Cartographica has extensive support for unicode and importing and exporting of files in other character formats. However, for some files, it's impossible to guess what character set was used in creating it.
For some file formats, there are markers that indicate the character set. However, unfortunately, these are often not used. For example, it is common to find ESRI Shapefiles that are encoded as Local OEM (LDID 57 for those who are into the dbf file format) and which have had their data encoded in just about any format.
Some formats are nicely self-describing, such as KML, but most of them are legacy formats and don't do a good job of indicating how they are encoded.
For shapefiles, the solution is simple. To specify the character set for the file when you import it, choose File > Import Vector Data… and then choose your desired character set from the menu at the bottom of window.
Internal Cartographica Use
Internally, Cartographica always uses UTF-8. If you import data and then modify it (or otherwise end up changing the file from a reference to an included file), the data will be stored in UTF-8 inside of the Cartographica Map Set document.