On 07/09/2012 10:24 PM, Owen Watson wrote: > Would --local-encoding=UTF-8 fix it?
Unlikely. IIRC, that changes how wget behaves in terms of deciding how to translate non-ascii URLs (IRIs) on the command-line, and I think how it saves non-ascii file names, but I don't believe it will modify file contents. Basically, your best bet is to run an equivalent to that Unix pipeline: something that can run through all the archived files, and correct their meta http-equiv stuff. If it were me, I'd probably install the "find" and "sed" commands so I could run that exact pipeline (except the single-quotes would have to be double-quotes in a dos pipeline, I think). Maybe install msys or some other kit that provides such commands. Cygwin's overkill, I'm sure... Note that what it would NOT do, was correct files that don't SPECIFY their character encoding... Firefox always assumes latin1 if nobody tells it otherwise. Probably because that's what it's supposed to do, according to the HTML specification, IIRC. However, Google Chrome at least, appears to auto-detect the content's encoding. Perhaps FF has a config that supports this too (or you could force FF to use UTF-8). Good luck! -mjc
