Error when using neural network rules

[cattaneod]

When getting suggestions based on neural network rules, it gives me the following error:

[Error - 12:46:12] Request textDocument/codeAction failed. Message: Internal error. Code: -32603 java.util.concurrent.CompletionException: java.lang.NullPointerException at java.util.concurrent.CompletableFuture.encodeThrowable(Unknown Source) at java.util.concurrent.CompletableFuture.completeThrowable(Unknown Source) at java.util.concurrent.CompletableFuture.uniApply(Unknown Source) at java.util.concurrent.CompletableFuture$UniApply.tryFire(Unknown Source) at java.util.concurrent.CompletableFuture.postComplete(Unknown Source) at java.util.concurrent.CompletableFuture.complete(Unknown Source) at org.eclipse.lsp4j.jsonrpc.RemoteEndpoint.handleResponse(RemoteEndpoint.java:212) at org.eclipse.lsp4j.jsonrpc.RemoteEndpoint.consume(RemoteEndpoint.java:193) at org.eclipse.lsp4j.jsonrpc.json.StreamMessageProducer.handleMessage(StreamMessageProducer.java:194) at org.eclipse.lsp4j.jsonrpc.json.StreamMessageProducer.listen(StreamMessageProducer.java:94) at org.eclipse.lsp4j.jsonrpc.json.ConcurrentMessageProcessor.run(ConcurrentMessageProcessor.java:113) at java.util.concurrent.Executors$RunnableAdapter.call(Unknown Source) at java.util.concurrent.FutureTask.run(Unknown Source) at java.util.concurrent.ThreadPoolExecutor.runWorker(Unknown Source) at java.util.concurrent.ThreadPoolExecutor$Worker.run(Unknown Source) at java.lang.Thread.run(Unknown Source) Caused by: java.lang.NullPointerException

However, the suggestions still work normally, just the console give that error.

[valentjn]

@catta202000 The stack trace only tells us there is a NullPointerException somewhere, so I need to reproduce this. I've never used neural network rules. Could you add steps to reproduce this (what do I have to install, to which directory did you set ltex.additionalRules.neuralNetworkModel, etc.) or point me to some instructions?

[cattaneod]

Ok, my settings are:

"ltex.additionalRules.languageModel": "C:\\LanguageTool\\ngrams", "ltex.additionalRules.word2VecModel": "C:\\LanguageTool\\word2vec", "ltex.additionalRules.motherTongue": "it", "ltex.language": "en-GB",

Both n-grams and word2vec are the models downloaded from the languagetool website.

A sentence that gives the error:
"And the taken."

[valentjn]

Ah, I didn't realize that word2vec is also based on neural networks. The stack trace is a bit longer for me:

Caused by: java.lang.NullPointerException
	at org.bsplines.languagetool_languageserver.LanguageToolLanguageServer$1.lambda$codeAction$0(LanguageToolLanguageServer.java:268)
	at java.util.concurrent.CompletableFuture.uniApply(CompletableFuture.java:602)
	... 13 more

match.getSentence() seems to be null, which is due to the match being constructed by the word2vec rules with a constructor of RuleMatch that does not take an AnalyzedSentence (which has been deprecated since LT 4.0). Hence, no quick fixes are offered, so it wasn't only an error message in the log.

The workaround is to not support ignoring word2vec matches until AnalyzedSentence is provided.

[valentjn]

Fix released in 4.7.3.