adopt more conventional directory structure

17 files changed, 0 insertions, 5662 deletions

diff --git a/mod/doc/context/third/rst/documentation.rst b/mod/doc/context/third/rst/documentation.rst
deleted file mode 100644
index 621528d..0000000
--- a/mod/doc/context/third/rst/documentation.rst
+++ /dev/null
@@ -1,691 +0,0 @@
-========================
-Features Not Implemented
-========================
-Nesting
-*******
-Proper nesting. So far only lists support real nested structures.
-There's no way you could have real paragraphs or bulleted lists
-inside table cells. The problem is that with true nesting some
-jobs like the dissection of tables would have to be moved from
-the formatter to the parser. If you feel you need thoroughly
-nested structures -- e.g. grid tables in footnotes or bullet lists
-inside simple tables inside enumerations inside quotations inside
-footnotes -- you should consider including |CONTEXT| code as
-substitution directives. (OTOH docutils' new and old LaTeX
-formatter seems to have problems with tables in footnotes as
-well. Not to mention its preference for enclosing random nested
-structures in ``quote``-environments.)
-
-Should you find yourself in desparate need of tables or whatever
-structures inside footnotes then I might agree to find a solution
-if you ask.
-
-Hyperlinks
-**********
-The hyperlink implementation should be fine in general use if you
-avoid certain situations.
-
-- Never ever call your hyperlink targets ``anon_#``, where ``#``
-  stands for  any integer. Just don't do it, OK? Great.
-
-- Referencing a structure element like a section heading by means
-  of an *empty link block* does work. However, if the element in
-  question requests a page break (e.g. the vanilla
-  ``\chapter{#1}`` command), the reference will link to the
-  previous page instead and become useless. You can avoid this
-  behaviour by referencing the section directly or by targetting
-  the first paragraph in the section instead.
-
-- Link chaining does not work with internal references. This is
-  considered a low-priority bug and will be addressed during the
-  next big hyperlink overhaul.
-
-=====
-Usage
-=====
-Invocation from the Command Line
-********************************
-|rstcontext| is integrated into the ``mtxrun`` command as a
-script, which relies, naturally, on the Lua interpreter of
-|LUATEX|. Therefore, |rstcontext| might not run at all on other
-Lua installations, at least not without modification of the
-source. Fortunately, every |CONTEXT| user is equipped with
-|LUATEX| nowadays so this dependency should be trivial.
-
-To generate |CONTEXT| code from a |rst| document named
-``infile.rst``, call ``mtxrun``: ::
-
-     $mtxrun --script rst --if=infile.rst --of=outfile.tex
-
-You should now have a file ``outfile.tex`` that is ready to be
-run by |CONTEXT|. With some exceptions the generated code is
-downward compatible with MkII, thus it does not matter for a
-start whether you decide to test it with ``texexec`` or
-``context``.
-
-The resulting |TEX| file has rather a basic layout, if at all.
-This is intentional as you are expected to include it in a
-document after specifying your own setups.
-An example for prepended setups can be found in the environment
-for this manual (``mod/doc/context/third/rst/manual.tex``).
-
-.. caution::
-    The output of |rstcontext| automatically inserts necessary
-    setups for the components found in the input. Therefore, the
-    ``\starttext`` and ``\stoptext`` commands are part of the
-    output and may not be specified in your setups file.
-    For now you have to use the |CONTEXT| command 
-    ``\appendtoks <token> \to \starttext`` to add content like
-    title pages and indices to the result. This mechanism works
-    reliable as long as you have an eye on the order in which the
-    tokens are given. Again, have a look at ``manual.tex`` to get
-    an impression how useful this can be. User hooks for these
-    and other common constructs are thought of but have yet to be
-    implemented.
-
-To build the documentation, first create a temporary directory
-somewhere safe. Then copy or symlink the Lua files from
-``mod/tex/context/third/rst/`` and the manual source there as
-well: ::
-
-    $mkdir tmp; cd tmp
-    $ln -s ../mod/doc/context/third/rst/documentation.rst .
-
-Now run |rstcontext| on the main documentation file as follows: ::
-
-    $mtxrun --script rst --if=documentation.rst --of=doc.tex
-
-Now run |CONTEXT| on the layout file: ::
-
-    $context ../mod/doc/context/third/rst/manual.tex
-
-This will include the generated code after a couple of setups --
-voilà, you have successfully built ``manual.pdf``. (Note that the
-commands you have to issue in each of the steps vary across
-different OS. In the literal form the example might only work on
-Linux or POSIX compliant systems.)
-
-Module
-******
-
-A provisional module for MkIV is included (``t-rst.mkiv``).
-Actually, the converter was thought of as a module for direct
-rendering of |rst| input initially, but certain objections
-diverted me from this path.
-
--   *Typography*. It’s all about the details. No matter how good your
-    converter is, auto-generated code will not reach |TEX|’s
-    omnipotence and flexibility. |rstcontext| is a tool to
-    generate raw material for your typesetting job, not a
-    typesetting system in itself.
-
--   *Testing*. Never underestimate the insights gained from reading
-    the resulting |CONTEXT| file. Quite some effort has been
-    undertaken to make it human-readable, especially the setups.
-
--   *MkII*. I’m not an MkII user at all, and compatibility with
-    it is not a primary objective for |rstcontext|.
-    However, an effort has been made to keep the output essentially
-    MkII clean. Do not expect Unicode to work without
-    precautions.
-
-During the development readability of the generated code was
-alway one of the main goals of |rstcontext|. Quite some computing
-effort is made to reflow even simple things as paragraphs into
-a shape understandable by more than only the |TEX| machine.
-If you should at one point decide that your project is
-ripe for the typographical finish and you want to add local
-changes in form of |TEX| code only, you should be able to use the
-output of |rstcontext| as starting point.
-
-However, using the module may have advantages when testing. There
-is a usage example in ``moduletest.tex``, introducing the macro
-``\\typesetRSTfile``.  Another example in ``hybridtest.tex``
-demonstrates the |CONTEXT| command ``\\RST`` as well as the
-corresponding environment.
-
-To install the module simply copy the files into your local |TEX|
-tree, i.e. if the minimals reside in ``~/context/``, you would
-issue the following line: :: 
-
-    $cp -r ./mod/* ~/context/tex/texmf-local/
-
-Then rebuild the filename database running ``context
---generate``. The module should be ready for use now.
-
-RST projects
-************
-
-In addition to the simple command ``\\typesetRSTfile`` the module
-also provides means for handling multiple |rst| input files. This
-is achieved by so-called *inclusions*. An inclusion has to be
-defined first, using the macro ``\\defineRSTinclusion``, which
-receives up to three arguments in brackets. The first one
-specifies the *identifier* by which the inclusion will be
-referred to afterwards (cf. |CONTEXT|’s ``\\useURL`` command). The
-second argument, which is mandatory as well, takes the file to be
-associated with an inclusion. Finally, optional setups can be
-passed to the parser via the third argument (cf. the section on
-`Tabs`_). E.g.: ::
-
-    \usemodule[rst]
-    \defineRSTinclusion [first][inc-first.rst]
-    \defineRSTinclusion[second][inc-second.rst][expandtab=true,shiftwidth=8]
-    \defineRSTinclusion [third][inc-third.rst]
-
-Those inclusions are afterwards accessible *within* the
-``\\[start|stop]project`` environment, and can be dereferenced by
-``\\RSTinclusion``, which takes the identifier as a single
-argument in brackets: ::
-
-    \startRSTproject
-    \RSTinclusion [first]
-    \RSTinclusion[second]
-    \RSTinclusion [third]
-    \stopRSTproject
-
-Within the project environment, |rstcontext| allows for arbitrary
-|CONTEXT| markup.
-
-=========
-Examples
-=========
-
-|rstcontext| was developed for the largest part by going through
-the |rst| specification_ step by step and tested against the
-examples given both in the spec and in the `quick reference`_.
-Therefore you should refer to those examples first (and drop me a
-note immediately if any of them stopped working).
-All kinds of text blocks and inline markup have been implemented
-with the exception of anything mentioned in the section on
-`Features Not Implemented`_.
-Some of them that I have not found a real-world usage for (such
-as *definition lists*) do not yet have a presentable output --
-there is room for improvements that should be supplied by
-somebody who actually uses those features.
-
-.. _specification: http://docutils.sourceforge.net/docs/ref/rst/restructuredtext.html
-.. _quick reference: http://docutils.sourceforge.net/docs/user/rst/quickref.html
-
-Block Quotes
-************
-
-The *block quote* syntax is fully supported, including
-attributions. For instance, the next snippet: ::
-
-    Some people have made the mistake of seeing Shunt’s work as a
-    load of rubbish about railway timetables, but clever people
-    like me, who talk loudly in restaurants, see this as a
-    deliberate ambiguity, a plea for understanding in a
-    mechanized world.
-
-    --- Gavin Millarrrrrrrrrr on Neville Shunt
-
-gets you a neatly indented quotation, typeset in a slightly
-smaller font magnitude.
-
-    Some people have made the mistake of seeing Shunt’s work as a
-    load of rubbish about railway timetables, but clever people
-    like me, who talk loudly in restaurants, see this as a
-    deliberate ambiguity, a plea for understanding in a
-    mechanized world.
-
-    --- Gavin Millarrrrrrrrrr on Neville Shunt
-
-Don’t forget proper indentation.
-
-Numbered List
-*************
-
-Save for nesting lists are fully implemented in |rstcontext|.
-The following code typesets a triple-nested list with different
-kinds of bulleting / numbering: ::
-
-    i   First order list, first entry.
-
-    ii  First order list, second entry.
-
-    iii First order list, third entry.
-
-        -   Second order list, first entry.
-
-            #   Third order list, first entry.
-
-            #   Third order list, second entry.
-
-            #   Third order list, third entry.
-                Real nesting rules!
-
-        -   Second order list, second entry.
-
-    iv  First order list, fourth entry.
-
-    v   First order list, fifth entry.
-
-The result looks like this:
-
-i   First order list, first entry.
-
-ii  First order list, second entry.
-
-iii First order list, third entry.
-
-    -   Second order list, first entry.
-
-        #   Third order list, first entry.
-        #   Third order list, second entry.
-        #   Third order list, third entry.
-            Real nesting rules!
-
-    -   Second order list, second entry.
-
-iv  First order list, fourth entry.
-
-v   First order list, fifth entry.
-
-.. caution:: 
-    Don’t forget the blank lines between list items.
-
-Line Blocks
-***********
-
-Line blocks are a convenient environment for parts of the text
-that need to preserve line breaks and indentation. This makes it
-the first choice for most kinds of poems: ::
-
-    | When does a dream begin? 
-    |   Does it start with a goodnight kiss? 
-    |       Is it conceived or simply achieved?
-    | When does a dream begin? 
-    |
-    | When does a dream begin? 
-    |   Is it born in a moment of bliss? 
-    |       Or is it begun when two hearts are one?
-    | When does a dream exist? 
-    |
-    | The vision of you appears somehow 
-        Impossible to resist 
-    | But I'm not imagining seeing you 
-        For who could have dreamed of this? 
-    |
-    | When does a dream begin? 
-    |   When reality is dismissed? 
-    |       Or does it commence when we lose all pretence?
-    | When does a dream begin?
-
-Indentation, continued lines, etc. should work out without
-problems:
-
-| When does a dream begin? 
-|   Does it start with a goodnight kiss? 
-|       Is it conceived or simply achieved?
-| When does a dream begin? 
-|
-| When does a dream begin? 
-|   Is it born in a moment of bliss? 
-|       Or is it begun when two hearts are one?
-| When does a dream exist? 
-|
-| The vision of you appears somehow 
-      Impossible to resist 
-| But I'm not imagining seeing you 
-      For who could have dreamed of this? 
-|
-| When does a dream begin? 
-|   When reality is dismissed? 
-|       Or does it commence when we lose all pretence?
-| When does a dream begin?
-
-
-==========
-Directives
-==========
-Admonitions
-************
-The following admonition directives have been implemented:
-
-Caution
--------
-The *caution* directive results in the text being prefixed by one
-“dangerous bend” symbol in order to resemble the “wizards only”
-passages of the TeXbook.
-For example, the directive: ::
-
-    .. caution:: White mice do worse in experiments than grey mice.
-
-will result in the following:
-
-.. caution:: White mice do worse in experiments than grey mice.
-
-Danger
-------
-Similar to the *caution* directive, the *danger* directive
-prefixes the given text with two “dangerous bends” giving it the
-look of Knuths’s “esoteric” annotations.
-
-.. danger:: Be nice to the parser: 
-    Don’t forget to align paragraphs that end a literal
-    block!
-
-
-Images
-******
-Including pictures is easy using the *image* directive: simply
-supply it the name of the image file as in ``.. image:: cow``.
-If the format is supported by |CONTEXT| the suffix can be
-neglected.
-
-The placement of images can be controlled via a set of optional
-arguments, each of which has to be specified on single line in
-``key: value`` style: ::
-
-    .. image:: cow
-        width: hsize
-        caption: A generic Dutch cow.
-
-This will place your image somewhere close to the spot where you
-defined it. (The placement parameter to ``placefigure`` will be
-set to ``here`` by default.)
-
-.. image:: cow
-    cow.pdf
-    width: hsize
-    alt: A generic Dutch cow (*bos primigenius taurus*).
-
-The supported parameters are ``width``
-(alias: ``size``), ``caption`` and ``scale``.
-The *width* parameter accepts the values ``hsize`` 
-(alias: ``fit``, ``broad``) or ``normal``.
-Alternatively, the *scale* parameter allows for arbitrary
-manipulation of the desired magnification; it defaults to ``1``
-(unscaled).
-The value passed as *caption* parameter will be used in as the
-caption text of the image.
-
-.. |CONTEXT| ctx:: \CONTEXT
-.. |TEX| ctx:: \TeX
-.. |PDFTEX| ctx:: \PDFTEX
-.. |LUATEX| ctx:: \LUATEX
-.. |rstcontext| ctx:: \bgroup\em rst\egroup\kern.5pt\CONTEXT
-.. |rst| ctx:: \bgroup\rm re\egroup\bgroup\ss Structured\egroup\bgroup\rm Text\egroup
-.. |LATEX| ctx:: \LATEX
-
-.. _outline: http://docutils.sourceforge.net/docs/ref/rst/restructuredtext.html
-.. _docutils: http://docutils.sourceforge.net/
-.. _Pandoc: http://johnmacfarlane.net/pandoc/
-
-Containers
-**********
-
-Upon request |rstcontext| now supports another kind of
-directive, namely containers_.
-Due to their being defined explicitly in terms of HTML,
-*containers* lack a corresponding construct in |CONTEXT| (or
-|TEX| for that matter).
-Some parts of |CONTEXT| (e. g. ``\\framed``) come quite close with
-respect to functionality as well as generality.
-However, none of the candidates alone covers the entire spectrum
-of functionality that containers_ are supposed to.
-For that reason the implementation leaves them essentially
-undefined.
-
-If an explicit name is specified, then the ``container``
-directive maps to the environment of that name.
-Anonymous containers are interpreted as a |TEX| group.
-Any text block inside the element is treated as ordinary
-paragraph.
-In below example the content will be handled as if between
-``\\startxyzzy`` and ``\\stopxyzzy``, where it is up to the user to
-define the *xyzzy* environment::
-
-    This is a paragraph.
-
-    .. container:: xyzzy
-
-        whatever
-
-        foo **bar** baz
-
-    This is another paragraph.
-
-The middle part translates to |CONTEXT| as follows::
-
-    \start[xyzzy]%
-    whatever
-
-    foo {\sc bar} baz
-    \stop
-
-Note that the ``\\start[foo]``/``\\stop``-environment is equivalent
-to ``\\startfoo``/``\\stopfoo``, except that the environment
-doesn’t actually need to be defined.
-
-.. caution::
-    Support for the *container* directive is considered
-    experimental.
-    Suggestions for improving or extending the current
-    implementation are always welcome.
-
-.. _containers: http://docutils.sourceforge.net/docs/ref/rst/directives.html#container
-
-=======================
-Substitution Directives
-=======================
-
-There are substitution directives for simple *replacing* and
-for insertion of |LUATEX|’s three languages: |mp|, Lua and,
-of course, |TEX|.
-
-.. |mp| replace:: \METAPOST
-
-Ordinary text replacement is done via the ``replace``
-substitution directive. E.g. in the main text you consistently
-use ``|replaceme|`` and have all its occurences substituted by
-``I wasn’t in the mood to write out this long sentence.``
-like in the next snippet:
-
-::
-
-    .. |replaceme| replace::
-        I wasn’t in the mood to write out this long sentence.
-
-The code insertions work similarly. You have to specify some
-phrase that gets substituted by the code you supply.
-E.g. this document accesses the fancy logos predefined in the
-|CONTEXT| core via substitutions: ::
-    
-    .. |CONTEXT| ctx:: \CONTEXT
-    .. |LUATEX| ctx:: \LUATEX
-
-Etc. pp. The respective directive names are ``ctx``, ``mp`` and
-``lua``. In order to get a |circle| drawn on spot, you would
-define a Metapost substitution:
-
-::
-
-    .. |circle| mp::
-        fill fullcircle scaled(8) withcolor blue;
-
-================
-Special Features
-================
-Text Roles
-**********
-
-The default *role* for interpreted text is *emphasis*.
-
-The role marker provides explicit access to formatting commands.
-The formatting routine for inline literals can be called with the
-role marker :literal:`literal`, strong emphasis likewise is
-achieved via the role marker :literal:`strong_emphasis`.
-
-Other roles that lack an equivalent among inline markup are
-``bold``, :ss:`ss` (alias :literal:`sans_serif`),
-``uppercase``, ``lowercase`` and colors.
-Color roles begin with the string ``color_`` (the underscore is
-compulsive), followed by either the string ``rgb_`` or a 
-`valid color name`__.
-An rgb vector is specified in decimal.
-Its values can be separated by either dashes or underscores.
-Thus, ``color_rgb_.3_.5_.8`` is a valid rgb expression, as is
-``color_rgb_0-1-0``.
-Unforturnately, the colon character ``:`` has to be escaped in
-color expressions, e.g. ``color_gray\:5``.
-
-__ http://wiki.contextgarden.net/Colors#Using_predefined_colors:_.5Csetupcolor
-
-For example, to give Mr. Neville Shunt’s work an apt
-typographic representation you can use these roles instead of
-the standard inline markup: ::
-
-    :color_rgb_.9_.2_.7:`Chuff`, chuff, :literal:`chuffwoooooch`,
-    woooooch! Sssssssss, sssssssss!  :uppercase:`Diddledum`,
-    `diddledum`, diddlealum.  :literal:`Toot`, toot. The train
-    :bold:`now` standing :color_gray\:5:`at` platform :ss:`eight,
-    tch`, tch, :color_rgb_0-1-0:`tch`,
-    :color_rgb_.5-.6-.2:`diddledum`, diddledum.
-    :lowercase:`Chuffff`, :strong_emphasis:`chuffffiTff`
-    eeeeeeeeeaaaaaaaaa :color_red:`Vooooommmmm`.
-
-which yields when passed through |rstcontext|:
-
-:color_rgb_.9_.2_.7:`Chuff`, chuff, :literal:`chuffwoooooch`,
-woooooch! Sssssssss, sssssssss!  :uppercase:`Diddledum`,
-`diddledum`, diddlealum.  :literal:`Toot`, toot. The train
-:bold:`now` standing :color_gray\:5:`at` platform :ss:`eight,
-tch`, tch, :color_rgb_0-1-0:`tch`,
-:color_rgb_.5-.6-.2:`diddledum`, diddledum. :lowercase:`Chuffff`,
-:strong_emphasis:`chuffffiTff` eeeeeeeeeaaaaaaaaa
-:color_red:`Vooooommmmm`.
-
-**************************
-Bibliography and Citations
-**************************
-
-.. caution::
-    Not much for now concerning the usage of Taco’s bib system.
-    It’s just that I use my own bibliography system and never
-    became sufficiently familiar with the standard |CONTEXT|
-    approach.  *If you feel that the current support should be
-    improved then feel free to contact me!* I will need somebody
-    for testing.
-
-When |rstcontext| first encounters a citation (``[texbook]_``) it
-automatically looks up a bibliography in the working directory by
-the name of ``\jobname``. E.g. with a main file ``manual.tex``
-bibtex will use the database called ``manual.bib``.  Symlinking
-your bibliography file in the local tree should suffice and you
-can keep whatever directory structure you prefer.  (Speaking for
-myself, bib data usually resides in its own subdirectory, so I’d
-use symlinks, too.)
-
-****
-Tabs
-****
-The |rst| specification requests that tabs (ASCII no 9) be
-treated as spaces_. Converting your tabs to spaces might be a
-good preparation for an |rstcontext| run. However, as of version
-123 |rstcontext| comes with built-in tab expansion. It can be
-enabled by supplying an optional argument to the
-``typesetRSTfile`` command: ::
-
-    \usemodule[rst]
-    \typesetRSTfile[expandtab=true,shiftwidth=4]{myfile.rst}
-
-The argument ``expandtab`` triggers a prepocessing step which
-expands all tabulation characters (``\t`` and ``\v``) into the
-correct amount of spaces. Optionally, the tab stop distance can
-be configured using the ``shiftwidth`` parameter, which defaults
-to 4.
-
-.. _spaces: http://docutils.sourceforge.net/docs/ref/rst/restructuredtext.html#whitespace
-
-===================
-About this software
-===================
-
-The docutils_ provide means to translate the extra-convenient
-markup language |rst| into various formats like PDF, HTML and
-|LATEX|, unfortunately omitting the One True Macro System:
-|CONTEXT|.
-
-As far as I am aware of it, there is some support for |rst| in
-Pandoc_ but as it relies on a rather large set of dependencies it
-proved very difficult (too difficult for me) to install on my
-favourite distribution.
-From the `interactive demo`__ I gather that support for |rst|’s
-language features is not very extensive and the result did not
-even come with proper setups.
-Additionally, it’s written in a language I am not familiar with
-and that does not make use of one the most awesome features of
-all the the extended capabilities |LUATEX| provides: the Lua
-interpreter.
-
-For quite some time I was thinking about how to implement an
-|rst| parser in |LUATEX|, until some discussion__ emerged on the
-|CONTEXT| mailing list that indicates a broader interest in
-convenient markup languages across the community.
-As the alternatives mentioned above don’t meet the expectations
-of a normal |CONTEXT| user, the initial step to write
-|rstcontext| was done.
-Handling most of the corner cases and usability features of |rst|
-proved in the end not nearly as easy as I imagined.
-
-__ http://johnmacfarlane.net/pandoc/try
-__ http://archive.contextgarden.net/message/20100814.051917.28caafcd.en.html
-
-.. caution::
-    |rstcontext| is experimental software and neither feature
-    complete nor thoroughly commented. Keep this in mind before you
-    start using it. Anything might still be subject to change, so
-    expect breakage *in case you start relying on exceptional
-    behaviour* (read: bugs) that does not conform to the |rst|
-    specification. Consider filing a bug report instead and wait for
-    me (the maintainer) to fix it, because regardless of how much
-    testing I do myself I alway run into the weirdest issues only 
-    during the actual deployment of the software. Thus, if you notice
-    that |rstcontext| does not adhere to the outline_ of |rst|
-    according to the Docutils documentation, very likely you have
-    discovered a corner case I was not aware of.
-
-.. |circle| mp::
-    fill fullcircle scaled(8) withcolor blue;
-
-
-=======
-License
-=======
-
-::
-
-    Copyright 2010-2013 Philipp Gesang. All rights reserved.
-
-    Redistribution and use in source and binary forms, with or
-    without modification, are permitted provided that the
-    following conditions are met:
-
-        1. Redistributions of source code must retain the above
-           copyright notice, this list of conditions and the
-           following disclaimer.
-
-        2. Redistributions in binary form must reproduce the
-           above copyright notice, this list of conditions and
-           the following disclaimer in the documentation and/or
-           other materials provided with the distribution.
