pmathml: an extension module for p-MathML

The module pmathml in http://gloss.bham.ac.ukmv/pmath allows presentation MathML to be entered in XML and XHTML documents in a fairly clean and convenient way. pmathml is an extension to xml.modes and so can be used for standalone MathML documents, or MathML embedded in XML or in XHTML. Two MV files http://gloss.bham.ac.ukmv/pmath/xml-pmathml.mv and http://gloss.bham.ac.ukmv/pmath/html-pmathml.mv are provided for these main applications.

The goals of pmathml are to provide for p-MathML input using as far as possible the standard entity names and standard 7-bit characters for input. That's not to say other unicode characters can't be used (they can) but if other unicode characters are used the automatic provision for choosing the correct math token element (mi, mn, mo, etc) may not work. (This could be provided later in a further extension. This is TO DO.)

A further extension, pmathx, provides some convenient extensions to pmathml but these are not standard and more likely to change later.

Math mode is initiated with math. MathML entities are given the prefix m: automatically, and this need not be typed. (TO DO: provide a way to change this prefix.)

The main mathtoken elements are available as you would expect, with standard GLOSS-xml syntax. For exampe, mn [234.456]. The elements mn, mi, mo, etc., take the usual attributes. However most characters and many element names have defaults, so mn, mi, mo, etc. can often be omitted. Thus math a is equivalent to math { mi [a] } and math 3.1415926535 is equivalent to math { mn [3.1415926535] }. A list of the main defaults follows.

elements a to z and A to Z are equivalent to { mi [NAME] }, where NAME is the name of the element.
unicode characters 0 to 9 are equivalent to { mn [CHAR] }, where CHAR is the character.
elements NAME with name corresponding to a standard MathML entity (see http://www.w3.org/2003/entities/) will be converted to character references and wrapped with mo, mn or mi.
unicode characters a to z and A to Z are equivalent to { mi [CHAR] }, where CHAR is the character.
(most) other unicode character will be wrapped with mo by default.
The b64 token = will be wrapped with mo.
int, fp and hex tokens will be wrapped with mn.
char and string tokens will be wrapped with ms.
eref and cref tokens will be wrapped with mo by default.

This behaviour can be changed by an extension module, hooking to token-dictionary in namespace http://gloss.bham.ac.uk/mv/pmath/pmathml.

Apart from these, the other usual p-MathML constructs should work as expected. (See http://www.w3.org/Math/). This includes:

mi mn mo mtext mspace ms mglyph mrow mfrac msqrt mroot
mstyle merror mpadded mphantom mfenced menclose
msub msup msubsup munder mover munderover mmulitiscripts mprescripts
mtable maligngroup malignmark maction mlabeledtr mtr mtd

author home contents next

This page is copyright. Web page design and creation by GLOSS.