View Full Version : Failed ePub Validator

12-08-2010, 07:44 AM
Hi! ThreePress Consulting sent me here after my document failed their ePub validator, and gave me obscure, exquisitely indecipherable error messages.

Would anyone be willing to help me? I converted my book to ePub using the SmashWords "Meatgrinder," so I have no clue what errors might have been created or transferred during the conversion process, and indeed, if truth be told, I have no clue what ePub even is.

Here's the error message I got;

1. ERROR: how-to-learn-french-for-free-on-the-internet.epub/tmp_55090552e15d51ee2c08122b339ed0f6_ZdNnzu.fixed. tidied.xfixed.html(1506): attribute "start" not allowed at this point; ignored

2. Check finished with warnings or errors!

If I have posted in the wrong area, would someone please direct me to the proper place? Thanks!

I hope someone can help me. I can't make anything out of the above error, nor from the "descriptions" of the errors that ThreePress provides


Nate the great
12-08-2010, 08:42 AM
Open the Epub in Sigil.

Look in that file on that line and see what's going on. Copy the several lines around the error and paste them here so we can see.

My bet is that some tags weren't converted properly from your source.

12-08-2010, 11:33 PM
Barry, can you give us more details about your source file?

You say you don't know what epub is - the main core of an epub is a html file, containing your text. And to get a valid epub file, your html file has to be valid xhtml 1.1. If it ain't valid, the whole epub will never be valid.

So I would recommend you use a validator from the internet to see if the source html file is really xhtml 1.1 valid.

I don't know this Meatgrinder (I only know a music band with that name). There are a few different converting programms, some very popular are Calibre and Sigil. But once you checked your html file for validity, you can give Meatgrinder another try.

12-09-2010, 09:12 AM
Thanks for responding. Meatgrinder is a tool used by Smashwords (ebook publisher) to convert a Word document to many different formats, including ePubs. So, I have no idea how proficient its xhtml conversion ability is. Would you recommend that I recreate the ebook in compliant xhtml, then convert it myself (using Calbre or Sigil)? I think I've answered my own question. Thanks so much for your response.


12-09-2010, 09:45 AM
I doubt calibre would yield results that would validate either. Kovid has rather strong feelings about that. (

I wouldn't worry too much about non-validation so long as you've tried it all the major epub software and it in fact works OK. That said, it is of course better to try to meet standards if you can. Beginning with compliant XHTML to begin with no doubt is a step in the right direction.

Nate the great
12-09-2010, 10:56 PM
Since you only have the one error the fastest way to fix it would be Sigil.

Heck, send me the file. I bet I can fix it in less than 3 minutes.