-
-    THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDER ``AS IS''
-    AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
-    LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND
-    FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT
-    SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY
-    DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
-    CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
-    PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
-    DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED
-    AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
-    LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
-    ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
-    ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
-
-
-.. vim:tw=65
diff --git a/mod/doc/context/third/rst/hybridtest.tex b/mod/doc/context/third/rst/hybridtest.tex
deleted file mode 100644
index bd509c0..0000000
--- a/mod/doc/context/third/rst/hybridtest.tex
+++ /dev/null
@@ -1,42 +0,0 @@
-\usemodule[rst][test=yes]
-\setuphead[chapter][page=no,style=bold]
-
-\def\RSTCTX{{\em rst}\kern.5pt\CONTEXT}
-\def\reST{{\rm re}{\ss Structured}{\rm Text}}
-
-\starttext
-
-\chapter{\RSTCTX\ Hybrid Documents}
-
-This example demonstrates the macro \type{\RST} which can be used
-to process \reST\ markup directly in a normal \CONTEXT\ document.
-
-\RST{
--------------
-This Chapter,
--------------
-
-… for instance, was given entirely in *reST* markup. Naturally,
-there are some :bold:`drawbacks` to expect when mixing markups:
-directives and hyperlink targets that have already been specified
-somewhere above the current section will *stay* accessible in
-later passages until you redefine them. Also, certain letters
-need to be thoroughly escaped in order for them to make it
-through to the *reST*-parser, e.g. *\\\{* (<left brace>), and
-you’ll have to be inventive to make a backslash
-(*\\letterbackslash*) pass through the parser.
-
-}
-
-\startRST
-
--------------
-Alternatively
--------------
-you may always use the matching environment ``\\[start|stop]RST``
-if you prefer.
-
-\stopRST
-
-
-\stoptext
diff --git a/mod/doc/context/third/rst/inc-first.rst b/mod/doc/context/third/rst/inc-first.rst
deleted file mode 100644
index 392649f..0000000
--- a/mod/doc/context/third/rst/inc-first.rst
+++ /dev/null
@@ -1,14 +0,0 @@
-=============
-First Chapter
-=============
-
-Mountaineer? *Mountaineer* where the devil are they, mound, mount... mountain...
-a mountaineer: ‘two men skilled in climbing mountains’. Jolly good, well you’re
-in. Congratulations, both of you.
-
-We’ll be leaving on January 22nd and taking the following routes. The A23s
-through Purleys down on the main roads near Purbrights avoiding Leatherheads and
-then taking the A231s entering Rottingdeans from the North. From Rottingdeans we
-go through Africa to Nairobis.  We take the South road out of Nairobis for about
-twelve miles and then ask.
-
diff --git a/mod/doc/context/third/rst/inc-second.rst b/mod/doc/context/third/rst/inc-second.rst
deleted file mode 100644
index eff120e..0000000
--- a/mod/doc/context/third/rst/inc-second.rst
+++ /dev/null
@@ -1,9 +0,0 @@
-==============
-Second Chapter
-==============
-Well Bruce, I heard the Prime Minister use it. S’hot enough to boil a monkey’s
-bum in ’ere, your Majesty, he said and she smiled quietly to herself.
-
-Now, Bruce teaches classical philosophy, Bruce teaches Hegelian philosophy, and
-Bruce here teaches logical positivism, and is also in charge of the sheepdip.
-
diff --git a/mod/doc/context/third/rst/inc-third.rst b/mod/doc/context/third/rst/inc-third.rst
deleted file mode 100644
index 67e6d5a..0000000
--- a/mod/doc/context/third/rst/inc-third.rst
+++ /dev/null
@@ -1,14 +0,0 @@
-=============
-Third Chapter
-=============
-You are hereby charged that on the 28th day of May 1970, you did wilfully,
-unlawfully, and with malice aforethought publish an alleged English-Hungarian
-phrasebook with intent to cause a breach of the peace. How do you plead?
-
-Not guilty.
-
-I quote an example. The Hungarian phrase meaning ‘Can you direct me to the
-station?’ is translated by the English phrase, ‘Please fondle my bum’.
-
-I wish to plead incompetence.
-
diff --git a/mod/doc/context/third/rst/inc.tex b/mod/doc/context/third/rst/inc.tex
deleted file mode 100644
index 7703fae..0000000
--- a/mod/doc/context/third/rst/inc.tex
+++ /dev/null
@@ -1,11 +0,0 @@
-\usemodule[rst]
-
-\defineRSTinclusion [first][inc-first.rst]
-\defineRSTinclusion[second][inc-second.rst]
-\defineRSTinclusion [third][inc-third.rst]
-
-\startRSTproject
-\RSTinclusion [first]
-\RSTinclusion[second]
-\RSTinclusion [third]
-\stopRSTproject
diff --git a/mod/doc/context/third/rst/manual.bib b/mod/doc/context/third/rst/manual.bib
deleted file mode 100644
index 638ed80..0000000
--- a/mod/doc/context/third/rst/manual.bib
+++ /dev/null
@@ -1,10 +0,0 @@
-@BOOK{texbook,
-    author      = {Donald E. Knuth},
-    title       = {The \TEX book},
-    year        = {1984},
-    address     = {Reading, Mass. et. al.},
-    location    = {Reading, Mass. et. al.},
-    series      = {Computers and typesetting vol. 1},
-    keywords    = {all}
-}
-
diff --git a/mod/doc/context/third/rst/manual.tex b/mod/doc/context/third/rst/manual.tex
deleted file mode 100644
index b3d63db..0000000
--- a/mod/doc/context/third/rst/manual.tex
+++ /dev/null
@@ -1,138 +0,0 @@
-\setuppapersize[A5][A5]
-
-\setupcombinedlist[content][interaction=all,]
-
-\setupinteraction[
-  state=start,
-  focus=standard,
-  color=darkcyan,
-  contrastcolor=\interactionparameter{color},
-]
-
-\setupindenting[yes,next,medium] % -> lead to the glue node error in mkiv
-
-\setuphead[chapter]   [style={\rm\sc},before={\blank[line,force]},after={\blank[2*line,force]}]
-\setuphead[section]   [style={\rm\it},before={\blank[line]}      ,after={\blank[line]}]
-\setuphead[subsection][style={\rm},before={\blank[line]}         ,after={\blank[line]}]
-
-\setupheads[indentnext=yes]
-\setupfloats[indentnext=yes]
-
-\setupbodyfont[latin-modern,10pt]
-\definefontfeature [textfigures]    [onum=yes]
-\definefontfeature [liningfigures]  [onum=no]
-\definefontfeature [texligatures]   [tlig=yes]
-\definefontfeature [notexligatures] [tlig=yes]
-\addfs{textfigures,texligatures}
-
-\setuptyping[
-  style={\addff{liningfigures,notexligatures}\ttx},
-]
-
-% title page
-\startbuffer[frontpage]
-\startstandardmakeup
-\raggedcenter
-\vfill
-{\setupbodyfont[,19pt]
-{\em Typesetting}
-\blank [2*big]
-{\tfc {\em re}{\ss Structured}{\em \kern-6ptText}}
-\blank [2*big]
-{\em with \CONTEXT}
-\blank [5*big]
-{\tfa A Manual for {\em rst}{\kern1pt\CONTEXT}}
-}
-\vfill
-\stopstandardmakeup
-\stopbuffer
-
-% additional information
-\startbuffer[author]
-\startstandardmakeup
-\vfill
-\framed [align=right,frame=off,topframe=on] {%
-\tfxx\ss\setupinterlinespace[small]\noindentation
-Copyright 2010--2013 Philipp Gesang, Radebeul.\par
-Mail any bug reports, patches or suggestions to\par
-\type{string.format("%s@%s.com", "phg42.2a", "gmail")}\par
-or pay a visit to \goto{my BitBucket home}[url(http://bitbucket.org/phg/)].\par
-}
-\stopstandardmakeup
-\stopbuffer
-
-
-% table of contents
-\startbuffer [toc]
-\setuppagenumbering[
-  state=start,
-  alternative=doublesided,
-  location=,
-]
-\setuplayout [grid=no]
-\setuptolerance [verytolerant]
-\setuptolerance [vertical,verytolerant]
-
-\newdimen \mypagenumdistance \mypagenumdistance .7em
-\def\fillercmd{\nobreak\hskip\mypagenumdistance}
-
-\definelistplacement[my_list_placement][command]#1#2#3%
-  {#2\fillercmd#3\hskip1.5em}
-
-\setuplist [chapter] [alternative=a,style=\word\sc]
-\setuplist [section] [alternative=my_list_placement,]
-
-\startsetups toc:subsection
-    \noindentation\placelist[subsection]
-\stopsetups
-
-\startsetups toc:section
-  \blank
-  \setupnarrower [middle=9em,]
-  \startnarrower[middle]\tfx
-  \noindentation\placelist[section][after=\setups{toc:subsection}]
-  \stopnarrower
-  \blank
-\stopsetups
-
-\title{Contents}
-
-\placelist [chapter] [criterium=text,after=\setups{toc:section}]
-
-% \setupheadertexts
-%   [{\tfx\sc\getmarking[chapter]}]   [{\tfx\bf \pagenumber}]
-%   [{\tfx\bf \pagenumber}]           [{\tfx{\em rst}{\kern.5pt\CONTEXT}}]
-% \completecontent
-\stopbuffer
-
-% something radically changed in mkiv
-\startluacode
-jobvariables           = job.variables or {}
-jobvariables.tobesaved = jobvariables.tobesaved or {}
-\stopluacode
-
-% list of publications
-\startbuffer[pubs]
-\setuplayout[grid=no]
-\setuptolerance[verytolerant]
-\setuptolerance[vertical,verytolerant]
-
-\completepublications
-\stopbuffer
-
-\appendtoks
-  \startfrontmatter
-    \getbuffer[frontpage]
-    \getbuffer[author]
-    \getbuffer[toc]
-  \stopfrontmatter
-\to \everystarttext
-
-%\prependtoks \getbuffer[pubs]      \to \everystoptext
-
-\setupwhitespace[none]
-\setuplayout[grid=verystrict]
-\setuptolerance[tolerant]
-\language[en]
-\setuppagenumbering[state=stop,location=]
-\input doc.tex
diff --git a/mod/doc/context/third/rst/moduletest.tex b/mod/doc/context/third/rst/moduletest.tex
deleted file mode 100644
index 76994e7..0000000
--- a/mod/doc/context/third/rst/moduletest.tex
+++ /dev/null
@@ -1,9 +0,0 @@
-%% Usage example
-% 1. load the module
-\usemodule[rst]
-% 2. add your setups
-\setuphead[chapter][page=no]
-\setupindenting[yes,medium,next]
-% 3. run the converter
-\typesetRSTfile{README.rst}
-
diff --git a/mod/scripts/context/lua/third/rst/mtx-t-rst.lua b/mod/scripts/context/lua/third/rst/mtx-t-rst.lua
deleted file mode 100644
index 6735b1d..0000000
--- a/mod/scripts/context/lua/third/rst/mtx-t-rst.lua
+++ /dev/null
@@ -1,64 +0,0 @@
-#!/usr/bin/env texlua
---------------------------------------------------------------------------------
---         FILE:  mtx-rst.lua
---        USAGE:  mtxrun --script rst --if=input.rst --of=output.tex 
---  DESCRIPTION:  context script interface for the reStructuredText module
--- REQUIREMENTS:  latest ConTeXt MkIV
---       AUTHOR:  Philipp Gesang (Phg), <megas.kapaneus@gmail.com>
---      CHANGED:  2013-03-27 00:25:32+0100
---------------------------------------------------------------------------------
---
-
-scripts     = scripts or { }
-scripts.rst = { }
-
-environment.loadluafile("rst_parser")
-
-local ea = environment.argument
-
-local helpinfo = [[
-===============================================================
-    The reStructuredText module, command line interface.
-    © 2010--2013 Philipp Gesang. License: 2-clause BSD.
-    Home: <https://bitbucket.org/phg/context-rst/>
-===============================================================
-
-USAGE:
-
-    mtxrun --script rst --if=input.rst --of=output.tex
-
-Mandatory arguments:
-
-    “infile.rst” is your input file containing reST markup.
-    “outfile.tex” is the target file that the TeX-code will be
-                  written to.
-
-Optional arguments:
-    --et=bool   “expandtab”, should tab chars (“\t”, “\v”) be
-                converted to spaces?
-    --sw=int    “shiftwidth”, tab stop modulo factor.
-
-===============================================================
-]]
-
-local application = logs.application {
-    name     = "mtx-rst",
-    banner   = "The reStructuredText module for ConTeXt, hg-rev 125+",
-    helpinfo = helpinfo,
-}
-
-scripts.rst.input  = ea("if")
-scripts.rst.output = ea("of")
-
-if scripts.rst.input and scripts.rst.output then
-    local expandtab  = ea("et") == "true" and true
-    local shiftwidth = ea("sw")
-    local debug      = ea("debug") == "true"
-    if expandtab  then thirddata.rst.expandtab  = true end
-    if shiftwdith then thirddata.rst.shiftwidth = tonumber(shiftwidth) end
-    if debug      then thirddata.rst_helpers.rst_debug = debug end
-    thirddata.rst.standalone(scripts.rst.input, scripts.rst.output)
-else
-    application.help()
-end
-
diff --git a/mod/tex/context/interface/third/t-rst.xml b/mod/tex/context/interface/third/t-rst.xml
deleted file mode 100644
index 68dcc1b..0000000
--- a/mod/tex/context/interface/third/t-rst.xml
+++ /dev/null
@@ -1,83 +0,0 @@
-<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
-
-<cd:interface xmlns:cd="http://www.pragma-ade.com/commands" 
-              name="context"
-              language="en" 
-              version="0.6; 2013-03-27 00:26:22+0100">
-
-  <cd:command name="typesetRSTfile" generated="no" file="t-rst.mkiv">
-    <cd:sequence>
-      <cd:variable value="typesetRSTfile"/>
-    </cd:sequence>
-    <cd:arguments>
-      <cd:assignments n="1" list="yes" optional="yes">
-        <cd:parameter name="expandtab">
-          <cd:constant type="false" default="false"/>
-          <cd:constant type="true"/>
-        </cd:parameter>
-        <cd:parameter name="shiftwidth">
-          <cd:constant type="cd:number" default="4"/>
-        </cd:parameter>
-        <cd:parameter name="stripBOM">
-          <cd:constant type="false" default="false"/>
-          <cd:constant type="true"/>
-        </cd:parameter>
-      </cd:assignments>
-      <cd:content n="2" optional="no"/>
-    </cd:arguments>
-  </cd:command>
-
-  <cd:command name="RST" generated="no" file="t-rst.mkiv">
-    <cd:sequence>
-      <cd:variable value="RST"/>
-    </cd:sequence>
-    <cd:arguments>
-      <cd:content n="1" optional="no"/>
-    </cd:arguments>
-  </cd:command>
-
-  <cd:command name="RST" generated="no" type="environment" file="t-rst.mkiv">
-    <cd:sequence>
-      <cd:variable value="RST"/>
-    </cd:sequence>
-  </cd:command>
-
-  <cd:command name="RSTproject" generated="no" type="environment" file="t-rst.mkiv">
-    <cd:sequence>
-      <cd:variable value="RST"/>
-    </cd:sequence>
-  </cd:command>
-
-  <cd:command name="defineRSTinclusion" generated="no" file="t-rst.mkiv">
-    <cd:sequence>
-      <cd:variable value="defineRSTinclusion"/>
-    </cd:sequence>
-    <cd:arguments>
-      <cd:content n="1" optional="yes" name="id"/>
-      <cd:content n="2" optional="yes" name="filename"/>
-      <cd:assignments n="3" list="yes" optional="yes">
-        <cd:parameter name="expandtab">
-          <cd:constant type="false" default="false"/>
-          <cd:constant type="true"/>
-        </cd:parameter>
-        <cd:parameter name="shiftwidth">
-          <cd:constant type="cd:number" default="4"/>
-        </cd:parameter>
-        <cd:parameter name="stripBOM">
-          <cd:constant type="false" default="false"/>
-          <cd:constant type="true"/>
-        </cd:parameter>
-      </cd:assignments>
-    </cd:arguments>
-  </cd:command>
-
-  <cd:command name="RSTinclusion" generated="no" file="t-rst.mkiv">
-    <cd:sequence>
-      <cd:variable value="RSTinclusion"/>
-    </cd:sequence>
-    <cd:arguments>
-      <cd:content n="1" optional="yes" name="id"/>
-    </cd:arguments>
-  </cd:command>
-
-</cd:interface>
diff --git a/mod/tex/context/third/rst/rst_context.lua b/mod/tex/context/third/rst/rst_context.lua
deleted file mode 100644
index c7e21fe..0000000
--- a/mod/tex/context/third/rst/rst_context.lua
+++ /dev/null
@@ -1,1316 +0,0 @@
-#!/usr/bin/env texlua
---------------------------------------------------------------------------------
---         FILE:  rst_context.lua
---        USAGE:  called by rst_parser.lua
---  DESCRIPTION:  Complement to the reStructuredText parser
---       AUTHOR:  Philipp Gesang (Phg), <phg42.2a@gmail.com>
---      CHANGED:  2013-03-26 22:46:17+0100
---------------------------------------------------------------------------------
---
---- TODO
----   Find an appropriate way to handle generic tables irrespective of the grid
----   settings. The problem is:
----   http://archive.contextgarden.net/message/20100912.112605.8a1aaf13.en.html
----   Seems we'll have to choose either the grid or split tables as default. Not
----   good.
-
-
-local helpers        = helpers        or thirddata and thirddata.rst_helpers
-local rst_directives = rst_directives or thirddata and thirddata.rst_directives
-
-local utf         = unicode.utf8
-local utflen      = utf.len
-local utflower    = utf.lower
-local utfupper    = utf.upper
-local iowrite     = io.write
-local tableconcat = table.concat
-
-local stringmatch  = string.match
-local stringgmatch = string.gmatch
-local stringgsub   = string.gsub
-
-local dbg_write = helpers.dbg_writef
-
-local C,  Cb, Cc, Cg, Cmt, Cp,
-      Cs, Ct, P,  R,  S,   V,  lpegmatch
-      = lpeg.C,  lpeg.Cb, lpeg.Cc, lpeg.Cg, lpeg.Cmt, lpeg.Cp,
-        lpeg.Cs, lpeg.Ct, lpeg.P,  lpeg.R,  lpeg.S,   lpeg.V,  lpeg.match
-
--- This one should ignore escaped spaces.
-do
-    local stripper = P{
-        [1] = "stripper",
-        stripper = V"space"^0 * C((V"space"^0 * (V"escaped" + V"nospace")^1)^0),
-        space    = S(" \t\v\n"),
-        nospace  = 1 - V"space",
-        escaped  = P"\\" * V"space"
-    }
-    function string.strip(str)
-        return lpegmatch(stripper, str) or ""
-    end
-end
-local stringstrip  = string.strip
-local stringformat = string.format
-
-local err = function(str)
-    if str then
-        iowrite("\n*[rstctx] Error: " .. str .. "\n\n")
-    end
-end
-
-local rst_context = thirddata.rst
-
-rst_context.collected_adornments = {}
-rst_context.last_section_level   = 0
-rst_context.anonymous_targets    = 0
-rst_context.anonymous_links      = {}
-
-rst_context.collected_references = {}
-rst_context.context_references   = {}
-rst_context.structure_references = {}
-rst_context.anonymous_set        = {}
-
-rst_context.substitutions        = {}
-rst_context.lastitemnumber       = 0  -- enumerations in RST allow arbitrary skips
-
-rst_context.current_footnote_number   = 0
-rst_context.current_symbolnote_number = 0
-
-function rst_context.addsetups(item)
-    local state = rst_context.state
-    state.addme[item] = state.addme[item] or true
-    return 0
-end
-
-function rst_context.footnote_reference (label)
-    local tf = rst_context.state.footnotes
-    if stringmatch(label, "^%d+$") then -- all digits
-        local c = tonumber(label)
-        return [[\\footnote{\\getbuffer[__footnote_number_]].. c .."]}"
-    elseif label == "#" then --autonumber
-        local rc = rst_context.current_footnote_number
-        rc = rc + 1
-        rst_context.current_footnote_number = rc
-        return [[\\footnote{\\getbuffer[__footnote_number_]].. rc .."]}"
-    elseif stringmatch(label, "^#.+$") then
-        local thelabel = stringmatch(label, "^#(.+)$")
-        return [[\\footnote{\\getbuffer[__footnote_label_]].. thelabel .."]}"
-    elseif label == "*" then
-        local rc = rst_context.current_symbolnote_number
-        rc = rc + 1
-        rst_context.current_symbolnote_number = rc
-        return [[\\symbolnote{\\getbuffer[__footnote_symbol_]].. rc .."]}"
-    else -- “citation reference” for now treating them like footnotes
-        rst_context.addsetups("citations")
-        return [[\\cite{]] .. label .. [[}]]
-    end
-end
-
-do
-    local w = S" \v\t\n" / "_"
-    local wp = Cs((w + 1)^1)
-    function rst_context.whitespace_to_underscore(str)
-        return  str and lpegmatch(wp, str) or ""
-    end
-end
-
---- So we can use crefs[n][2] to refer to the place where the reference was
---- created.
-local function get_context_reference (str)
-    local crefs = rst_context.context_references
-    local srefs = rst_context.structure_references
-    srefs[str] = true
-    refstring = "__target_" .. rst_context.whitespace_to_underscore(str)
-    crefs[#crefs + 1] = { refstring, str }
-    return refstring
-end
-
-function rst_context.emphasis (str)
-    return [[{\\em ]] .. str .. [[}]]
-end
-
-function rst_context.strong_emphasis (str)
-    return [[{\\sc ]] .. str .. [[}]]
-end
-
-function rst_context.literal (str)
-    return [[\\type{]] .. str .. [[}]]
-end
-
---- ROLES for interpreted text
-
-rst_context.roles = {}
-rst_context.roles.emphasis = rst_context.emphasis
-rst_context.roles.strong_emphasis = rst_context.strong_emphasis
-rst_context.roles.literal = rst_context.literal
-rst_context.roles.bold = function(str)
-    return [[{\\bold ]] .. str .. [[}]]
-end
-rst_context.roles.bf = rst_context.roles.bold
-
-rst_context.roles.italic = function(str)
-    return [[{\\italic ]] .. str .. [[}]]
-end
-rst_context.roles.it = rst_context.roles.italic
-
-rst_context.roles.sans = function(str)
-    return [[{\\ss ]] .. str .. [[}]]
-end
-rst_context.roles.sans_serif = rst_context.roles.sans
-rst_context.roles.ss         = rst_context.roles.sans
-
-rst_context.roles.uppercase = function(str)
-    return utfupper(str)
-end
-
-rst_context.roles.lowercase = function(str)
-    return utflower(str)
-end
-
-rst_context.roles.color = function(color, str)
-    local p = helpers.patterns
-    local definition = stringmatch(color, "^color_(.+)$")
-    if stringmatch(definition, "^rgb_") then -- assume rgb
-        local rgb = lpegmatch(p.rgbvalues, definition)
-        definition = stringformat([[r=%s,g=%s,b=%s]], rgb[1], rgb[2], rgb[3])
-    end
-    return stringformat([[\\colored[%s]{%s}]], definition, str)
-end
-
---------------------------------------------------------------------------------
---- Inofficial text roles for my private bib
---------------------------------------------------------------------------------
-
--- Afterthought:
--- Different citation commands are essentially typographical instructions:
--- they are concerned with the final representation of the data with respect to
--- a concrete implementation. Not the thing at all that would make reST
--- portable. But then its support for Python-style string escaping &c. ain’t at
--- all portable either. The problem is the same with XML written to be
--- processed with ConTeXt -- when processing the text directly in MkIV you’ll
--- always find yourself adding setups that allow fine-grained control of the
--- typeset output. At the same time those instructions directly contradict the
--- main reason for XML: to provide an application-independent data markup.
