Tables in ePub

[Tex2002ans]

Originally Posted by Hitch

Perhaps I'm misunderstanding something then.

How is it you expect to 'create a table' if not with an HTML editor like Sigil or Calibre? I'm unclear on what you're expecting. If you did use Word and exported to HTML, yes, you'd get an HTML table. Apparently, you don't wish to do that.

[...]

So, in short--you want/need HTML tables, but you seem to not want to use the tool that will export that, in HTML, to a file format that you can then use to make an ePUB.

Is that right?

That's how I'm understanding it too... baffling.

1. Tables are super important to me.

2. All these crappy programs I'm choosing to use don't handle tables...

Well, then that's on you! Ask those programs to add proper tables support or change your workflow.

- - -

Side Note: A similar situation happened with a book I worked on a few years ago.

The author was writing a Non-Fiction book, nearly a thousand footnotes + URLs + citations... and someone bamboozled him into trying Scrivener, because they told him:

"I hear it's the best at making books!"

... No.

He wasted lots of time trying to wrestle with Scrivener, trying to force it to do something it wasn't designed to do.

(Luckily, he fell back on Word + someone taught him how to properly use Styles! He became one of the rare 1%!!!)

Originally Posted by Hitch

Word, for all its alleged issues (which are non-existent if you clean the file properly and use Styles and headings, I might add) is a perfectly straightforward way to make HTML tables. I don't know if Libre Office does the same thing, but others here will know.

Yep. Same exact thing in LibreOffice (that's what I use to recreate/export trickier tables):

1. File > Export as XHTML
2. Clean up XHTML cruft
-- (Bring it down to barebones <table>, <td>, <th>, [...])
3. Copy/Paste HTML into the EPUB.

Because LibreOffice carries over a ton of inline fonts/widths/alignment/crap...

I usually just do these two regular expressions:

Find: style="[^"]+"
Replace: ***BLANK NOTHING IN HERE***

Find: class="[^"]+"
Replace: ***BLANK NOTHING IN HERE***

Example Table

First   Last
Joe     Brownstone
Tex     Testerson

LibreOffice XHTML:

Spoiler:

  <table border="0" cellspacing="0" cellpadding="0" class="ta1">
    <colgroup>
      <col width="99"/>
      <col width="99"/>
    </colgroup>

    <tbody>
      <tr class="ro1">
        <td style="text-align:left;width:0.889in; " class="ce1">
          <p>First</p>
        </td>

        <td style="text-align:left;width:0.889in; " class="ce1">
          <p>Last</p>
        </td>
      </tr>

      <tr class="ro1">
        <td style="text-align:left;width:0.889in; " class="Default">
          <p>Joe</p>
        </td>

        <td style="text-align:left;width:0.889in; " class="Default">
          <p>Brownstone</p>
        </td>
      </tr>

      <tr class="ro1">
        <td style="text-align:left;width:0.889in; " class="Default">
          <p>Tex</p>
        </td>

        <td style="text-align:left;width:0.889in; " class="Default">
          <p>Testerson</p>
        </td>
      </tr>
    </tbody>
  </table>

but after running those 2 regex, it's much more reasonable:

Spoiler:

  <table border="0" cellspacing="0" cellpadding="0">
    <colgroup>
      <col width="99"/>
      <col width="99"/>
    </colgroup>

    <tbody>
      <tr>
        <td>
          <p>First</p>
        </td>

        <td>
          <p>Last</p>
        </td>
      </tr>

      <tr>
        <td>
          <p>Joe</p>
        </td>

        <td>
          <p>Brownstone</p>
        </td>
      </tr>

      <tr>
        <td>
          <p>Tex</p>
        </td>

        <td>
          <p>Testerson</p>
        </td>
      </tr>
    </tbody>
  </table>

You can then strip down that HTML further:

Spoiler:

  <table>
    <thead>
      <tr>
        <th>First</th>
        <th>Last</th>
      </tr>
    </thead>
    
    <tbody>
      <tr>
        <td>Joe</td>
        <td>Brownstone</td>
      </tr>
      <tr>
        <td>Tex</td>
        <td>Testerson</td>
      </tr>
    </tbody>
  </table>

(Create a "Saved Search" Group in Sigil or Calibre, and this multi-step Word/LibreOffice <table> cleanup can be a one-button process.)

