Create an extension that initially hides example blocks with the role "result" and uses JavaScript to toggle the visibility of the content when a dedicated label (e.g., "View Result") is clicked or touched.

An example of this extension can be seen on the AsciiDoc Syntax Quick Reference page on the Asciidoctor website. Another example can be found in the Groovy documentation.

This could be really useful when writing documentation for a JavaScript library.
The extension will search for JavaScript source code blocks and add a link/button "Run this code in your browser" to execute the code and print the result in the browser.

I'm sure you are already working on it 😄
https://blog.twitter.com/2014/open-sourcing-twitter-emoji-for-everyone

//cc @lordofthejars

With the help of #56 and by removing the hardcoded noheader attribute I tried to add ToCs in the chunked html files to no avail.

It looks like the multipage-html5-converter extension breaks :toc: handling somehow.

Hi,

I'm trying to write a extension for placing videos from a Sign Language. It would be some thing like this:

libras.yaml:

"casa": https://www.youtube.com/watch?v=xjxjTMBoNjE

sample.adoc:

:sign-lang: libras

sign::casa[]

It would read casa from libras.yaml and translate to:

video::https://www.youtube.com/watch?v=xjxjTMBoNjE[]

It would be more complex, but this would be a start.

Can someone help me?

Here's what I have done so far:

• I think it would be a subclasse of BlockMacroProcessor, like gist-block-macro
• A made a copy of gist-block-macro and replace gist to sign:

(...)
class SignBlockMacro < Extensions::BlockMacroProcessor
  use_dsl

  named :sign
(...)

sample.adoc

= Sign Block Macro Extension

.Guard setup to live preview AsciiDoc output
sign::mojavelinux/5546622[]

• And I run it with -r:

asciidoctor-extensions-lab$ asciidoctor --trace -r lib/sign-languages/extension.rb lib/sign-languages/sample.adoc  
/home/santana/.rvm/rubies/ruby-2.2.1/lib/ruby/site_ruby/2.2.0/rubygems/core_ext/kernel_require.rb:54:in `require': cannot load such file -- lib/sign-languages/extension.rb (LoadError)
    from /home/santana/.rvm/rubies/ruby-2.2.1/lib/ruby/site_ruby/2.2.0/rubygems/core_ext/kernel_require.rb:54:in `require'

• It didn't work. So I have tests gist alone:

asciidoctor-extensions-lab$ asciidoctor --trace -r lib/gist-block-macro/extension.rb lib/gist-block-macro/sample.adoc 
/home/santana/.rvm/rubies/ruby-2.2.1/lib/ruby/site_ruby/2.2.0/rubygems/core_ext/kernel_require.rb:54:in `require': cannot load such file -- lib/gist-block-macro/extension.rb (LoadError)
    from /home/santana/.rvm/rubies/ruby-2.2.1/lib/ruby/site_ruby/2.2.0/rubygems/core_ext/kernel_require.rb:54:in `require'

• Since I don't know if gist is working, then I have tested an other BlockMacroProcessor extension, the TreeBlockMacro. I changed the sample file to directory of mine, and tested it:

asciidoctor-extensions-lab$ asciidoctor --trace -r lib/tree-block-macro/extension.rb lib/tree-block-macro/sample.adoc 
/home/santana/.rvm/rubies/ruby-2.2.1/lib/ruby/site_ruby/2.2.0/rubygems/core_ext/kernel_require.rb:54:in `require': cannot load such file -- lib/tree-block-macro/extension.rb (LoadError)
    from /home/santana/.rvm/rubies/ruby-2.2.1/lib/ruby/site_ruby/2.2.0/rubygems/core_ext/kernel_require.rb:54:in `require'

Can someone guide me? What I am doing wrong?

I think we should use https://github.com/zenorocha/clipboard.js

Inspiration: http://foundation.zurb.com/sites/docs/installation.html#command-line-tool

Asciidoctor version 1.5.2 mentions, in the help, an option to require ruby modules before running

    -r, --require LIBRARY            require the specified library before executing the processor (using require)
                                     may be specified more than once

I guess this can be used to load these extensions from the command line but I can't find any documentation about this. Especially it's not mentioned either here or here.

