Intrinsic Redundancy in Model-to-Text Transformation

Demonstration of using intrinsic redundancy to support the integration and preservation of hand-written content within generated files (including deleting and editing generated lines). The prototype has been implemented on top of the EGL model-to-text transformation language.

How to Run

• Clone this repository
• Download and install Java 17
• Download and install Maven
• Run mvn compile exec:java from the command line
• Open the generated gen/Alice.html file
• Add a hand-written <p>Alice is an experienced systems analyst.</p> line to it as follows

<h1>Alice's Tasks</h1>
<p>Alice is an experienced systems analyst.</p>

<ul>
  <li>Analysis</li>
  <li>Design</li>
</ul>


<!--PGgxPkFsaWNlJ3MgVGFza3M8L2gxPg==--PHVsPg==-CTxsaT5BbmFseXNpczwvbGk+-CTxsaT5EZXNpZ248L2xpPg==-PC91bD4=---->

• Edit acme.flexmi to assign Alice to the Implementation task as follows

<?nsuri psl?>
<project title="ACME">
	<person name="Alice"/>
	<person name="Bob"/>
	<task title="Analysis" start="1" dur="3">
		<effort person="Alice"/>
	</task>
	<task title="Design" start="4" dur="6">
		<effort person="Alice"/>
		<effort person="Bob"/>
	</task>
	<task title="Implementation" start="7" dur="3">
		<effort person="Bob" perc="50"/>
		<effort person="Alice" perc="50"/>
	</task>
</project>

• Re-run the generator using mvn compile exec:java
• Observe how gen/Alice.html now contains a new Implementation row but also retains the hand-written <h2>Alice is awesome</h2> line

Running the Benchmarks

• Go to the root folder of the repository
• Run mvn clean package to build the project
• Run java -jar target/intred.jar -rf json to run the benchmarks and produce a JSON file with the results
  • The produced JSON file can be visualised in https://jmh.morethan.io
• To run each benchmark just once (e.g. to get a quick estimate of its execution time), use the following command

java -jar target/intred.jar -rf json -f 1 -wi 0 -i 1

• To run a specific benchmark just once, use the following command

java -jar target/intred.jar 'intred.benchmarks.IntredBenchmarks.intredTransformationWithChangesBenchmark' -rf json -f 1 -wi 0 -i 1

Benchmarks for the Paper

• Run the regularEglTransformationBenchmark and the intredTransformationBenchmark together

java -jar target/intred.jar  'intred.benchmarks.IntredBenchmarks.intredTransformationBenchmark' 'intred.benchmarks.IntredBenchmarks.regularEglTransformationBenchmark' -rf json -f 2 -wi 3

• Run the intredTransformationWithChangesBenchmark on its own

java -jar target/intred.jar  'intred.benchmarks.IntredBenchmarks.intredTransformationWithChangesBenchmark' -rf json -f 2 -wi 3