- - -

Side Note: You can also use LibreOffice (or Excel) to transpose tables very easily.

(Flipping wide tables from "horizontal" to "vertical".)

Very important in ebooks, since you have "infinite" vertical space, but very limited horizontal. (You can scroll up/down forever, but it's a pain in the neck to go left/right if something flows off the screen.)

Transposing is discussed in the table threads I linked previously + the linked threads below.

Originally Posted by Artie

And I think this GIF is brilliant! Super! https://old.reddit.com/r/gifs/commen...less_terrible/

Originally Posted by Artie

I started this thread to figure out which is the best workflow for tables inside an EPUB file.

You could read all types of workflows (and tips/hiccups) here:

• 2020: "OCRing + EPUBing my first book: Tips?"
• 2019: "Workflow for simultaneous EPUB and PDF production?"
• 2014: "Delicate text digitalizing + scanning issues"

It pretty much boils down to these stages:

• Input format -> Intermediate cleanup -> Output format

for example:

• DOCX -> EPUB (Dirty) -> EPUB (Clean)
• InDesign -> EPUB (Dirty) -> EPUB (Clean)
• PDF -> [DOCX -> EPUB (Dirty)] -> EPUB (Clean)

Every conversion/stage is going to bring its own unique troubles:

• EPUB -> HTML
  
  • Like chapters from an ebook -> articles on a website.
    
  • How do you deal with links to other chapters?
    
  • Will your document transfer over cleanly? The website will usually have completely different CSS.
• HTML -> EPUB
  
  • Like articles from a website -> ebook.
    
  • Complex CSS (like CSS3 Counters) won't work in ebooks. You'll have to "hardcode" in <ol> lists.
    
  • Fonts, font sizes, and lots of other formatting won't carry over nicely, so you can't just copy/paste the CSS from the site.
    
• InDesign -> EPUB
  
  • InDesign exports "iBooks-friendly" EPUBs, but lots of that complex CSS doesn't work on other ereaders.
• PDF -> EPUB
  
  • Footnotes? Floating tables? Floating figures?
• [...]

Source (Master) Documents

And every person/publisher is going to have a preferred format.

They will make all their changes in the master document, then export to the other formats from there.

So you can have:

• "HTML-first" workflow
• "InDesign-first" workflow
• "Word-first" workflow
• "EPUB-first" workflow
• [...]

This also complicates the situation.

- - -

Side Note: Read about the "bifurcation" I discuss in the above threads.

Once you split the file into multiple formats, each correction/change will multiply the total amount of labor + potential errors.

If the author magically decides to make 100 extra changes since you created the ebook... now you have to correct:

• 100 in the Print
• 100 in the EPUB
• 100 in the HTML
• [...].

so you want to squash these corrections EARLY and as thoroughly as you can.

- - -

My Workflow

Personally, I go with an "EPUB-first" workflow.

I get everything into HTML+EPUB as soon as possible:

• InDesign? Get into EPUB.
• Word? Get into EPUB.
• LibreOffice? Get into EPUB.
• Google Docs? Get into DOCX, then EPUB.
  
  • Google Docs creates some of the ugliest HTML known to man.
• PDF? Finereader to OCR it, squash errors there, then get into EPUB.

then use my EPUB as the master source file for everything else:

Step 1. Anything -> EPUB
Step 2. Clean with Sigil/Calibre
Step 3. EPUB -> Anything

But if you work within a team, and require other source formats, you'll have to adjust.

Anyway, like Hitch said, Word/LibreOffice/GoogleDocs and tons of other writing tools already support HTML tables... so I don't see the issue.

Originally Posted by Artie

I don't want to use images for tables (and maths formulas).

Formulas. Images, sadly, are the only way:

• 2013: "Tutorial: Formulas to PNG"

MathML isn't well supported, so you have to fallback to images anyway.

But long-term thinking is key. You can create formulas in ways where you can easily export to MathML in the future.

(I currently use LaTeX as my source for formulas. This allows me to easily automate generation of new/higher-quality images + export to MathML in the future.)

Originally Posted by Artie

Likewise, I'm confused about how Calibre and Sigil may help me, can you please elaborate a little more on that @Kalleen?

Sigil + Calibre's Editor are EPUB editors.

