Context to epub: itemize

newer
Distance between the header and...

older
Blank space between \subsection &...

Axel Kielhorn

16 Feb 2015 16 Feb '15

5:17 p.m.

Hi, I'm trying to generate epub from ConTeXt and I couldn't get the itemize environment to work. I follow http://wiki.contextgarden.net/ePub and get the structure described but when I look into the xhtml, there is no itemize environment. (At least no with Safari, Firefox or Calibre.) When use vi I see something that looks like my items in a very complex form. I was expecting something like <ul> <li> </li> <li> </li> </ul> but i see <div class="itemgroup itemize level-1 symbol-1"> <div class="item"> <div class="itemtag">http://www.w3.org/1998/Math/MathML">•</div> <div class="itemcontent">Listenpunkt 1</div> </div> <div class="item"> <div class="itemtag">http://www.w3.org/1998/Math/MathML">•</div> <div class="itemcontent">Listenpunkt 2 <div class="itemgroup itemize level-2 symbol-2"> <div class="item"> <div class="itemtag">http://www.w3.org/1998/Math/MathML">−</div> <div class="itemcontent">Listenpunkt 2.1</div></div> <div class="item"> <div class="itemtag">http://www.w3.org/1998/Math/MathML">−</div> <div class="itemcontent">Listenpunkt 2.2</div></div></div></div> </div> Shouldn't the itemtag definition come from the default.css instead of using MathML? itemgroup[symbol="1"], div.itemgroup.symbol-1 { list-style-type : disc ; } I’m using: mtx-context | current version: 2015.02.03 23:55 and context --script epub --make epub_mwe to create the epub. Axel

Attachments:

epub_mwe.tex (application/octet-stream — 1.6 KB)

Show replies by date

Hans Hagen

16 Feb 16 Feb

7:55 p.m.

On 2/16/2015 5:17 PM, Axel Kielhorn wrote:

...

Hi,

I'm trying to generate epub from ConTeXt and I couldn't get the itemize environment to work.

I follow http://wiki.contextgarden.net/ePub and get the structure described but when I look into the xhtml, there is no itemize environment. (At least no with Safari, Firefox or Calibre.) When use vi I see something that looks like my items in a very complex form.

I was expecting something like

<ul> <li> </li> <li> </li> </ul>

but i see

<div class="itemgroup itemize level-1 symbol-1"> <div class="item"> <div class="itemtag">http://www.w3.org/1998/Math/MathML">•</div> <div class="itemcontent">Listenpunkt 1</div> </div> <div class="item"> <div class="itemtag">http://www.w3.org/1998/Math/MathML">•</div> <div class="itemcontent">Listenpunkt 2 <div class="itemgroup itemize level-2 symbol-2"> <div class="item"> <div class="itemtag">http://www.w3.org/1998/Math/MathML">−</div> <div class="itemcontent">Listenpunkt 2.1</div></div> <div class="item"> <div class="itemtag">http://www.w3.org/1998/Math/MathML">−</div> <div class="itemcontent">Listenpunkt 2.2</div></div></div></div> </div>

Shouldn't the itemtag definition come from the default.css instead of using MathML?

itemgroup[symbol="1"], div.itemgroup.symbol-1 { list-style-type : disc ; }

I’m using: mtx-context | current version: 2015.02.03 23:55

and

context --script epub --make epub_mwe

to create the epub.

the bullets come from math fonts (currently, awaiting additional symbols in the text fonts) Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | voip: 087 875 68 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------

Axel Kielhorn

17 Feb 17 Feb

7:11 a.m.

Am 16.02.2015 um 19:55 schrieb Hans Hagen :

...

On 2/16/2015 5:17 PM, Axel Kielhorn wrote:

...
Hi,

I'm trying to generate epub from ConTeXt and I couldn't get the itemize environment to work.

I follow http://wiki.contextgarden.net/ePub and get the structure described but when I look into the xhtml, there is no itemize environment. (At least no with Safari, Firefox or Calibre.) When use vi I see something that looks like my items in a very complex form.

