Read and Write Entire GSSHA Projects

Last Updated: July 30, 2014

Each of the GSSHA model files can be read or written in the same manner that was illustrated with the project file. Each has a read() and a write() method. However, the gsshapy.orm.ProjectFile class has several additional methods that can be used to read and write all or some of the GSSHA model files simultaneously. These methods are:

  • readProject()
  • readInput()
  • readOutput()
  • writeProject()
  • writeInput()
  • wrtieOutput()

The names are fairly self explanatory, but more detailed explanations are provided in the API documentation for the gsshapy.orm.ProjectFile class. In this tutorial we will learn how to read an entire project and write it back to file.

Read All Files

Create a new session for this part of the tutorial, but use the same database:

>>> from gsshapy.lib import db_tools as dbt
>>> sqlalchemy_url = dbt.sqlalchemy_url = dbt.init_postgresql_db(username='gsshapy', host='localhost', database='gsshapy_tutorial', port='5432', password='pass')
>>> all_session = dbt.create_session(sqlalchemy_url)

Instantiate a new gsshapy.orm.ProjectFile object:

>>> from gsshapy.orm import ProjectFile
>>> projectFileAll = ProjectFile()

Invoke the readProject() method to read all supported GSSHA input and output files into the database:

>>> readDirectory = '/path_to/tutorial-data'
>>> filename = 'parkcity.prj'
>>> projectFileAll.readProject(directory=readDirectory, projectFileName=filename, session=all_session)

The process of reading all the files into the database takes a moment, so be patient. The readInput() and readOutput methods can be used to read only input files or output files, respectively. If the task you are using GsshaPy for is related to pre-processing the input files, you may want to use the readInput() method to save a little time on overhead. Similarly, if the task you are performing is related to post-processing the output files you may find the readOutput() method useful.

If you feel adventurous, you could use psql or PGAdminIII to investigate the database. Many of the tables will now be populated with data.

Write All Files

Now that all of the files have been read into the database, we can write them back out to file. Retrieve the project file from the database and invoke the writeProject() method:

>>> newProjectFileAll = all_session.query(ProjectFile).filter( == 2).one()
>>> writeDirectory = '/path_to/tutorial-data/write'
>>> name = 'mymodel'
>>> newProjectFileAll.writeProject(session=all_session, directory=writeDirectory, name=name)


We filter our query using the project file id of 2, because this is the second project file we have read in during this set of tutorials. If you end up reading in several projects, you can easily change this to another id to retrieve the GSSHA project you desire.

All of the files that were read into the database should be written to file in the write directory of the tutorial files. For the files that use the project name prefix as filename convention, the prefix has been changed to match the name supplied by the user (‘mymodel’ if you followed the tutorial exactly). Like the read methods, there are two other write methods that can be used to write only the input files or only the output files: writeInput() and writeOutput(), respectively. Use writeInput() when you want to write the model out to execute a simulation.