Looking for Volunteers to help update Sigil Users Guide

[BeckyEbook]

I have a question, by the way:
When merging xhtml files, does Sigil have to create this entry?

I ask this because every time after merging the files epubcheck shouts me warnings.

If this anchor has to be created for some reason (I commented on this line and didn't notice any problems) then maybe you can at least surround it with paragraph tags.

[DNSB]

I think that those IDs are used to update the navigation document. ePub3 does not complain about the naked <a> tags but ePub2 does so wrapping them in <p> tags would be good.

[KevinH]

I will push a change to wrap it in p tags.

[KevinH]

Update:

The more I think about this, wrapping it in p tags will take up a line/space on the page that was previously not in the epub.

Would something like:

<p id="blah" style="display:none" hidden="hidden"></p>

be better since it would not be visible on the finished page taking up no space but still serve the same purpose as a zero space a anchor point

Thoughts?

[KevinH]

And the current html spec as used in epub3 does allow an anchor a tag to be a direct child of the body tag as it is considered "flow content".

So for epub3, the current anchor tag is correct.

I have no idea why epubcheck under epub2 would consider this an error but perhaps this is something new for epubcheck?

[KevinH]

So I think the right approach here is to leave as it is now for epub3 and use something like the following for epub2:

<p id="blah" style="display:none" hidden="hidden"></p>

Would that work for you?

[BeckyEbook]

Theoretically, yes.
However, I recognize this code because it reminds me of your code from the TOCSaver plugin.
There, too, there is a problem (only in epub2) that epubcheck considers the hidden attribute invalid.

Col: 53: ERROR(RSC-005): Error while parsing file: attribute "hidden" not allowed here; expected attribute "class", "dir", "lang", "title" or "xml:lang"

So I would suggest something along these lines:

            QString version = sink_resource->GetEpubVersion();
            if (version.startsWith("3")) {
                new_bodies.append("  <p id=\"" + section_id + "\" style=\"display:none\" hidden=\"hidden\"></p>\n" + updated_bodies[bookpath]);
            } else {
                new_bodies.append("  <p id=\"" + section_id + "\" style=\"display:none\"></p>\n" + updated_bodies[bookpath]);
            }

I built Sigil just now and it passes epubcheck for epub2 and epub3.


I know your approach to epub2, however, in Poland at least 80-90% of new titles in bookstores are still epub2. The volume of epub3 files is gradually increasing, but this is happening extremely slowly.

P.S. In the modified TOCSaver plugin I use a similar condition (in three places in the code) and it works very well.

[KevinH]

Great!
I will use anchor tags only for epub3 and use what you have tested for epub2 (ie. without the hidden attribute).

Thanks!

I will update the TOCSaver plugin to do the same.