I was expecting something like

<ul> <li> </li> <li> </li> </ul>

but i see

<div class="itemgroup itemize level-1 symbol-1"> <div class="item"> <div class="itemtag">http://www.w3.org/1998/Math/MathML">•</div> <div class="itemcontent">Listenpunkt 1</div> </div> <div class="item"> <div class="itemtag">http://www.w3.org/1998/Math/MathML">•</div> <div class="itemcontent">Listenpunkt 2 <div class="itemgroup itemize level-2 symbol-2"> <div class="item"> <div class="itemtag">http://www.w3.org/1998/Math/MathML">−</div> <div class="itemcontent">Listenpunkt 2.1</div></div> <div class="item"> <div class="itemtag">http://www.w3.org/1998/Math/MathML">−</div> <div class="itemcontent">Listenpunkt 2.2</div></div></div></div> </div>

Shouldn't the itemtag definition come from the default.css instead of using MathML?

itemgroup[symbol="1"], div.itemgroup.symbol-1 { list-style-type : disc ; }

I’m using: mtx-context | current version: 2015.02.03 23:55

and

context --script epub --make epub_mwe

to create the epub.

the bullets come from math fonts (currently, awaiting additional symbols in the text fonts)

Yes I know. Let me rephrase the question: Should the symbols used in epub be identical to those used in the pdf or should the css determine the look of the symbols? Especially when one considers that most current ebook readers can't display MathML. Even if I substitute the Symbols: \definesymbol[uni1][•] \definesymbol[uni2][–] \definesymbol[uni3][*] \definesymbol[uni4][·] \setupitemgroup [itemize] [1] [symbol=uni1] \setupitemgroup [itemize] [2] [symbol=uni2] \setupitemgroup [itemize] [3] [symbol=uni3] \setupitemgroup [itemize] [4] [symbol=uni4] I get a result that neither ADE nor Calibre interprets as a list. (Wrapping it in <ul> </ul> doesn't help.) Axel

Keith J. Schultz

8:24 a.m.

New subject: Epub format Was: Re: Context to epub: itemize

Hi Axel, All, You have mentioned that most current ebook readers can not display MathML. Well, the problem lies in the epub guidelines for the readers. The problem is even worse. many readers do not even implement the full HTML5 standard. though that is the standard used in the latest epub stadnard. The fact is to be considered epub readers the do not have to, it is a problem of the epub reader definition. The problem is what should ConTeXt output? Should it output code that implements according to the full epub standard OR JUST what is handled by most epub readers! IMHO : ConTeXt should not output XHMTL for epub, but HMTL5. Yes, yes, I know XHMTL is part of HMTL5. Another question for me is if ConTeXt should also output code for the iBooks epub format, (Basicaly epub, but some extras). There are thing you can do in/with ConTeXt that would work with iBooks that do not in „ordinary“ epub books. Then, there is the Kindle format one might want to consider! Another, question which needs investigation is if the ConTeXt output for epub, also, implementing the fallbacks for missing features which is required? Just some added thoughts. Basically, you can only expect ConTeXts epub output to be a starting point that then has to be tweak. regards Keith.

...

Am 17.02.2015 um 07:11 schrieb Axel Kielhorn :

[...] Yes I know. Let me rephrase the question:

Should the symbols used in epub be identical to those used in the pdf or should the css determine the look of the symbols? Especially when one considers that most current ebook readers can't display MathML.

Even if I substitute the Symbols:

\definesymbol[uni1][•] \definesymbol[uni2][–] \definesymbol[uni3][*] \definesymbol[uni4][·]

\setupitemgroup [itemize] [1] [symbol=uni1] \setupitemgroup [itemize] [2] [symbol=uni2] \setupitemgroup [itemize] [3] [symbol=uni3] \setupitemgroup [itemize] [4] [symbol=uni4]

I get a result that neither ADE nor Calibre interprets as a list. (Wrapping it in <ul> </ul> doesn't help.)

