Validation Package for our EMMSAD'20 Paper on Systematic Architecture Reconstruction of Microservice-Based Software Systems

Reviewing the Reconstructed Models of Lakeside Mutual

---

If you are interested in reviewing the reconstruction models of Lakeside Mutual expressed in our modeling languages and in the form after finishing Activity 6 of our SAR modeling method as described in the paper, please follow these steps:

1. Clone this repository to your harddrive.
2. Download an Eclipse version pre-bundled with the plugins of our modeling languages from the GitHub repository's release tab.
3. Start the downloaded Eclipse version and import the files contained in the Reconstruction Models folder of the cloned repository.
4. The reconstruction models are contained in the folders customer-core, customer-management-backend, customer-self-service-backend, policy-management-backend (each of which reflects one of the reconstructed backend microservices of Lakeside Mutual), and technology (which reflects the refactored technology models as mentioned in the paper). The intermediate folder may be ignored. The extensions for the different types of models are as follows:
   • .data: Domain models.
   • .mapping: Mapping models.
   • .operation: Operation models.
   • .services: Microservice models.
   • .technology: Technology models.

Please note that the source code of Lakeside Mutual, on which we based the validation of our SAR modeling method, is also contained in this repository. It is placed in the Validation LM Source Code folder.

The source code of our modeling languages may be found in this GitHub repository.

Retrieving the Data Mentioned in the Validation of Lakeside Mutual

---

Please follow these steps to retrieve the data presented in the Validation section for the Lakeside Mutual application (e.g., number of lines of code (LOC) of the input file set, number of reconstructed domain concepts, etc.):

1. Clone this repository to your harddrive.
2. Run the execute_activities_and_derive_tex_vars.sh bash script to retrieve the data of the SAR modeling method's activities. Please note that we do not provide a version of the scripts for the Windows operating system. However, the raw script outputs can be found in the Produced Result Files folder of the cloned repository (these files have the .out extension).
   Please note that the script simply invokes the sub-scripts activity-1-loc-analysis.sh, activity-2-domain-analysis.sh, and activity-3-service-analysis.sh on the top-level of this repository, as well as csv2tex_vars.py in the Produced Result Files folder of this repository. For the sub-scripts to work, the following prerequisites must be fulfilled:
   • activity-1-loc-analysis.sh: Requires the cloc utility. We used version 1.84 on Linux Mint 19.3.
   • activity-2-domain-analysis.sh: Requires Java 11 or higher, because the script executes a preliminary static analyzer (cf. the static-analyzer-standalone.jar in this repository). If you have Java 11 installed, but it is not your default Java environment, please adapt the export_java_home.sh script in the cloned repository as described in its source code.
   • activity-3-service-analysis.sh: The same prerequisites as for activity-2-domain-analysis.shapply.
   • csv2tex_vars.py: Requires Python 3. We used version 3.6.9 on Linux Mint 19.3.
3. Besides its direct output, which partially reflects the data mentioned in the Validation section of our paper, the execute_activities_and_derive_tex_vars.sh script will produce the following files in the Produced Result Files folder of the cloned repository:
   • activity-1-loc-analysis.csv: Contains LOC information for the Lakeside Mutual source code, which we examined and that is also part of this repository in the Validation LM Source Code folder.
   • activity-2-domain-analysis.csv: Contains raw data related to the Domain Modeling SAR activity of the Lakeside Mutual validation. These data will be used to derive TeX variables from them leveraging the csv2tex_vars.py Python script. The resulting TeX variables were effectively included to produce our submitted paper via pdflatex.
   • activity-3-service-analysis.csv: Like activity-2-domain-analysis.csv, but with raw data related to the Service Modeling SAR activity of the Lakeside Mutual validation.
   • tex_vars.tex: Contains the TeX variables derived from the previously mentioned CSV files via csv2tex_vars.py.

If for technical reasons the scripts do not work, please refer to the Produced Result Files folder in the cloned repository. It contains raw outputs of executing each of the mentioned scripts on Linux Mint 19.3. The raw outputs are stored in the files with the .out extension.