We need a test dataset

[timlinux]

Problem

We are going to be writing a lot of tests and we need to be able to validate when things are working. To do this we should have a reasonable test dataset which allows us to exercise all of the functionality that the plugin provides. The test dataset should be accompanied with a table showing how many of each feature type we should find in the dataset. For example:

Feature type	Count
Well	5
Sink	3
Branch	10
Confluence	2
Pseudo nodes	10
Self Intersections	2
Segment centers	50
Watershed	5

If more that one dataset is provided, an additional column per dataset should be provided which enumerates the number of each feature type for that layer, and the entire matrix should include at least one example of each type.

Note: According to the specification, only the feature types listed in the table above will be identified.

Proposed solution

Ask client for a validation dataset for the stream network modelling tool.

• [ ] This should include at least one each of the features we are to identify automatically on the network.
• [ ] The dataset should be preferably small as possible to allow tests to run quickly while still exercising all of our rules.
• [ ] Several datasets would be ok if they cannot provide one single one that represents all features.
• [ ] The provided dataset needs to be redistributable as it will form the basis for our unit tests.

  Expected outcome

We should have a well defined test dataset to use as the basis for validation of functionality. All functionality shall be deemed to work if it works on the test dataset provided for this issue.

[timlinux]

The client has supplied this, closing.