Hans Hagen

9:16 p.m.

New subject: Epub format Was: Re: Context to epub: itemize

On 2/17/2015 8:24 AM, Keith J. Schultz wrote:

...

Hi Axel, All,

You have mentioned that most current ebook readers can not display MathML.

Well, the problem lies in the epub guidelines for the readers.

The problem is even worse. many readers do not even implement the full HTML5 standard. though that is the standard used in the latest epub stadnard. The fact is to be considered epub readers the do not have to, it is a problem of the epub reader definition.

The problem is what should ConTeXt output? Should it output code that implements according to the full epub standard OR JUST what is handled by most epub readers!

IMHO : ConTeXt should not output XHMTL for epub, but HMTL5. Yes, yes, I know XHMTL is part of HMTL5.

And html 5 is just html with some extras (and assuming javascript i guess) ... marketing.

...

Another question for me is if ConTeXt should also output code for the iBooks epub format, (Basicaly epub, but some extras).

Isn't a book supposed to be independent and kind of portable + long term visible? So no way that a context epub will be specific for a device. (I probably can't even get it on / test it on my outdated ipad 1 anyway.)

...

There are thing you can do in/with ConTeXt that would work with iBooks that do not in „ordinary“ epub books.

Then, there is the Kindle format one might want to consider!

Another lock-in device.

...

Another, question which needs investigation is if the ConTeXt output for epub, also, implementing the fallbacks for missing features which is required?

Context outputs three varants: regular xml, xhtml, and stupified xhtml with div/class tagging. The last one is supposed to work on all devices as it doesn't demands anything beyond css.

...

Just some added thoughts.

Basically, you can only expect ConTeXts epub output to be a starting point that then has to be tweak.

Indeed, as it's all xml one can process it into something 'better'. There is no way we can adapt to every change in specs, so the best we can do is something generic. Hans

...

regards Keith.

...
Am 17.02.2015 um 07:11 schrieb Axel Kielhorn mailto:tex@axelkielhorn.de>:

[...] Yes I know. Let me rephrase the question:

Should the symbols used in epub be identical to those used in the pdf or should the css determine the look of the symbols? Especially when one considers that most current ebook readers can't display MathML.

Even if I substitute the Symbols:

\definesymbol[uni1][•] \definesymbol[uni2][–] \definesymbol[uni3][*] \definesymbol[uni4][·]

\setupitemgroup [itemize] [1] [symbol=uni1] \setupitemgroup [itemize] [2] [symbol=uni2] \setupitemgroup [itemize] [3] [symbol=uni3] \setupitemgroup [itemize] [4] [symbol=uni4]

I get a result that neither ADE nor Calibre interprets as a list. (Wrapping it in <ul> </ul> doesn't help.)

___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://tex.aanhet.net archive : http://foundry.supelec.fr/projects/contextrev/ wiki : http://contextgarden.net ___________________________________________________________________________________

Keith Schultz

10:01 p.m.

New subject: Epub format Was: Re: Context to epub: itemize

...

Am 17.02.2015 um 21:16 schrieb Hans Hagen :

On 2/17/2015 8:24 AM, Keith J. Schultz wrote:

...
Hi Axel, All,

[snip, snip]

...
IMHO : ConTeXt should not output XHMTL for epub, but HMTL5. Yes, yes, I know XHMTL is part of HMTL5.

And html 5 is just html with some extras (and assuming javascript i guess) ... marketing. The html 5 standard is far more. For one is makes flash, java and more obsolete. Far more than a marketing gag!

...
Another question for me is if ConTeXt should also output code for the iBooks epub format, (Basicaly epub, but some extras).

Isn't a book supposed to be independent and kind of portable + long term visible? So no way that a context epub will be specific for a device. (I probably can't even get it on / test it on my outdated ipad 1 anyway.) I was thinking more along the line of the spine.

