Google Docs to EPUB?

[wlashack]

Hello,
someone has experience with converting Google Documents to EPUB format?
I have problems with all bulleted and numbered lists. They do not appear in ebook readers.

No matter if I use Google Docs option "Download as EPUB" right in the menu or if I try to open it in Sigil, the results are always the same — no bullets, no numbers.

I also tried the Sigil plugin for opening Google Docs (saved as HTML) but the same results.

Can someone help? Thanks.

[BeckyEbook]

Drastic method: remove all css styles from the xhtml file header. Usually enough.

[Tex2002ans]

Originally Posted by wlashack

I have problems with all bulleted and numbered lists. They do not appear in ebook readers.

You have to open the EPUB using Sigil (or Calibre's Editor) and see the actual HTML+CSS.

If you could go into the book and grab the HTML of a broken list and show the CSS, perhaps we could help some more.

It's very likely they are using CSS that isn't supported on ereaders (content: counter, list-style-type, :before, etc. etc.).

Side Note: Support for lists in ereaders are........ poor. Which is why many of us recommend hardcoding numbers into the text itself:

<p>1. One</p>
<p>2. Two</p>
<p>3. Three</p>

instead of:

<ol>
<li>One</li>
<li>Two</li>
<li>Three</li>
</ol>

1. One
2. Two
3. Three

[Maui]

But this does not explain why the bullets/numbering are gone/no longer visible. in your first case there should not be such problems. In your second case, at least Sigil should be able to display them.

But: without having an example here this can only be guessing with a lot of possible root causes.

[wlashack]

Originally Posted by Maui

But this does not explain why the bullets/numbering are gone/no longer visible. in your first case there should not be such problems. In your second case, at least Sigil should be able to display them.

But: without having an example here this can only be guessing with a lot of possible root causes.

I agree,
here is the Google Document I need to convert to epub format properly, so you can try: https://docs.google.com/document/d/1...it?usp=sharing

I tried to remove all styles as suggested by BeckyEbook but it did not leed to the result.

I also tried to open it in Calibre and then use the button Convert Ebook. The result is now most close to what I need to get.

Thanks.

[Maui]

Thats a webpage in a language i dont understand, so its not obvious to me how you are content to sigil.

[Tex2002ans]

Originally Posted by wlashack

I agree,
here is the Google Document I need to convert to epub format properly, so you can try: https://docs.google.com/document/d/1...it?usp=sharing

I tried to remove all styles as suggested by BeckyEbook but it did not leed to the result.

I also tried to open it in Calibre and then use the button Convert Ebook. The result is now most close to what I need to get.

.... yeah, the code out of that Google Doc uses an abomination of CSS3 (counter-increment + counter-reset) and nested ordered lists.

Many ereaders don't support CSS3.

It also uses the HTML start and value element to renumber the lists, which doesn't carry over to many EPUB readers (they are very buggy and start from 1 in many cases).

Also, it tries to use the CSS list-style-type to try to specify lowercase letters + roman numerals, which just doesn't work well on ereaders.

The only way to get this to work on ebooks is to hardcode the letters/numbers instead.

Before (I ran it through Calibre to make it a tiny more readable... the original Google Docs -> EPUB was... horrifying):

Spoiler:

<h4 id="id_1g8uqbvmo1o" class="block_18">Nic mi nechybí (vše potřebné mi bylo dáno)</h4>
	<ol class="list_">
	<li value="18" class="block_13"><span class="text_10">Narodil jsem se přesně takový, jaký mám být</span><span class="text_9">. Byl jsem optimálně přednastaven.</span></li>
</ol>
	<ol class="list_1">
	<li class="block_11"><span class="text_5">Vypadám i funguji přesně tak, jak mám</span><span class="text_9">. Není na mně co měnit.</span></li>
</ol>
	<ol class="list_1">
	<li class="block_11"><span class="text_5">Mám všechny potřebné talenty</span><span class="text_9">. Neumím jen to, co potřebovat nebudu.</span></li>
</ol>
	<ol class="list_2">
	<li class="block_12"><span class="text_9">Život se sám stará, abych v sobě vše potřebné rozvinul (není co vymýšlet).</span></li>
</ol>
	<ol class="list_1">
	<li value="2" class="block_11"><span class="text_5">Mé touhy a role jsou nastaveny v souladu s mými talenty a schopnostmi</span><span class="text_9">. Je tím zajištěno, že mě bude můj život naplňovat.</span></li>
</ol>

