Guillermo Gutiérrez

We need to discuss whether it would be safe to fix this bug in case users rely on JavaRosa's current behavior, even if it's not up to spec. @tiritea commented...

Hi, @joeflack4! I believe everything you're reporting is actually what we should expect. JR depends on SLF4J for logging but it doesn't include any specific implementation of SLF4J. This lets...

> I still continued to see messages until I included the slf4j no-op dependency in my package. I wonder why... That's expected too: the api is not an implementation of...

There's something that I can't understand what purpose it serves in the `TreeReference.parent()` implementation: ```java public TreeReference parent(TreeReference baseReference) { if (isAbsolute()) { return this.clone(); } else { TreeReference newRef...

With a small tweak we could '../../../foo'.parent('bar/baz') into '../foo', which is weird but feels like something that should be valid.

Copied from https://github.com/opendatakit/javarosa/pull/350#issuecomment-415514714 > For `parent`, we discussed that the `null` is because of the limitation described in "However, if this ref has 'parent' steps (..), it can only be...

Just a refresher on latest work on this front: We have improved the test suite around the DAG, which includes specific cases of cycles in the DAG due to self-references...

So, I've tried this in a test and it turns out that `[position() = 1]` works as expected. The error I'm actually getting is that `XPathPathExpr` doesn't know what to...

I need to wrap my head around this and understand it. I'll try to say something as soon as possible :P

@lognaturel affirms that "I believe Gareth had found something in the XForms spec that suggested that expressions should be evaluated regardless of relevance. This has performance implications, certainly.". @tiritea, could...