1 or more Approvers, which is provided by the Data Set at Working Change Set update or before the Pending Change Set promotion through a call to the Data Set API which takes all the record field changes and returns the list of Approval Groups and Approvers in the ChangeIn specified format.
  • Approvers - Members of an approval group. Human or role accounts. This could also be a Service based authorization, through API call-back.

Security, Owners, and Actions

The Change Management can invoke scripts against data for owners, groups to provide security and proper approvals, and can invoke scripts as actions to perform pre- and post- commit work at various levels.

Command Line Usage

Schema Actions:

  info                                       Print info on current schema directory
  init <path>                                Initialize a path for new schemas
  schema create <schema>                     Create a schema interactively
  schema export <schema> <source>            Export a database schema from a source
  schema update <schema> <source> <target>   Migrate schema/data from source to target
  data export <schema> <source>              Export all the data from the schema/source
  data import <schema> <source>              Import data into the schema/source

Primary Data Actions:

  put <schema> <source> <json>        Put JSON data into a Schema instance
  get <schema> <source> <json>        Get Schema instance records from JSON keys
  filter <schema> <source> <json>     Filter Schema instance records
  delete <schema> <source> <json>     Delete records from Schema instance

Additional Data Actions:

  action config version_change_management <target_schema>  Configure Version and Change Management
  action populate schema_into_db <target_schema>           Populate Schema Into DB

Options:

  -d <path>, --dir=<path>             Directory for SchemaMan data/conf/schemas
                                          (Default is current working directory)
  -y, --yes                           Answer Yes to all prompts

  -h, -?, --help                      This usage information
  -v, --verbose                       Verbose output

Docs