W3C home > Mailing lists > Public > www-math@w3.org > September 2015

Re: lmoustache/rmoustache as stretchy delimiters

From: William F Hammond <hammond@csc.albany.edu>
Date: Tue, 01 Sep 2015 13:31:53 -0400
To: www-math@w3.org
Message-ID: <i7io7u59me.fsf@hilbert.math.albany.edu>
Bruce Miller writes:

...
> On 08/30/2015 07:30 PM, David Carlisle wrote:
...
>> the TeXBook says:
>>
>> \danger ......
>> You can also use ^|\lgroup| and ^|\rgroup|,
>> which are constructed from braces without the middle parts; and
>> ^|\lmoustache| and ^|\rmoustache|, ^^{moustaches}
>> which give you the top and bottom halves of large braces. For example,
>> here are the |\Big| and |\bigg| versions of....
>
> That quote justifies the dubious assignment to the unicode bracket
> fragments; they apparently were intended to be merely fragments, rather
> than stand-alone delimiters. OTOH, they're also declared as \delimiter,
> so they can be used stand-alone with the expectation of stretchiness.
>
> Independently, of course, of whether anyone _should_ do that....

First, let me say I don't recall ever seeing the moustaches used as
standalone delimiters.

Second, there is this bit in http://unicode.org/Public/UNIDATA/NamesList.txt
(version 8.0):

------
     @               Bracket pieces
     23B0    UPPER LEFT OR LOWER RIGHT CURLY BRACKET SECTION
             = left moustache
     23B1    UPPER RIGHT OR LOWER LEFT CURLY BRACKET SECTION
             = right moustache
------

which may be seen as an acknowledgement of the TeXBook.

Third, it breaks the design of XML (and SGML) when a
processor or a renderer makes decisions on nuggets of CDATA.
(SDATA, which is excluded from XML, was for that.)  The
better way to handle delimiters for pMathML is to use the
attributes of <mfenced>, ignoring the spec directive that
<mfenced> should always be resolved in a certain way to an
<mrow>.

Fourth, relative to various experiments in rendering pMathML
solely with CSS (as CSS continues to become more capable),
there is an idea I raised a year ago at TUG 2014 that it can
be useful to think of delimiters as border decorations.  To
the extent that one might be able in the years ahead to go
forward with this, it will work better with <mfenced> than
with the mrow equivalent.

                            -- Bill
Received on Tuesday, 1 September 2015 17:32:18 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 1 September 2015 17:32:18 UTC