Ehhhh I think this is out of scope for Asciidoctor personally unless we figure out a way to make the backends/templates more generic and figure out a way to generate PDF's from that. 👎 I think we'd probably not be adding much value here.

from asciidoctor.

Obviously the impact will be different for different audiences. The documentation teams I've talked to that want to adopt Asciidoctor consider this a must have. The feature will certainly take time to evolve as a whole, but I feel very strongly that this is in scope for Asciidoctor.

from asciidoctor.

We should use the following option to control this output:

-f FORMAT, --format=FORMAT

I'd also like to stick with fop for now, so the --fop option will be implied.

Also, add this option to keep the artifacts.

-k, --keep-artifacts 

from asciidoctor.

The icons dir must be set explicitly for fop to insert the admonition icons:

--icons-dir=images/icons

from asciidoctor.

Looking forward to built-in PDF output support :-)

from asciidoctor.

This would be a killer feature! 👍

from asciidoctor.

sandoc provides a standalone solution today (written in Scala, uses the Asciidoctor Java integration).

http://discuss.asciidoctor.org/Announcement-Markup-Converter-Sandoc-0-1-td319.html

I still want this to be in Asciidoctor core, but libraries such as sandoc will provide nice alternatives that may open new doors.

from asciidoctor.

I'd really love to have PDF output provided by Asciidoctor directly rather than by another third-party solution that I'd have to integrate in my build, add even more dependencies and so on :-(

from asciidoctor.

Totally agreed. I'll be getting to this one soon, at least a first stab at
it. It will probably require a few iterations to find what we like best.

from asciidoctor.

I'm sure you seen this one @mojavelinux, but if not, let's take a good hard look at prawn. I just skimmed their manual, its only 130 page PDF they create! You can find the manual at http://prawn.majesticseacreature.com/manual.pdf. Sure it may not have templating just yet, man that would be a powerful thing to add (eventually). I also installed it to see what it took, here's the installation:

gem install prawn
Fetching: Ascii85-1.0.2.gem (100%)
Fetching: ruby-rc4-0.1.5.gem (100%)
Fetching: hashery-2.1.0.gem (100%)
Fetching: ttfunk-1.0.3.gem (100%)
Fetching: afm-0.2.0.gem (100%)
Fetching: pdf-reader-1.3.3.gem (100%)

  ********************************************

  v1.0.0 of PDF::Reader introduced a new page-based API. There are extensive
  examples showing how to use it in the README and examples directory.

  For detailed documentation, check the rdocs for the PDF::Reader,
  PDF::Reader::Page and PDF::Reader::ObjectHash classes.

  The old API is marked as deprecated but will continue to work with no
  visible warnings for now.

  ********************************************

Fetching: prawn-0.12.0.gem (100%)

  ********************************************


  A lot has changed since 0.8.4

  Please read the changelog for details:

  https://github.com/sandal/prawn/wiki/CHANGELOG


  ********************************************

Successfully installed Ascii85-1.0.2
Successfully installed ruby-rc4-0.1.5
Successfully installed hashery-2.1.0
Successfully installed ttfunk-1.0.3
Successfully installed afm-0.2.0
Successfully installed pdf-reader-1.3.3
Successfully installed prawn-0.12.0
7 gems installed
Installing ri documentation for Ascii85-1.0.2...
Installing ri documentation for ruby-rc4-0.1.5...
Installing ri documentation for hashery-2.1.0...
Installing ri documentation for ttfunk-1.0.3...
Installing ri documentation for afm-0.2.0...
Installing ri documentation for pdf-reader-1.3.3...
Installing ri documentation for prawn-0.12.0...
Installing RDoc documentation for Ascii85-1.0.2...
Installing RDoc documentation for ruby-rc4-0.1.5...
Installing RDoc documentation for hashery-2.1.0...
Installing RDoc documentation for ttfunk-1.0.3...
Installing RDoc documentation for afm-0.2.0...
Installing RDoc documentation for pdf-reader-1.3.3...
Installing RDoc documentation for prawn-0.12.0...