--- Typesetting XML (and now reST) with TeX, you will always end up writing TeX
--- code disguised in XML brackets. (Btw. the docutils reST specification has
--- the same kind of inclination to HTML -- some of its components don’t even
--- have a meaning save in HTML peculiarities.) If you strive to avoid this
--- *and* would like to have decent typesetting, you should use the
--- automatically generated TeX code as a starting point for the actual
--- typesetting job. Wish it was possible to have both -- the data in a
--- universal form and the output in the Optimal Typesetting System -- but
--- that’s a dream for now. If you really read these musings, then prove me
--- wrong if you can! Or go tell those digital publishers and their willing
--- subordinates, the authors, who think they can save a few pennys,
--- substituting the typesetter and editor by some fancy software. Keep in mind
--- that zapf.tex is not just random dummy text. </rant>
-
-function rst_context.roles.ctsh(str) -- shorthand
-    rst_context.addsetups("citator")
-    return [[\\ctsh{]] .. str .. [[}]]
-end
-
-function rst_context.roles.ctas(str) -- short cite
-    rst_context.addsetups("citator")
-    return [[\\ctas{]] .. str .. [[}]]
-end
-
-function rst_context.roles.ctau(str) -- author only
-    rst_context.addsetups("citator")
-    return [[\\ctau{]] .. str .. [[}]]
-end
-
-function rst_context.roles.cttt(str) -- title only
-    rst_context.addsetups("citator")
-    return [[\\cttt{]] .. str .. [[}]]
-end
-
-function rst_context.roles.ctay(str) -- author year
-    rst_context.addsetups("citator")
-    return [[\\ctay{]] .. str .. [[}]]
-end
-
-function rst_context.roles.ctfu(str) -- full cite
-    rst_context.addsetups("citator")
-    return [[\\ctfu{]] .. str .. [[}]]
-end
-
-function rst_context.roles.nocite(str) -- nocite
-    rst_context.addsetups("citator")
-    return [[\\nocite[]] .. str .. [=[]]=]
-end
-
---------------------------------------------------------------------------------
---- End citator roles
---------------------------------------------------------------------------------
-
---------------------------------------------------------------------------------
---- Experimental roles.
---------------------------------------------------------------------------------
-
---- Feature request by Philipp A.
-function rst_context.roles.math(str)
-    return [[\\mathematics{]] .. str .. [[}]]
-end
-
---------------------------------------------------------------------------------
---- End roles
---------------------------------------------------------------------------------
-
-function rst_context.interpreted_text (...)
-    local tab = { ... }
-    local role, str
-    role = stringmatch(tab[1], "^:(.*):$") or stringmatch(tab[3], "^:(.*):$")
-    str  = tab[2]
-
-    if not role then -- implicit role
-        role = "emphasis"
-    end
-
-    if stringmatch(role, "^color_") then
-        return rst_context.roles.color(role, str)
-    end
-
-    return rst_context.roles[role](str)
-end
-
-function rst_context.link_standalone (str)
-    return "\n"
-        .. [[\\goto{\\hyphenatedurl{]] .. str .. [[}}[url(]] .. str .. [=[)]]=]
-end
-
-function rst_context.reference (str)
-    rst_context.addsetups("references")
-    str = stringmatch(str, "^`?([^`]+)`?_$")
-    return [[\\RSTchoosegoto{__target_]] .. rst_context.whitespace_to_underscore(str) .. "}{"
-            .. str .. "}"
-end
-
-function rst_context.anon_reference (str)
-    rst_context.addsetups("references")
-    str = stringmatch(str, "^`?([^`]+)`?__$")
-    rst_context.anonymous_links[#rst_context.anonymous_links+1] = str
-    link = "__target_anon_" .. #rst_context.anonymous_links
-    return stringformat([[\\RSTchoosegoto{%s}{%s}]], link, str)
-end
-
-local whitespace = S" \n\t\v"
-local nowhitespace = 1 - whitespace
-local removewhitespace = Cs((nowhitespace^1 + Cs(whitespace / ""))^0)
-
-function rst_context.target (tab)
-    rst_context.addsetups("references")
-    --local tab = { ... }
-    local  refs = rst_context.collected_references
-    local arefs = rst_context.anonymous_set
-    local target = tab[#tab] -- Ct + C could be clearer but who cares
-    tab[#tab] = nil
-
-    local function create_anonymous ()
-        rst_context.anonymous_targets = rst_context.anonymous_targets + 1
-        return { "anon_" .. rst_context.anonymous_targets, rst_context.anonymous_targets }
-    end
-
-    local insert = ""
-
-    if target == "" then -- links here
-        for _, id in next, tab do
-            insert = insert .. "\n\\reference[__target_" .. id .. "]{}"
-        end
-    else
-        for i=1,#tab do
-            local id = tab[i]
-            if id == "" then -- anonymous
-                local anon = create_anonymous()
-                id, arefs[anon[1]] = anon[1], anon[2]
-            else
-                local tmp = tab[i]
-                tmp = stringgsub(tmp, "\\:",":")
-                tmp = stringmatch(tmp, "`?([^`]+)`?")
-                id = tmp
-                --id = tab[i]:gsub("\\:",":"):match("`?([^`]+)`?") -- deescaping
-            end
-            if id then
-                refs[id] = refs[id] or target
-            end
-        end
-    end
-
-    return insert
-end
-
-function rst_context.inline_internal_target (str)
-    return "\\\\reference[__target_" .. rst_context.whitespace_to_underscore(str) .."]{}"
-end
-
-function rst_context.substitution_reference (str, underscores)
-    local sub = ""
-    rst_context.addsetups "substitutions"
-    if underscores == "_" then -- normal reference
-        sub = sub .. [[\\reference[__target_]] .. rst_context.whitespace_to_underscore(stringstrip(str)) .. "]{}"
-    elseif underscores == "__" then -- normal reference
-        rst_context.anonymous_targets = rst_context.anonymous_targets + 1
-        sub = sub .. [[\\reference[__target_anon_]] .. rst_context.anonymous_targets .. "]{}"
-    end
-    return sub .. [[{\\RSTsubstitution]] .. stringgsub(str, "%s", "") .. "}"
-end
-
-do
-    -- see catc-sym.tex
-    local escape_me = {
-        ["&"]   = [[\letterampersand ]],
-        ["$"]   = [[\letterdollar ]],
-        ["#"]   = [[\letterhash ]],
-        ["^"]   = [[\letterhat ]],
-        ["_"]   = [[\letterunderscore ]],
-    }
-
-    local chars
-    for chr, repl in next, escape_me do
-        chars = chars and chars + (P(chr) / repl) or P(chr) / repl
-    end
-
-    local p_escape = P{
-        [1]      = Cs((V"skip"
-                 --+ V"literal" -- achieved via gsub later
-                 + chars
-                 + 1)^1),
-        skip1    = P"\\starttyping" * (1 - P"\\stoptyping")^1,
-        balanced = P"{" * (V"balanced" + (1 - P"}"))^0 * P"}",
-        skip2    = P"\\type" * V"balanced",
-        skip3    = P"\\mathematics" * V"balanced",
-        skip     = V"skip1" + V"skip2" + V"skip3",
-        --literal  = Cs(P"\\" / "") * 1
-    }
-
-    function rst_context.escape (str)
-        str = stringgsub(str, "\\(.)", "%1")
-        return lpegmatch(p_escape, str)
-    end
-end
-
-function rst_context.joinindented (tab)
-    return tableconcat (tab, "")
-end
-
-local corresponding = {
-    ['"'] = '"',
-    ["'"] = "'",
-    ["{"] = "}",
-    ["("] = ")",
-    ["["] = "]",
-    ["<"] = ">",
-}
-
-local inline_parser = P{
-    [1] = "block",
-
-    block = Cs(V"inline_as_first"^-1 * (V"except" + V"inline_element" + V"normal_char")^0),
-
-    inline_element = V"precede_inline"
-                   * Cs(V"inline_do_elements")
-                   * #V"succede_inline"
-                   + V"footnote_reference"
-                   ,
-
-    -- Ugly but needed in case the first element of a paragraph is inline
-    -- formatted.
-    inline_as_first = V"inline_do_elements" * #V"succede_inline",
-
-    except = P"\\starttyping" * (1 - P"\\stoptyping")^1 * P"\\stoptyping"
-           + V"enclosed"
-           ,
-
-    inline_do_elements = V"strong_emphasis"
-                       + V"substitution_reference"
-                       + V"anon_reference"
-                       + V"inline_literal"
-                       + V"reference"
-                       + V"emphasis"
-                       + V"interpreted_text"
-                       + V"inline_internal_target"
-                       + V"link_standalone"
-                       ,
-
-    precede_inline = V"spacing"
-                   + V"eol"
-                   + -P(1)
-                   + S[['"([{<-/:]]
-                   + P"‘" + P"“" + P"’" + P"«" + P"¡" + P"¿"
-                   + V"inline_delimiter"
-                   + P"„", -- not in standard Murkin reST
-
-    succede_inline = V"spacing"
-                   + V"eol"
-                   + S[['")]}>-/:.,;!?\]]
-                   + P"’" + P"”" + P"»"
-                   + V"inline_delimiter"
-                   + -P(1)
-                   + P"“" -- non-standard again but who cares
-                   ,
-
-    enclosed = V"precede_inline"^-1
-             * Cg(V"quote_single" + V"quote_double" + V"leftpar", "lastgroup")
-             * V"inline_delimiter"
-             * Cmt(C(V"quote_single" + V"quote_double" + V"rightpar") * Cb("lastgroup"), function(s, i, char, oldchar)
-                    return corresponding[oldchar] == char
-                end)
-             * V"succede_inline"^-1
-             * -V"underscore"
-             ,
-
-    space = P" ",
-    whitespace = (P" " + Cs(P"\t") / "        " + Cs(S"\v") / " "),
-    spacing = V"whitespace"^1,
-
-    eol = P"\n",
-    --inline_delimiters = P"‐" + P"‑" + P"‒" + P"–" + V"emdash" + V"space",        -- inline markup
-    inline_delimiter = P"‐" + P"‑" + P"‒" + P"–" + V"emdash" + V"space"
-                     + V"bareia"
-                     + V"asterisk"
-                     + V"bar"
-                     + V"lbrack" + V"rbrack"
-                     ,        -- inline markup
-    asterisk = P"*",
-    quote_single = P"'",
-    quote_double = P'"',
-    double_asterisk = V"asterisk" * V"asterisk",
-    bareia = P"`",
-    backslash = P"\\",
-    bar = P"|",
-    double_bareia = V"bareia" * V"bareia",
-    escaped_bareia = (Cs(V"backslash") / "" * V"bareia") + 1,
-    colon = P":",
-    escaped_colon = (Cs(V"backslash") / "" * V"colon") + 1,
-    semicolon = P";",
-    underscore = P"_",
-    double_underscore = V"underscore" * V"underscore",
-    dot = P".",
-    interpunct = P"·",
-    comma = P",",
-    dash = P"-",
-    emdash = P"—",
-    ellipsis   = P"…" + P"...",
-    exclamationmark = P"!",
-    questionmark = P"?",
-    interrobang = P"‽",
-    double_dash = V"dash" * V"dash",
-    triple_dash = V"double_dash" * V"dash",
-    hyphen = P"‐",
-    dashes = V"dash" + P"‒" + P"–" + V"emdash" + P"―",
-
-    lparenthesis = P"(",
-    rparenthesis = P")",
-    lbrack  = P"[",
-    rbrack  = P"]",
-    lbrace  = P"{" / [[{\\letterleftbrace}]],
-    rbrace  = P"}" / [[{\\letterrightbrace}]],
-    less    = P"<",
-    greater = P">",
-    leftpar  = V"lparenthesis" + V"lbrack" + V"lbrace" + V"less",
-    rightpar = V"rparenthesis" + V"rbrack" + V"rbrace" + V"greater",
-
-    normal_char = V"lbrace" + V"rbrace" + V"lbrack" + V"rbrack" -- escape those if in input
-                + 1
-                ,
-
-    --groupchars = S"()[]{}",
-    groupchars = V"leftpar" + V"rightpar",
-    apostrophe = P"’" + P"'",
-
-    guillemets = P"«" + P"»",
-    quotationmarks= P"‘" + P"’" + P"“" + P"”",
-    solidus= P"⁄",
-    slash = P"/",
-
-    gartenzaun = P"#",
-    digit  = R"09",
-    letter = R"az" + R"AZ",
-
-    punctuation = V"apostrophe"
-                + V"colon"
-                + V"comma"
-                + V"dashes"
-                + V"dot"
-                + V"ellipsis"
-                + V"exclamationmark"
-                + V"guillemets"
-                + V"hyphen"
-                + V"interpunct"
-                + V"interrobang"
-                + V"questionmark"
-                + V"quotationmarks"
-                + V"semicolon"
-                + V"slash"
-                + V"solidus"
-                + V"underscore"
-                ,
-
-    emphasis        = (V"asterisk" - V"double_asterisk")
-                    * Cs((V"normal_char" - V"spacing" - V"eol" - V"asterisk")
-                       * ((V"normal_char" - (V"normal_char" * V"asterisk"))^0
-                        * (V"normal_char" - V"spacing" - V"eol" - V"asterisk"))^-1)
-                    * V"asterisk"
-                    / rst_context.emphasis,
-
-    strong_emphasis = V"double_asterisk"
-                    * Cs((V"normal_char" - V"spacing" - V"eol" - V"asterisk")
-                       * ((V"normal_char" - (V"normal_char" * V"double_asterisk"))^0
-                        * (V"normal_char" - V"spacing" - V"eol" - V"asterisk"))^-1)
-                    * V"double_asterisk"
-                    / rst_context.strong_emphasis,
-
-    inline_literal  = V"double_bareia"
-                    * C ((V"escaped_bareia" - V"spacing" - V"eol" - V"bareia")
-                       * ((V"escaped_bareia" - (V"normal_char" * V"double_bareia"))^0
-                        * (V"escaped_bareia" - V"spacing" - V"eol" - V"bareia"))^-1)
-                    * V"double_bareia"
-                    / rst_context.literal,
-
-    interpreted_single_char = (V"normal_char" - V"spacing" - V"eol" - V"bareia") * #V"bareia",
-    interpreted_multi_char  = (V"normal_char" - V"spacing" - V"eol" - V"bareia") * (V"normal_char" - (1 * V"bareia"))^0 * (1 - V"spacing" - V"eol" - V"bareia"),
-
-    interpreted_text = C(V"role_marker"^-1)
-                     * (V"bareia" - V"double_bareia")
-                     * C(V"interpreted_single_char" + V"interpreted_multi_char")
-                     * V"bareia"
-                     * C(V"role_marker"^-1)
-                     / rst_context.interpreted_text,
-
-    role_marker = V"colon" * (V"backslash" * V"colon" + V"letter" + V"digit" + V"dash" + V"underscore" + V"dot")^1 * V"colon",
-
-    link_standalone = C(V"uri")
-                    / rst_context.link_standalone,
-
-    anon_reference = Cs(V"anon_phrase_reference" + V"anon_normal_reference")
-              / rst_context.anon_reference,
-
-    anon_normal_reference = C((1 - V"underscore" - V"spacing" - V"eol" - V"punctuation" - V"groupchars")^1) * V"double_underscore",
-
-    anon_phrase_reference = (V"bareia" - V"double_bareia")
-                          * C((1 - V"bareia")^1)
-                          * V"bareia" * V"double_underscore"
-                          ,
-
-    reference = Cs(V"normal_reference" + V"phrase_reference")
-              / rst_context.reference,
-
-    normal_reference = (1 - V"underscore" - V"spacing" - V"eol" - V"punctuation" - V"groupchars")^1 * V"underscore",
-
-    phrase_reference = (V"bareia" - V"double_bareia")
-                     * C((1 - V"bareia")^1)
-                     * V"bareia" * V"underscore"
-                     ,
-
-    footnote_reference = V"lbrack"
-                       * Cs(V"footnote_label" + V"citation_reference_label")
-                       * V"rbrack"
-                       * V"underscore"
-                       / rst_context.footnote_reference
-                       ,
-
-    footnote_label = V"digit"^1
-                   + V"gartenzaun" * V"letter"^1
-                   + V"gartenzaun"
-                   + V"asterisk"
-                   ,
-
-    citation_reference_label = V"letter" * (1 - V"rbrack")^1,
-
-    inline_internal_target = V"underscore"
-                           * V"bareia"
-                           * Cs((1 - V"bareia")^1)
-                           * V"bareia"
-                           / rst_context.inline_internal_target
-                           ,
-
-    substitution_reference = V"bar"
-                           * C((1 - V"bar")^1)
-                           * V"bar"
-                           * C((V"double_underscore" + V"underscore")^-1)
-                           / rst_context.substitution_reference
-                           ,
-
---------------------------------------------------------------------------------
--- Urls
---------------------------------------------------------------------------------
-    uri = V"url_protocol" * V"url_domain" * V"url_path_char"^0,
-
-    url_protocol = (P"http" + P"ftp" + P"shttp" + P"sftp") * P"://",
-    url_domain_char = 1 - V"dot" - V"spacing" - V"eol" - V"punctuation",
-    url_domain = V"url_domain_char"^1 * (V"dot" * V"url_domain_char"^1)^0,
-    url_path_char = R("az", "AZ", "09") + S[[-_.!~*'()/]],
-}
-
-rst_context.inline_parser = inline_parser
-
-function rst_context.paragraph (data)
-    local str
-    if not data then
-        return ""
-    elseif type(data) == "table" then
---        str = #data > 1 and  helpers.string.wrapat(lpegmatch(inline_parser, tableconcat(data, " ")), 65) 
---                        or   inline_parser:match(data[1])
-        if #data > 1 then
-            str = helpers.string.wrapat(
-                lpegmatch(inline_parser, tableconcat(data, " "))
-                , 65)
-        else
-            str = lpegmatch(inline_parser, data[1])
-        end
-    else
-        str = data
-    end
-    return stringformat([[
-
-\\startparagraph
-%s
-\\stopparagraph
-]], str)
-end
-
-local sectionlevels = {
-    [1] = "chapter",
-    [2] = "section",
-    [3] = "subsection",
-    [4] = "subsubsection",
-    [5] = "subsubsubsection",
-}
-
-local function get_line_pattern (chr)
-    return P(chr)^1 * (-P(1))
-end
-
-function rst_context.section (...)  -- TODO general cleanup; move validity
-    local tab = { ... }             -- checking to parser.
-    local section, str = true, ""
-    local adornchar 
-    local ulen = utflen
-    if #tab == 3 then -- TODO use unicode length with ConTeXt
-        adornchar = tab[1]:sub(1,1)
-        section = ulen(tab[1]) >= ulen(tab[2])
-        str = stringstrip(tab[2])
-    else -- no overline
-        adornchar = tab[2]:sub(1,1)
-        section = ulen(tab[1]) <= ulen(tab[2])
-        str = tab[1]
-    end
-
-    if section then -- determine level
-        local level = rst_context.last_section_level
-        local rca = rst_context.collected_adornments
-        if rca[adornchar] then
-            level = rca[adornchar]
-        else
-            level = level + 1
-            rca[adornchar] = level
-            rst_context.last_section_level = level
-        end
-
-        ref = get_context_reference (str)
-
-        str = stringformat("\n\\\\%s[%s]{%s}\n", sectionlevels[level], ref, str)
-    else
-        return [[{\\bf fix your sectioning!}\\endgraf}]]
-    end
-
-    return section and str or ""
-end
-
--- Prime time for the fancybreak module.
-function rst_context.transition (str)
-    rst_context.addsetups("breaks")
-    --return "\\fancybreak\n"
-    return "\\fancybreak{$* * *$}\n"
-end
-
-function rst_context.bullet_marker(str)
-    return "marker"
-end
-
--- This one should ignore escaped spaces.
-do
-    local stripper = P{
-        [1] = "stripper",
-        stripper = V"space"^0 * C((V"space"^0 * V"nospace"^1)^0),
-        space    = S(" \t\v\n"),
-        escaped  = P"\\" * V"space",
-        nospace  = V"escaped" + (1 - V"space"),
-    }
-    function stringstrip(str)
-        return lpegmatch(stripper, str) or ""
-    end 
-end
-
-local enumeration_types = {
-    ["*"] = "*", -- unordered bulleted
-    ["+"] = "*",
-    ["-"] = "*",
-    ["•"] = "*",
-    ["‣"] = "*",
-    ["⁃"] = "*",
-
-    ["#"] = "n", -- numbered lists and conversion
-    ["A"] = "A",
-    ["a"] = "a",
-    ["I"] = "R",
-    ["i"] = "r",
-}
-
--- \setupitemize[left=(, right=), margin=4em, stopper=]
-
-local stripme   = S"()."
-local dontstrip = 1 - stripme
-local itemstripper = stripme^0 * C(dontstrip^1) * stripme^0
-
-local function parse_itemstring(str)
-    local offset = nil
-    local setup = ",fit][itemalign=flushright,"
-    if stringmatch(str, "^%(") then
-        setup = setup .. [[left=(,]]
-    end
-    if stringmatch(str, "%)$") then
-        setup = setup .. [[right=)]]
-    end
-    if stringmatch(str, "%.$") then
-        setup = setup .. [[stopper={.\\space}]]
-    end
-    local num = stringmatch(str, "^%d")
-    if num then
-        -- http://thread.gmane.org/gmane.comp.tex.context/61728/focus=61729
-        setup = setup .. ",start=" .. num
-        str = "n"
-    end
-
-    str = lpegmatch(itemstripper, str)
-    str = enumeration_types[str] or str
-    return { setup = setup, str = str }
-end
-
-function rst_context.startitemize(str)
-    local setup = ""
-    local result = ""
-    str = stringstrip(str)
-
-    local listtype = enumeration_types[str] or parse_itemstring(str)
-
-    if type(listtype) == "table" then
-        setup = listtype.setup
-        listtype = listtype.str
-    end
-
-    result = [[
-\\startitemize[]] .. listtype .. setup .. [[]
-]] 
-    return result
-end
-
-local last_item = {} -- stack
-local current_itemdepth = 0
-function rst_context.stopitemize(str)
-    last_item[current_itemdepth] = nil
-    current_itemdepth = current_itemdepth - 1
-    return str .. [[
-\\stopitemize
-]]
-end
-
-function rst_context.bullet_item (tab)
-    local li = last_item
-    -- The capture of the first item has the \startitemize as 
-    -- *second* element in the array.
-    local content  = #tab == 2 and tab[2] or tab[3]
-    local startstr = #tab == 3 and tab[2] or nil
-    local itemtype = tab[1]
-    local result = startstr or ""
-    if startstr then
-        current_itemdepth = current_itemdepth + 1
-        li[current_itemdepth] = itemtype
-    elseif li[current_itemdepth] then
-        if helpers.list.successor(itemtype, li[current_itemdepth]) then
-            -- just leave it alone
-        elseif helpers.list.greater(itemtype, li[current_itemdepth]) then
-            local itemnum = tonumber(stringstrip(itemtype)) or helpers.list.get_decimal(itemtype)
-            result = result .. stringformat([[
-\\setnumber[itemgroup:itemize]{%s}
-]], itemnum)
-        end
-        li[current_itemdepth] = itemtype
-    end
-
-    return result .. [[
-
-\\item ]] .. lpegmatch(inline_parser, content) .. [[
-
-]]
-end
-
---------------------------------------------------------------------------------
--- Definition lists 
---------------------------------------------------------------------------------
--- TODO define proper setups (probably bnf-like and some narrower for def-paragraphs)
-
-function rst_context.deflist (list)
-    rst_context.addsetups("deflist")
-
-    local deflist = [[
-\\startRSTdefinitionlist
-]] 
-    for nd=1, #list do
-        local item = list[nd]
-        local term = item[1]
-        local nc = 2
-        local tmp = [[
-
-  \\RSTdeflistterm{]] .. stringstrip(term) .. "}"
-        if #item > 2 then
-            while nc < #item do
-                tmp = tmp .. [[
-
-  \\RSTdeflistclassifier{]] .. stringstrip(item[nc]) .. "}"
-                nc = nc + 1
-            end
-        end
-        tmp = tmp .. [[
-
-  \\RSTdeflistdefinition{%
-]]
-        local final = item[#item]
-        for np=1, #final do
-            local par = final[np]
-            tmp = tmp .. [[
-    \\RSTdeflistparagraph{%
-]] .. lpegmatch(inline_parser, par) .. "}\n"
-        end
-        tmp = tmp .. "  }"
-        deflist = deflist .. tmp
-    end
-    return deflist .. [[
-
-\\stopRSTdefinitionlist
-]]
-end
-
---------------------------------------------------------------------------------
--- Field lists
---------------------------------------------------------------------------------
-
--- TODO Do something useful with field lists. For now I'm not sure what as the
--- bibliography directives from the reST specification seem to make sense only
--- when using docinfo and, after all, we have .bib files that are portable.
