Re: Question about defining a math macro

[Aditya Mahajan <adityam@umich.edu>]

On Sat, 24 Feb 2007, Hans Hagen wrote:

> Aditya Mahajan wrote:
>> On Sat, 24 Feb 2007, Taco Hoekwater wrote:
>>
>>
>>> Aditya Mahajan wrote:
>>>
>>>> Hi,
>>>>
>>>> Can I define a math macro, say \EXP that does the following:
>>>>
>>>> 1) \EXP{something} gives me, say E(something)
>>>> 2) \EXP_a^b{something} gives my E_a^b(something)
>>>>
>>>> That is, I want to write a macro that will only be used in the math
>>>> mode, that does something with its argument. But, I also want it to
>>>> handle _ and ^ if they are present.
>>>>
>>>> The only method that I can think of is to check _ or ^ with
>>>> \ifnextcharacter and go through all 5 cases (nothing, only _, only ^,
>>>> _ followed by ^, ^ followed by _) but this seems very inefficient. Is
>>>> there a better way?
>>>>
>>> Actually, there is, by using a bit of black magic in the
>>> macro definition:
>>>
>>>   \def\EXP#1#{\bgroup\def\EXParg{#1}\doEXP}
>>>   \def\doEXP#1{\mfunction{E}\EXParg({\rm #1})\egroup}
>>>
>>> That extra hash mark has the effect of putting everything
>>> upto the next explicit left brace into #1.
>>>
>>
>> Wow Taco. This is great.
>>
> some applications of this mechanism can be found in the context source code (supp-box etc)

Yes, those were parts of the core that I could never understand. Now, 
I do understand them a little.

Thanks,
Aditya