Obscure ePub validation error

[brahms]

Hi, and thanks for your help.

I've built the ePub from scratch using Notepad++ and I'm using dummy content files to get the package working.

I have only one remaining error when I use the online epub validator at http://validator.idpf.org/, but I can't see the cause.

The error is:
./tmp/uploads/1455200484959_0000117833/inputFile/mybook5.epub/OEBPS/content.opf(-1,-1)

There was an error when parsing the EPUB version: Version not found.

My content.opf starts like this (I've replaced confidential info below with XXXX):

<?xml version="1.0" encoding="utf-8"?>
<package version="2.0" xmlns="http://www.idpf.org/2007/opf" unique-identifier="BookId">
<metadata xmlns:dc="http://purl.org/dc/elements/1.1/" xmlnspf="http://www.idpf.org/2007/opf">

<!--Required Metadata-->
<dc:title>XXXXXX</dc:title>
<dc:language>en-gb</dc:language>
<dc:identifier id="BookId" opf:scheme="ISBN">urn:isbn:XXXXXXX</dc:identifier> <!--Use the Same for the toc.ncx file -->
<dc:creator opf:role="aut">XXXXXXX</dc:creator>
<dcublisher>XXXXXX</dcublisher>
<dc:date>2016-03-14</dc:date>

<meta name="cover" content="My_Cover_ID"/> <!--Required for KindleGen-->

etc etc

<metadata>

I see above that code "colon o" got interpreted as an icon above. The original looks ok.

Thanks for your help

[Toxaris]

It looks alright from what I can see. Did you store the opf with UTF-8 encoding?

[brahms]

Originally Posted by Toxaris

It looks alright from what I can see. Did you store the opf with UTF-8 encoding?

Thanks for replying.

No, ANSI. I've just tested saving as utf-8 and still the same error. Which should it be? UTF-8 renders a hyphen as x97 in the <dc:description>.

Dave

[JSWolf]

Download epubcheck 4.0.1 and try that. There is no indication what version validator is being use online.

[brahms]

Thanks JS; I'll give that a shot later and get back.

[Toxaris]

Originally Posted by brahms

Thanks for replying.

No, ANSI. I've just tested saving as utf-8 and still the same error. Which should it be? UTF-8 renders a hyphen as x97 in the <dc:description>.

Dave

Always use UTF-8 as file encoding. That is easy enough with Notepad++ and as you already specify that the content is UTF-8, you must ensure that it actually is. If characters change is you change the encoding, reenter the required character.

[brahms]

Thanks Toxaris. Will do.

A how-to book i'm using says encode the toc.ncx in ANSI. Is that correct?

[Toxaris]

Originally Posted by brahms

Thanks Toxaris. Will do.

A how-to book i'm using says encode the toc.ncx in ANSI. Is that correct?

No. To prevent issues ensure that all text files (so, including toc.ncx) are encoded in UTF-8.

[brahms]

Many thanks guys. The problem was caused by the wrong encoding of some files. Notepad++ seems to remember the last file's encoding setting so if you resave a file without checking it can re=encode.

Or something like that.

Again, many thanks.

Dave

[JSWolf]

In some older version ADE, I've seen display errors due to not being encoded as UTF-8 BOM.