-
-function rst_context.field_list (str)
-    rst_context.addsetups("fieldlist")
-    return [[
-
-\\startRSTfieldlist]] .. str .. [[\\eTABLEbody\\stopRSTfieldlist
-]]
-end
-
-function rst_context.field_name (str)
-    return [[\\fieldname{]] .. str .. [[}]]
-end
-
-function rst_context.field_body (str)
-    return [[\\fieldbody{]] .. lpegmatch(inline_parser, str) .. [[}]]
-end
-
-function rst_context.field (tab)
-    local name, body = tab[1], tab[2]
-    return stringformat([[
-
-    \\RSTfieldname{%s}
-    \\RSTfieldbody{%s}
-]], name, lpegmatch(inline_parser, body))
-end
-
-function rst_context.line_comment (str)
-    return "% " .. str
-end
-
-function rst_context.block_comment (str)
-    return stringformat([[
-
-\iffalse %% start block comment
-%s\fi %% stop block comment
-]], str)
-end
-
-function rst_context.option_list (str)
-    return [[
-\\setupTABLE[c][first] [background=color, backgroundcolor=grey, style=\tt]
-\\setupTABLE[c][each]  [frame=off]
-\\setupTABLE[r][each]  [frame=off]
-\\bTABLE[split=yes,option=stretch]
-\\bTABLEhead
-\\bTR
-  \\bTH  Option \\eTH
-  \\bTH  Description \\eTH
-\\eTR
-\\eTABLEhead
-\\bTABLEbody
-]] .. lpegmatch(inline_parser, str) .. [[
-
-\\eTABLEbody
-\\eTABLE
-]]
-end
-
-function rst_context.option_item (tab)
-    return stringformat([[\\bTR\\bTC %s \\eTC\\bTC %s \\eTC\\eTR
-]], tab[1], tab[2])
-end
-
-function rst_context.test(str)
-    return ":"
-end
-
-function rst_context.literal_block (str, included)
-    local indent = P" "^1
-    local stripme = #str
-    for line in stringgmatch(str, "[^\n]+") do
-        -- setting to the lowest indend of all lines
-        local idt = lpegmatch(indent, line)
-        if line and idt then
-            stripme = idt < stripme and idt or stripme
-        end
-    end
-
-    local strip = P{
-        [1] = "strip",
-        strip = Cs(V"line"^1),
-        eol = P"\n",
-        restofline = (1 - V"eol")^0,
-        stop = Cs(V"eol" * P" "^0) * -P(1) / "", -- remove trailing blank lines
-        line = Cs(V"restofline" * (V"stop" + V"eol")) / function (line)
-            return #line > stripme and line:sub(stripme) or line
-        end,
-    }
-
-    str = lpegmatch(strip, str)
-    str = [[
-
-\starttyping[lines=hyphenated]
-]] .. str .. [[
-
-\stoptyping
-]]
-    if included then -- escaping can ruin your day
-        str = str:gsub("\\", "\\\\")
-    end
-    return str
-end
-
-function rst_context.included_literal_block (str)
-    return rst_context.literal_block(str, true)
-end
-
-function rst_context.line_block (str)
-    rst_context.addsetups("lines")
-    return [[
-
-\\startlines
-]] .. lpegmatch(inline_parser, str) .. [[\\stoplines
-]]
-end
-
-function rst_context.line_block_line(str)
-    str = str:gsub("\n", " ")
-    return str .. "\n"
-end
-
-function rst_context.line_block_empty()
-    return "\n"
-end
-
-function rst_context.block_quote (tab)
-    rst_context.addsetups("blockquote")
-    local str = [[
-\\startlinecorrection
-\\blank[small]
-\\startblockquote
-]] .. lpegmatch(inline_parser, tab[1]) .. [[
-
-\\stopblockquote
-]]
-
-    return tab[2] and str .. [[
-\\blank[small]
-\\startattribution
-]] .. lpegmatch(inline_parser, tab[2]) .. [[
-\\stopattribution
-\\blank[small]
-\\stoplinecorrection
-]]  or str .. [[
-\\blank[small]
-\\stoplinecorrection
-]] 
-end
-
---function rst_context.table (str)
-    --return [[
---\\startlinecorrection
---]] .. str .. [[
-
---\\stoplinecorrection
---]]
---end
-
-function rst_context.grid_table (tab)
-    local body = ""
-    local nr = 1
-    local head
-    if tab.has_head then
-        head = [[
-\\setupTABLE[c][each]  [frame=off]
-\\setupTABLE[r][each]  [frame=off]
-%\\startlinecorrection
-\\bTABLE[split=repeat,option=stretch]
-\\bTABLEhead
-]]
-        while nr <= tab.head_end do
-            local r = tab.rows[nr]
-            local isempty = true
-            for n=1, #r do
-                local cell = r[n]
-                if cell.variant == "normal" then
-                    isempty = false
-                    break
-                end
-            end
-
-            if not isempty then
-                local row = [[\\bTR]]
-                for n=1, #r do
-                    local c = r[n]
-                    if not (c.parent or
-                            c.variant == "separator") then
-                        local celltext = lpegmatch(inline_parser, c.stripped)
-                        if c.span.x or c.span.y then
-                            local span_exp = "["
-                            if c.span.x then
-                                span_exp = span_exp .. "nc=" .. c.span.x .. ","
-                            end
-                            if c.span.y then
-                                span_exp = span_exp .. "nr=" .. c.span.y
-                            end
-                            celltext  = span_exp .. "] " .. celltext
-
-                        end
-
-                        row = row .. "\n  " .. [[\\bTH ]] .. celltext .. [[\\eTH]]
-                    end
-                end
-                head = head .. row .. "\n" .. [[\\eTR]] .. "\n"
-            end
-            nr = nr + 1
-        end
-        head = head .. [[
-\\eTABLEhead
-\\bTABLEbody
-]] 
-    else
-        head = [[
-\\setupTABLE[c][each]  [frame=off]
-\\setupTABLE[r][each]  [frame=off]
-%\\startlinecorrection
-\\bTABLE[split=repeat,option=stretch]
-\\bTABLEbody
-]] 
-    end
-    while nr <= #tab.rows do
-        local r = tab.rows[nr]
-        local isempty = true
-        for n=1, #r do
-            local cell = r[n]
-            if cell.variant == "normal" then
-                isempty = false
-                break
-            end
-        end
-
-        if not isempty then
-            local row = [[\\bTR]]
-            for n=1, #r do
-                local c = r[n]
-                if not (c.parent or
-                        c.variant == "separator") then
-                    local celltext = lpegmatch(inline_parser, c.stripped)
-                    if c.span.x or c.span.y then
-                        local span_exp = "["
-                        if c.span.x then
-                            span_exp = span_exp .. "nc=" .. c.span.x .. ","
-                        end
-                        if c.span.y then
-                            span_exp = span_exp .. "nr=" .. c.span.y
-                        end
-                        celltext  = span_exp .. "] " .. celltext
-
-                    end
-
-                    row = row .. "\n  " .. [[\\bTC ]] .. celltext .. [[\\eTC]]
-                end
-            end
-            body = body .. row .. "\n" .. [[\\eTR]] .. "\n"
-        end
-        nr = nr + 1
-    end
-    local tail = [[
-\\eTABLEbody
-\\eTABLE
-%\\stoplinecorrection
-]]
-    return head .. body .. tail
-end
-
-
-function rst_context.simple_table(tab)
-    local head
-    local nr = 1
-    if tab.head_end then
-        head = [[
-\\setupTABLE[c][each]  [frame=off]
-\\setupTABLE[r][each]  [frame=off]
-%\\startlinecorrection
-\\bTABLE[split=yes,option=stretch]
-\\bTABLEhead
-]]
-        while nr <= tab.head_end do
-            local row = tab[nr]
-            if not row.ignore then
-                dbg_write(">hr>" .. #row)
-                head = head .. [[\\bTR]]
-                for nc=1, #row do
-                    local cell = row[nc]
-                    dbg_write("%7s | ", cell.content)
-                    local celltext = lpegmatch(inline_parser, cell.content)
-                    if cell.span then
-                        head = head .. stringformat([=[\\bTH[nc=%s]%s\\eTH]=], cell.span.x, celltext or "")
-                    else
-                        head = head .. [[\\bTH ]] .. celltext .. [[\\eTH]]
-                    end
-                end
-                dbg_write("\n")
-                head = head .. "\\\\eTR\n"
-            end
-            nr = nr + 1
-        end
-
-        head = head .. [[
-\\eTABLEhead
-\\bTABLEbody
-]] 
-    else
-        head = [[
-\\setupTABLE[c][each]  [frame=off]
-\\setupTABLE[r][each]  [frame=off]
-%\\startlinecorrection
-\\bTABLE[split=yes,option=stretch]
-\\bTABLEbody
-]] 
-    end
-    local tail = [[
-\\eTABLEbody
-\\eTABLE
-%\\stoplinecorrection
-]]
-    local body = ""
-    while nr <= #tab do
-        local row = tab[nr]
-        if not row.ignore then
-            dbg_write(">tr>" .. #row)
-            body = body .. [[\\bTR]]
-            for nc=1, #row do
-                local cell = row[nc]
-                dbg_write("%7s | ", cell.content)
-                local celltext = lpegmatch(inline_parser, cell.content)
-                if cell.span then
-                    body = body .. stringformat([=[\\bTC[nc=%s]%s\\eTC]=], cell.span.x, celltext or "")
-                else
-                    body = body .. [[\\bTC ]] .. celltext .. [[\\eTC]]
-                end
-            end
-            dbg_write("\n")
-            body = body .. "\\\\eTR\n"
-        end
-        nr = nr + 1
-    end
-    return head .. body .. tail
-end
-
-function rst_context.footnote (label, content)
-    local tf = rst_context.state.footnotes
-    rst_context.addsetups("footnotes")
-    if stringmatch(label, "^%d+$") then -- all digits
-        tf.numbered[tonumber(label)] =
-            rst_context.escape(lpegmatch(inline_parser, content))
-    elseif label == "#" then --autonumber
-        repeat -- until next unrequested number
-            tf.autonumber = tf.autonumber + 1
-        until tf.numbered[tf.autonumber] == nil
-        tf.numbered[tf.autonumber] =
-            rst_context.escape(lpegmatch(inline_parser, content))
-    elseif stringmatch(label, "^#.+$") then
-        local thelabel = stringmatch(label, "^#(.+)$")
-        tf.autolabel[thelabel] =
-            rst_context.escape(lpegmatch(inline_parser, content))
-    elseif label == "*" then
-        rst_context.addsetups("footnote_symbol")
-        tf.symbol[#tf.symbol+1] =
-            rst_context.escape(lpegmatch(inline_parser, content))
-    else -- “citation reference” treated like ordinary footnote
-        repeat -- until next unrequested number
-            tf.autonumber = tf.autonumber + 1
-        until tf.numbered[tf.autonumber] == nil
-        tf.numbered[tf.autonumber] =
-            rst_context.escape(lpegmatch(inline_parser, content))
-    end
-    return ""
-end
-
---- hack to differentiate inline images
-local special_substitutions = {
-    image = "inline_image",
-}
-
-function rst_context.substitution_definition (subtext, directive, data)
-    local special = special_substitutions[directive]
-    if special then
-        --- override; pass data directly
-        directive = special
-    else
-        local tmp
-        if data.first ~= "" then
-            tmp = { data.first }
-        else
-            tmp = { }
-        end
-        data.first = nil
-        for i=1, #data do -- paragraphs
-            local current = tableconcat(data[i], "\n")
-            --current = lpegmatch(inline_parser, current)
-            --current = rst_context.escape(current)
-            tmp[#tmp+1] = current
-        end
-        data = tableconcat(tmp, "\n\n")
-        data = stringstrip(data)
-    end
-    subtext = stringgsub(subtext, "%s", "")
-    rst_context.substitutions[subtext] = { directive = directive,
-                                           data      = data }
-    return ""
-end
-
--- not to be confused with the directive definition table rst_directives
-function rst_context.directive(directive, data)
-    local fun = rst_directives[directive]
-    if fun then
-        rst_context.addsetups("directive")
-        local result = ""
-        result = fun(data)
-        return result
-    end
-    return ""
-end
-
--- vim:ft=lua:sw=4:ts=4:expandtab
diff --git a/mod/tex/context/third/rst/rst_directives.lua b/mod/tex/context/third/rst/rst_directives.lua
deleted file mode 100644
index f5572b7..0000000
--- a/mod/tex/context/third/rst/rst_directives.lua
+++ /dev/null
@@ -1,381 +0,0 @@
-#!/usr/bin/env texlua
---------------------------------------------------------------------------------
---         FILE:  rst_directives.lua
---        USAGE:  called by rst_parser.lua
---  DESCRIPTION:  Complement to the reStructuredText parser
---       AUTHOR:  Philipp Gesang (Phg), <phg42.2a@gmail.com>
---      CHANGED:  2013-06-03 18:52:35+0200
---------------------------------------------------------------------------------
---
-
-local helpers = helpers or thirddata and thirddata.rst_helpers
-
---------------------------------------------------------------------------------
--- Directives for use with |substitutions|
---------------------------------------------------------------------------------
-
-local rst_directives     = { }
-thirddata.rst_directives = rst_directives
-local rst_context        = thirddata.rst
-
-local lpegmatch      = lpeg.match
-local stringformat   = string.format
-local stringstrip    = string.strip
-local tableconcat    = table.concat
-local tableflattened = table.flattened
-local type           = type
-
---rst_directives.anonymous     = 0
-rst_directives.images        = {}
-rst_directives.images.done   = {}
-rst_directives.images.values = {}
-
-
-rst_directives.images.keys = {
-    ["width"]   = "width",
-    ["size"]    = "width",
-    ["caption"] = "caption",
-    ["alt"]     = "caption",
-    ["scale"]   = "scale",
-}
-
-rst_directives.images.values.scale = function (orig)
-    -- http://wiki.contextgarden.net/Reference/en/useexternalfigure
-    -- scale=1000 is original size; to get 72%, use scale=720.
-    return tonumber(orig) * 1000
-end
-
-rst_directives.images.values.width = {
-    ["fit"]    = "\\hsize",
-    ["hsize"]  = "\\hsize",
-    ["broad"]  = "\\hsize",
-    ["normal"] = "local",
-    ["normal"] = "local",
-}
-
--- we won't allow passing arbitrary setups to context
-local permitted_setups = {
-    "width",
-    "scale"
-}
-
-local function img_setup (properties)
-    local result = ""
-    for _, prop in next, permitted_setups do
-        if properties[prop] then
-            result = result .. prop .. "=" .. properties[prop] .. ","
-        end
-    end
-    if result ~= "" then
-        result = "[" .. result .. "]"
-    end
-    return result
-end
-
-local collect_image_properties = function (data)
-    local image_directives  = rst_directives.images
-    local p_keyval          = helpers.patterns.colon_keyval
-    local properties        = { }
-
-    data = tableflattened(data)
-    for i=1, #data do
-        local str = stringstrip(data[i])
-        local key, val = lpegmatch(p_keyval, str)
-        if key and val then
-            key = image_directives.keys[key] -- sanitize key expression
-            local valtype = type(image_directives.values[key])
-            if valtype == "table" then
-                val = image_directives.values[key][val]
-            elseif valtype == "function" then
-                val = image_directives.values[key](val)
-            end
-            properties[key] = val
-        end
-    end
-    return properties
-end
-
---- ordinary image directives are converted to floats
-
-local float_image = function (data)
-    rst_context.addsetups "image"
-    local inline_parser = rst_context.inline_parser
-    local properties
-    local anon          = false
-    local rdi           = rst_directives.images
-    local hp            = helpers.patterns
-    local caption       = ""
-    local name          = ""
-
-    if data.name then
-        name = stringstrip(data.name)
-        data.name   = nil
-    else
-        if next(data[1]) then
-            name = data[1][1]
-        end
-    end
-
-    --rd.anonymous = rd.anonymous + 1
-    --anon = true -- indicates a nameless picture
-    --name = "anonymous" .. rd.anonymous
-
-    properties = collect_image_properties(data)
-
-    if properties.caption then
-        caption = lpegmatch(inline_parser, properties.caption)
-        caption = rst_context.escape(caption)
-    end
-
-    properties.setup = img_setup(properties) or ""
-    local img = ""
---    local images_done = rdi.done
---    if not anon then -- TODO: implement?
---        if not images_done[name] then
---            img = img .. stringformat([[
---
---\useexternalfigure[%s][%s][]%%
---]], name, data)
---        images_done[name] = true
---        end
---        img = img .. stringformat([[
---\def\RSTsubstitution%s{%%
---  \placefigure[here]{%s}{\externalfigure[%s]%s}%%
---}
---]], name, rst_context.escape(lpegmatch(inline_parser, properties.caption)), name, properties.setup)
---    else -- image won't be referenced but used instantly
-    img = stringformat(
-        "\n\\placefigure[here]{%s}{\\externalfigure[%s]%s}",
-        caption,
-        name,
-        properties.setup)
---    end
-    return img
-end
-
---- inline substitutions are converted to bare external figures
-local inline_image = function (name, data)
-    rst_context.addsetups "image"
-    local filename  = data.first
-    local p_keyval  = helpers.patterns.colon_keyval
-    local properties
-
-    if not filename then --- garbage, ignore
-        return ""
-    end
-    data.first = nil
-    filename   = stringstrip(filename)
-    properties = collect_image_properties(data)
-
-    local scheme  = "\n\\def\\RSTsubstitution%s{\n  \\externalfigure[%s]%s%%\n}\n"
-    local options = ""
-    if next(properties) then
-        local tmp = { }
-        tmp[#tmp+1] = "["
-        for key, value in next, properties do
-            tmp[#tmp+1] = key
-            tmp[#tmp+1] = "={"
-            tmp[#tmp+1] = rst_context.escape(value)
-            tmp[#tmp+1] = "},"
-        end
-        tmp[#tmp+1] = "]"
-        options = tableconcat(tmp)
-    end
-    return stringformat(scheme, name, filename, options)
-end
-
-rst_directives.image        = float_image
-rst_directives.inline_image = inline_image
-
-rst_directives.caution = function(data)
-    local inline_parser = rst_context.inline_parser
-    rst_context.addsetups("dbend")
-    rst_context.addsetups("caution")
-    local text = { }
-    for i=1, #data do -- paragraphs
-        local current = tableconcat(data[i], "\n")
-        current = lpegmatch(inline_parser, current)
-        current = rst_context.escape(current)
-        text[i] = current
-    end
-    return stringformat([[
-\startRSTcaution
-%s
-\stopRSTcaution
-]], tableconcat(text, "\n\n"))
-end
-
-rst_directives.danger = function(data)
-    local inline_parser = rst_context.inline_parser
-    rst_context.addsetups("dbend")
-    rst_context.addsetups("danger")
-    local text = { }
-    for i=1, #data do -- paragraphs
-        local current = tableconcat(data[i], "\n")
-        current = lpegmatch(inline_parser, current)
-        current = rst_context.escape(current)
-        text[i] = current
-    end
-    return stringformat([[
-\startRSTdanger
-%s
-\stopRSTdanger
-]], tableconcat(text, "\n\n"))
-end
-
--- http://docutils.sourceforge.net/docs/ref/rst/directives.html
-rst_directives.DANGER = function(data)
-    local inline_parser = rst_context.inline_parser
-    local text = { }
-    for i=1, #data do -- paragraphs
-        local current = tableconcat(data[i], "\n")
-        current = lpegmatch(inline_parser, current)
-        current = rst_context.escape(current)
-        text[i] = current
-    end
-    return stringformat([[
-
-%% The Rabbit of Caerbannog
-\startlinecorrection
-\blank[force,big]
-\framed[frame=on,
-        corner=round,
-        rulethickness=5pt,
-        align=middle,
-        width=\hsize,
-        frameoffset=.5em,
-        backgroundoffset=1em,
-        background=color,
-        backgroundcolor=red,
-        foreground=color,
-        foregroundcolor=black]{%%
-  \language[en-gb]\tfb\bf
-  Follow only if ye be men of valour, for the entrance to this cave is guarded
-  by a creature so foul, so cruel that no man yet has fought with it and lived.
-  Bones of full fifty men lie strewn about its lair. So, brave knights, if you
-  do doubt your courage or your strength, come no further, for death awaits you
-  all with nasty, big, pointy teeth.%%
-  \blank[force,big]
-  %s%%
-}
-\blank[force,big]
-\stoplinecorrection
-]], tableconcat(text, "\n\n"))
-end
-
-rst_directives.mp = function(name, data)
-    local mpcode = stringformat([[
-\startreusableMPgraphic{%s}
-%s
-\stopreusableMPgraphic
-]], name, data)
-    mpcode = mpcode .. stringformat([[
-\def\RSTsubstitution%s{%%
-  \reuseMPgraphic{%s}%%
-}
-]], name, name)
-    return mpcode
-end
-
---- There’s an issue with buffers leaving trailing spaces due to their
---- implementation.
---- http://archive.contextgarden.net/message/20111108.175913.1d994624.en.html
-rst_directives.ctx = function(name, data)
-    local ctx = stringformat([[
-
-\startbuffer[%s]
-%s\stopbuffer
-\def\RSTsubstitution%s{%%
-  \getbuffer[%s]\removeunwantedspaces%%
-}
-]], name, data, name, name)
-    return ctx
-end
-
-rst_directives.lua = function(name, data)
-    local luacode = stringformat([[
-
-\startbuffer[%s]
-\startluacode
-%s
-\stopluacode
-\stopbuffer
-\def\RSTsubstitution%s{%%
-  \getbuffer[%s]\removeunwantedspaces%%
-}
-]], name, data, name, name)
-    return luacode
-end
-
---------------------------------------------------------------------------------
---- Experimental math directive
---------------------------------------------------------------------------------
-
-rst_directives.math = function (name, data)
-    data = data or name
-    local formula
-    if type(data) == "table" then
-        local last, i = #data, 1
-        while i <= last do
-            local line = stringstrip(data[i])
-            if line and line ~= "" then
-                formula = formula and formula .. " " .. line or line
-            end
-            i = i + 1
-        end
-    end
-    return stringformat([[
-\startformula
-%s
-\stopformula
-]], formula)
-end
-
---------------------------------------------------------------------------------
---- End math directive
---------------------------------------------------------------------------------
-
-rst_directives.replace = function(name, data)
-    return stringformat([[
-
-\def\RSTsubstitution%s{%s}
-]], name, data)
-end
-
---------------------------------------------------------------------------------
---- Containers.