These let you powerfully edit the HTML... while also making your EPUB-life infinitely easier by taking care of all the EPUB-specific things... like:

• splitting chapter files
• reordering chapters by drag/drop
• renaming files + updating all internal links
• generating Table of Contents
• inserting/attaching Cover files
• Metadata Editor
  
  • Add title, author, publisher, ISBN, publication date, [...]
• [...]

along with tons of other super helpful tools/enhancements:

• Spellcheck Lists
  
  • All words in a book displayed in a searchable list. Lets you mass catch/correct errors instead of going through red squigglies one-by-one.
    
  • Also lets you know you have the proper HTML languages marked.
• Saved Searches
  
  • Lets you run whole groups of S&R or Regular Expressions at the push of a button.
    
    • Like my "find split paragraphs" or my "12-step Finereader Cleanup"
• Reports
  
  • List all images used
    
    • Image used 0 times? Get rid of it.
      
    • Low quality image you accidentally left in there? Whoops.
  • List all URLs within a book
    
    • Amazing for catching broken/mismatching links.
  • Show all used/unused CSS classes
    
    • Can even remove all unused classes for you.
  • Show all used characters
    
    • Have lots of obscure Unicode like Gibbous Moons 🌔🌖???
      
    • Accidentally left in "Vulgar Fractions" ¼???
      
    • Used Cyrillic letter 'С' instead of English 'C'?
      
    • Used dash-like characters like the MINUS SIGN??? You can easily spot that.
• Plugins
  
  • epubcheck to make sure your ebooks are valid + good for sale.
    
  • Smarten your punctuation to turn those 'dumb quotes' into “curly”
• [...]

[phillipgessert]

I don't know much about Ulysses, but it appears to be a Markdown editor, and Markdown does support tables. Tables in Markdown are a bit limited, though--for example, I don't think you can run a cell across multiple rows or columns.

[j.p.s]

Originally Posted by phillipgessert

I don't know much about Ulysses, but it appears to be a Markdown editor, and Markdown does support tables. Tables in Markdown are a bit limited, though--for example, I don't think you can run a cell across multiple rows or columns.

It's a moot point that markdown tables are limited because it looks like Ulysses doen't support tables.

https://www.markdownguide.org/tools/ulysses/

[jmurphy]

Originally Posted by j.p.s

It's a moot point that markdown tables are limited because it looks like Ulysses doen't support tables.

https://www.markdownguide.org/tools/ulysses/

Not directly, but:
https://blog.ulysses.app/tables-with...es-and-marked/

[phillipgessert]

Originally Posted by j.p.s

It's a moot point that markdown tables are limited because it looks like Ulysses doen't support tables.

https://www.markdownguide.org/tools/ulysses/

Interesting. Doesn't even support HTML.

[jmurphy]

Originally Posted by phillipgessert

Interesting. Doesn't even support HTML.

No, that means it doesn't support embedded HTML tags in Markdown.
It does allow exporting Markdown as HTML, but since it's full Markdown support is so limited, it still doesn't help.

[phillipgessert]

Originally Posted by jmurphy

No, that means it doesn't support embedded HTML tags in Markdown.
It does allow exporting Markdown as HTML, but since it's full Markdown support is so limited, it still doesn't help.

Sorry, I did get that (that’d be wild if a markdown editor didn’t support html export) but I was surprised it doesn’t even support embedded html as you noted. OP, I don’t think you will find a solution to this problem that doesn’t include moving on from Ulysses.

[j.p.s]

At some point, one has to wonder why not write markdown directlly in any text editor one likes and directly convert that to EPUB with pandoc.

[Artie]

Hi all,

My last post has been in hold for review. When I wrote it, I reduced my workflow to only two apps: Ulysses and Calibre.

Now I have advanced a little more, and for very basic tables, where the width is extremely precious, I am capable to handle with Ulysses only.

I have managed to embed an HTML table into Ulysses, using the Raw block code. It exports to HTML and EPUB.

I am capable to view the table in my Kindle Voyage. CSS is not understood, but the content of the table is visible. In both Ulysses and Calibre, CSS works and the preview looks gorgeous.

@Tex2002ans*I'm afraid I hate Word or any app that work with styles. I prefer Markdown instead. Thank you.

@phillipgessert*@j.p.s*@jmurphy Thank you for understanding that I prefer to work with Markdown.
You have shared an interesting
resource:*https://www.markdownguide.org/tools/

So I assume my next step now is to select a markdown editor with table support that exports to EPUB. Will take me time...