To be consistent with the <name>::<target>[<attributes>] convention, I think we should change the placement of arguments so the data is <target> and the chart type is the first positional attribute. I think chart type should default to "line" if not specified.

chart::sample.csv[line]

chart::sample.csv[bar]

chart::sample.csv[step]

chart::sample.csv[spline]

I think we should also support positional attributes for x and y label as well as size.

chart::<data-uri>[<type>,<x-label>,<y-label>,<width>,<height>]

We should also support a block form where the data is specified in a literal block:

[chart,line]
....
January,February,March,April,May,June,July
28,48,40,19,86,27,90
65,59,80,81,56,55,40
....

It will be necessary to register separate extensions for the block macro and block forms, but they can share common code.

Hi,

I want to build chart-block-macro, after build I'm getting only

/* Generated by Opal 0.6.3 */
(function($opal) {
  var $a, $b, TMP_1, self = $opal.top, $scope = $opal, nil = $opal.nil, $breaker = $opal.breaker, $slice = $opal.slice;

  self.$require_relative("chart-block-macro/extension");
  return ($a = ($b = $scope.Extensions).$register, $a._p = (TMP_1 = function(){var self = TMP_1._s || this, $a;

  if ((($a = self.$document()['$basebackend?']("html")) !== nil && (!$a._isBoolean || $a == true))) {
      return self.$block_macro($scope.ChartBlockMacro)
      } else {
      return nil
    }}, TMP_1._s = self, TMP_1), $a).call($b);
})(Opal);

What can be the reason?

I want to add simple tree diagrams to a text. Is possible to add an easy tree generation feature.

E.g.

- level 1 (A)
-- child of level A (B)
---  child of B
-- 2nd child of A
- 2nd child of root

Let's ruin the internet. R U with me?

In many documents it is needed to have list of figures and list of tables right below table of contents. Would be good to have this feature in asciidoctor-pdf

I think it's an extension I need, but I'm not sure how to go about making one. Here's what I need:

I am converting a text book to various e-formats and the book has several in-text devices. Up till now I have been using the built-in admonitions and just created a new stylesheet to use different icons to match what I need. However, it seems to me that rather than subverting the existing admonitions it would be better to create new custom blocks for this project. For example, I need a block that looks and behaves exactly like an admonition except that it should be called thinkspot rather than tip, caution, warning, or note.

Can someone give me a step-by-step on what to do?

Thanks!

As an alternative to the currently CSS-based approach, allow the salt diagram from plantuml to be used (via Asciidoctor Diagram) to generate the file tree diagram for the tree block macro.

See https://github.com/asciidoctor/asciidoctor-diagram/blob/master/lib/asciidoctor-diagram/plantuml/extension.rb#L70

That way writers could ensure that theirs documents respect conventions and best practices. @mojavelinux I'm sure you already have a list of best practices ? 😄

Then it would be great to let users configure rules with a simple config file.

I'm using the multipage converter to create chunked output. I ran into some issues trying to get the following markup to render properly:

##### Example title

pipeline.<name>.example::

Example text
+
*Required.*
+
.Example
----
# pipeline configuration
pipeline.example.quorum = \
----

I'd only get the title of this in the output. I think this can be addressed by the reparenting function here: https://github.com/asciidoctor/asciidoctor-extensions-lab/blob/master/lib/multipage-html5-converter.rb#L60. I think it maybe sufficient to add something like:

      if block.context == :dlist
        block.parent = parent
      end

after the recursive call to reparent. Adding that I started getting the output that I expected. Not sure if this is the right approach.

Running:

asciidoctor -r ./lib/showcomments-preprocessor.rb lib/showcomments-preprocessor/sample.adoc

gives me:

/Users/max/code/asciidoctor-extensions-lab/lib/showcomments-preprocessor.rb:4:in `<top (required)>': uninitialized constant Extensions (NameError)
	from /Users/max/.rvm/rubies/ruby-2.3.3/lib/ruby/site_ruby/2.3.0/rubygems/core_ext/kernel_require.rb:55:in `require'
	from /Users/max/.rvm/rubies/ruby-2.3.3/lib/ruby/site_ruby/2.3.0/rubygems/core_ext/kernel_require.rb:55:in `require'
	from /Users/max/.rvm/gems/ruby-2.3.3/gems/asciidoctor-1.5.6.1/lib/asciidoctor/cli/options.rb:230:in `block in parse!'
	from /Users/max/.rvm/gems/ruby-2.3.3/gems/asciidoctor-1.5.6.1/lib/asciidoctor/cli/options.rb:228:in `each'
	from /Users/max/.rvm/gems/ruby-2.3.3/gems/asciidoctor-1.5.6.1/lib/asciidoctor/cli/options.rb:228:in `parse!'
	from /Users/max/.rvm/gems/ruby-2.3.3/gems/asciidoctor-1.5.6.1/lib/asciidoctor/cli/options.rb:33:in `parse!'
	from /Users/max/.rvm/gems/ruby-2.3.3/gems/asciidoctor-1.5.6.1/lib/asciidoctor/cli/invoker.rb:22:in `initialize'
	from /Users/max/.rvm/gems/ruby-2.3.3/gems/asciidoctor-1.5.6.1/bin/asciidoctor:12:in `new'
	from /Users/max/.rvm/gems/ruby-2.3.3/gems/asciidoctor-1.5.6.1/bin/asciidoctor:12:in `<top (required)>'
	from /Users/max/.rvm/gems/ruby-2.3.3/bin/asciidoctor:23:in `load'
	from /Users/max/.rvm/gems/ruby-2.3.3/bin/asciidoctor:23:in `<main>'
	from /Users/max/.rvm/gems/ruby-2.3.3/bin/ruby_executable_hooks:15:in `eval'
	from /Users/max/.rvm/gems/ruby-2.3.3/bin/ruby_executable_hooks:15:in `<main>'

Create an extension to detect merge conflicts in AsciiDoc source files.

With large commits, it's easy to miss some conflict and accidentally commit it. Example here:

http://gist.asciidoctor.org/?e4b59c5f6b0a85689859

When this happens in source code, the compiler or a test will normally tell you. If Asciidoctor could detect the pattern and tell you (aka fail the build), that would be great. This could likely be handled by a Preprocessor extension.

Create either an extension or an add-on script that extracts the revision history from the git history.

The idea is to only store the latest revision number and corresponding remark in the document itself. Whenever the revision number is changed, you commit the change. Then, you can get a full history of revision numbers and corresponding remarks by walking the git history and looking for when then revision number changes. Create a catalog, then output it to a temporary docinfo file. Finally, convert as normal with the docinfo enabled.

This is a very minor issue, since it's got such an easy workaround, but I'm reporting it anyway.

If I have a doc with conditional formatting based on ifdef::[], the script ignores the attributes set in the file. They have to be passed via the command line. I would expect the following file to give me the english version when passed through the script, but it just gives me blank output. If I do ruby asciidoc-coalescer.rb -a en file.ad, however, I get the correct output.

:en:

ifdef::en[]
== My Document
English version

include::[some-file-en.ad]
endif::en[]

ifdef::pt[]
== Meu Documento
Versão em portugûes

include::[some-file-pt.ad]
endif::pt[]

I tried creating an extension that would generate HTML ready to be pasted into a Wordpress post in source form (minimal HTML).

For it to work without setting the --no-header-footer in addition to the -r flag I would have liked to set the :header_footer = false option directly when the extension is registered.

I've tried several ways and it didn't work. Including subclassing and overloading initialize and trying to access the @options field and add the option to it (but the hash is frozen).

I need guidance. Here's the code with attempts commented out: https://github.com/asciidoctor/asciidoctor-extensions-lab/compare/master...obilodeau:wordpress-extension?expand=1

Example:

RUBY_ENGINE == 'opal' ? (require 'man-inline-macro/extension') : (require_relative 'chrome-inline-macro/extension')

We can revert this change as soon as Opal 0.7.0 is released.

It might be interesting to integrate Asciidoctor with RubyPants to handle the conversion of straight quotes to smart quotes. Although there is support for smart quotes in the AsciiDoc syntax, this extension would convert them implicitly.

See https://github.com/jmcnevin/rubypants

When using extension(.js) in the chrome extension, assets are stored within the extension package and the URL is chrome-extension://<EXTENSION_ID>/.

I think it would be nice to have a standard way to overload assets path.

Using the JSON format will allow to control how the data is appended to the document. The embedded script is using document.write which can cause issue (see asciidoctor/asciidoctor.js#99)

Can you put testing in place so these are readily potable and usable from AJ version also.

Ala Markdown, create an extension that parses document attributes from the bottom of the document. We want to support this feature in core, but it can be implemented for Asciidoctor 1.5.0 as a Preprocessor extension.