After:

Spoiler:

<h4 id="id_1g8uqbvmo1o">Nic mi nechybí (vše potřebné mi bylo dáno)</h4>

<div class="list">
	<p class="listone">18. <b>Narodil jsem se přesně takový, jaký mám být</b>. Byl jsem optimálně přednastaven.</p>

		<p class="listtwo">a. <b>Vypadám i funguji přesně tak, jak mám</b>. Není na mně co měnit.</p>

		<p class="listtwo">b. <b>Mám všechny potřebné talenty</b>. Neumím jen to, co potřebovat nebudu.</p>

			<p class="listthree">i. Život se sám stará, abych v sobě vše potřebné rozvinul (není co vymýšlet).</p>

		<p class="listtwo">c. <b>Mé touhy a role jsou nastaveny v souladu s mými talenty a schopnostmi</b>. Je tím zajištěno, že mě bude můj život naplňovat.</p>
</div>

with ~ this CSS:

Spoiler:

div.list {
	margin-top: 1em;
	margin-bottom: 1em;
}
p.listone {
	text-align: left;
	text-indent: -2em;
	margin-left: 2em;
}

p.listtwo {
	text-align: left;
	text-indent: -2em;
	margin-left: 4em;
}

p.listthree {
	text-align: left;
	text-indent: -2em;
	margin-left: 6em;
}

[wlashack]

Originally Posted by Maui

Thats a webpage in a language i dont understand, so its not obvious to me how you are content to sigil.

Oh sorry, it is a Google Docs webpage so I thought everyone sees it the language according to his system preferences. Please go to the left of the top bar and select "Soubor" (File) and from the dropdown choose "Stáhnout jako (Download as) and select the last option "Publikace EPUB (.epub)".

It is the dorect link to the Document so you can download it. I wanted you to see the original file, this is why I posted a link to the Google Document.

I also attached the .epub and .PDF of the document, if you have trouble seeing the Google Doc.

[DiapDealer]

@Tex2002ans has the right of it. There's list-related css there that simply will never work on many, many epub rendering systems.

[wlashack]

Originally Posted by Tex2002ans

.... yeah, the code out of that Google Doc uses an abomination of CSS3 (counter-increment + counter-reset) and nested ordered lists.

Many ereaders don't support CSS3.

It also uses the HTML start and value element to renumber the lists, which doesn't carry over to many EPUB readers (they are very buggy and start from 1 in many cases).

Also, it tries to use the CSS list-style-type to try to specify lowercase letters + roman numerals, which just doesn't work well on ereaders.

Thanks for the info, it helped me a lot in understanding what is going on. Meantime I found that when I Download Google Doc as EPUB, then Open it in Calibre and pust "Convert ebook" button here, Save the autoconverted .epub file and open it in Sigil, it looks almost perfect (only some issues to polish like left indent of numbered lists, big space above subhead paragraphs...; I hope it could be done automatically by some rule in HTML).

I use Moon+ Reader for Android to read ebooks and this application can display "original formatting". So it can displays the document the same way as Sigil, with all the bullets and numbers. But for some reason it can display only one chapter at a time in this mode so to "turn page" to another part of the document, you need to switch to "normal" view, open chapter and switch to "original formatting" again. A little bit frustrating.

