Im been working in a document converter, from docx to xml-jats using node-js with javascrypt. I tried to narrow it down to the specific docx template used for the journal im working with. But im been stucked for a week. Right now the problem is that the xml are generated as mere headers and the htmls are pretty much fine but the subtitles are generated oddly, i dont even know what thinks should i attach. this version of the code in particular does read the xml but generates it very oddly mixing everything. Im really desperate, im sorry if im not being clear enough. the deadline from this is in just one month.