TeX, HTML and other text formats use special codes and commands to indicate special characters that may not be available on your keyboard. For spell-checking these codes need to be translated into ANSI characters because the dictionaries are ANSI based.
When saving a file it can be useful to translate the ANSI characters back to their special codes. You can switch this feature on or off in the "Options" tab sheet. If you haven't specified all the commands you use for ANSI characters, words containing codes will be misinterpreted and usually split into several words. If you see this happening, you should revise the accent file (see the "Format" tab sheet).
The accents file is format dependent. For the TEX format the file TEXACCENTS.4SPELL is used (filename syntax is: "format"ACCENTS.4SPELL). This file could look like this:
\"u=ë \'u=ú \`u=ù \^u=ûN.B. remember \"u is something different then \"{u} or {\"u} or "u
The HTML equivalent would look like this:
ë=ë ú=ú ù=ù û=ûThe file states that the characters to the left of the equal sign are the code that has to be replaced by the ANSI code specified to the right of the equal sign.