9. Glossary¶
- BSP
- A board support package supported by a test target
- BSP model
A combination of BSPs on which a test target works.
For example, the Arduino 101 x86 and arc BSPs that can work independently or together, the following BSP models would be supported: x86, arc and x86+arc.
- hash
- In the context of tcf run, a hash is a four character code that uniquely identifies the test case name and the target where it ran.
- herd
- Group of TCF servers that are meant to be used together, they all provide redundancy for targets of the same style.
- RunID
- A unique identifier for a run of testcases that groups the output and results of testcases. When specified, it will be prefixed to testcase output to clarify when it was generated.
- site
- Server herds can be split into sites, which is a subgroup of servers of the same herd that are physically located in the same place and might be sharing physical infrastructure
- static testcase
- A test case that doesn’t run on remote targets but in the local host where the test runner is executing.
- NUT
- Network Under Test or test network, to which test targets are connected. Usually isolated from the test infrastructure itself.
- POS
- Provisioning OS; operating system running on the test target that is used to provision the target’s drive.
- test case
A test case is a flow to evaluate a single requirement, feature or functionality.
A test case can be a program or a script
- test target
Something that can be used to run a test case; this can be a group of actual HW units, not just a single one. It can be a computer, a light switch or a toaster.
Each test target has a name, a type and it might support zero or more BSPs (think for example, a board that has an x86 and an ARC processor, like the Arduino 101); as well it has a list of tags and values that describe its capabilities.
- target controller
- A piece of software that understands how to talk to a target
- test target broker
- A service/machine to which one or more test targets are connected. It offers an abstracted API to manipulate and operate them; the API is implemented via target-specific target controllers.
- target broker
- See test target broker.
- target driver
- Same as target controller
- test finder
- A piece of software that locates test cases by looking at tag strings that define them as such.
- test case finder
- Same as test finder
- test runner
- Linux machine that can drive the running and execution of test cases or command a target hardware to execute test cases.
- test case driver
this is the entity that runs on the test runner that understands how different test cases can be run.
eg: knows how to launch a @static test case
eg: knows how to interpret the output of a cetain type of test cases to decide what it is
- ttbd
- See test target broker
- run ID
- An identification tag for a particular run of a sequence of test cases in a set of targets