[DiapDealer]

Originally Posted by wlashack

I use Moon+ Reader for Android to read ebooks and this application can display "original formatting". So it can displays the document the same way as Sigil, with all the bullets and numbers. But for some reason it can display only one chapter at a time in this mode so to "turn page" to another part of the document, you need to switch to "normal" view, open chapter and switch to "original formatting" again. A little bit frustrating.

That's one of the very reasons I gave up trying to work with Moon+. It's fantastic for someone who wants to diddle every single aspect of how ebooks look (without editing the epub), but not so good at "original format."

[wlashack]

Originally Posted by DiapDealer

That's one of the very reasons I gave up trying to work with Moon+. It's fantastic for someone who wants to diddle every single aspect of how ebooks look (without editing the epub), but not so good at "original format."

Have you a better tip how to read "original format" on Android (and even on Mac)?

[DiapDealer]

Originally Posted by wlashack

Have you a better tip how to read "original format" on Android (and even on Mac)?

"Better" is extremely subjective, of course. But my go-to epub reading app on Android is Bookari Premium (formerly Mantano). It uses Adobe's RMSDK rendering engine for EPUB2 and Readium's for EPUB3.

[wlashack]

Thanks, I give it a try!

[Tex2002ans]

Originally Posted by wlashack

I use Moon+ Reader for Android to read ebooks and this application can display "original formatting". So it can displays the document the same way as Sigil, with all the bullets and numbers.

No, Moon+ is the exact opposite of "Original Formatting". What they do is throw away almost everything in the book's CSS, and overwrite it with Moon+'s settings. They also don't follow any sort of standard, and try to smush ALL document formats into Moon+'s way of doing things.

As DiapDealer said, Bookari is one of the standards-compliant readers on Android (it uses the same engine as the EPUB ereaders).

Gitden is also a pretty decent Android EPUB2/3 reader.

Originally Posted by wlashack

But for some reason it can display only one chapter at a time in this mode so to "turn page" to another part of the document, you need to switch to "normal" view, open chapter and switch to "original formatting" again. A little bit frustrating.

... mostly because the code generated from that Google Doc is horrifying. There is a giant block of hundreds of CSS classes all smushed together into a single line, and who knows what is in there:

.lst-kix_5lzjr43cw69v-3>li{counter-increment:lst-ctn-kix_5lzjr43cw69v-3}ul.lst-kix_b56rwjgjsre-8{list-style-type:none}.lst-kix_2dpmixc6z6cd-8>li{counter-increment:lst-ctn-kix_2dpmixc6z6cd-8}ol.lst-kix_yfjv4tvanyan-1.start{counter-reset:lst-ctn-kix_yfjv4tvanyan-1 0}.lst-kix_2dpmixc6z6cd-2>li:before{content:"" counter(lst-ctn-kix_2dpmixc6z6cd-2,lower-roman) ". "}.lst-kix_jsljq9wvyxl9-7>li{counter-increment:lst-ctn-kix_jsljq9wvyxl9-7}.lst-kix_2dpmixc6z6cd-3>li:before{content:"" counter(lst-ctn-kix_2dpmixc6z6cd-3,decimal) ". "}.lst-kix_2dpmixc6z6cd-4>li:before{content:"" counter(lst-ctn-kix_2dpmixc6z6cd-4,lower-latin) ". "}.lst-kix_2dpmixc6z6cd-5>li:before{content:"" counter(lst-ctn-kix_2dpmixc6z6cd-5,lower-roman) ". "}ul.lst-kix_b56rwjgjsre-5{list-style-type:none}.lst-kix_xhtrpcql5mbs-7>li:before{content:"\0025cb "}.lst-kix_2dpmixc6z6cd-8>li:before{content:"" counter(lst-ctn-kix_2dpmixc6z6cd-8,lower-roman)
[...]

I'm assuming what's causing the odd page breaks is the page-break-after:avoid floating around in there.

Side Note: The CSS alone is about 6 times larger than your actual text... talk about a bloated mess!