...
There are thing you can do in/with ConTeXt that would work with iBooks that do not in „ordinary“ epub books.

Then, there is the Kindle format one might want to consider!

Another lock-in device.

...
Another, question which needs investigation is if the ConTeXt output for epub, also, implementing the fallbacks for missing features which is required?

Context outputs three varants: regular xml, xhtml, and stupified xhtml with div/class tagging. The last one is supposed to work on all devices as it doesn't demands anything beyond css. If that is the case it is not very useful for a decent workflow. There would be to much to tweak to make good looking output.

I have not investigated how it handles dimensions, widths, font sizes, spacing, etc. These are important in order to create output that adjust proper to the devices capabilities. One should avoid absolute values. Of course if we want to emulate the printed output then we could simply create a pdf or very page as pdf and put in an epub wrapper. Not asking for anything, just mentioning it. reagrds Keith.

Hans Hagen

18 Feb 18 Feb

5:54 p.m.

New subject: Epub format Was: Re: Context to epub: itemize

On 2/17/2015 10:01 PM, Keith Schultz wrote:

...

...
Am 17.02.2015 um 21:16 schrieb Hans Hagen mailto:pragma@wxs.nl>:

On 2/17/2015 8:24 AM, Keith J. Schultz wrote:

...
Hi Axel, All,

[snip, snip]

...
IMHO : ConTeXt should not output XHMTL for epub, but HMTL5. Yes, yes, I know XHMTL is part of HMTL5.

And html 5 is just html with some extras (and assuming javascript i guess) ... marketing. The html 5 standard is far more. For one is makes flash, java and more obsolete. Far more than a marketing gag!

well, if javascript is used instead it's still fluent ... every time i played with e.g. jquery or libraries i had to make sure i used the latest and had to adapt code .. there will be something new (and maybe even different) every decade (if only because that generates money)

...

...
...
Another question for me is if ConTeXt should also output code for the iBooks epub format, (Basicaly epub, but some extras).

Isn't a book supposed to be independent and kind of portable + long term visible? So no way that a context epub will be specific for a device. (I probably can't even get it on / test it on my outdated ipad 1 anyway.) I was thinking more along the line of the spine.

ah, direct neural injection ...

...

...
...
There are thing you can do in/with ConTeXt that would work with iBooks that do not in „ordinary“ epub books.

Then, there is the Kindle format one might want to consider!

Another lock-in device.

...
Another, question which needs investigation is if the ConTeXt output for epub, also, implementing the fallbacks for missing features which is required?

Context outputs three varants: regular xml, xhtml, and stupified xhtml with div/class tagging. The last one is supposed to work on all devices as it doesn't demands anything beyond css. If that is the case it is not very useful for a decent workflow. There would be to much to tweak to make good looking output.

i'm not sure what you have in mind but context is a typesetting system and as such cannot produce an epub simply because epub is a blob of xml (html) with a css packages in a zip

...

I have not investigated how it handles dimensions, widths, font sizes, spacing, etc.

structure normally doesn't carry that information

...

These are important in order to create output that adjust proper to the devices capabilities. One should avoid absolute values.

there are hardly any values output ... but one can of course extend the css with font and color directives, probably different from the ones used in the pdf

...

Of course if we want to emulate the printed output then we could simply create a pdf or very page as pdf and put in an epub wrapper.

i've seen impressive examples of pdf turned html (huge html files btw) (fyi: i see pdf and html as independent things) (and i don't believe in reflow of pdf: it's either a design for made up pages or a flexible adaptive one as in html; and till now i never wished a paper book to reflow itself anyway but maybe if i really would use an epub device ... who knows ... i don't own ebooks currently as i cannot share docs freely then like i share books and i consider ebooks to be way too expensive for fun) Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | voip: 087 875 68 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------

Keith Schultz

19 Feb 19 Feb

10:27 a.m.

