groff
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Groff] eqn converts "lambda" to "&lamda;"


From: Ted Harding
Subject: Re: [Groff] eqn converts "lambda" to "&lamda;"
Date: Tue, 15 Dec 2015 17:01:54 -0000 (GMT)

Hi Ingo,
You have replied to a message from Brian McGuinness, 13/12/2015.
I do not seem to have received such a message myself, nor can
I find it in the groff archive (though your reply is there).
Maybe it was a personal message to you from Brian, and you copied
your reply to the groff list.

If so, could you forward also Brian's original mressage to the list?
It looks as though the parts you did not retain in your reply may
raise other points of interest to us!

With best wishes,
Ted.


On 15-Dec-2015 12:10:17 Ingo Schwarze wrote:
> Hi Brian,
> 
> McGuinness, Brian wrote on Sun, Dec 13, 2015 at 07:15:21PM -0500:
> 
>> I was experimenting with eqn because I think it
>> will prove very useful for generating MathML
>> code easily.
> 
> Groff is certainly a good choice for processing eqn(1)
> input, in particular when aiming for typeset output,
> for example PostScript or PDF.
> I haven't experimented with GNU eqn MathML output, though.
> 
> For converting eqn input to MathML output, you might also
> wish to look at mandoc as an alternative: http://mdocml.bsd.lv/
> It may or may not be sufficient or better for your purpose,
> and i would be interested in any feedback.
> 
> One thing that may or may not be relevant for your use case is
> speed.  Here are example numbers from my OpenBSD-current system:
> 
>   $ cd /usr/X11R6/man/man3/
>   $ time eqn -TMathML glMultMatrix.3 | groff -man -Txhtml > /dev/null 2>&1
>  0m17.67s real     0m15.92s user     0m01.45s system
>   $ time mandoc -Thtml glMultMatrix.3 > /dev/null
>  0m00.00s real     0m00.01s user     0m00.00s system
>   $ time sh -c 'for f in `jot 1000`; do mandoc -Thtml glMultMatrix.3 \
>     > /dev/null; done'
>  0m07.25s real     0m02.83s user     0m03.33s system
> 
> Mandoc does HTML5 and MathML by default and does not need any options
> or preprocessors.
> 
> Being faster by a factor of more than 2000 (two thousand)
> may be relevant for some applications.
> 
> Yours,
>   Ingo
> 

-------------------------------------------------
E-Mail: (Ted Harding) <address@hidden>
Date: 15-Dec-2015  Time: 17:01:50
This message was sent by XFMail
-------------------------------------------------


reply via email to

[Prev in Thread] Current Thread [Next in Thread]