Create jGraphT with Graalvm POC

farhan5248 / sheep-dog-tools

Eclipse and Maven plug-ins to help manual testers support developers adopting bdd and tdd using lean principles and model based testing

GNU General Public License v3.0

0 stars 0 forks source link

Create jGraphT with Graalvm POC #31

Closed farhan5248 closed 9 months ago

farhan5248 commented 10 months ago

start with this tutorial https://jgrapht.org/guide/HelloJGraphT

farhan5248 commented 10 months ago

I've gone through the jGraphT and jGraphX demos. I've also drawn out the graph models I want to have. Will work on creating a tests to convert a table (list of list) into graph model. I'll need one test for for the first two models and another for the last model type.

farhan5248 commented 10 months ago

Need tests for

[x] - single row table to graph
[x] - multi-row table to graph
[ ] - empty cell table to graph

Generate test automation using mvn org.farhan:mbt-maven-plugin:cucumber-to-uml -Dtag=debug

Add runner class

Make coverage methods

[x] - path coverage, every edge combination
[ ] - edge coverage, pick unpicked edges unless there are none
[ ] - vertex coverage, pick unpicked vertices unless there are none
[ ] - tag coverage, do path coverage, sort paths by most tags covered. keep the one with the most tags. keep paths which have tags not in the first one and "untag" these paths for tags in the first one. Sort these paths and now pick the first again. repeat until there are no tagged paths left to sort.

farhan5248 commented 10 months ago

After tag coverage, move some of the test methods into the main code to make wrappers for jGraphT to be used in the mbt-transformer. Then move the code + tests into the mbt-transformer.

farhan5248 commented 9 months ago

Integrated mbt-generator and validator into mbt-graph

farhan5248 commented 9 months ago

Next step is to integrate mbt-graph into the transformer

Install the mbt-graph jar
Update the xtext pom, fix broken refs
Install the xtext jar
Update the mbt-transformer pom, fix broken refs

farhan5248 commented 9 months ago

Integrate AsciiDoctor to create graph models from pages of sections of tables. 1 adoc file corresponds to 1 feature file for easy reverse engineering Each adoc has a top level feature section aka the rule, subsections for Background or Scenarios, sub-sub-section for Steps, each of those sub-sub-sections have one or more tables for the fields. Why 1 or more and not just one? Because it makes it easier to read. For now I'll just start with one though.