Now, if somebody can point me to a resource for CSS for tables for EPUB, would be great.

Each and one of the answers of this thread has helped me to arrive to this conclusion. Maybe I'm a bit slow, but I don't care. I couldn't have arrived to this point without you. Again, many thanks.

[Turtle91]

… or simply build it in html in the first place. In Sigil I have a blank table template set as a clip. One click gives me:

- I prefer working with tables in this compact format -

<table>
  <thead>
    <tr><th></th><th></th><th></th></tr>
  </thead>
  <tbody>
    <tr><td></td><td></td><td></td></tr>
    <tr><td></td><td></td><td></td></tr>
    <tr><td></td><td></td><td></td></tr>
  </tbody>
  <tfoot>
    <tr><td></td><td></td><td></td></tr>
  </tfoot>
</table>

- although some people prefer to see it like this - 

<table>
  <thead>
    <tr>
      <th></th>
      <th></th>
    </tr>
  </thead>
  <tbody>
    <tr>
      <td></td>
      <td></td>
    </tr>
    <tr>
      <td></td>
      <td></td>
    </tr>
  </tbody>
  <tfoot>
    <tr>
      <td></td>
      <td></td>
    </tr>
  </tfoot>
</table>

Then a few copy/paste/delete to refine the number of columns/rows and if I need header/footer rows. Then I can easily insert class/colspan as needed.

This literally takes a minute to do.

Now I just type in the data.
Style (as needed) using a css sheet (shame to any who use inline styles - shame!)

Result: perfectly legit table in my ePub!

[Turtle91]

Originally Posted by Artie

Hi all,

…

It exports to HTML and EPUB.

…

@Tex2002ans*I'm afraid I hate Word or any app that work with styles. I prefer Markdown instead. Thank you.

…

Now, if somebody can point me to a resource for CSS for tables for EPUB, would be great.

I get the feeling that you think ePub format is different than HTML… ePub IS just a zip folder of strict HTML files (as mentioned above). Using markdown to convert to html/ePub is like writing a book in Latin then using Google translate to print in English… it just doesn’t work well.

You need to spend the very little bit of time it takes to learn the language of ePub (html/css) if you want to produce ePubs.

Here is a link to a CSS/HTML resource.

Cheers!

[Quoth]

Originally Posted by j.p.s

At some point, one has to wonder why not write markdown directlly in any text editor one likes and directly convert that to EPUB with pandoc.

I scribe the bits directly on the platter.

[Hitch]

Originally Posted by Artie

Hi all,


@Tex2002ans*I'm afraid I hate Word or any app that work with styles. I prefer Markdown instead. Thank you.

Didn't you say that you'd worked with HTML for years? Given that Word (and LO, OO, etc.)'s "styles" are nothing but CSS...doesn't that seem a little odd to you? It's not like Styles came out of the blue. They exist quite simply because they are CSS. CSS that you control, not some invisible man behind the curtain.

Seems bizarrely peculiar to me. I mean..you're happier hand-writing CSS, along with hand-writing HTML, in markdown, rather than simply creating those selfsame styles in Application X, and then exporting them to a stylesheet?

Whatever. Not sure how you don't experience the frisson of cognitive dissonance there, but...it's your (manual) workflow.

Hitch

[phillipgessert]

Originally Posted by Artie

So I assume my next step now is to select a markdown editor with table support that exports to EPUB. Will take me time...

Any text editor can serve as a markdown editor, but I imagine few will export to epub directly. You will probably wind up either converting the *.md to epub via a tool like pandoc, mentioned upthread; or by using any number of tools to convert to HTML (the intended purpose of markdown) and then importing that HTML file into the epub editor of your choosing.

[Hitch]

Originally Posted by phillipgessert

Any text editor can serve as a markdown editor, but I imagine few will export to epub directly. You will probably wind up either converting the *.md to epub via a tool like pandoc, mentioned upthread; or by using any number of tools to convert to HTML (the intended purpose of markdown) and then importing that HTML file into the epub editor of your choosing.

If I knew how to put a headbanging gif in here...I would. (Not at YOU Phillip, dear boy.)

Something like this: https://tenor.com/view/cat-head-bang...d-gif-16831747

Hitch