On https://issues.apache.org/jira/browse/TIKA-3816, Jason Guo posted an example file where Tika is not extracting text from a table element. The issue is that there's an SDT at the same level as the table row. It looks like we can get a list of sdts from the table object in the underlying bean. We should probably add a wrapper for these underlying bits? At the very least, we should add the example file so that we load the sdtrow into our beans in the general ooxml-schemas.
I added r1904079 but more work needs to be done to get the XWPFWordExtractor to CTSdtRow content.
I also added r1904081 This has test code that walks the Std xml objects so more classes will end up in poi-ooxml-lite (that jar is populated with poi-ooxml-full classes that are needed for our unit tests to run. I don't really use the XWPF code so don't really want to spend more time on this. Seems to me that XWPFTable needs to be extended so that it looks at the Std data as well as the CTRow and related cell data.