PPM Pro Command Line client

This project is a small CLI interface to PPM Pro. It has the ability to set daily hours on each activity and run a report of hours for a given timesheet.

It has the ability to automatically look up daily hours from an external source via the get_hours() hook, although this funcionality is left unimplemented by default as there is no standard way of doing this.

It is suitable for running out of cron to automatically set hours each day.

Requirements/Usage

The tool is written in Python and requires the availability of a supported Python3 interpreter. All the modules required should be part of a typical base install.

You will need to modify the value of the org_name constant near the top of the file. This should be the same as the first part of your PPM Pro URL (e.g. https://foo.ppmpro.com would use org_name foo).

Authentication

Per https://success.planview.com/Planview_PPM_Pro/150_PPM_Pro_Administrator_Documentation/015_System_Settings/010_Authentication and https://success.planview.com/Planview_PPM_Pro/Reports_and_Dashboards/020_Reports/040_OData_Setup#section_2 if your instance of PPM Pro uses SAML (which apparently is their recommendation and I suspect what most sites implement), you do not currently appear to be able to use an API key for authentication even if you follow Planview's instructions for generating one in the app. This appears to be an internal limitation on the site URL, as the token does work as described for the OData endpoints.

As a result, for now the only way to authenticate appears to be to present a valid session cookie. The tool expects this cookie to be stored locally on the filesystem, by default in ~/.ppmpro.session, although that location can be overridden with a command line flag.

The user will need to log in to the webapp the first time and obtain a suitable cookie from their browser storage. A valid session cookie will be of the form:

RKVM_SID=some-uuid-string

Known Issues

• Due to the way authentication is handled (see above), you will need to make sure the tool is called regularly to keep your stored session active. Once an hour seems to work. Calling the tool without any arguments issues a simple API GET request that is sufficient for this.

• The get_hours() method implementation is left as an exercise to the reader. The example implementation simply divides a target daily number of hours evenly among all activities. This probably does not reflect reality.