---------------------------------------------------------------------------------
-
---- *data*:
----     { [1]  -> directive name,
----       [>1] -> paragraphs }
-
-rst_directives.container = function(data)
-    local inline_parser = rst_context.inline_parser
-    local tmp = { }
-    for i=1, #data do -- paragraphs
-        local current = tableconcat(data[i], "\n")
-        current = lpegmatch(inline_parser, current)
-        current = rst_context.escape(current)
-        tmp[i] = current
-    end
-    local content = tableconcat(tmp, "\n\n")
-    local name = data.name
-    if name and name ~= "" then
-        name = stringstrip(data.name)
-        return stringformat([[
-\start[%s]%%
-%s%%
-\stop
-]], name, content)
-    else
-        return stringformat([[
-\begingroup%%
-%s%%
-\endgroup
-]], content)
-    end
-end
-
--- vim:ft=lua:sw=4:ts=4:expandtab
-
diff --git a/mod/tex/context/third/rst/rst_helpers.lua b/mod/tex/context/third/rst/rst_helpers.lua
deleted file mode 100644
index 97d4dd7..0000000
--- a/mod/tex/context/third/rst/rst_helpers.lua
+++ /dev/null
@@ -1,657 +0,0 @@
-#!/usr/bin/env texlua
---------------------------------------------------------------------------------
---         FILE:  rst_helpers.lua
---        USAGE:  called by rst_parser.lua
---  DESCRIPTION:  Complement to the reStructuredText parser
---       AUTHOR:  Philipp Gesang (Phg), <phg42.2a@gmail.com>
---      CHANGED:  2013-03-26 23:55:04+0100
---------------------------------------------------------------------------------
---
-
-local P, R, S, V, lpegmatch
-    = lpeg.P, lpeg.R, lpeg.S, lpeg.V, lpeg.match
-
-local C,   Carg, Cb, Cc, Cg,
-      Cmt, Cp,   Cs, Ct 
-    = lpeg.C,   lpeg.Carg, lpeg.Cb, lpeg.Cc, lpeg.Cg,
-      lpeg.Cmt, lpeg.Cp,   lpeg.Cs, lpeg.Ct
-
-local helpers
-helpers       = thirddata.rst_helpers
-helpers.table = {}
-helpers.cell  = {}
-
-local utf    = unicode.utf8
-local utflen = utf.len
-
-local stringstrip  = string.strip
-local stringformat = string.format
-
-function helpers.dbg_writef(...)
-    if helpers.rst_debug then
-        io.write(stringformat(...))
-    end
-end
-
-local dbg_write = helpers.dbg_writef
-
-helpers.patterns  = {}
-
-do
-    local p = helpers.patterns
-    p.dash   = P"-"
-    p.equals = P"="
-    p.plus   = P"+"
-    p.bar    = P"|"
-    p.eol    = P"\n"
-    p.last   = -P(1)
-    p.space  = P" "
-
-    p.dash_or_equals = p.dash + p.equals
-
-    p.celldelim   = p.bar + p.plus
-    p.cellcontent = (1 - p.celldelim)
-    p.cell        = p.celldelim * C((1 - p.celldelim)^1) * #p.celldelim
-    p.cell_line   = p.plus * p.dash^1 * #p.plus
-    p.dashesonly  = p.dash^1  * p.last
-    p.spacesonly  = p.space^1 * p.last
-
-    p.col_start = Cp() * p.dash_or_equals^1
-    p.col_stop  = p.dash_or_equals^1 * Cp()
-    p.column_starts = Ct(p.col_start * ( p.space^1 * p.col_start)^1)
-    p.column_stops  = Ct(p.col_stop  * ( p.space^1 * p.col_stop)^1)
-
-    p.st_headsep = p.equals^1 * (p.space^1 * p.equals^1)^1
-    p.st_colspan = p.dash^1 * (p.space^1 * p.dash^1)^0 * p.space^0 * p.last
-    p.st_span_starts = Ct(Cp() * p.dash^1 * (p.space^1 * Cp() * p.dash^1)^0)
-    p.st_span_stops  = Ct(p.dash^1 * Cp() * (p.space^1 * p.dash^1 * Cp())^0)
-
-
-    p.cells = P{
-        [1] = "cells",
-        cells = p.celldelim 
-              * (C(V"in_cell")
-               * (V"matchwidth" * C(V"in_cell"))^1),
-
-        in_cell = p.cellcontent^1
-                + (p.dash - p.cellcontent)^1,
-
-        matchwidth = Cmt(C(p.celldelim) * Carg(1), function(s,i,del, layout)
-                         local pos = 1
-                         local lw  = layout.widths
-                         for n=1, #lw do
-                             pos = pos + lw[n] + 1
-                             if (i - 1) == pos then return true end
-                         end
-                         return false
-                     end),
-    }
-
-    p.sep_line = p.plus * (p.dash^1   * p.plus)^1 * p.last
-    p.sep_head = p.plus * (p.equals^1 * p.plus)^1 * p.last
-
-    p.sep_part = ((1 - p.cell_line)^0 * p.cell_line) - p.sep_line
-
-    p.new_row = p.sep_line + p.sep_head + p.sep_part
-
-    p.whitespace = S" \t\v\r\n"^1
-    p.strip = p.whitespace^0 * C((1 - (p.whitespace * p.last))^1) * p.whitespace^0 * p.last
-
-
-    local colon = P":"
-    local escaped_colon = P"\\:"
-    local nocolon = (escaped_colon + (1 - colon))^1
-    p.colon_right = nocolon * colon
-    p.colon_keyval = colon^-1 * C(nocolon) * colon * p.space^1 * C((1 - (p.space^0 * P(-1)))^1)
-
-    -- color expression matching for text roles
-    local digit = R"09"
-    local dot   = P"."
-    local colvalue = digit * dot * digit^1
-                   + digit
-                   + dot * digit^1
-    local coldelim = P"_" + P"-"
-    p.rgbvalues = P"rgb_"
-                * Ct( C(colvalue) * coldelim * C(colvalue) * coldelim * C(colvalue) )
-end
-
-function helpers.cell.create(raw, n_row, n_col, parent, variant)
-    local p = helpers.patterns
-    local cell = {}
-    cell.stripped = raw and p.strip:match(raw) or ""
-    cell.content  = raw
-    cell.width    = raw and utflen(raw) or 0
-    cell.bytes    = raw and #raw or 0
-    cell.variant  = "normal" -- [normal|separator|y_continue|x_continue]
-    cell.pos      = {}
-    cell.pos.x    = n_col
-    cell.pos.y    = n_row
-    cell.span     = {}
-    cell.span.x   = 1
-    cell.span.y   = 1
-    cell.parent   = parent
-    return cell
-end
-
-function helpers.cell.get_x_span(content, layout, init)
-    local acc = 0
-    local lw = layout.widths
-    for n=init, #lw do
-        acc = acc + lw[n] + 1
-        if utflen(content) + 1 == acc then 
-            return n - init
-        end
-    end
-    return false
-end
-
-
--- Extending a cell by 1 cell horizontally.
-function helpers.cell.add_x (cell)
-    cell.span.x = cell.span.x + 1
-end
-
-
-local function set_layout (line)
-    local p = helpers.patterns
-    local layout = {}
-    local slice = Ct((p.plus * C(p.dash^1) * #p.plus)^1)
-
-    layout.widths = {}
-    layout.slices = {}
-    local elms = lpegmatch(slice, line)
-    for n=1, #elms do
-        local elm = elms[n]
-        layout.widths[n] = #elm
-        layout.slices[n] =  elm
-    end
-    return layout
-end
-
-function helpers.table.create(raw)
-    local newtab = {}
-    newtab.rows  = {}
-    newtab.layout = set_layout(raw[1])
-
-    local p = helpers.patterns
-
-    newtab.resolve_parent = function(row, col, array)
-        local array = array or newtab.rows
-        local cell  = array[row][col]
-        local par_row, par_col = row, col
-        if cell.parent then
-            par_row, par_col = newtab.resolve_parent(cell.parent.y, cell.parent.x)
-        end
-        return par_row, par_col
-    end
-
-    newtab.__init = function()
-        local hc = helpers.cell
-        local rowcount = 0
-        local newtablayout = newtab.layout
-        for nr=1, #raw do
-            local row = raw[nr]
-            newtab.rows[nr] = {}
-            local this_row = newtab.rows[nr]
-            this_row.sepline = p.sep_line:match(row)
-            this_row.sephead = p.sep_head:match(row)
-            this_row.seppart = p.sep_part:match(row)
-            if this_row.sephead then
-                newtab.has_head = true
-                newtab.head_end = nr
-            end
-
-            local splitted = { p.cells:match(row, 1, newtablayout) }
-            local pos_layout, pos_row = 1, 1
-            local make_empty = {}
-            make_empty.n, make_empty.parent = 0, nil
-
-            while pos_layout <= #newtablayout.widths do
-                local splitpos = splitted[pos_layout]
-                local layoutwidth = newtablayout.widths[pos_layout]
-                local span = 1
-                local this
-
-                if make_empty.n > 0 then
-                    make_empty.n = make_empty.n - 1
-                    this = hc.create("", nr, pos_layout, make_empty.parent)
-                    this.parent  = make_empty.parent
-                    p_row, p_col = newtab.resolve_parent(this.parent.y, this.parent.x)
-                    local thisparent = newtab.rows[p_row][p_col]
-                    if this_row.sepline or this_row.sephead or
-                        newtab.rows[p_row][p_col].variant == "separator" then
-                        this.variant = "separator"
-                    else
-                        this.variant = "empty1"
-                    end
-                else
-                    local cellwidth = utflen(splitpos)
-                    if cellwidth > layoutwidth then
-                        span = span + hc.get_x_span(splitpos, newtablayout, pos_layout)
-                    end
-                    pos_row = pos_row + span
-                    this = hc.create(splitpos, nr, pos_layout, nil)
-                    if p.dashesonly:match(splitpos) or
-                        this_row.sepline or this_row.sephead then
-                        this.variant = "separator"
-                    end
-                    this.span.x = span
-                    make_empty.n = span - 1
-                    make_empty.parent = span > 1 and { y = nr, x = pos_layout } or nil
-                end
-
-                this_row[pos_layout] = this
-                pos_layout = pos_layout + 1
-            end -- while
-        end -- for loop over rows
-
-        local oldrows = newtab.rows
-        local newrows = oldrows
-        for nc=1, #newtablayout.widths do
-            local width = newtablayout.widths[nc]
-            -- this is gonna be extremely slow but at least it's readable
-            local newrow
-            local currentrow = 1
-            for nr=1, #newrows do
-                local row = newrows[nr]
-                local cell = row[nc]
-                dbg_write("nc: %s, nr:%2s | %9s | ", nc, nr,cell.variant)
-                if  row.sepline or row.sephead
-                    or p.dashesonly:match(cell.content)
-                    or cell.variant == "separator" then -- separator; skipping and beginning new row
-                    newrows[nr][nc] = cell
-                    currentrow = currentrow + 1
-                    newrow = true
-                    dbg_write("new >%24s< ", cell.stripped)
-                    if cell.parent then dbg_write("parent |") else dbg_write("no par |") end
-                else
-                    dbg_write("old >%24s< ", cell.stripped)
-                    if cell.parent then dbg_write("parent |") else dbg_write("no par |") end
-                    if newrow then
-                        newrows[nr][nc] = cell
-                        currentrow = currentrow + 1
-                    else -- continuing parent
-
-                        local par_row, par_col
-                        local parent
-                        if cell.parent then
-                            par_row, par_col = newtab.resolve_parent(cell.parent.y, cell.parent.x, newrows)
-                            dbg_write(" use %s,%2s | ", par_col, par_row)
-                        else -- Using vertical predecessor.
-                            par_row, par_col = newtab.resolve_parent(nr-1,nc, newrows)
-                            dbg_write(" new %s,%2s | ", par_col, par_row)
-                        end
-                        parent = newrows[par_row][par_col]
-
-                        if newrows[nr].seppart then
-                            dbg_write("span++")
-                            parent.span.y   = parent.span.y + 1
-                        end
-
-                            parent.content  = parent.content  .. cell.content
-                            parent.stripped = parent.stripped .. " " .. cell.stripped
-                            cell.variant = "empty2"
-                        cell.parent  = { x = par_col, y = par_row }
-                    end
-                    newrow = false
-                end
-                dbg_write("\n")
-                newrows[nr][nc] = cell
-            end -- for loop over rows
-        end -- for loop over columns
-        --newtab.rows = oldrows
-        newtab.rows = newrows
-    end
-
-    newtab.__init()
-
---[[
-    newtab.__draw_debug = function()
-        for nr=1, #newtab.rows do
-            local row = newtab.rows[nr]
-            for nc=1, #row do
-                local cell = row[nc]
-                local field = cell.variant:sub(1,7)
-                if cell.parent then
-                    field = field .. string.format(" %s,%2s",cell.parent.x, cell.parent.y)
-                end
-                dbg_write("%12s | ", field)
-            end
-            dbg_write("\n")
-        end
-    end
---]]
-
-    return newtab
-end
-
-
-
-function helpers.table.resolve_parent (row, col, array)
-    local cell = array[row][col]
-    local par_row, par_col = row, col
-    if cell.parent then
-        par_row, par_col = self.resolve_parent(cell.parent.y, cell.parent.x)
-    end
-    return par_row, par_col
-end
-
-
--- Check the column boundaries of a simple table.
-function helpers.get_st_boundaries (str)
-    local p_column_starts = helpers.patterns.column_starts
-    local p_column_stops  = helpers.patterns.column_stops
-    local starts, stops, slices, elms = { }, { }, { }, nil
-
-    elms = lpegmatch(p_column_starts, str)
-    for n=1, #elms do
-        local elm = elms[n]
-        slices[n] = { start = elm }
-        starts[elm] = true
-    end
-
-    elms = lpegmatch(p_column_stops, str)
-    for n=1, #elms do
-        local elm = elms[n]
-        slices[n]["stop"]  = elm
-        stops[elm] = true
-    end
-    return { starts = starts, stops = stops, slices = slices }
-end
-
-function helpers.table.simple(raw)
-    local rows = {}
-    local multispans = {}
-    local bounds = helpers.get_st_boundaries(raw[1])
-    local p = helpers.patterns
-
-    for nr=1, #raw do
-        local row = raw[nr]
-        local newrow = {}
-        if not p.st_headsep:match(row) and
-           not p.st_colspan:match(row) then
-            local starts, stops = {}, {}
-            local check_span = false
-            if p.st_colspan:match(raw[nr+1]) then  -- expect spans over several columns
-                starts = p.st_span_starts:match(raw[nr+1])
-                stops  = p.st_span_stops :match(raw[nr+1])
-                check_span = true
-            else
-                for ncol=1, #bounds.slices do
-                    local slice = bounds.slices[ncol]
-                    starts[ncol] = slice.start
-                    stops [ncol] = slice.stop
-                end
-            end
-
-            for nc=1, #starts do
-                local start = starts[nc]
-                -- last column can exceed layout width
-                local stop = nc ~= #starts and stops[nc] or #row
-                local cell = {
-                    content = "",
-                    span   = { x = 1, y = 1 },
-                }
-                cell.content = stringstrip(row:sub(start, stop))
-                if check_span then
-                    local start_at, stop_at
-                    for ncol=1, #bounds.slices do
-                        local slice = bounds.slices[ncol]
-                        if slice.start == start then
-                            start_at = ncol
-                        end
-                        if start_at and
-                           not (ncol == #bounds.slices) then
-                            if slice.stop == stop then
-                                stop_at = ncol
-                                break
-                            end
-                        else -- last column, width doesn't matter
-                            stop_at = ncol
-                        end
-                    end
-                    cell.span.x = 1 + stop_at - start_at
-                end
-                newrow[nc] = cell
-            end
-        elseif p.st_colspan:match(row) then
-            newrow.ignore = true
-        elseif not rows.head_end    and
-                nr > 1 and #raw > nr then -- ends the header
-            rows.head_end = nr
-            newrow.head_sep = true
-            newrow.ignore = true
-        else
-            newrow.ignore = true
-        end
-        rows[nr] = newrow
-    end
-
-    for nr=1, #rows do
-        local row = rows[nr]
-        if not row.ignore and row[1].content == "" then
-            row.ignore = true
-            for nc=1, #row do
-                local cell = row[nc]
-                local par_row, par_col = helpers.table.resolve_parent(nr - 1, nc, rows)
-                parent = rows[par_row][par_col]
-                parent.content = parent.content .. " " .. cell.content
-                cell.content = ""
-            end
-
-        end
-    end
-
-    return rows
-end
-
-helpers.list = {}
-
-do
-    local c = {}
-    c.roman = S"ivxlcdm"^1
-    c.Roman = S"IVXLCDM"^1
-    c.alpha = R"az" - P"i" - P"v" - P"x" - P"l"
-    c.Alpha = R"AZ" - P"I" - P"V" - P"X" - P"L"
-    c.digit = R"09"^1
-    c.auto  = P"#"
-
-    local stripme   = S" ()."
-    local dontstrip = 1 - stripme
-    local itemstripper = stripme^0 * C(dontstrip^1) * stripme^0
-
-    local con = function (str)
-        str = itemstripper:match(str)
-        for conv, pat in next, c do
-            if pat:match(str) then
-                return conv
-            end
-        end
-        return false
-    end
-    helpers.list.conversion = con
-
-    local rnums = {
-        i = 1,
-        v = 5,
-        x = 10,
-        l = 50,
-        c = 100,
-        d = 500,
-        m = 1000,
-    }
-
-    local function roman_to_arab (str)
-        local n = 1
-        local curr, succ
-        local max_three = { }
-        local value = 0
-        while n <= #str do
-            if curr and curr == max_three[#max_three] then
-                if #max_three >= 3 then
-                    return "Not a number"
-                else
-                    max_three[#max_three+1] = curr
-                end     
-            else    
-                max_three = { curr }
-            end     
-
-            curr = rnums[str:sub(n,n)] or 1
-
-            n = n + 1
-            succ = str:sub(n,n)
-
-            if succ and succ ~= "" then
-                succ = rnums[succ]
-                if curr < succ then
-                    --n = n + 1
-                    --value = value + succ - curr
-                    value = value  - curr
-                else    
-                    value = value + curr
-                end     
-            else    
-                value = value + curr
-            end     
-        end     
-        return value
-    end
-    helpers.list.roman_to_arab = roman_to_arab
-
-    local suc = function (str, old)
-        str, old = itemstripper:match(str), itemstripper:match(old)
-        local n_str, n_old = tonumber(str), tonumber(old)
-        if n_str and n_old then -- arabic numeral
-            return n_str == n_old + 1
-        end
-
-        local con_str, con_old = con(str), con(old)
-        if con_str == "alpha"  or
-           con_str == "Alpha" then
-            return str:byte() == old:byte() + 1
-        else -- “I'm a Roman!” - “A woman?” - “No, *Roman*! - Au!” - “So your father was a woman?”
-            if not (str:lower() == str  or
-                    str:upper() == str) then -- uneven cased --> fail
-                return false
-            end
-
-            local trc = thirddata.rst.state.roman_cache
-            n_str = trc[str] or nil
-            n_old = trc[old] or nil
-            if not n_str then
-                n_str = roman_to_arab(str:lower())
-                trc[str] = n_str
-            end
-            if not n_old then
-                n_old = roman_to_arab(old:lower())
-                trc[old] = n_old
-            end
-            return n_str == n_old + 1 
-        end
-    end
-    helpers.list.successor = suc
-
-    local greater = function (str, old)
-        str, old = itemstripper:match(str), itemstripper:match(old)
-        local n_str, n_old = tonumber(str), tonumber(old)
-        if n_str and n_old then -- arabic numeral
-            return n_str > n_old
-        end
-
-        local con_str, con_old = con(str), con(old)
-        if con_str == "alpha"  or
-           con_str == "Alpha" then
-            return str:byte() > old:byte()
-        else
-            if not (str:lower() == str  or
-                    str:upper() == str) then -- uneven cased --> fail
-                return false
-            end
-
-
-            local trc = thirddata.rst.state.roman_cache
-            n_str = trc[str] or nil
-            n_old = trc[old] or nil
-            if not n_str then
-                n_str = roman_to_arab(str:lower())
-                trc[str] = n_str
-            end
-            if not n_old then
-                n_old = roman_to_arab(old:lower())
-                trc[old] = n_old
-            end
-            return n_str > n_old
-        end
-    end
-    helpers.list.greater = greater
-
-    local gd = function(str)
-        str = itemstripper:match(str)
-        local value
-        local con_str = con(str)
-        if con_str == "alpha"  or
-           con_str == "Alpha" then
-            return str:byte()
-        else
-            if not (str:lower() == str  or
-                    str:upper() == str) then
-                return false
-            end
-
-            local trc = thirddata.rst.state.roman_cache
-            n_str = trc[str] or nil
-            if not n_str then
-                n_str = roman_to_arab(str:lower())
-                trc[str] = n_str
-            end
-            return n_str
-        end
-    end
-
-    helpers.list.get_decimal = gd
-end
-
-helpers.string = {}
-
-do
-    --- This grammar inside the function is slightly faster than the
-    --- same as an upvalue with the value of “width” repeatedly given
-    --- via lpeg.Carg(). This holds for repeated calls as well.
-    local ulen = utflen
-    function helpers.string.wrapat (str, width)
-        local width = width or 65
-        local linelength = 0
-        local wrap = P{
-            [1] = "wrapper",
-
-            wrapper       = Cs(V"nowhitespace"^0 * (Cs(V"wrapme") + V"other")^1),
-            whitespace    = S" \t\v" + P"\n" / function() linelength = 0 end,
-            nowhitespace  = 1 - V"whitespace",
-            typing        = P[[\\type{]]  * (1 - P"}")^0 * P"}",
-            typingenv     = P[[\\starttyping]] * (1 - P[[\\stoptyping]])^0 * P[[\\stoptyping]],
-            ignore        = V"typing" + V"typingenv",
-            --- the initial whitespace of the “other” pattern must not
-            --- be enforced (“^1”) as it will break the exceptions
-            --- (“ignore” pattern)! In general it is better to have the
-            --- wrapper ignore some valid breaks than to not have it
-            --- matching some valid strings at all.
-            other         = Cmt(V"whitespace"^0 * (V"ignore" + (1 - V"whitespace")^1), function(s,i,w)
-                                   linelength = linelength + ulen(w)
-                                   return true
-                               end),
-            wrapme = Cmt(V"whitespace"^1 * (1 - V"whitespace" - V"ignore")^1, function(s,i,w)
-                        local lw = ulen(w)
-                        if linelength + lw > width then
-                            linelength = lw
-                            return true
-                        end
-                        return false
-                    end) / function (word) return "\n" .. word:match("[^%s]+") end,
-        }
-
-        local reflowed = wrap:match(str)
-        return reflowed
-    end
-end
-
diff --git a/mod/tex/context/third/rst/rst_parser.lua b/mod/tex/context/third/rst/rst_parser.lua
deleted file mode 100644
index e633899..0000000
--- a/mod/tex/context/third/rst/rst_parser.lua
+++ /dev/null
@@ -1,1605 +0,0 @@
-#!/usr/bin/env texlua
---------------------------------------------------------------------------------
---         FILE:  rst_parser.lua
---        USAGE:  refer to doc/documentation.rst
---  DESCRIPTION:  https://bitbucket.org/phg/context-rst/overview
---       AUTHOR:  Philipp Gesang (Phg), <phg42.2a@gmail.com>
---      VERSION:  0.6
---      CHANGED:  2014-02-28 06:49:01+0100
---------------------------------------------------------------------------------
---
-
-
-thirddata             = thirddata or { }
-thirddata.rst         = { }
-thirddata.rst_helpers = { }
-
-environment.loadluafile"rst_helpers"
-environment.loadluafile"rst_directives"
-environment.loadluafile"rst_setups"
-environment.loadluafile"rst_context"
-
-local rst                   = thirddata.rst
-local helpers               = thirddata.rst_helpers
-local optional_setups       = thirddata.rst_setups
-
-rst.strip_BOM               = true
-rst.expandtab               = true
-rst.shiftwidth              = 4
-rst.crlf                    = true
-helpers.rst_debug           = false
-
-local utf                   = unicode.utf8
-
-local ioopen                = io.open
-local iowrite               = io.write
-local stringfind            = string.find
-local stringformat          = string.format
-local stringgsub            = string.gsub
-local stringlen             = string.len
-local stringmatch           = string.match
-local stringstrip           = string.strip
-local stringsub             = string.sub
-local utflen                = utf.len
-
-local warn
-do
-    local ndebug = 0
-    warn = function(str, ...)
-        if not helpers.rst_debug then return false end
-        ndebug = ndebug + 1
-        local slen = #str + 3
-        --str = "*["..str.."]"
