Error Generating Artifacts

[tuc-fmay]

Hallo all,

I checked out the current master branch.

It builds through but problems arising when generating the language artifacts of SysML.xtext.

It results in the ollowing error:

As a result, alternative(s) 4 were disabled for that input
error(201): ../org.omg.sysml.xtext/src-gen/org/omg/sysml/xtext/parser/antlr/internal/InternalSysML.g:10056:2: The following alternatives can never be matched: 6

295564 [main] INFO  text.InternalParserSplitterFragment  - Reading src-gen/org/omg/sysml/xtext/parser/antlr/internal/InternalSysMLParser.java
295570 [main] ERROR text.InternalParserSplitterFragment  - java.nio.file.NoSuchFileException: src-gen/org/omg/sysml/xtext/parser/antlr/internal/InternalSysMLParser.java
295704 [main] INFO  text.xtext.generator.XtextGenerator  - Generating common infrastructure
295792 [main] ERROR mf.mwe2.launch.runtime.Mwe2Launcher  - Problems running workflow org.omg.sysml.xtext.GenerateSysML: 
[ERROR]: GeneratorException: (Element: -UNKNOWN-; Reported by: XtextGenerator)
	 org.eclipse.emf.common.util.WrappedException: java.lang.reflect.InvocationTargetException

It seems to be a problem finding the content for the InternalParserSplitterFragment.

I actually have some serious understanding issues :) Why is the so called template for splitting up the parser in the src-gen folder?
Why the splitting is neccessary is understandable for me.

Any hints or thoughts are welcome!

Thank you very much!

[seidewitz]

The problem is that there was an error generating the parser. Since the parser was not generated, there was not anything to for the splitter to split.

I don't know why you got an error generating the SysML parser. The parser on master was successfully generated and tested before being pushed to the repository. No one else has reported any problem. Did you make any change in the grammar?

Even if you made no change in the grammar, you may need to regenerate KerMLExpressions.xtext and KerML.xtext before SysML.xtext. Note, though, that you don't actually have to regenerate the parsers before using the implementation. All the generated files are already provided in the repository.

(The need for the splitter is described in the "Technical Update" section at the end of the description of PR #608.)

[tuc-fmay]

Thank you for your quick reply.

I also thought it might be a project setup issue because no one else had similar problems so far.

After your reply, I investigated a little bit further ...

It actually worked after raising the maximum heap size.

Just a hint for anyone which might stumble across that.