It's hard to navigate the repository because some extensions are defined an an extension.rb file and some are defined in the main extension file. Add links to each extension listed in the README to navigate to each of the following files:

• registration
• extension code
• sample

In some case, the registration and extension code are the same file.

Create an extension that inserts shy hyphens into text by passing the text through a library such as text-hyphen. The result is that justified text can be hyphenated at logical locations so that the spread of words is more even.

The output will work with the existing styles in the default Asciidoctor stylesheet. Here's a screenshot that demonstrates how it would look:

I don't know if it matters, but in the file emoji-inline-macro/extension.rb, the class has a wrong name, it's EmojiBlockMacro, it should be EmojiInlineMacro instead.

RunKit is a tool to try Node.js in the browser. It is wired to Node package registry – npm.

I imagine something like this:

[source,javascript,runkit="nodeVersion:6"]
----
const lodash = require("lodash");

lodash.camelCase('get content');
----

becomes interactive as https://runkit.com/oncletom/58bc55b33cdd1a0013ae1a89 by pressing the Run in browser button (like the View Results extension).

<script src="https://embed.runkit.com"></script>
<script>
document.querySelectorAll('[data-runkit]').forEach(el => {
  RunKit.createNotebook({
    element: el,
    source: el.querySelector("code").textContent
  });
});
</script>

// ...

<pre class="javascript" data-runkit data-runkit-version="6"><code>
const lodash = require("lodash");

lodash.camelCase('get content');
</code></pre>

RunKit documentation is available here: https://runkit.com/docs/embed

cc @Mogztter

Links are generated as if the output was a single file, using local anchors.
They should reference anchors inside the chunked files instead.

https://github.com/gionkunz/chartist-js

Looks great!

Conditional statements are not working when I use 'mathematical-treeprocessor' extension.
The "hello world" string doesn't apear when using 'preprossor extension'.

:enable-string:

ifdef::enable-string[]
hello world!
endif::[]

Conditional statements are working well when I delete mathematical-preprocessor code on 'mathematical-treeprocessor.rb'.

Extensions.register do
  # if @document.basebackend? 'html'
  //inline_macro MathematicalInlineMacro
  treeprocessor MathematicalTreeprocessor
end

I've tested several other preprocessors, it's same on other preprocessors make same problem too.

Conditional statements which looks for CLI attributes works well.
Only conditional satements which looks for DOCUMENT attributes are not working.

Contents

• PROBLEM
• EXAMPLE
• HTML Output
• Code

@mojavelinux, could you take a look at the below?

PROBLEM: Add offsets for subsections

In the example below a modification to sectnumoffset-treeprocessor.rb is used to offset subsection
numbers in the first section that is processed. I would like to insert a parent section at the head
of the document and eliminate the text == Outer Test\n so that the first line of the
HTML output is

<h3 id="_inner_test_a">2.3. Inner Test A</h3>

Help is needed from the Ascii Doctor.

EXAMPLE:

require 'asciidoctor-latex'
source = "== Outer Test\n=== Inner Test A\n$a^2 = 1$\n\n=== Inner Test B\nfoo\n\n== Outer again\nho ho ho!\n\n=== Yikes!"

puts Asciidoctor.convert(source, backend: 'html5', attributes: 'sectnums sectnumoffset=1  subsectnumoffset=2')

HTML output:

[~/dev/asciidoctor-latex/work] ruby-2.2.2 $ ruby y.rb   
subsectnumoffset: 2
Insert parent section at 'head' of document with offset 1
<div class="sect1">
<h2 id="_outer_test">2. Outer Test</h2>
<div class="sectionbody">
<div class="sect2">
<h3 id="_inner_test_a">2.3. Inner Test A</h3>
<div class="paragraph">
<p>\(a^2 = 1\)</p>
</div>
</div>
<div class="sect2">
<h3 id="_inner_test_b">2.4. Inner Test B</h3>
<div class="paragraph">
<p>foo</p>
</div>
</div>
</div>
</div>
<div class="sect1">
<h2 id="_outer_again">3. Outer again</h2>
<div class="sectionbody">
<div class="paragraph">
<p>ho ho ho!</p>
</div>
<div class="sect2">
<h3 id="_yikes">3.1. Yikes!</h3>

</div>
</div>
</div>

CODE:

Extensions.register do
  # A treeprocessor that increments each level-1 section number by the value of
  # the `sectnumoffset` attribute.
  #
  # In addition, if `subsectnumoffset` is defined and greater than zero,
  # the numbers of subsections in the first section encountered will
  # be incremented by the offset.
  #
  # The numbers of all subsections will be
  # incremented automatically since those values are calculated dynamically.
  #
  # Run using:
  #
  # asciidoctor -r ./lib/sectnumoffset-treeprocessor.rb -a sectnums -a sectnumoffset=1 lib/sectnumoffset-treeprocessor/sample.adoc
  #
  #
  treeprocessor do
    process do |document|
      if (document.attr? 'sectnums') && (sectnumoffset = (document.attr 'sectnumoffset', 0).to_i) > 0
        subsectnumoffset = (document.attr 'subsectnumoffset', 0).to_i
        warn "subsectnumoffset: #{subsectnumoffset}".red
        section_count = 0
        if subsectnumoffset > 0
          warn "Insert parent section at 'head' of document with offset #{sectnumoffset}".cyan
        end
        ((document.find_by context: :section) || []).each do |sect|
          next unless sect.level <= 2
          if sect.level == 1
            section_count += 1
            sect.number += sectnumoffset
          elsif sect.level == 2 && section_count == 1
            sect.number += subsectnumoffset
          end
        end
      end
      nil
    end
  end
end

http://bramp.github.io/js-sequence-diagrams/

I am planning to write a preprocessor to parse attributes from a yaml file as described in 566. In order to maximize the use of the config file, I am planing to implement a theme feature from the config file, which contains common attributes (overidable) and theme specific attributes.
Example:

linkcss: true
copycss: true
stylesdir: styles
themes:
   bright:
      stylesheet: asciidoctor
      source-highlighter: highlight.js
      highlightjs-theme: atom
   dark:
      stylesheet: asciidoctor-dark
      source-highlighter: highlight.js
      highlightjs-theme: monokai

Then the cli command is asciidoctor --config-file=conf.yml --theme=dark -a stylesdir=new_style input.adoc.

The order of the attributes is cli > theme > common. It's easier to implement theme > common, but for cli > theme, config-file and theme, the preprocessor need to read the command params.

The current gist-block-macro extension is not cross compiling well with Opal.
I don't know the reason behind but when I replaced registry.document with @document everything work as expected.

Given that other extensions favor @document over registry.document I think it's a good idea to stick with @document (shorter).

We need a README for this repository that covers (at least) the following information:

1. What extensions are available here
2. How to load an extension from the extension lab

We should also cover how these extensions are organized to help people create new ones.

This is a follow-up from the question posed in #35.

It would be more verbose to rename files inside lib/*.rb to include registration in their names:

lib/chart-block-macro-registration.rb
lib/chrome-inline-macro-registration.rb
lib/copyright-footer-postprocessor-registration.rb
lib/emoji-inline-macro-registration.rb
(...)

It would avoid errors like #45.

Add a macro for inserting GitHub / BaseCamp emoji.

http://www.emoji-cheat-sheet.com/

It would be interested to demonstrate how to create an extension for AsciiDoc that ran through all the source listings and verified the expected output. This is an inside-out-approach from pulling snippets from a test suite. Different circumstances call for different approaches.

Since the extension has to run the assertions, it will likely be language-specific. We should focus on Ruby or JavaScript in the example.

Here's the inspiration for this extension (though it uses really quirky syntax which I'd like to improve on).

http://www.jspatterns.com/unit-testing-in-asciidoc/

Create a DocinfoProcessor extension for embedding JavaScript files in generated HTML output.

For example:

:javascripts: myjavascript.js

Converts into :

<script type='text/javascript' src='myjavascript.js'></script>

The javascripts attribute can be a comma-separated list of paths. We may also want to specify the location of the JavaScript file explicitly.

:header_javascripts: ...
:footer_javascripts: ...

The default should probably be header.

Integrate with Roaster (https://github.com/forge/roaster) to allow Java source code to be included inside a listing block by its structure instead of by line numbers or tagged regions.

Add a block extension to demonstrate how to output a chart using the content of the block as the data. One idea is using chart.js or d3.js.

The idea is to be able to generate a changelog:

We could even support multiple SCM. For instance, if we want to display the Git changelog of a project, between master and version 1.0.0:

changelog:git[master..v1.0.0]

OK, couldn't find a better title... Here is what I want to do:
I am writing technical machine (not software :-) ) documentation using asciidoctor on atom. The machines are described by elements. With the element it is suitable to define the required maintenance work. Could be a small table, an admonition, ... whatever.

