MobileRead Forums - View Single Post - encoding problem with mobi converted to epub

ldolse · 08-14-2009, 07:39 AM

Hi,

I'm seeing a problem with a .mobi file I picked up from Amazon. I converted it to epub for use on the Sony, and I'm seeing accented characters get screwed up. For example:

Code:

Provençal

The 'ç' will be displayed as gibberish in both the Calibre viewer as well as the Sony 505. Other accented characters do the same thing.

The original mobi appears to be fine, the mobi doc looks ok in Stanza and the iphone kindle app (but not in Calibre viewer).

What's really weird is that when I extracted the epub to copy/paste the screwed up text I discovered that Calibre had actually converted the text correctly, at least at the individual xhtml level. The xhtml files appears to be saved as UTF-8 with no BOM, and the <head> section specifies UTF-8 as the encoding as well.

So it seems something else in the epub isn't being set correctly causing the final encoding output to go awry. Any ideas as to what it could be?

Edit:
The screwed up text does actually display correctly on Desktop ADE as well, but not in Calibre viewer or mobile ADE on the 505.

08-14-2009, 07:39 AM	#1
ldolse Wizard Posts: 1,337 Karma: 123455 Join Date: Apr 2009 Location: Malaysia Device: PRS-650, iPhone	encoding problem with mobi converted to epub Hi, I'm seeing a problem with a .mobi file I picked up from Amazon. I converted it to epub for use on the Sony, and I'm seeing accented characters get screwed up. For example: Code: Provençal The 'ç' will be displayed as gibberish in both the Calibre viewer as well as the Sony 505. Other accented characters do the same thing. The original mobi appears to be fine, the mobi doc looks ok in Stanza and the iphone kindle app (but not in Calibre viewer). What's really weird is that when I extracted the epub to copy/paste the screwed up text I discovered that Calibre had actually converted the text correctly, at least at the individual xhtml level. The xhtml files appears to be saved as UTF-8 with no BOM, and the <head> section specifies UTF-8 as the encoding as well. So it seems something else in the epub isn't being set correctly causing the final encoding output to go awry. Any ideas as to what it could be? Edit: The screwed up text does actually display correctly on Desktop ADE as well, but not in Calibre viewer or mobile ADE on the 505. Last edited by ldolse; 08-14-2009 at 07:42 AM.