Aaron Grewell wrote: > On Wednesday 05 April 2006 11:38, Matt Kettler wrote: >> Sounds like you've got something other than spaces/tabs on that line (ie: >> unprintable escape codes), or the file was created on a windows machine >> with windows end-of-line format but is being read by a *nix machine. > > Ah, that's it. According to file(1) the thing ended up in UTF-8. That's a > side effect of my spreadsheet addiction, I was using OpenOffice Calc (on > Linux) to edit the file. I found a program that will do the conversion to > ASCII, so this should be pretty easy to fix once I get that built. > >> In general the line should be perfectly valid, so there must be something >> else going on. Although I will comment that 6.5 is a rather insane choice >> of score for BAYES_60, it should not create a parse error. > > It's high, yes, but Bayes is quite accurate here. We're pretty FP-sensitive, > so if it was causing a problem I would know by now. >
If the line isn't being parsed, it's not causing any FP problems because it's not being used.
