I'm almost sure that it is a unicode character like non-breaking-space (U+00A0), which shows up as  (U+00C2) when you have set your display to iso-8859 instead of UTF-8 because the bytes to encode U+00A0 in UTF-8 are \xC2\xA0.
There is no way to tell from your problem description where in the process you fail.
In reply to Re: What's with the weird accent "a" after I parse a table?
by Tux
in thread What's with the weird accent "a" after I parse a table?
by jdlev
| For: | Use: | ||
| & | & | ||
| < | < | ||
| > | > | ||
| [ | [ | ||
| ] | ] |