MobileRead Forums - View Single Post

Barb-B · 11-09-2012, 10:11 PM

It has to do with the specified character set. If you use mobi-unpack tool to disassemble it to its component parts, you can open the main html file with notepad or wordpad and see what character encoding has been used. Then when you convert in calibre, you can supply that character set name in the "Look & Feel" section of the conversion dialog.

My experience has been that nearly every time, it is iso8859-1. So now I just try that first, before bothering with taking anything apart. On the rare occasion it doesn't work, then you can still dig into the mobi file to see what it really is.

Also, you're best off NOT trying to make the changes in Sigil, which assumes by default that everything is UTF-8. Once you've opened the file with Sigil and then saved any changes. You're pretty much stuck with the bad characters, until you do as cybmole suggests, and manually edit them. Ugh.

11-09-2012, 10:11 PM	#3
Barb-B Tech appreciator Posts: 23 Karma: 10 Join Date: Jul 2012 Device: Onyx Boox Kepler Pro (wow I love this reader!)	It has to do with the specified character set. If you use mobi-unpack tool to disassemble it to its component parts, you can open the main html file with notepad or wordpad and see what character encoding has been used. Then when you convert in calibre, you can supply that character set name in the "Look & Feel" section of the conversion dialog. My experience has been that nearly every time, it is iso8859-1. So now I just try that first, before bothering with taking anything apart. On the rare occasion it doesn't work, then you can still dig into the mobi file to see what it really is. Also, you're best off NOT trying to make the changes in Sigil, which assumes by default that everything is UTF-8. Once you've opened the file with Sigil and then saved any changes. You're pretty much stuck with the bad characters, until you do as cybmole suggests, and manually edit them. Ugh. Last edited by Barb-B; 11-09-2012 at 10:15 PM.