-        str = stringformat("*[%4d][%s]", ndebug, str)
-        local arglst = { ... }
-        for i=1, #arglst do
-            local current = arglst[i]
-            if 80 - i * 8 - slen < 0 then
-                local indent = ""
-                for i=1, slen do
-                    indent = indent .. " "
-                end
-                str = str .. "\n" .. indent
-            end
-            str = str .. stringformat(" |%6s", stringstrip(tostring(current)))
-        end
-        iowrite(str .. " |\n")
-        return 0
-    end
-end
-
-local C,   Cb, Cc, Cg,
-      Cmt, Cp, Cs, Ct
-    = lpeg.C,   lpeg.Cb, lpeg.Cc, lpeg.Cg,
-      lpeg.Cmt, lpeg.Cp, lpeg.Cs, lpeg.Ct
-
-local P, R, S, V, lpegmatch
-    = lpeg.P, lpeg.R, lpeg.S, lpeg.V, lpeg.match
-
-local utf = unicode.utf8
-
-local state          = {}
-thirddata.rst.state  = state
-
-state.depth          = 0
-state.bullets        = {}  -- mapping bullet forms to depth
-state.bullets.max    = 0
-state.lastbullet     = ""
-state.lastbullets    = {}
-state.roman_cache    = {}  -- storing roman numerals that were already converted
-state.currentindent  = ""  -- used in definition lists and elsewhere
-state.previousindent = ""  -- for literal blocks included in paragraphs to restore the paragraph indent
-state.currentwidth   = 0   -- table layout
-state.currentlayout  = {}  -- table layout
-state.previousadorn  = nil -- section underlining and overlining
-
-state.footnotes            = {}
-state.footnotes.autonumber = 0
-state.footnotes.numbered   = {}
-state.footnotes.labeled    = {}
-state.footnotes.autolabel  = {}
-state.footnotes.symbol     = {}
-
-state.addme                = {}
-
-do
-    local first_adornment = ""
-    local valid_adornment = P {
-        [1] = "adorncheck",
-        adorncheck  = V"check_first" * V"check_other"^1 * -P(1),
-        check_first = Cmt(V"adornment_char", function(_,_, first)
-                            first_adornment = first
-                            return true
-                        end)
-                    ,
-        check_other = Cmt(V"adornment_char", function(_,_, char)
-                            local prev = first_adornment
-                            return char == prev
-                        end)
-                    ,
-        adornment_char = S[[!"#$%&'()*+,-./:;<=>?@[]^_`{|}~]] + P[[\\]],
-    }
-    state.valid_adornment = valid_adornment
-end
-
-local enclosed_mapping = {
-    ["'"] = "'",
-    ['"'] = '"',
-    ["("] = ")",
-    ["["] = "]",
-    ["{"] = "}",
-    ["<"] = ">",
-}
-
-local utfchar = P{ -- from l-lpeg.lua, modified to use as grammar
-    [1] = "utfchar",
-    utf8byte      = R("\128\191"),
-    utf8one       = R("\000\127"),
-    utf8two       = R("\194\223") * V"utf8byte",
-    utf8three     = R("\224\239") * V"utf8byte" * V"utf8byte",
-    utf8four      = R("\240\244") * V"utf8byte" * V"utf8byte" * V"utf8byte",
-    utfchar       = V"utf8one" + V"utf8two" + V"utf8three" + V"utf8four",
-}
-
-
-
-local rst_parser = P {
-    [1] = V"document",
-
-    document = V"blank_line"^0 * Cs(V"block"^1),
-
---------------------------------------------------------------------------------
--- Blocks
---------------------------------------------------------------------------------
-
-    block = V"explicit_markup"
-          + Cs(V"section")     / rst.escape
-          + V"target_block"
-          + V"literal_block"
-          + Cs(V"list")        / rst.escape
-          + Cs(V"line_block")  / rst.escape
-          + Cs(V"table_block") / rst.escape
-          + V"transition"    --/ rst.escape
-          + V"comment_block"
-          + Cs(V"block_quote") / rst.escape
-          + Cs(V"paragraph")   / rst.escape
-          ,
-
---------------------------------------------------------------------------------
--- Explicit markup block
---------------------------------------------------------------------------------
-
-    explicit_markup_start = V"double_dot" * V"whitespace",
-
-    explicit_markup = V"footnote_block"
-                    + V"directive_block"
-                    + V"substitution_definition"
-                    ,
-
-    explicit_markup_block = V"explicit_markup"^1
-                          ,
-
---------------------------------------------------------------------------------
--- Directives block
---------------------------------------------------------------------------------
-
-    directive_block = V"directive"
-                    --* (V"blank_line"^-1 * V"directive")^0
-                    * V"end_block"
-                    ,
-
-    directive = V"explicit_markup_start"
-              * C(((V"escaped_colon" + (1 - V"colon" - V"eol"))
-                 - V"substitution_text")^1) --> directive name
-              * V"double_colon"
-              * Ct(V"directive_block_multi" + V"directive_block_single") --> content
-              / rst.directive
-              ,
-
-    directive_block_multi = Cg((1 - V"eol")^0, "name") -- name
-                          * V"eol"
-                          * V"blank_line"^0 -- how many empty lines are permitted?
-                          * V"directive_indented_lines"
-                          ,
-
-    directive_block_single = V"whitespace"^1 * Ct(C((1 - V"eol")^1)) * V"eol",
-
---------------------------------------------------------------------------------
--- Substitution definition block
---------------------------------------------------------------------------------
-
-    substitution_definition = V"explicit_markup_start"
-                            * V"substitution_text"
-                            * V"whitespace"^1
-                            * C((1 - V"colon" - V"space" - V"eol")^1) -- directive
-                            * V"double_colon"
-                            * Ct(V"data_directive_block")
-                            * V"end_block"^-1
-                            / rst.substitution_definition
-                            ,
-
-    substitution_text = V"bar"
-                      * C((1 - V"bar" - V"eol")^1)
-                      * V"bar"
-                      ,
-
-    data_directive_block = V"data_directive_block_multi"
-                         + V"data_directive_block_single"
-                         ,
-    data_directive_block_single = Ct(C((1 - V"eol")^0)) * V"eol",
-
-    data_directive_block_multi  = Cg((1 - V"eol")^0, "first") * V"eol"
-                                * V"directive_indented_lines"
-                                ,
-
-    directive_indented_lines = Ct(V"directive_indented_first"
-                                * V"directive_indented_other"^0)
-                             * (V"blank_line"^1 * Ct(V"directive_indented_other"^1))^0
-                             ,
-
-
-    directive_indented_first = Cmt(V"space"^1, function(s,i,indent)
-                                    warn("sub-i", #indent, i)
-                                    state.currentindent = indent
-                                    return true
-                                end)
-                             * C((1 - V"eol")^1) * V"eol"
-                             ,
-
-    directive_indented_other = Cmt(V"space"^1, function(s,i,indent)
-                                    warn("sub-m",
-                                      #state.currentindent <= #indent,
-                                      #indent,
-                                      #state.currentindent,
-                                      i)
-                                    return #state.currentindent <= #indent
-                                end)
-                             * C((1 - V"eol")^1) * V"eol"
-                             ,
-
-
---------------------------------------------------------------------------------
--- Explicit markup footnote block
---------------------------------------------------------------------------------
-
-    footnote_block = V"footnote"^1 * V"end_block",
-
-    footnote = V"explicit_markup_start"
-             * (V"footnote_marker" + V"citation_reference_label")
-             * C(V"footnote_content")
-             * (V"blank_line" - V"end_block")^-1
-             / rst.footnote
-             ,
-
-    footnote_marker = V"lsquare" * C(V"footnote_label") * V"rsquare" * V"whitespace"^0
-                    ,
-
-    citation_reference_label = V"lsquare" * C(V"letter" * (1 - V"rsquare")^1) * V"rsquare" * V"whitespace"^0,
-
-    footnote_label = V"digit"^1
-                   + (V"gartenzaun" * V"letter"^1)
-                   + V"gartenzaun"
-                   + V"asterisk"
-                   ,
-
-    footnote_content = V"footnote_long" -- single line
-                     + V"footnote_simple"
-                     ,
-
-    footnote_simple = (1 - V"eol")^1 * V"eol"
-                    ,
-
-    footnote_long = (1 - V"eol")^1 * V"eol"
-                  * V"footnote_body"
-                  ,
-
-    footnote_body = V"fn_body_first"
-                  * (V"fn_body_other" + V"fn_body_other_block")^0
-                  ,
-
-    fn_body_first = Cmt(V"space"^1, function(s, i, indent)
-                        warn("fn-in", true, #indent)
-                        state.currentindent = indent
-                        return true
-                    end)
-                  * (1 - V"eol")^1 * V"eol"
-                  ,
-
-    fn_matchindent = Cmt(V"space"^1, function(s, i, indent)
-                        local tc = state.currentindent
-                        warn("fn-ma", tc == indent, #tc, #indent, i)
-                        return tc == indent
-                    end)
-                   ,
-
-    fn_body_other = V"fn_body_other_regular"
-                  * (V"blank_line" * V"fn_body_other_regular")^0
-                  ,
-
-    fn_body_other_regular = V"fn_matchindent"
-                          * (1 - V"eol")^1 * V"eol"
-                          ,
-
-    -- TODO find a way to get those to work in footnotes!
-    fn_body_other_block = V"line_block"
-                        + V"table_block"
-                        + V"transition"
-                        + V"block_quote"
-                        + V"list"
-                        ,
-
---------------------------------------------------------------------------------
--- Table block
---------------------------------------------------------------------------------
-
-    table_block = V"simple_table"
-                + V"grid_table"
-                ,
-
---------------------------------------------------------------------------------
--- Simple tables
---------------------------------------------------------------------------------
-
-    simple_table = Ct(V"st_first_row"
-                    * V"st_other_rows")
-                 * V"end_block"
-                 / function (tab)
-                     return rst.simple_table(helpers.table.simple(tab))
-                 end
-                 ,
-
-    st_first_row = V"st_setindent"
-                 * C(V"st_setlayout")
-                 * V"space"^0
-                 * V"eol"
-                 ,
-
-    st_setindent = Cmt(V"space"^0, function(s, i, indent)
-                        warn("sta-i", "true",  #indent, "set", i)
-                        state.currentindent = indent
-                        return true
-                    end)
-                 ,
-
-    st_matchindent = Cmt(V"space"^0, function(s, i, indent)
-                          warn("sta-m", state.currentindent == indent, #indent, #state.currentindent, i)
-                          return state.currentindent == indent
-                      end)
-                   ,
-
-    st_setlayout = Cmt((V"equals"^1) * (V"spaces" * V"equals"^1)^1, function(s, i, layout)
-                        local tc = state.currentlayout
-                        warn("sta-l", #layout, "set", "", i)
-                        tc.raw = layout
-                        tc.bounds = helpers.get_st_boundaries(layout)
-                        return true
-                    end)
-                 ,
-
-    st_other_rows = (V"st_content"^1 * V"st_separator")^1,
-
-    st_content = V"blank_line"^-1
-               * C(V"st_matchlayout"),
-
-    st_matchlayout = -#V"st_separator" * Cmt((1 - V"eol")^1, function (s, i, content)
-                        -- Don't check for matching indent but if the rest is
-                        -- fine then the line should be sane. This allows
-                        -- cells starting with spaces.
-                        content = stringsub (content, #state.currentindent)
-                        local tcb = state.currentlayout.bounds
-                        local n = 1
-                        local spaces_only = P" "^1
-                        while n < #tcb.slices do
-                            local from = tcb.slices[n]  .stop
-                            local to   = tcb.slices[n+1].start
-                            local between = lpegmatch (spaces_only, content, from)
-                            if not between then -- Cell spanning more than one row.
-                                -- pass
-                                warn("sta-c", "span", from, to, i)
-                            elseif not (between >= to) then
-                                warn("sta-c", "false", from, to, i)
-                                return false
-                            end
-                            n = n + 1
-                        end
-                        warn("sta-c", "true", #tcb.slices, "", i)
-                        return true
-                     end)
-                     * V"eol"
-                   ,
-
-    st_separator = V"st_matchindent"
-                 * C(V"st_normal_sep" + V"st_colspan_sep")
-                 * V"eol"
-                 ,
-
-    st_normal_sep = Cmt((V"equals"^1) * (V"spaces" * V"equals"^1)^1, function(s, i, layout)
-                        warn("sta-s", state.currentlayout.raw == layout, #layout, #state.currentlayout.raw, i)
-                        return state.currentlayout.raw == layout
-                    end)
-                  ,
-
-    st_colspan_sep = Cmt(V"dash"^1 * (V"spaces" * V"dash"^1)^0, function(s, i, layout)
-                         local tcb = state.currentlayout.bounds
-                         local this = helpers.get_st_boundaries (layout)
-                         local start_valid = false
-                         for start, _ in next, this.starts do
-                             if tcb.starts[start] then
-                                 start_valid = true
-                                 local stop_valid = false
-                                 for stop, _ in next, this.stops do
-                                     if tcb.stops[stop] then -- bingo
-                                         stop_valid = true
-                                     end
-                                 end
-                                 if not stop_valid then
-                                     warn("sta-x", stop_valid, #layout, #state.currentlayout.raw, i)
-                                     return false
-                                 end
-                             end
-                         end
-                         warn("sta-x", start_valid, #layout, #state.currentlayout.raw, i)
-                         return start_valid
-                     end)
-                   ,
-
-
---------------------------------------------------------------------------------
--- Grid tables
---------------------------------------------------------------------------------
-
-    grid_table = Ct(V"gt_first_row"
-                  * V"gt_other_rows")
-               * V"blank_line"^1
-               / function(tab)
-                   return rst.grid_table(helpers.table.create(tab))
-               end
-               ,
-
-    gt_first_row = V"gt_setindent"
-                 * C(V"gt_sethorizontal")
-                 * V"eol"
-                 ,
-
-    gt_setindent = Cmt(V"space"^0, function(s, i, indent)
-                        warn("tab-i", true, #indent, "set", i)
-                        state.currentindent = indent
-                        return true
-                    end)
-                 ,
-
-    gt_layoutmarkers = V"table_intersection" + V"table_hline" + V"table_header_hline",
-
-    gt_sethorizontal = Cmt(V"gt_layoutmarkers"^3, function (s, i, width)
-                             warn("tab-h", "width", "true", #width, "set", i)
-                             state.currentwidth = #width
-                             return true
-                         end)
-                     ,
-
-    gt_other_rows = V"gt_head"^-1
-                  * V"gt_body"
-                  ,
-
-    gt_matchindent = Cmt(V"space"^0, function (s, i, this)
-        local matchme = state.currentindent
-        warn("tab-m", "indent", #this == #matchme, #this, #matchme, i)
-        return #this == #matchme
-    end)
-    ,
-
-
-    gt_cell = (V"gt_content_cell" + V"gt_line_cell")
-    * (V"table_intersection" + V"table_vline")
-    ,
-
-    gt_content_cell = ((1 - V"table_vline" - V"table_intersection" - V"eol")^1),
-
-    gt_line_cell = V"table_hline"^1,
-
-    gt_contentrow = V"gt_matchindent"
-                   * C((V"table_intersection" + V"table_vline")
-                     * V"gt_cell"^1)
-                   * V"whitespace"^-1 * V"eol"
-                  ,
-
-    gt_body = ((V"gt_contentrow" - V"gt_bodysep")^1 * V"gt_bodysep")^1,
-
-    gt_bodysep = V"gt_matchindent"
-               * C(Cmt(V"table_intersection"
-                     * (V"table_hline"^1 * V"table_intersection")^1, function(s, i, separator)
-                          local matchme = state.currentwidth
-                          warn("tab-m", "body", #separator == matchme, #separator, matchme, i)
-                          return #separator == matchme
-                      end))
-               * V"whitespace"^-1 * V"eol"
-               ,
-
-    gt_head = V"gt_contentrow"^1
-            * V"gt_headsep"
-            ,
-
-    gt_headsep = V"gt_matchindent"
-               * C(Cmt(V"table_intersection"
-                    * (V"table_header_hline"^1 * V"table_intersection")^1, function(s, i, separator)
-                          local matchme = state.currentwidth
-                          warn("tab-s", "head", #separator == matchme, #separator, matchme, i)
-                          return #separator == matchme
-                      end))
-               * V"whitespace"^-1 * V"eol"
-               ,
-
-
---------------------------------------------------------------------------------
--- Block quotes
---------------------------------------------------------------------------------
-
-    block_quote = Ct(Cs(V"block_quote_first"
-                   * V"block_quote_other"^0
-                   * (V"blank_line" * V"block_quote_other"^1)^0)
-                   * (V"blank_line"
-                   *  Cs(V"block_quote_attri"))^-1)
-                * V"end_block"
-                / rst.block_quote
-                ,
-
-    block_quote_first = Cmt(V"space"^1, function (s, i, indent)
-                             warn("bkq-i", #indent, "", indent, "", i)
-                             state.currentindent = indent
-                             return true
-                         end) / ""
-                      * -V"attrib_dash"
-                      * (1 - V"eol")^1
-                      * V"eol"
-                      ,
-
-    block_quote_other = Cmt(V"space"^1, function (s, i, indent)
-                            warn("bkq-m", #indent, #state.currentindent,
-                                           indent,  state.currentindent, i)
-                            return state.currentindent == indent
-                        end) / ""
-                      * -V"attrib_dash"
-                      * (1 - V"eol")^1
-                      * V"eol"
-                      ,
-
-    block_quote_attri = V"block_quote_attri_first"
-                      * V"block_quote_attri_other"^0,
-
-    block_quote_attri_first = Cmt(V"space"^1 * V"attrib_dash" * V"space", function (s, i, indent)
-                                   local t = state
-                                   warn("bqa-i", utflen(indent), #t.currentindent,
-                                                 indent,         t.currentindent, i)
-                                   local ret = stringmatch (indent, " *") == t.currentindent
-                                   t.currentindent = ret and indent or t.currentindent
-                                   return ret
-                               end) / ""
-                            * (1 - V"eol")^1
-                            * V"eol"
-                            ,
-
-    block_quote_attri_other = Cmt(V"space"^1, function (s, i, indent)
-                                  warn("bqa-m", #indent, utflen(state.currentindent),
-                                                 indent,  state.currentindent, i)
-                                  return utflen(state.currentindent) == #indent
-                              end) / ""
-                            * (1 - V"eol")^1
-                            * V"eol"
-                            ,
-
---------------------------------------------------------------------------------
--- Line blocks
---------------------------------------------------------------------------------
-
-    line_block = Cs(V"line_block_first"
-                  * (V"line_block_other"
-                   + V"line_block_empty")^1)
-               --* V"blank_line"
-               * V"end_block"
-               / rst.line_block
-               ,
-
-    line_block_marker = V"space"^0 * V"bar" * V"space",
-
-    line_block_empty_marker = V"space"^0 * V"bar" * V"space"^0 * V"eol",
-
-
-    line_block_first = Cmt(V"line_block_marker", function(s, i, marker)
-                            warn("lbk-i", #marker, "", marker, "", i)
-                            state.currentindent = marker
-                            return true
-                        end) / ""
-                     * V"line_block_line"
-                     ,
-
-    line_block_empty = Cmt(V"line_block_empty_marker", function(s, i, marker)
-                            warn("lbk-e", #marker, #state.currentindent, marker, state.currentindent, i)
-                            marker = stringgsub (marker, "|.*", "| ")
-                            return state.currentindent == marker
-                        end) / ""
-                     / rst.line_block_empty
-                     ,
-
-    line_block_other = Cmt(V"line_block_marker", function(s, i, marker)
-                            warn("lbk-m", #marker, #state.currentindent, marker, state.currentindent, i)
-                            return state.currentindent == marker
-                        end) / ""
-                     * V"line_block_line"
-                     ,
-
-    line_block_line = Cs((1 - V"eol")^1
-                       * V"line_block_cont"^0
-                       * V"eol")
-                    / rst.line_block_line
-                    ,
-
-    line_block_cont = (V"eol" - V"line_block_marker")
-                    * Cmt(V"space"^1, function(s, i, spaces)
-                            warn("lbk-c", #spaces, #state.currentindent, spaces, state.currentindent, i)
-                            return #spaces >= #state.currentindent
-                        end) / ""
-                    * (1 - V"eol")^1
-                    ,
-
---------------------------------------------------------------------------------
--- Literal blocks
---------------------------------------------------------------------------------
-
-    literal_block = V"literal_block_marker"
-                    * Cs(V"literal_block_lines")
-                    * V"end_block"
-                    / rst.literal_block,
-
-    literal_block_marker = V"double_colon" * V"whitespace"^0 * V"eol" * V"blank_line",
-
-    literal_block_lines = V"unquoted_literal_block_lines"
-                        + V"quoted_literal_block_lines"
-                        ,
-
-    unquoted_literal_block_lines = V"literal_block_first"
-                                 * (V"blank_line"^-1 * V"literal_block_other")^0
-                                 ,
-
-    quoted_literal_block_lines =  V"quoted_literal_block_first"
-                               * V"quoted_literal_block_other"^0 -- no blank lines allowed
-                               ,
-
-    literal_block_first = Cmt(V"space"^1, function (s, i, indent)
-                        warn("lbk-f", #indent, "", "", i)
-                        if not indent or
-                            indent == "" then
-                            return false
-                        end
-                        if state.currentindent and #state.currentindent < #indent then
-                            state.currentindent = state.currentindent .. " "
-                            return true
-                        else
-                            state.currentindent = " "
-                            return true
-                        end
-                    end)
-                   * V"rest_of_line"
-                   * V"eol",
-
-    literal_block_other = Cmt(V"space"^1, function (s, i, indent)
-                        warn("lbk-m",
-                             #indent,
-                             #state.currentindent,
-                             #indent >= #state.currentindent,
-                             i)
-                        return #indent >= #state.currentindent
-                    end)
-                   * V"rest_of_line"
-                   * V"eol"
-                   ,
-
-    quoted_literal_block_first = Cmt(V"adornment_char", function (s, i, indent)
-                        warn("qlb-f", #indent, indent, "", i)
-                        if not indent    or
-                            indent == "" then
-                            return false
-                        end
-                        state.currentindent = indent
-                        return true
-                    end)
-                   * V"rest_of_line"
-                   * V"eol"
-                   ,
-
-    quoted_literal_block_other = Cmt(V"adornment_char", function (s, i, indent)
-                        warn("qlb-m",
-                             #indent,
-                             #state.currentindent,
-                             #indent >= #state.currentindent,
-                             i)
-                        return #indent >= #state.currentindent
-                    end)
-                   * V"rest_of_line"
-                   * V"eol",
-
---------------------------------------------------------------------------------
--- Lists
---------------------------------------------------------------------------------
-
-    list = (V"option_list"
-          + V"bullet_list"
-          + V"definition_list"
-          + V"field_list")
-         - V"explicit_markup_start"
-         ,
-
---------------------------------------------------------------------------------
--- Option lists
---------------------------------------------------------------------------------
-
-    option_list = Cs((V"option_list_item"
-                   * V"blank_line"^-1)^1)
-                /rst.option_list,
-
-    option_list_item = Ct(C(V"option_group")
-                        * Cs(V"option_description"))
-                     / rst.option_item,
-
-    option_description = V"option_desc_next"
-                       + V"option_desc_more"
-                       + V"option_desc_single",
-
-    option_desc_single = V"space"^2
-                       --* V"rest_of_line"
-                       * (1 - V"eol")^1
-                       * V"eol",
-
-    option_desc_more = V"space"^2
-                     * (1 - V"eol")^1
-                     * V"eol"
-                     * V"indented_lines"
-                     * (V"blank_line" * V"indented_lines")^0,
-
-    option_desc_next = V"eol"
-                     * V"indented_lines"
-                     * (V"blank_line" * V"indented_lines")^0,
-
-    option_group = V"option"
-                 * (V"comma" * V"space" * V"option")^0,
-
-    option = (V"option_posixlong"
-            + V"option_posixshort"
-            + V"option_dos_vms")
-            * V"option_arg"^-1,
-
-    option_arg = (V"equals" + V"space")
-               * ((V"letter" * (V"letter" + V"digit")^1)
-                + (V"angle_left" * (1 - V"angle_right")^1 * V"angle_right")),
-
-    option_posixshort = V"dash" * (V"letter" + V"digit"),
-
-    option_posixlong = V"double_dash"
-                     * V"letter"
-                     * (V"letter" + V"digit" + V"dash")^1,
-
-    option_dos_vms = V"slash"
-                   * V"letter"^1,
-
---------------------------------------------------------------------------------
--- Field lists (for bibliographies etc.)