There's NO NATIVE ANYTHING there! I think we may have found a winner.

from asciidoctor.

Prawn looks stellar, but it's going to be a long run solution. Despite the fact that we all despise the DocBook XSL templates (mostly because they are...XSL), there is years of layout work in there. We can throw something together quickly with Prawn, but it's going to take a while to cover all the layouts handled by DocBook XSL + XSL-FO.

That's why, in the meantime, I put effort into creating a bootstrap script, fopdf*, that handles all tasks to set up the toolchain for generating PDFs from AsciiDoc DocBook output using DocBook XSL and Apache FOP. All you have to do is run a Unix shell script or Windows batch file and watch the magic happen.

You can find out more about the project on its GitHub project page:

https://github.com/mojavelinux/asciidoctor-fopdf

The plan is to integrate this tool into Asciidoctor. But first, I think we need to give it a chance to breathe and collect some feedback. That's why I'm proposing to move the completion of this issue to the 1.5.0 release. Consider fopdf to be phase 1 of the implementation.

Btw, the research done for fopdf is great news for the Asciidoctor Java integration. After all, Apache FOP is a Java application. There's a good chance that the shell script won't be needed and that Asciidoctor Java can handle bundling all the assets that fopdf identified and also handle executing the toolchain. The fopdf tool is only necessary, then, when using Asciidoctor from the CLI or Ruby (non-JRuby).

* I'm considering renaming it fopub (intended to be read as it looks)

from asciidoctor.

Btw, we still have to figure out what the arguments to the Asciidoctor CLI or API will be to get to PDF. I propose using a backend alias, which behind the scenes handles these two phases:

1. Convert from AsciiDoc to DocBook (-b docbook)
2. Run fopdf on the DocBook

Something like:

asciidoctor -b fopdf sample.adoc

I choose "fopdf" rather than "pdf" because I'm reserving the latter for the long-term, native solution (i.e., Prawn).

from asciidoctor.

We can try in Asciidoctor Java Integration without any problem, I can intercept fopdf option and do the required changes. Only one small problem, is Asciidoctor-fopdf in Maven repository? For what I have seen is not there, so we will need to resolve this dependency locally. No problem at all of course.
Do you think we could add this feature for Asciidoctor-Java-Integration 1.4.0, or we can create an alternative branch.

WDYT?

from asciidoctor.

Native PDF support would be awesome!

from asciidoctor.

I think we are getting closer to this as a reality. The way I envision this working is that we publish Asciidoctor PDF as a gem (asciidoctor-pdf). Then, the renderer in that gem associates itself with the "pdf" backend. When Asciidoctor looks for a renderer for the "pdf" backend, it will find the registered Asciidoctor PDF renderer and delegate to it. That would enable us to use the asciidoctor cli without having the Asciidoctor PDF code base in Asciidoctor core (which would otherwise make it a monolithic project).

In other words, we want to accomplish this through modularity.

from asciidoctor.

Note that we could do the same with "fopub" as well...though the integration is a bit more tricky...it can be done.

from asciidoctor.

Hi, I'm new to asciidoc and asciidoctor (as of yesterday). I'm about to convert a book of mine to a number of e-formats, namely mobi, epub, and pdf. I don't seem to be able to find a single pipeline to go from asciidoctor source files to each of the three output formats cleanly. If I use the stem block for math, asciidoctor-fopub and asciidoctor-epub3 won't handle them. The book doesn't have much math so, at a pinch, I could get by with LaTeXing the equations and inserting them as .pngs.

I was going to use Pandoc and markdown but I really like the existing blocks and stylesheets offered by asciidoctor and the -epub3 extension seems to do a really good job of making ibook-compatible epub files.

