epub validation question

[DNSB]

Originally Posted by Quoth

I wonder is Vellum really bad or are they just clueless at using it?

Vellum takes a docx input file (Word, LO Writer, Scrivener, Pages, etc.) and converts it to an ePub file (I think mobi is an alternate output but don't actually know anyone who has used it). The major issues some people seem to have with it is that it has two stylesheets, one of which is dedicated to media queries and that it loves using masses of <div>s. OTOH, it is pretty consistent and so easy to clean up. I have a collection of saved searches which automate pretty much the whole procedure.

The ePub output can be aimed at Kindle for upload to Amazon or ePub for everywhere else. The main difference being that the Amazon specific media queries are not present in the ePub for elsewhere.

[theducks]

My main complaints:
Huge spacing on titlepage (overflows onto second screen. design for Print or Hires screens?)
Uses letterspacing which kepub mishandles the space between words (makes it really small)
Chapter43 not Chapter 43
I have saved searches to correct both (I just comment out that in the CSS)

[JSWolf]

Originally Posted by Quoth

I wonder is Vellum really bad or are they just clueless at using it?

Vellum really is that bad. It should be banned.

[odamizu]

Vellum drives me nuts. If there is an easy way to get rid of the excessive divs with the multitude of classes and what that does to the css stylesheet, I would love

I use Sigil with TagMechanic to manually reduce the number of divs one by one and cssRemoveUnusedSelectors to clean up the css, but it's an arduous, manual process for me.

I'm also perplexed by the media queries. I usually just delete the entire media query stylesheet then manually edit the main stylesheet to my liking. Again, an arduous, manual process.

Advice, recommendations, instruction, tips — all greatly appreciated.

[JSWolf]

Originally Posted by odamizu

Vellum drives me nuts. If there is an easy way to get rid of the excessive divs with the multitude of classes and what that does to the css stylesheet, I would love

I use Sigil with TagMechanic to manually reduce the number of divs one by one and cssRemoveUnusedSelectors to clean up the css, but it's an arduous, manual process for me.

I'm also perplexed by the media queries. I usually just delete the entire media query stylesheet then manually edit the main stylesheet to my liking. Again, an arduous, manual process.

Advice, recommendations, instruction, tips — all greatly appreciated.

I do it similarly with Vellum made eBooks. As for the excessive chapter header divs, I dump them all as they aren't needed.

[DNSB]

Originally Posted by odamizu

Vellum drives me nuts. If there is an easy way to get rid of the excessive divs with the multitude of classes and what that does to the css stylesheet, I would love

I use Sigil with TagMechanic to manually reduce the number of divs one by one and cssRemoveUnusedSelectors to clean up the css, but it's an arduous, manual process for me.

I'm also perplexed by the media queries. I usually just delete the entire media query stylesheet then manually edit the main stylesheet to my liking. Again, an arduous, manual process.

Advice, recommendations, instruction, tips — all greatly appreciated.

Since Vellum is pretty consistent, I have a collection of saved searches and automations in Sigil to do the clean up.

1. Replace the media queries stylesheet with a stylesheet with about 5 entries
2. Run the HTML saved search/replace set
3. Run the first automation collection
4. Run the CSS cleanup saved search/replace set
5. Manually run CSS reformat (multiple lines per style) to clean up the blank lines in the stylesheet
6. Run the second automation collection
7. Run the Afters saved search/replace set
8. Use Tag Mechanic to remove blank <div> and <span>
9. Run the second automation to check everything is okay

Some oddities are that the HTML search/replace set replaces any &lt; and &gt; in the HTML files with #lt; and #gt; since the CustomCleanerPlus plugin I use with Sigil is not happy with them. The Afters reverses that change. One author loves using "&lt;&lt; text string &gt;&gt;" in their books to indicate non-verbal communication between two characters. << do you believe in telepathy? >>

[odamizu]

Originally Posted by JSWolf

I do it similarly with Vellum made eBooks. As for the excessive chapter header divs, I dump them all as they aren't needed.

Yes, those chapter header divs are ones that I delete, too.

Originally Posted by DNSB

Since Vellum is pretty consistent, I have a collection of saved searches and automations in Sigil to do the clean up.

1. Replace the media queries stylesheet with a stylesheet with about 5 entries
2. Run the HTML saved search/replace set
3. Run the first automation collection
4. Run the CSS cleanup saved search/replace set
5. Manually run CSS reformat (multiple lines per style) to clean up the blank lines in the stylesheet
6. Run the second automation collection
7. Run the Afters saved search/replace set
8. Use Tag Mechanic to remove blank <div> and <span>
9. Run the second automation to check everything is okay

Some oddities are that the HTML search/replace set replaces any &lt; and &gt; in the HTML files with #lt; and #gt; since the CustomCleanerPlus plugin I use with Sigil is not happy with them. The Afters reverses that change. One author loves using "&lt;&lt; text string &gt;&gt;" in their books to indicate non-verbal communication between two characters. << do you believe in telepathy? >>

Thanks DNSB. I must admit that's pretty overwhelming to me, but I appreciate the roadmap