I'm trying to figure out how character encoding works in Boomi. I have a process that only works if I set the atom -Dfile option to UTF-8. Unfortunately, that breaks other processes that rely on the default, windows-1252. The file that I'm processing is an XML file that contains ISO-Latin-1 characters
After removing the -Dfile setting, I've tried encoding conversion after the file is read using a process shape to UTF-8, and I've tried using the encoding option on a map profile setting UTF-8. Neither of these work. Downstream I get errors indicating illegal characters in other shapes (e.g. a process shape that attempts to combine XML documents).
Is there a complete and thorough, and perhaps instructive, guide to how Boomi handles character encoding. The user guide is very terse. There's a lot of information on the forum on character encoding but it's disjointed, not comprehensive and not instructive.
What magic is the -Dfile option doing? How is it different from using explicit conversion with process shape? Is Java behind the scenes messing with me?
To add to my misery, I have to check the lengths of XML elements that contain ISO-Latin characters and load the XML document into a SQL server [2008 R2] XML type column. I made that work but that's journey down hack street.
Bottom line, everything works if -Dfile is set to UTF-8. Everything is sideways if its the default.