The Electrolyte Database (EDB) stores data about chemical species. It is accessed through a Python API to return well-defined Python objects.
The data are stored in a database called MongoDB. MongoDB is open-source and free software, so you can install and use the database locally. You may also use our open cloud-hosted version of the database. See the MongoDB installation instructions <install-mongodb> for more details.
The native storage format for MongoDB is a JSON object, which MongoDB calls a "document". The expected structure of the EDB data is defined by a JSON Schema. Most users will not need to deal with the MongoDB documents, as there is a Python data API <edb-data-api> for searching the database and building IDAES "config blocks" from the components and reactions.
The EDB is intended to support some known workflows out of the box, with lower-level functions available when these are not sufficient. Example workflows can be seen in the how to guides from the link above.
Connect to the database and create, read, update and delete its contents.
watertap.edb.db_api
Data models for components and reactions, including conversion to IDAES config objects.
watertap.edb.data_model
edb
The edb command-line program lets you interact with the database and the data schemas from a terminal.
The work of the program is all done by subcommands.
--help
Show options and subcommands
-v, --verbose
Increase verbosity
-q, --quiet
Increase quietness
edb-load
Load JSON records into the EDB.
-f, --file FILENAME
File to load [required]
-t, --type [componentbase]
Type of records [required]
-u, --url TEXT
Database connection URL
-d, --database TEXT
Database name
--validate / -n, --no-validate
Turn on or off validation of input
-b, --bootstrap
Bootstrap a new database by loading in the standard base data.
Dump JSON records from the EDB to a file.
-f, --file FILENAME
File to create (will overwrite existing files!) [required]
-t, --type [componentbase]
Type of records (MongoDB collection name)
-u, --url TEXT
Database connection URL
-d, --database TEXT
Database name
Show JSON schemas, in raw or readable forms, for the different record types.
-f, --file FILENAME
Write output to this file instead of printing to the screen
-o, --format [jsonhtml|html-js]
Output format
-t, --type [component|reaction]
Type of records [required]
-u, --url TEXT
Database connection URL
-d, --database TEXT
Database name
The EDB data is encoded in JSON. Naturally, the expected form of the records is specified as JSON Schema. There are schemas for the component and reaction records. Currently, there is no schema for base data (this will change soon, though).
html
latex or epub or text
html
latex or epub or text
The EDB data for the most common base configs are made available in the standard bootstrap bundled with WaterTAP. Those bases are listed below:
| Base | Description |
|---|---|
| default_thermo | Default ThermoConfig: contains only AqueousPhase and uses FTPx state vars |
| thermo_Liq_FpcTP | ThermoConfig: contains only AqueousPhase and uses FpcTP state vars |
| thermo_Liq_Sol_FpcTP | ThermoConfig: contains AqueousPhase + SolidPhase and uses FpcTP state vars |
| thermo_Liq_Vap_FpcTP | ThermoConfig: contains AqueousPhase + VaporPhase and uses FpcTP state vars |
| reaction | ReactionConfig: Blank template for reaction configs |
The naming convention for the thermo bases is as follows: (i) each section of the name is broken up by the underscore character (_), (ii) the first word in the name is always thermo to denote the type of base, (iii) the last word in the name always denotes the state_vars, and (iv) each word in between denotes the set of phases for the configuration file.