New subject: Epub format Was: Re: Context to epub: itemize

Hi Hans, I have been playing with the on using LaTeX and later LuaLaTeX for create epub or ebook. Though I have not got around to. The idea was to set enviorments, macros and commands that would depending on a switch that would output the need HTML and epub structure or call the normal *TeX ones. In ConTeXt maybe the mode-system could be used. Like I said I have not investigated deeply, yet. regards Keith.

...

Am 18.02.2015 um 17:54 schrieb Hans Hagen :

On 2/17/2015 10:01 PM, Keith Schultz wrote:

...
...
Context outputs three varants: regular xml, xhtml, and stupified xhtml with div/class tagging. The last one is supposed to work on all devices as it doesn't demands anything beyond css.

If that is the case it is not very useful for a decent workflow. There would be to much to tweak to make good looking output.

i'm not sure what you have in mind but context is a typesetting system and as such cannot produce an epub simply because epub is a blob of xml (html) with a css packages in a zip

Aditya Mahajan

17 Feb 17 Feb

11:37 p.m.

New subject: Epub format Was: Re: Context to epub: itemize

On Tue, 17 Feb 2015, Hans Hagen wrote:

...

Context outputs three varants: regular xml, xhtml, and stupified xhtml with div/class tagging. The last one is supposed to work on all devices as it doesn't demands anything beyond css.

The XML generated by ConTeXt is very well structured. So, an option is to simply provide XLST stylesheets that transform ConTeXT-XML to XHTML/HTML5/whatever. Such XLST transformation will be better than have ConTeXt chase a moving target. Aditya

Henning Hraban Ramm

18 Feb 18 Feb

9:14 a.m.

New subject: Epub format Was: Re: Context to epub: itemize

Am 2015-02-18 um 04:37 schrieb Aditya Mahajan :

...

On Tue, 17 Feb 2015, Hans Hagen wrote:

...
Context outputs three varants: regular xml, xhtml, and stupified xhtml with div/class tagging. The last one is supposed to work on all devices as it doesn't demands anything beyond css.

The XML generated by ConTeXt is very well structured. So, an option is to simply provide XLST stylesheets that transform ConTeXT-XML to XHTML/HTML5/whatever.

Such XLST transformation will be better than have ConTeXt chase a moving target.

See e.g. http://wiki.contextgarden.net/Epub_Sample I’ll try to update the pages about ConTeXt’s current export soon, since I have a new book in the pipeline. Greetlings, Hraban --- http://www.fiee.net http://wiki.contextgarden.net https://www.cacert.org (I'm an assurer)

Keith J. Schultz

3:10 p.m.

New subject: Epub format Was: Re: Context to epub: itemize

Hi Hraban, Good to know. I will investigate at another time and give feed back. I am in the middle of moving, so I do not have much time. We ought to think of making a export specific to epub in the long run! regards Keith.

...

Am 18.02.2015 um 09:14 schrieb Henning Hraban Ramm :

Am 2015-02-18 um 04:37 schrieb Aditya Mahajan :

...
On Tue, 17 Feb 2015, Hans Hagen wrote:

...
Context outputs three varants: regular xml, xhtml, and stupified xhtml with div/class tagging. The last one is supposed to work on all devices as it doesn't demands anything beyond css.

The XML generated by ConTeXt is very well structured. So, an option is to simply provide XLST stylesheets that transform ConTeXT-XML to XHTML/HTML5/whatever.

Such XLST transformation will be better than have ConTeXt chase a moving target.

See e.g. http://wiki.contextgarden.net/Epub_Sample

I’ll try to update the pages about ConTeXt’s current export soon, since I have a new book in the pipeline.

Greetlings, Hraban --- http://www.fiee.net http://wiki.contextgarden.net https://www.cacert.org (I'm an assurer)

___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://tex.aanhet.net archive : http://foundry.supelec.fr/projects/contextrev/ wiki : http://contextgarden.net ___________________________________________________________________________________

Henning Hraban Ramm

3:21 p.m.