---------------------------------------------------------------------------------
-
-    field_list = Cs(V"field"
-                  * (V"blank_line"^-1 * V"field")^0)
-               * V"end_block"
-               / rst.field_list,
-
-    field = Ct(V"field_marker"
-             * V"whitespace"
-             * V"field_body")
-          / rst.field,
-
-    field_marker = V"colon"
-                 * C(V"field_name")
-                 * V"colon",
-
-    field_name = (V"escaped_colon" + (1 - V"colon"))^1,
-
-    field_body = V"field_single" + V"field_multi",
-
-    field_single = C((1 -V"eol")^1) * V"eol",
-
-    field_multi = C((1 - V"eol")^0 * V"eol"
-                  * V"indented_lines"^-1),
-
---------------------------------------------------------------------------------
--- Definition lists
---------------------------------------------------------------------------------
-
-    definition_list = Ct((V"definition_item" - V"comment")
-                      * (V"blank_line" * V"definition_item")^0)
-                    * V"end_block"
-                    / rst.deflist
-                    ,
-
-    definition_item = Ct(C(V"definition_term")
-                       * V"definition_classifiers"
-                       * V"eol"
-                       * Ct(V"definition_def"))
-                    ,
-
-    definition_term = #(1 - V"space" - V"field_marker")
-                    * (1 - V"eol" - V"definition_classifier_separator")^1
-                    ,
-
-    definition_classifier_separator = V"space" * V"colon" * V"space",
-
-    definition_classifiers = V"definition_classifier"^0,
-
-    definition_classifier = V"definition_classifier_separator"
-                          * C((1 - V"eol" - V"definition_classifier_separator")^1)
-                          ,
-
-    definition_def = C(V"definition_firstpar") * C(V"definition_par")^0
-                   ,
-
-    definition_indent = Cmt(V"space"^1, function(s, i, indent)
-                            warn("def-i", #indent, #state.currentindent, indent == state.currentindent, i)
-                            state.currentindent = indent
-                            return true
-                        end),
-
-    definition_firstpar = V"definition_parinit"
-                        * (V"definition_parline" - V"blank_line")^0
-                        ,
-
-    definition_par = V"blank_line"
-                   * (V"definition_parline" - V"blank_line")^1
-                   ,
-
-    definition_parinit = V"definition_indent"
-                       * (1 - V"eol")^1
-                       * V"eol"
-                       ,
-
-    definition_parline = V"definition_match"
-                       * (1 - V"eol")^1
-                       * V"eol"
-                       ,
-
-    definition_match = Cmt(V"space"^1, function (s, i, this)
-                            warn("def-m", #this, #state.currentindent, this == state.currentindent, i)
-                            return this == state.currentindent
-                        end),
-
---------------------------------------------------------------------------------
--- Bullet lists and enumerations
---------------------------------------------------------------------------------
-
-    -- the next rule handles enumerations as well
-    bullet_list = V"bullet_init"
-                * (V"blank_line"^-1 * (V"bullet_list" + V"bullet_continue"))^1
-                * V"bullet_stop"
-                * Cmt(Cc(nil), function (s, i)
-                    local t = state
-                    warn("close", t.depth)
-                    t.bullets[t.depth] = nil -- “pop”
-                    t.depth = t.depth - 1
-                    t.lastbullet = t.lastbullets[t.depth]
-                    return true
-                end),
-
-    bullet_stop = V"end_block" / rst.stopitemize,
-
-    bullet_init = Ct(C(V"bullet_first") * V"bullet_itemrest")
-                / rst.bullet_item
-                ,
-
-    bullet_first = #Cmt(V"bullet_indent", function (s, i, bullet)
-                        local t = state
-                        local oldbullet = t.bullets[t.depth]
-                        local n_spaces = lpegmatch(P" "^0, bullet)
-                        warn("first",
-                            t.depth,
-                            (t.depth == 0 and n_spaces >= 1) or
-                            (t.depth >  0 and n_spaces >  1),
-                            bullet,
-                            oldbullet,
-                            helpers.list.conversion(bullet))
-
-                        if t.depth == 0 and n_spaces >= 1 then -- first level
-                            t.depth = 1             -- “push”
-                            t.bullets[1] = bullet
-                            t.lastbullet = bullet
-                            t.bullets.max = t.bullets.max < t.depth and t.depth or t.bullets.max
-                            return true
-                        elseif t.depth > 0 and n_spaces > 1 then    -- sublist (of sublist)^0
-                            if n_spaces >= utflen(oldbullet) then
-                                t.lastbullets[t.depth] = t.lastbullet
-                                t.depth = t.depth + 1
-                                t.bullets[t.depth] = bullet
-                                t.lastbullet = bullet
-                                t.bullets.max = t.bullets.max < t.depth and t.depth or t.bullets.max
-                                return true
-                            end
-                        end
-                        return false
-                    end)
-                    * V"bullet_indent"
-                    / rst.startitemize
-                    ,
-
-    bullet_indent = V"space"^0 * V"bullet_expr" * V"space"^1,
-
-    bullet_cont  = Cmt(V"bullet_indent", function (s, i, bullet)
-                        local t = state
-                        local conversion = helpers.list.conversion
-                        warn("conti",
-                                t.depth,
-                                bullet == t.bullets[t.depth],
-                                bullet,
-                                t.bullets[t.depth],
-                                t.lastbullets[t.depth],
-                                conversion(t.lastbullet),
-                                conversion(bullet)
-                                )
-
-                        if utflen(t.bullets[t.depth]) ~= utflen(bullet) then
-                            return false
-                        elseif not conversion(bullet) and t.bullets[t.depth] == bullet then
-                            return true
-                        elseif conversion(t.lastbullet) == conversion(bullet) then -- same type
-                            local autoconv  = conversion(bullet) == "auto"
-                            local greater   = helpers.list.greater  (bullet, t.lastbullet)
-                            t.lastbullet = bullet
-                            return autoconv or successor or greater
-                        end
-                    end),
-
-    bullet_continue = Ct(C(V"bullet_cont") * V"bullet_itemrest")
-                    /rst.bullet_item
-                    ,
-
-    bullet_itemrest = C(V"bullet_rest"                               -- first line
-                       * ((V"bullet_match" * V"bullet_rest")^0        -- any successive lines
-                        * (V"blank_line"
-                         * (V"bullet_match" * (V"bullet_rest" - V"bullet_indent"))^1)^0))
-                    ,
-                         --                                     ^^^^^^^^^^^^^
-                         --                                     otherwise matches bullet_first
-
-    bullet_rest = (1 - V"eol")^1 * V"eol",  -- rest of one line
-
-    bullet_next  = V"space"^1
-                 ,
-
-    bullet_match = Cmt(V"bullet_next", function (s, i, this)
-                         local t = state
-                         warn("match",
-                                t.depth,
-                                stringlen(this) == utflen(t.bullets[t.depth]),
-                                utflen(t.bullets[t.depth]), stringlen(this) )
-                         return stringlen(this) == utflen(t.bullets[t.depth])
-                     end)
-                 ,
-
-    bullet_expr = V"bullet_char"
-                + (P"(" * V"number_char" * P")")
-                +        (V"number_char" * P")")
-                + (V"number_char" * V"dot") * #V"space"
-                + (V"number_char" * #V"space")
-                ,
-
-    number_char = V"roman_numeral"
-                + V"Roman_numeral"
-                + P"#"
-                + V"digit"^1
-                + R"AZ"
-                + R"az"
-                ,
-
---------------------------------------------------------------------------------
--- Transitions
---------------------------------------------------------------------------------
-
-    transition_line = C(V"adornment_char"^4),
-
-    transition = V"transition_line" * V"eol"
-               * V"end_block"
-               / rst.transition
-               ,
-
---------------------------------------------------------------------------------
--- Sectioning
---------------------------------------------------------------------------------
-
-    section_adorn = V"adornment_char"^1,
-
-    section = ((V"section_text" * V"section_once")
-             + (V"section_before" * V"section_text" * V"section_after"))
-            / rst.section
-            * (V"end_block" + V"blank_line")
-            ,
-
-    -- The whitespace handling after the overline is necessary because headings
-    -- without overline aren't allowed to be indented.
-    section_before = C(Cmt(V"section_adorn", function(s,i, adorn)
-                          local adorn_matched = lpegmatch (state.valid_adornment, adorn)
-                          state.previousadorn = adorn
-                          warn ("sec-f", adorn_matched,
-                                stringsub (adorn, 1,2) .. "...", "", i)
-                          if adorn_matched then
-                              return true
-                          end
-                          return false
-                      end))
-                   * V"whitespace"^0
-                   * V"eol"
-                   * V"whitespace"^0
-                   ,
-
-    section_text = C((1 - V"space" - V"eol") * (1 - V"eol")^1) * V"eol",
-
-    section_after = C(Cmt(V"section_adorn", function(s,i, adorn)
-                         local tests = false
-                         if lpegmatch (state.valid_adornment, adorn) then
-                           tests = true
-                         end
-                         if state.previousadorn then
-                             tests = tests and adorn == state.previousadorn
-                         end
-                         warn ("sec-a", tests, stringsub (adorn, 1,2) .. "…", "", i)
-                         state.previousadorn = nil
-                         return tests
-                     end))
-                    * V"whitespace"^0
-                    ,
-
-    section_once = C(Cmt(V"section_adorn", function(s,i, adorn)
-                         local tests = false
-                         if lpegmatch (state.valid_adornment, adorn) then
-                           tests = true
-                         end
-                         warn ("sec-o", tests, stringsub (adorn, 1,2) .. "…", "", i)
-                         state.previousadorn = nil
-                         return tests
-                     end))
-                    * V"whitespace"^0
-                    ,
-
---------------------------------------------------------------------------------
--- Target Blocks
---------------------------------------------------------------------------------
-
-    tname_normal = C((V"escaped_colon" + 1 - V"colon")^1)
-                 * V"colon",
-
-    tname_bareia = C(V"bareia"
-                    * (1 - V"eol" - V"bareia")^1
-                    * V"bareia")
-                 * V"colon",
-
-    target_name = V"double_dot"
-                * V"space"
-                * V"underscore"
-                * (V"tname_bareia" + V"tname_normal"),
-
-    target_firstindent = V"eol" * Cg(V"space"^1, "indent"),
-
-    target_nextindent  = V"eol" * C(V"space"^1),
-
-    target_indentmatch = Cmt(V"target_nextindent" -- I ♡ LPEG!
-                           * Cb("indent"), function (s, i, a, b)
-                                return a == b
-                            end),
-
-    target_link  = ( V"space"^0 * V"target_firstindent"
-                 * Ct(C(1 - V"whitespace" - V"eol")^1
-                    * (V"target_indentmatch"
-                     * C(1 - V"whitespace" - V"eol")^1)^0)
-                 * V"eol" * #(1 - V"whitespace" - "eol")) / rst.joinindented
-                 + C((1 - V"eol")^1) * V"eol" * #(V"double_dot" + V"double_underscore" + V"eol")
-                 + (1 - V"end_block")^0 * Cc(""),
-
-    target       = Ct((V"target_name" * (V"space"^0 * V"eol" * V"target_name")^0)
-                 * V"space"^0
-                 * V"target_link")
-                 / rst.target,
-
-    anonymous_prefix = (V"double_dot" * V"space" * V"double_underscore" * V"colon")
-                     + (V"double_underscore")
-                     ,
-
-    anonymous_target = V"anonymous_prefix"
-                     * V"space"^0
-                     * Ct(Cc"" * V"target_link")
-                     / rst.target
-                     ,
-
-    target_block = (V"anonymous_target" + V"target")^1
-                 * V"end_block",
-
---------------------------------------------------------------------------------
--- Paragraphs * Inline Markup
---------------------------------------------------------------------------------
-
-    paragraph = Ct(V"par_first"
-                 * V"par_other"^0) / rst.paragraph
-              * V"end_block"
-              * V"reset_depth"
-              ,
-
-    par_first = V"par_setindent"
-              * C((1 - V"literal_block_shorthand" - V"eol")^1)
-              * (V"included_literal_block" + V"eol")
-              ,
-
-    par_other = V"par_matchindent"
-              * C((1 - V"literal_block_shorthand" - V"eol")^1)
-              * (V"included_literal_block" + V"eol")
-              ,
-
-    par_setindent = Cmt(V"space"^0, function (s, i, indent)
-                        warn("par-i", #indent, "", "", i)
-                        state.previousindent = state.currentindent
-                        state.currentindent = indent
-                        return true
-                    end),
-
-    par_matchindent = Cmt(V"space"^0, function (s, i, indent)
-                          warn("par-m", state.currentindent == indent, #indent, #state.currentindent, i)
-                          return state.currentindent == indent
-                      end),
-
-    included_literal_block = V"literal_block_shorthand"
-                           * V"literal_block_markerless"
-                           * Cmt(Cp(), function (s, i, _)
-                                  warn("par-s", "", #state.previousindent, #state.currentindent, i)
-                                  state.currentindent = state.previousindent
-                                  return true
-                              end)
-                           ,
-
-    literal_block_shorthand = Cs((V"colon" * V"space" * V"double_colon"
-                                + V"double_colon")
-                             * V"whitespace"^0
-                             * V"eol"
-                             * V"blank_line")
-                             -- The \unskip is necessary because the lines of a
-                             -- paragraph get concatenated from a table with a
-                             -- space as separator. And the literal block is
-                             -- treated as one such line, hence it would be
-                             -- preceded by a space. As the ":" character
-                             -- always  follows a non-space this should be a
-                             -- safe, albeit unpleasant, hack. If you don't
-                             -- agree then file a bug report and I'll look into
-                             -- it.
-                             / "\\\\unskip:"
-                            ,
-
-    literal_block_markerless = Cs(V"literal_block_lines")
-                             * V"blank_line"
-                             / rst.included_literal_block
-                             ,
-
-    -- This is needed because lpeg.Cmt() patterns are evaluated even
-    -- if they are part of a larger pattern that doesn’t match. The
-    -- result is that they confuse the nesting.
-    -- Resetting the current nesting depth at every end of block
-    -- should be safe because this pattern always matches last.
-    reset_depth = Cmt(Cc("nothing") / "", function (s,i, something)
-                        state.depth = 0
-                        warn("reset", "", state.depth, #state.currentindent, i)
-                        return true
-                    end)
-                ,
-
---------------------------------------------------------------------------------
--- Comments
---------------------------------------------------------------------------------
-
-    comment_block = V"comment"
-                  * V"end_block"^-1
-                  ,
-
-    comment = V"double_dot" / ""
-            * (V"block_comment" + V"line_comment")
-            ,
-
-    block_comment = V"whitespace"^0
-                  * Cs((1 - V"eol")^0 * V"eol"
-                     * V"indented_lines")
-                  / rst.block_comment,
-
-    line_comment = V"whitespace"^1
-                 * C((1 - V"eol")^0 * V"eol")
-                 / rst.line_comment
-                 ,
-
---------------------------------------------------------------------------------
--- Generic indented block
---------------------------------------------------------------------------------
-
-    indented_lines = V"indented_first"
-                   * (V"indented_other"^0
-                    * (V"blank_line" * V"indented_other"^1)^0)
-                   ,
-
-    indented_first = Cmt(V"space"^1, function (s, i, indent)
-                        warn("idt-f", indent, i)
-                        state.currentindent = indent
-                        return true
-                    end) / ""
-                   * (1 - V"eol")^1
-                   * V"eol"
-                   ,
-
-    indented_other = Cmt(V"space"^1, function (s, i, indent)
-                        warn("idt-m", #indent, #state.currentindent, #indent == #state.currentindent, i)
-                        return indent == state.currentindent
-                    end) / ""
-                   * (1 - V"eol")^1
-                   * V"eol"
-                   ,
-
---------------------------------------------------------------------------------
--- Urls
---------------------------------------------------------------------------------
-    uri             = V"url_protocol" * V"url_domain" * (V"slash" * V"url_path")^0,
-
-    url_protocol    = (P"http" + P"ftp" + P"shttp" + P"sftp") * P"://",
-    url_domain_char = 1 - V"dot" - V"spacing" - V"eol" - V"punctuation",
-    url_domain      = V"url_domain_char"^1 * (V"dot" * V"url_domain_char"^1)^0,
-    url_path_char   = R("az", "AZ", "09") + S"-_.!~*'()",
-    url_path        = V"slash" * (V"url_path_char"^1 * V"slash"^-1)^1,
-
---------------------------------------------------------------------------------
--- Terminal Symbols and Low-Level Elements
---------------------------------------------------------------------------------
-
-    asterisk          = P"*",
-    backslash         = P"\\",
-    bar               = P"|",
-    bareia            = P"`",
-    slash             = P"/",
-    solidus           = P"⁄",
-    equals            = P"=",
-
-    --- Punctuation
-    -- Some of the following are used for markup as well as for punctuation.
-
-    apostrophe        = P"’" + P"'",
-    comma             = P",",
-    colon             = P":",
-    dot               = P".",
-    interpunct        = P"·",
-    semicolon         = P";",
-    underscore        = P"_",
-    dash              = P"-",
-    emdash            = P"—",
-    hyphen            = P"‐",
-    questionmark      = P"?",
-    exclamationmark   = P"!",
-    interrobang       = P"‽",
-    lsquare           = P"[",
-    rsquare           = P"]",
-    ellipsis          = P"…" + P"...",
-    guillemets        = P"«" + P"»",
-    quotationmarks    = P"‘" + P"’" + P"“" + P"”",
-
-    period            = V"dot",
-    double_dot        = V"dot" * V"dot",
-    double_colon      = V"colon" * V"colon",
-    escaped_colon     = V"backslash" * V"colon",
-    double_underscore = V"underscore" * V"underscore",
-    double_dash       = V"dash" * V"dash",
-    triple_dash       = V"double_dash" * V"dash",
-    attrib_dash       = V"triple_dash" + V"double_dash" + V"emdash", -- begins quote attribution blocks
-    dashes            = V"dash" + P"‒" + P"–" + V"emdash" + P"―",
-
-
-
-    punctuation = V"apostrophe"
-                + V"colon"
-                + V"comma"
-                + V"dashes"
-                + V"dot"
-                + V"ellipsis"
-                + V"exclamationmark"
-                + V"guillemets"
-                + V"hyphen"
-                + V"interpunct"
-                + V"interrobang"
-                + V"questionmark"
-                + V"quotationmarks"
-                + V"semicolon"
-                + V"slash"
-                + V"solidus"
-                + V"underscore"
-                ,
-
-    -- End punctuation
-
-    letter       = R"az" + R"AZ",
-    digit        = R"09",
-
-    space        = P" ",
-    spaces       = V"space"^1,
-    whitespace   = (P" " + Cs(P"\t") / "        " + Cs(S"\v") / " "),
-    spacing      = V"whitespace"^1,
-    blank_line   = V"whitespace"^0 * V"eol",
-
-    rest_of_line = (1 - V"eol")^1,
-
-    eol          = S"\r\n",
-    eof          = V"eol"^0 * -P(1),
-
-    end_block    = V"blank_line"^1 * V"eof"^-1
-                 + V"eof"
-                 ,
-
-    -- diverse markup character sets
-    adornment_char     = S[[!"#$%&'()*+,-./:;<=>?@[]^_`{|}~]] + P[[\\]], -- headings
-    bullet_char        = S"*+-" + P"•" + P"‣" + P"⁃",                    -- bullet lists
-
-    roman_numeral      = S"ivxlcdm"^1,
-    Roman_numeral      = S"IVXLCDM"^1,
-
-    angle_left         = P"<",
-    angle_right        = P">",
-    gartenzaun         = P"#",
-
-    table_intersection = P"+",
-    table_hline        = V"dash",
-    table_vline        = V"bar",
-    table_header_hline = P"=",
-}
-
---- 225 rules at 2014-02-28 with lpeg 0.12 and Luatex 0.78.3
---lpeg.print(rst_parser)
---lpeg.ptree(rst_parser)
---os.exit()
-
-local file_helpers = { }
-
-function file_helpers.strip_BOM (raw)
-    if stringmatch (raw, "^\239\187\191") then
-        return stringsub (raw, 4)
-    end
-    return raw
-end
-
---- Tab expansion: feature request by Philipp A.
-do
-    local shiftwidth = rst.shiftwidth
-    local stringrep  = string.rep
-    local position   = 1
-
-    local reset_position     = function ()  position = 1 return "\n" end
-    local increment_position = function (c) position = position + 1 return c end
-    local expand_tab         = function ()
-        local expand = (shiftwidth - position) % shiftwidth + 1
-        position     = position + expand
-        return stringrep(" ", expand)
-    end
-
-    local tab      = S"\t\v" / expand_tab
-    local utfchar  = utfchar / increment_position
-    local eol      = P"\n"   / reset_position
-    local p_expand = Cs((tab + eol + utfchar)^1)
-
-    function file_helpers.expandtab (raw)
-        position = 1
-        return lpegmatch (p_expand, raw)
-    end
-end
-
---- Spotted by Philipp A.
-function file_helpers.insert_blank (raw)
-    if not stringfind (raw, "\n%s$") then
-        return raw .. "\n\n"
-    end
-    return raw
-end
-
-function file_helpers.crlf (raw)
-    if stringfind (raw, "\r\n") then
-        return stringgsub (raw, "\r\n", "\n")
-    end
-    return raw
-end
-
-local function load_file (name)
-    f = assert(ioopen(name, "r"), "Not a file!")
-    if not f then return 1 end
-    local tmp = f:read("*all")
-    f:close()
-
-    local fh = file_helpers
-    if thirddata.rst.strip_BOM then
-        tmp = fh.strip_BOM(tmp)
-    end
-    if thirddata.rst.crlf then
-        tmp = fh.crlf(tmp)
-    end
-    if thirddata.rst.expandtab then
-        tmp = fh.expandtab(tmp)
-    end
-    return fh.insert_blank(tmp)
-end
-
-local function save_file (name, data)
-    f = assert(ioopen(name, "w"), "Could not open file "..name.." for writing! Check its permissions")
-    if not f then return 1 end
-    f:write(data)
-    f:close()
-    return 0
-end
-
-local function get_setups (inline)
-    local optional_setups = optional_setups
-    local setups = ""
-    if not inline then
-        setups = setups .. [[
-%+-------------------------------------------------------------+%
-%|                           Setups                            |%
-%+-------------------------------------------------------------+%
-% General                                                       %
-%---------------------------------------------------------------%
-
-]]
-    end
-
-    setups = setups .. [[
-\setupcolors[state=start]
-%% Interaction is supposed to be handled manually.
