A generator is an algorithm that decides how to traverse a model. Different generators will generate different test sequences, and they will navigate in different ways.
RANDOM
Navigate through the model in a completely random manor. Also called "Drunkard’s walk", or "Random walk".
The algorithm selects an out-edge from a vertex by random, and repeats the process in the next vertex.
A_STAR
Will try to generate the shortest possible test sequence through a model. This is very good when trying to find the fastest path with complete coverage in a model. The downside is that it only works on smaller models. If used on large models, mbt will take a lot of CPU time, computing the shortest path. See Wikipedia article: A* search algorithm
SHORTEST_NON_OPTIMIZED
This is a compromise between A_STAR and RANDOM. The algorithm works as follows:
- Choose an edge not yet visited by random.
- Select the shortest path to that edge using Dijkstra's algorithm (See Wikipedia: Dijkstra's algorithm)
- Walk that path, and mark all those edges as visited.
When reaching the selected edge in step 1, start all over, repeating steps 1->4.
Pros: The algorithm works well an very large models, and generates reasonably short sequences.
Cons: When used in a Extended FSM, it is quite possible to end up in a situation, where the only edges left to choose from, are protected by guards. This means that MBT enters an internal infinite loop. Also, be sure to have a rolling log file. Or else, your file system will be all filled up. (Below is an example how log4j should be configured. This is deafult in mbt.properties)
log4j.appender.LOG=org.apache.log4j.RollingFileAppender log4j.appender.LOG.File=logs/mbt.log log4j.appender.LOG.MaxFileSize=20MB log4j.appender.LOG.MaxBackupIndex=5
LIST
Generates a list of all edges and vertices in a graph. If the label of a vertex or an edge occurs more than once in the model, it will only be printed once.
See also How to list all unique labels in a graph.
STUB
Generates source code, with the use of templates.
See also How to generate source code from a model.
REQUIREMENTS
Generates a list of all requirements found in a model.
See also How to generate list of requirements from a model.