TOC generating problem in Sigil

[DiapDealer]

Nope. Never mind. Opening an html5 document still opens as an epub2. The doctype gets converted. No worries.

[KevinH]

No actually the code already existed in ImportHTML.cpp to handle creating the nav if needed.
I just added the code to create a default ncx and somehow ended up with two entries in the manifest for the same file.

I just pushed fixes for that. So we should now be okay (no crashing anymore).

That said the default NCX we create has a bad reference to Start Text/Section1.xhtml but that goes away when you Create the Table Of Contents.

It will take some diggng to find out the name and location of the first xhtml imported and to fixup the NCX code so it is always correct. I will look into adding that last piece over the weekend sometime.

None of this matters for AddExisting...

[KevinH]

All should now be fine in master for ImportHTML into an epub.

[harriduc]

I would like to pledge for keeping / repairing the HTML import:
I produce epub2 on Win10 to read on my tolino.
I carefully prepare the xhtml files. 00.xhtml is the title page of the epub containing all metadata that import into the epub.
My procedure: I open 00.xhtml then add all the chapters (01.xhtml ...) than save as epub. This worked perfectly until 0.9.12 (I think). With later versions Sigil crashed, but after opening the epub a couple of times the toc.ncx was produced. Ennerving but worked. Now Sigil 1.0.0 won't open the defective epub anymore .
Is there a different option to load prepared metadata?
Need I go back to 0.9.12?
Any suggestion is very appeciated
Btw. Sigil is great!! I wanted to create correct epubs but keep control over it. That drove me away from Calibre.

Thanks

[DiapDealer]

We already fixed it. The default ncx will be generated when opening html files. It will be in the next release of Sigil.

But I don't understand what you mean about importing prepared metadata to an epub via opening xhtml files. To my knowledge, Sigil doesn't parse xhtml for epub metadata. Epub metadata is strictly part of the opf file.

[KevinH]

It will try to read what metadata it can from html files that are imported but that is really a leftover from days when html metadata was more prevalent.

[harriduc]

This is part of my xhtml header:

<link rel='schema.DC' href='http://purl.org/dc/elements/1.1/' />
<meta name='DC.title' content='Titel' />
<meta name='DC.creator' content='Autor' />
<meta name='DC.description' content="Text" />
<meta name='DC.publisher' content='fanfiction.net' />
<meta name='DC.date' content='1958-07-23' />
<meta name='DC.language' content='en' />
<meta name='DC.contributor' content='Harri' />

It is much easier for me to handle than the metadata editor as the metadata are predefined (the same in all epubs). They appear in metadata editor as if I had entered there. Its just convenience.