I guess what I'm saying is that from my perspective which, longer term will involve using asciidoctor to generate handouts for classes I teach, I would need it to have a seamless pipeline that handles pdf, html, ebook, mobi and math (now called 'stem'?) in one tool. At the moment if I want math I seem to be restricted to vanilla html output as the other tools for generating epub and pdf lag behind the main asciidoctor project.

from asciidoctor.

I've been using and math for some time now. Works beautifully
with no imbedded images.

My configuration: asciidoctor-1.5.0.rc.5 with ruby 2.1.1.

Attached is a file to test your configuration. Below the test file is a screen
shot of what you should see in the browser. (I made the browser window
really small for this).

I've been experimenting with asciidoctor-pdf, but haven't been able
to get a working configuration yet. Of course you can print the html
to pdf.

Jim
@epsilon2718
http://square-the-circle.com

File: test_math.ad

Usage: asciidoctor test_math.ad, then view test_math.html in a browser

:stem: latexmath

The Pythagorean theorem: (a^2 + b^2 = c^2)

The fundamental theorem of calculus:

[
\frac{d}{dx}\int_0^x f(s),ds = f(x)

]

On Aug 6, 2014, at 8:58 AM, Paul Vickers [email protected] wrote:

Hi, I'm new to asciidoc and asciidoctor (as of yesterday). I'm about to convert a book of mine to a number of e-formats, namely mobi, epub, and pdf. I don't seem to be able to find a single pipeline to go from asciidoctor source files to each of the three output formats cleanly. If I use the stem block for math, asciidoctor-fopub and asciidoctor-epub3 won't handle them. The book doesn't have much math so, at a pinch, I could get by with LaTeXing the equations and inserting them as .pngs.

I was going to use Pandoc and markdown but I really like the existing blocks and stylesheets offered by asciidoctor and the -epub3 extension seems to do a really good job of making ibook-compatible epub files.

I guess what I'm saying is that from my perspective which, longer term will involve using asciidoctor to generate handouts for classes I teach, I would need it to have a seamless pipeline that handles pdf, html, ebook, mobi and math (now called 'stem'?) in one tool. At the moment if I want math I seem to be restricted to vanilla html output as the other tools for generating epub and pdf lag behind the main asciidoctor project.

—
Reply to this email directly or view it on GitHub.

from asciidoctor.

Thanks for that. Yes, I managed to get the latexmath stems working, but not, as you say, with the pdf converter or the asciidoctor-epub3 converters.

from asciidoctor.

I guess what I'm saying is that from my perspective which, longer term will involve using asciidoctor to generate handouts for classes I teach, I would need it to have a seamless pipeline that handles pdf, html, ebook, mobi and math (now called 'stem'?) in one tool.

That's precisely the mission we are currently pursuing. We just haven't finished the race yet :)

At the moment if I want math I seem to be restricted to vanilla html output as the other tools for generating epub and pdf lag behind the main asciidoctor project.

Once we get Asciidoctor EPUB3 and PDF to their initial stable releases, they will no longer lag behind because they will be handled as core converters (i.e., backends). We just had a lot of catching up to do.

from asciidoctor.

Btw, Asciidoctor PDF can now be installed and used as a gem.

$ gem install asciidoctor-pdf --pre
$ asciidoctor-pdf README.adoc

Keep in mind it's currently alpha. You can track the progress by following the Asciidoctor PDF project.

https://github.com/asciidoctor/asciidoctor-pdf

Asciidoctor PDF also works in AsciidoctorJ.

You can find an example of Asciidoctor PDF in action in a Maven build here:

https://github.com/asciidoctor/asciidoctor-maven-examples/tree/master/asciidoc-to-pdf-example

To use it currently requires a manual installation, but we envision it will be packaged in the asciidoctorj-pdf jar to make it as simple to use as AsciidoctorJ.

from asciidoctor.

Now that Asciidoctor PDF has been published, I'm going to close this issue and handle issues using the asciidoctor-pdf (or asciidoctor-fopub) issue tracker.

from asciidoctor.