New subject: Epub format Was: Re: Context to epub: itemize

Am 2015-02-18 um 20:10 schrieb Keith J. Schultz :

...

Hi Hraban,

Good to know. I will investigate at another time and give feed back. I am in the middle of moving, so I do not have much time.

We ought to think of making a export specific to epub in the long run!

For me, the regular export XML plus XSL conversion is the way to go; the „HTML“ is rather unusable. The epub script is nice and all, it was important for me to get me running, but I guess I won’t use it in my final workflow. Greetlings, Hraban --- http://www.fiee.net http://wiki.contextgarden.net https://www.cacert.org (I'm an assurer)

Hans Hagen

5:44 p.m.

New subject: Epub format Was: Re: Context to epub: itemize

On 2/18/2015 3:10 PM, Keith J. Schultz wrote:

...

Hi Hraban,

Good to know. I will investigate at another time and give feed back. I am in the middle of moving, so I do not have much time.

We ought to think of making a export specific to epub in the long run!

The export is not really geared at that, as it's a representation of the rendering (although structured). The additional html output can be used for previewing and to what extend it's okay depends on the intended use. Normally one starts from a neutral source (xml or tex) and then produces different output. So, for specific epub (or whatever) one can either start with the original code (xml) or manipulate what comes out of the export. The export is kind of standard so one can build all kind of scripts to convert it. There is no universal epub (rendering) solution, just as there is no universal pdf layout. Of course the idea is that the default export is good enough for simple documents. What more gets provided depends on needs (and currently at least I have no need for epub other than a dedicated pdf suitable for displays). Hans

...

regards Keith.

...
Am 18.02.2015 um 09:14 schrieb Henning Hraban Ramm :

Am 2015-02-18 um 04:37 schrieb Aditya Mahajan :

...
On Tue, 17 Feb 2015, Hans Hagen wrote:

...
Context outputs three varants: regular xml, xhtml, and stupified xhtml with div/class tagging. The last one is supposed to work on all devices as it doesn't demands anything beyond css.

The XML generated by ConTeXt is very well structured. So, an option is to simply provide XLST stylesheets that transform ConTeXT-XML to XHTML/HTML5/whatever.

Such XLST transformation will be better than have ConTeXt chase a moving target.

See e.g. http://wiki.contextgarden.net/Epub_Sample

I’ll try to update the pages about ConTeXt’s current export soon, since I have a new book in the pipeline.

Greetlings, Hraban --- http://www.fiee.net http://wiki.contextgarden.net https://www.cacert.org (I'm an assurer)

___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://tex.aanhet.net archive : http://foundry.supelec.fr/projects/contextrev/ wiki : http://contextgarden.net ___________________________________________________________________________________

___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://tex.aanhet.net archive : http://foundry.supelec.fr/projects/contextrev/ wiki : http://contextgarden.net ___________________________________________________________________________________

Hans Hagen

5:55 p.m.

New subject: Epub format Was: Re: Context to epub: itemize

On 2/17/2015 11:37 PM, Aditya Mahajan wrote:

...

On Tue, 17 Feb 2015, Hans Hagen wrote:

...
Context outputs three varants: regular xml, xhtml, and stupified xhtml with div/class tagging. The last one is supposed to work on all devices as it doesn't demands anything beyond css.

The XML generated by ConTeXt is very well structured. So, an option is to simply provide XLST stylesheets that transform ConTeXT-XML to XHTML/HTML5/whatever.

Such XLST transformation will be better than have ConTeXt chase a moving target.

indeed, and the context wiki can provide useful examples Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | voip: 087 875 68 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------

3594

Age (days ago)

3597

Last active (days ago)

List overview

Download

13 comments

6 participants

participants (6)

Aditya Mahajan
Axel Kielhorn
Hans Hagen
Henning Hraban Ramm
Keith J. Schultz
Keith Schultz

Context to epub: itemize

tags

participants (6)