-%%\setupinteraction[state=start,focus=standard,color=darkgreen,contrastcolor=darkgreen]
-\setupbodyfontenvironment [default]  [em=italic]
-\sethyphenatedurlnormal{:=?&}
-\sethyphenatedurlbefore{?&}
-\sethyphenatedurlafter {:=/-}
-
-\doifundefined{startparagraph}{% -->mkii
-  \enableregime[utf]
-  \let\startparagraph\relax
-  \let\stopparagraph\endgraf
-}
-
-]]
-    for item, _ in next, state.addme do
-        local f = optional_setups[item]
-        setups = f and setups .. f() or setups
-    end
-    if not inline then
-        setups = setups .. [[
-
-
-%+-------------------------------------------------------------+%
-%|                            Main                             |%
-%+-------------------------------------------------------------+%
-
-\starttext
-]]
-    end
-    return setups
-end
-
-function thirddata.rst.standalone (infile, outfile)
-    local testdata = load_file(infile)
-    if testdata == 1 then return 1 end
-
-    local processeddata = lpegmatch (rst_parser, testdata)
-    local setups = get_setups(false)
-
-    processeddata = setups .. processeddata .. [[
-
-\stoptext
-
-%+-------------------------------------------------------------+%
-%|                       End of Document                       |%
-%+-------------------------------------------------------------+%
-
-% vim:ft=context:tw=65:shiftwidth=2:tabstop=2:set expandtab
-]]
-
-    if processeddata then
-        save_file(outfile, processeddata)
-    else
-        return 1
-    end
-    return 0
-end
-
-local p_strip_comments
-do
-    local Cs, P = lpeg.Cs, lpeg.P
-    local percent = P"%"
-    local eol     = P"\n"
-    local comment = percent * (1 - eol)^0 * eol / "\n"
-    p_strip_comments = Cs((comment + 1)^0)
-end
-
-function thirddata.rst.do_rst_file(fname)
-    local raw_data   = load_file(fname)
-    local processed  = lpegmatch (rst_parser, raw_data)
-    local setups     = get_setups(false)
-    local tmp_file   = tex.jobname .. "–rst_temporary.tex.tmp"
-
-    if processed then
-        processed = lpegmatch (p_strip_comments, setups..processed.."\n\\stoptext\n")
-        save_file(tmp_file, processed)
-        context.input("./"..tmp_file)
-    end
-end
-
-local rst_inclusions = { }
-local rst_incsetups  = { }
-function thirddata.rst.do_rst_inclusion (iname, fname)
-    local raw_data   = load_file(fname)
-    local processed  = lpegmatch (rst_parser, raw_data)
-    local setups     = get_setups(true)
-
-    local incnr    = #rst_incsetups  + 1
-    local tmp_file = tex.jobname .. stringformat("–rst_inclusion-%d.tex.tmp", incnr)
-
-    if processed then
-        processed = lpegmatch (p_strip_comments, processed)
-        save_file(tmp_file, processed)
-        rst_inclusions[iname] = tmp_file
-        rst_incsetups[#rst_incsetups +1] = setups
-    end
-end
-
-function thirddata.rst.do_rst_setups ()
-    local out = table.concat(rst_incsetups)
-    --context(out) --- why doesn’t this work?
-    local tmp_file = tex.jobname .. "–rst_setups.tex.tmp"
-    save_file(tmp_file, out)
-    context.input(tmp_file)
-end
-
-function thirddata.rst.get_rst_inclusion (iname)
-    if rst_inclusions[iname] then
-        context.input(rst_inclusions[iname])
-    else
-        context(stringformat([[{\bf File for inclusion “%s” not found.}\par ]], iname))
-    end
-end
-
-function thirddata.rst.do_rst_snippet(txt)
-    local processed  = lpegmatch (rst_parser, txt)
-    local setups     = get_setups(true)
-    local tmp_file   = tex.jobname .. "–rst_temporary.tex.tmp"
-
-    if processed then
-        warn("·cs·",txt)
-        processed = lpegmatch (p_strip_comments, setups..processed)
-        save_file(tmp_file,processed)
-        context.input("./"..tmp_file)
-    else
-        warn("·cs·",txt)
-        context.par()
-        context("{\\bf context-rst could not process snippet.\\par}")
-        context.type(txt)
-        context.par()
-    end
-end
-
-local usage_info = [[
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-                           rstConTeXt
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-Functionality has been moved, the reST converter can now be
-accessed via mtxrun:
-
-    $mtxrun --script rst
-
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-]]
-
-local function main()
-    iowrite("\n"..usage_info.."\n")
-    return -1
-end
-
-if not (context or scripts) then
-    return main()
-end
-
---- vim:tw=79:et:sw=4:ts=8:sts=4
diff --git a/mod/tex/context/third/rst/rst_setups.lua b/mod/tex/context/third/rst/rst_setups.lua
deleted file mode 100644
index 31f314e..0000000
--- a/mod/tex/context/third/rst/rst_setups.lua
+++ /dev/null
@@ -1,377 +0,0 @@
-#!/usr/bin/env texlua
---------------------------------------------------------------------------------
---         FILE:  rst_setups.lua
---        USAGE:  called by rst_parser.lua
---  DESCRIPTION:  Complement to the reStructuredText parser
---       AUTHOR:  Philipp Gesang (Phg), <phg42.2a@gmail.com>
---      CHANGED:  2013-06-03 18:52:29+0200
---------------------------------------------------------------------------------
---
-
-local optional_setups   = { }
-thirddata.rst_setups    = optional_setups
-local rst_directives    = thirddata.rst_directives
-local rst_context       = thirddata.rst
-
-local stringformat      = string.format
-local stringstrip       = string.strip
-local stringgsub        = string.gsub
-
-function optional_setups.footnote_symbol ()
-    local setup = [[
-%---------------------------------------------------------------%
-% Footnotes with symbol conversion                              %
-%---------------------------------------------------------------%
-\definenote[symbolnote][footnote]
-\setupnote [symbolnote][way=bypage,numberconversion=set 2]
-]]
-    return setup
-end
-
-function optional_setups.footnotes ()
-    local tf = rst_context.state.footnotes
-    local fn = [[
-
-%---------------------------------------------------------------%
-% Footnotes                                                     %
-%---------------------------------------------------------------%
-]]
-    local buffer = [[
-
-%% %s
-\startbuffer[%s]
-%s\stopbuffer
-]]
-    
-    for nf, note in next, tf.numbered do
-        fn = fn .. stringformat(buffer, "Autonumbered footnote", "__footnote_number_"..nf, note)
-    end
-    for nf, note in next, tf.autolabel do
-        fn = fn .. stringformat(buffer, "Labeled footnote", "__footnote_label_"..nf, note)
-    end
-    for nf, note in next, tf.symbol do
-        fn = fn .. stringformat(buffer, "Symbol footnote", "__footnote_symbol_"..nf, note)
-    end
-    return fn
-end
-
-function optional_setups.references ()
-    local refs  = rst_context.collected_references
-    local crefs = rst_context.context_references
-    local arefs = rst_context.anonymous_set
-    
-    local function urlescape (str)
-        return str:gsub("#", "\\#")
-    end
-
-    local function resolve_indirect (r)
-        if r and r:match(".*_$") then -- pointing elsewhere
-            local look_me_up = r:match("^`?([^`]*)`?_$")
-            local result = resolve_indirect (refs[look_me_up])
-            if result then
-                return result
-            else
-                if rst_context.structure_references[look_me_up] then
-                    -- Internal link, no useURL etc.
-                    return false
-                end
-            end
-        end
-        return r
-    end
-
-    local refsection = [[
-
-%---------------------------------------------------------------%
-% References                                                    %
-%---------------------------------------------------------------%
-
-]]
-    local references = {}
-    local ref_keys   = {}
-    for ref, target in next, refs do
-        ref_keys[#ref_keys+1] = [[__target_]] .. rst_context.whitespace_to_underscore(ref)
-        target = resolve_indirect(target)
-        if target ~= false then
-            ref_text = ref
-            if arefs[ref_text] then
-                ref_text = rst_context.anonymous_links[tonumber(arefs[ref_text])]
-            end
-            references[#references+1] = stringformat([[
-\useURL[__target_%s] [%s] []   [%s] ]], rst_context.whitespace_to_underscore(ref), urlescape(target), ref_text)
-        end
-    end
-    refsection = refsection .. table.concat(references, "\n")
-    -- this is needed in order to select the right reference command later
-    refsection = refsection .. "\n\n" .. [[\def \RSTexternalreferences{]] .. table.concat(ref_keys, ",") .. [[}
-
-% #1 target name, #2 link text
-\def\RSTchoosegoto#1#2{%
-  \rawdoifinsetelse{#1}{\RSTexternalreferences}%
-    {\from[#1]}%
-    {\goto{#2}[#1]}%
-}
-]]
-
-    return refsection
-end
-
-function optional_setups.substitutions ()
-    local directives = rst_directives
-    local substitutions = [[
-
-%---------------------------------------------------------------%
-% Substitutions                                                 %
-%---------------------------------------------------------------%
-]]
-    local rs = rst_context.substitutions
-    for name, content in next, rs do
-        local id, data = content.directive, content.data
-        local directive = directives[id]
-        if directive then
-            substitutions = substitutions .. directive(name, data)
-        else
-            err(id .. " does not exist.")
-        end
-    end
-    return substitutions
-end
-
-function optional_setups.directive ()
-    --local dirstr = [[
-
---%---------------------------------------------------------------%
---% Directives                                                    %
---%---------------------------------------------------------------%
---]]
-    --return dirstr
-    return ""
-end
-
-function optional_setups.blockquote ()
-    return [[
-
-%---------------------------------------------------------------%
-% Blockquotes                                                   %
-%---------------------------------------------------------------%
-\setupdelimitedtext  [blockquote][style={\tfx}] % awful placeholder
-\definedelimitedtext[attribution][blockquote]
-\setupdelimitedtext [attribution][style={\tfx\it}]
-]]
-end
-
-function optional_setups.deflist ()
-    return [[
-
-%---------------------------------------------------------------%
-% Definitionlist                                                %
-%---------------------------------------------------------------%
-\def\startRSTdefinitionlist{
-  \bgroup
-  \def      \RSTdeflistterm##1{{\bf ##1}}
-  \def\RSTdeflistclassifier##1{\hbox to 1em{\it ##1}}
-  \def\RSTdeflistdefinition##1{%
-    \startnarrower[left]
-    ##1%
-    \stopnarrower}
-  \def\RSTdeflistparagraph ##1{%
-    \startparagraph{%
-      \noindentation ##1
-    \stopparagraph}
-  }
-}
-
-\let\stopRSTdefinitionlist\egroup
-]]
-end
-
-function optional_setups.lines ()
-    return [[
-
-%---------------------------------------------------------------%
-% Lines environment (line blocks)                               %
-%---------------------------------------------------------------%
-
-\setuplines[%
-  space=on,%
-  before={\startlinecorrection\blank[small]},%
-  after={\blank[small]\stoplinecorrection},%
-]
-]]
-end
-
-function optional_setups.breaks ()
-    return [[
-
-%---------------------------------------------------------------%
-% Fancy transitions                                             %
-%---------------------------------------------------------------%
-
-% Get Wolfgang’s module at <https://bitbucket.org/wolfs/fancybreak>.
-\usemodule[fancybreak]
-\setupfancybreak[symbol=star]
-]]
-end
-
-function optional_setups.fieldlist ()
-    return [[
-
-%---------------------------------------------------------------%
-% Fieldlists                                                    %
-%---------------------------------------------------------------%
-
-\def\startRSTfieldlist{%
-  \bgroup%
-  \unexpanded\def\RSTfieldname##1{\bTR\bTC ##1\eTC}
-  \unexpanded\def\RSTfieldbody##1{\bTC ##1\eTC\eTR}
-%
-  \setupTABLE[c][first] [background=color, backgroundcolor=grey, style=\bf]
-  \setupTABLE[c][2]     [align=right]
-  \setupTABLE[c][each]  [frame=off]
-  \setupTABLE[r][each]  [frame=off]
-  \bTABLE[split=yes,option=stretch]
-  \bTABLEhead
-  \bTR
-   \bTH  Field       \eTH
-   \bTH  Body        \eTH
-  \eTR
-  \eTABLEhead
-  \bTABLEbody
-}
-
-\def\stopRSTfieldlist{%
-  %\eTABLEbody % doesn't work, temporarily moved to rst_context.field_list()
-  \eTABLE
-  \egroup%
-}
-]]
-end
-
-function optional_setups.dbend ()
-    -- There's just no reason for not providing this.
-    optional_setups.dbend_done = true
-    return [[
-%---------------------------------------------------------------%
-% Dangerous bend                                                %
-%---------------------------------------------------------------%
-
-\loadmapfile [manfnt.map]
-\definefontsynonym [bends] [manfnt]
-
-\def\GetSym#1{\getglyph{bends}{\char#1}}
-
-\startsymbolset [Dangerous Bends]
-    \definesymbol [dbend]       [\GetSym{127}]
-    \definesymbol [lhdbend]     [\GetSym{126}]
-    \definesymbol [lhdbend]     [\GetSym{0}]
-\stopsymbolset
-
-\setupsymbolset [Dangerous Bends]
-
-]]
-end
-
-function optional_setups.caution ()
-    local result = ""
-    --if not optional_setups.dbend_done then
-        --result = result .. optional_setups.dbend()
-    --end
-    return result .. [[
-%---------------------------------------------------------------%
-% Caution directive                                             %
-%---------------------------------------------------------------%
-
-\usemodule[lettrine]
-
-\setbox0=\hbox{\symbol[dbend]}
-\newskip\RSTbendskip
-\RSTbendskip=\wd0
-\advance\RSTbendskip by 1em % These two lines should add
-\advance\RSTbendskip by 1pt % 13.4pt in mkiv and 13.14983pt in mkii
-                            % to make the indent equal to the indent
-                            % of the “danger” directive.
-                            % (2*(width)dbend + (kern)1pt + 1em
-
-\def\startRSTcaution{%
-\startparagraph
-\dontleavehmode\lettrine[Lines=2,Raise=.6,Findent=\RSTbendskip,Nindent=0pt]{\symbol[dbend]}{}%
-}
-
-\let\stopRSTcaution\stopparagraph
-
-]]
-
-end
-
-function optional_setups.danger ()
-    local result = ""
-    --if not optional_setups.dbend_done then
-        --result = result .. optional_setups.dbend()
-    --end
-    return result .. [[
-%---------------------------------------------------------------%
-% Danger directive                                              %
-%---------------------------------------------------------------%
-
-\usemodule[lettrine]
-
-\def\startRSTdanger{%
-\startparagraph
-\lettrine[Lines=2,Raise=.6,Findent=1em,Nindent=0pt]{\symbol[dbend]\kern 1pt\symbol[dbend]}{}%
-}
-
-\let\stopRSTdanger\stopparagraph
-
-]]
-
-end
-
-function optional_setups.citations ()
-    local cit = [[
-%---------------------------------------------------------------%
-% Citations                                                     %
-%---------------------------------------------------------------%
-\setupbibtex[database=\jobname]
-]]
-    
-
-    return cit
-end
-
-function optional_setups.citator ()
-    local cit = [[
-%---------------------------------------------------------------%
-% Citator Options                                               %
-%---------------------------------------------------------------%
-\usemodule[citator]
-\loadbibdb{\jobname.bib}
-\setupcitator[sortmode=authoryear]
-\setupcite[mainmode=authoryear]
-
-\startbuffer[bibliography]
-\chapter{References}
-\setupbodyfont[small]
-\bibbykey{shorthand}{all}{author}
-\stopbuffer
-
-\prependtoks \getbuffer[bibliography] \to \everystoptext
-]]
-
-    return cit
-end
-
-function optional_setups.image ()
-    local image = [[
-
-%---------------------------------------------------------------%
-% images                                                        %
-%---------------------------------------------------------------%
-\setupexternalfigure[location={local,global,default}]
-
-]]
-    return image
-end
-
-return optional_setups
-
--- vim:ft=lua:sw=4:ts=4:expandtab:tw=80
diff --git a/mod/tex/context/third/rst/t-rst.mkiv b/mod/tex/context/third/rst/t-rst.mkiv
deleted file mode 100644
index 48801b5..0000000
--- a/mod/tex/context/third/rst/t-rst.mkiv
+++ /dev/null
@@ -1,241 +0,0 @@
-%D \module [
-%D         file=t-rst,
-%D      version=0.6 ‘It’s the Arts’,
-%D        title=\CONTEXT\ User Module,
-%D     subtitle=reStructuredText,
-%D       author=Philipp Gesang,
-%D         date=\currentdate,
-%D    copyright=Philipp Gesang,
-%D      license=2-clause BSD,
-%D ]
-
-%M \usemodule  [rst]
-%M \usemodule  [int-load]
-%M \loadsetups [t-letterspace.xml]
-
-%C Read the license conditions in the file \type{COPYING}.
-
-%M \definecolor [gutenred] [x=bf221f] % rubrication from digitized Göttingen Gutenberg bible
-%M \setupinteraction [contrastcolor=gutenred,color=gutenred]
-%M
-%M \define\beautifyshowsetups{%
-%M   \unexpanded \def \setupnumfont  {\rm}%
-%M   \unexpanded \def \setuptxtfont  {\rm}%
-%M   \unexpanded \def \setupintfont  {\rm\sc\Word}%
-%M   \unexpanded \def \setupvarfont  {\rm\it}%
-%M   \unexpanded \def \setupoptfont  {\rm\it}%
-%M   \unexpanded \def \setupalwcolor {gutenred}%
-%M   \unexpanded \def \setupoptcolor {gutenred}%
-%M   \defineframedtext [setuptext] [
-%M     frame=off,
-%M     background=color,
-%M     backgroundcolor=gray:2,
-%M     width=\hsize,
-%M     height=fit,
-%M     align=right,
-%M     offset=0.75em,
-%M   ]%
-%M }
-%M 
-%M \let \Oldshowsetup \showsetup
-%M 
-%M \define [1] \showsetup {%
-%M   \bgroup \beautifyshowsetups%
-%M     \Oldshowsetup{#1}%
-%M   \egroup%
-%M }
-
-\writestatus{loading}{ConTeXt User Module / reStructuredText}
-
-\unprotect
-
-\startinterface all
-  \setinterfacevariable {RST} {RST}
-\stopinterface
-
-\definenamespace [\v!RST] [
-  type=module,
-  comment=reStructuredText module,
-  version=0.6,
-  name=\v!RST,
-  style=\v!no,
-  command=\v!yes,
-  setup=\v!list,
-  parent=\v!RST,
-]
-
-%D Loading the reStructuredText parser.
-\ctxloadluafile{rst_parser}
-
-%D Easy way to define a global test setting. Activated
-%D by \type{\usemodule[rst][test=yes]}.
-
-\startmoduletestsection
-  \ctxlua{thirddata.rst_helpers.rst_debug = true}
-\stopmoduletestsection
-
-%D To process inline reST markup we’ll have to reset all catcodes
-%D except for grouping, escaping and cs arguments.
-
-\newcatcodetable   \RST_catcodes
-\startcatcodetable \RST_catcodes
-  \catcode`\^^I = 12
-  \catcode`\^^M = 12
-  \catcode`\^^L = 12
-  \catcode`\    = 12
-  \catcode`\^^Z = 12
-  \catcode`\\   =  0
-  \catcode`\%   = 12
-  \catcode`\#   =  6
-  \catcode`\_   = 12
-  \catcode`\^   = 12
-  \catcode`\&   = 12
-  \catcode`\|   = 12
-  \catcode`\{   =  1
-  \catcode`\}   =  2
-  \catcode`\~   = 12
-  \catcode`\$   = 12
-\stopcatcodetable
-
-%D \section {User-level Commands}
-%D
-%D \subsection{Typesetting reST-Files}
-%D
-%D \macros
-%D    {typesetRSTfile}
-%D
-%D This command loads and processes an \type{*.rst} file.
-%D All necessary setups for the elements to be used (e.g. tables)
-%D have to be specified {\em before} this macro is called.
-%D As \type{\typesetRSTfile} is intended to process a single file
-%D only, it will handle \type{\start|stoptext} automatically.
-%D Thus, the user should never supply any of these manually,
-%D neither before nor after \type{\typesetRSTfile}.
-%D
-%D We now handle rogue utf-8 byte order marks on demand, just set
-%D the optional parameter \type{stripBOM} to {\em true}.
-%D
-%D There also is an option \type{expandtab} to convert tabs
-%D (ascii 0x09) to indents prior to converting reST input. The
-%D expansion width defaults to {\em 4} and can be configured
-%D through the parameter \type{shiftwidth} (takes an integer).
-%D
-%D \showsetup{typesetRSTfile}
-
-\def\do_typeset_RST_file[#1]#2{%
-  \iffirstargument
-    \getparameters[RST][#1]%
-    \doifdefined{RSTstripBOM}  {\ctxlua{thirddata.rst.strip_BOM  = \RSTstripBOM}}%
-    \doifdefined{RSTexpandtab} {\ctxlua{thirddata.rst.expandtab  = \RSTexpandtab}}%
-    \doifdefined{RSTshiftwidth}{\ctxlua{thirddata.rst.shiftwidth = \RSTshiftwidth}}%
-  \fi
-  \ctxlua{thirddata.rst.do_rst_file("#2")}%
-}
-
-\def\typesetRSTfile{%
-  \dosingleempty\do_typeset_RST_file%
-}
-
-%D \subsection{Typesetting Inline Snippets}
-%D
-%D reST markup can be handy in situations where \CONTEXT\ markup
-%D would result in unappropriately verbose source code, e.g. when
-%D typesetting tables with simple layout.
-%D
-%D \macros
-%D    {RST,startRST}
-%D
-%D The environment \type{\[start|stop]RST} and the macro
-%D \type{\RST} allow access to reST-parser from inside a
-%D \CONTEXT\ document when the module is loaded.
-%D
-%D \showsetup{RST}
-%D \showsetup{startRST}
-
-% Wolfgang’s code below.
-
-\unexpanded \def \startRST{%
-  \begingroup
-  \setcatcodetable \RST_catcodes
-  \do_start_RST%
-}
-
-\let \stopRST \relax
-
-\def \do_start_RST#1\stopRST{%
-  \endgroup%
-  \ctxlua{thirddata.rst.do_rst_snippet(\!!bs#1\!!es)}%
-}
-
-\def \RST{%
-  \begingroup
-  \setcatcodetable \RST_catcodes
-  \do_RST%
-}
-
-\def \do_RST#1{%
-  \endgroup%
-  \ctxlua{thirddata.rst.do_rst_snippet(\!!bs#1\!!es)}%
-}
-
-%D \subsection{Including multiple reST files}
-%D
-%D \macros
-%D    {defineRSTinclusion,startRSTproject,RSTinclusion}
-%D
-%D When content is split among multiple files, these macros allow
-%D for including them in arbitrary order. Note that setups from
-%D previous includes, unless overwritten, will remain defined.
-%D \type{\defineRSTinclusion} takes three arguments: the first one
-%D will be the identifier that can be used to refer to the actual
-%D inclusion, which is specified via the second argument as a
-%D filename. The third optional argument receives the usual
-%D setups \type{stripBOM}, \type{expandtab} and
-%D \type{shiftwidth}.
-%D
-%D Defined inclusions can be typeset only within an the
-%D \type{\startRSTproject} environment using the macro
-%D \type{RSTinclusion}. Between those inclusion all kinds of
-%D \TEX\ code except for \type{\starttext} and \type{\stoptext}
-%D are permitted.
-
-\def\do_define_RST_inclusion[#1][#2][#3]{%
-  \ifthirdargument
-    \getparameters[RST][#3]%
-    \doifdefined{RSTstripBOM}  {\ctxlua{thirddata.rst.strip_BOM  = \RSTstripBOM}}%
-    \doifdefined{RSTexpandtab} {\ctxlua{thirddata.rst.expandtab  = \RSTexpandtab}}%
-    \doifdefined{RSTshiftwidth}{\ctxlua{thirddata.rst.shiftwidth = \RSTshiftwidth}}%
-  \fi
-  \ifsecondargument
-    \ctxlua{thirddata.rst.do_rst_inclusion("#1", "#2")}%
-  \fi%
-}
-
-\def\defineRSTinclusion{%
-  \dotripleempty\do_define_RST_inclusion%
-}
-
-\def\do_RST_inclusion[#1]{%
-  \iffirstargument
-    \ctxlua{thirddata.rst.get_rst_inclusion("#1")}%
-  \fi%
-}
-
-\def\do_RST_setups{%
-  \ctxlua{thirddata.rst.do_rst_setups()}%
-}
-
-\def\startRSTproject{
-  \begingroup
-  \def\RSTinclusion{\dosingleempty\do_RST_inclusion}
-  \do_RST_setups
-  \starttext%
-}
-
-\def\stopRSTproject{
-  \stoptext \endgroup \endinput
-}
-
-\protect \endinput
-
-% vim:ft=context:sw=2:ts=2