While for the writer it is suitable to have the maintenance described with the element, it is more suitable for the maintenance person to have a maintenance summary list at the end of the document.

I came up with tags and a self-including document, see code snippet below. Any better idea?

= Title

== Motor

Content not part of the maintenance tag

|===
| Element | Maintenance Instruction | Frequency

// tag::Maintenance[]
| icon:Maintenance[] <<Motor>> Oil | Change Motor Oil, use for salad | Daily
// end::Maintenance[]
|===

== Extruder

|===
| Element | Maintenance Instruction | Frequency

// tag::Maintenance[]
| icon:Maintenance[] <<Extruder>> Screw | Make nice & shiny | Before go to sleep
// end::Maintenance[]
|===


== Maintenance Summary

|===
| Element | Maintenance Instruction | Frequency

include::Source.adoc[tag=Maintenance]

|===

Been trying to get a compilation path that will render a document with math in it across html, pdf, epub, and mobi. Managed to get mathoid-treeprocessor installed after some tinkering.

Using the sample .adoc file:

= _Precompiled_ Math!

[stem#equation1]
++++
"d"(bbp,bbq) = sqrt((p_1 - q_1)^2 + (p_2 - q_2)^2)
++++

and the command:

asciidoctor  -r asciidoctor-pdf -r ./lib/mathoid-treeprocessor  -b pdf test.adoc --trace

I get the following output:

asciidoctor: WARNING: could not embed image: /Volumes/SSD Disk/git Repos/test/equation1.svg; Duplicate attribute "xlink:href"
Line: 23
Position: 7155
Last 80 unconsumed characters:

The generated svg file is not an equation but text containing an error message "This page contains the following errors..."

Running phantomjs main.js with the query string http://localhost:16000/?q=x^2&type=asciimath generates a web page in the browser saying:

This page contains the following errors:

error on line 1 at column 1252: Attribute xlink:href redefined
Below is a rendering of the page up to the first error.

The page source is:

<svg xmlns:xlink="http://www.w3.org/1999/xlink" viewBox="-22 -915.1157009464328 932.0389244992066 953.2641499988939" style="width: 2.167ex; height: 2.167ex; vertical-align: -0.135ex; margin: 1px 0px; position: static;" xmlns="http://www.w3.org/2000/svg"><defs id="MathJax_SVG_glyphs"><path id="STIXWEBMAINI-78" stroke-width="10" d="M243 355l12 -57c70 107 107 143 151 143c24 0 41 -15 41 -37c0 -21 -14 -36 -34 -36c-19 0 -28 17 -52 17c-18 0 -54 -44 -98 -121c0 -7 2 -21 8 -45l32 -134c7 -28 16 -41 30 -41c13 0 24 10 47 40c9 12 13 18 21 28l15 -9c-58 -90 -84 -114 -122 -114 c-32 0 -47 18 -59 68l-29 119l-88 -119c-44 -59 -64 -68 -95 -68s-50 16 -50 42c0 20 14 36 34 36c9 0 19 -4 32 -11c10 -6 20 -9 26 -9c11 0 30 19 51 49l82 116l-28 124c-14 60 -21 68 -46 68c-8 0 -20 -2 -39 -7l-18 -5l-3 16l11 4c61 22 94 29 117 29 c25 0 37 -18 51 -86Z"></path><path id="STIXWEBMAIN-32" stroke-width="10" d="M474 137l-54 -137h-391v12l178 189c94 99 130 175 130 260c0 91 -54 141 -139 141c-72 0 -107 -32 -147 -130l-21 5c21 117 85 199 208 199c113 0 185 -77 185 -176c0 -79 -39 -154 -128 -248l-165 -176h234c42 0 63 11 96 67Z"></path></defs><g stroke="black" fill="black" stroke-width="0" transform="matrix(1 0 0 -1 0 0)"><use xlink:href="#STIXWEBMAINI-78" xlink:href="#STIXWEBMAINI-78"></use><use transform="scale(0.7071067811865476)" xlink:href="#STIXWEBMAIN-32" x="640" y="583" xlink:href="#STIXWEBMAIN-32"></use></g></svg>

Any ideas?