two-mode: make-local-variable and cpu consumption

two-mode: make-local-variable and cpu consumption

[Peter Münster]

Hello,

I've two problems with two-mode.el:

1.) make-local-variable does not work with default-mode and second-modes:
After (make-local-variable 'default-mode) and
(make-local-variable 'second-modes) new assignments with setq don't have
any effect.

2.) When the current buffer is in two-mode-mode, the cpu usage of emacs is
100%.

Could anybody help please?

Thanks in advance for any hints!
Regards, Peter

-- 
http://pmrb.free.fr/contact/

Re: two-mode: make-local-variable and cpu consumption

[Lennart Borgman (gmail)]

Peter Münster wrote:
> Hello,
> 
> I've two problems with two-mode.el:
> 
> 1.) make-local-variable does not work with default-mode and second-modes:
> After (make-local-variable 'default-mode) and
> (make-local-variable 'second-modes) new assignments with setq don't have
> any effect.
> 
> 2.) When the current buffer is in two-mode-mode, the cpu usage of emacs is
> 100%.
> 
> Could anybody help please?


Hi Peter,

Perhaps you want to try MuMaMo:

   http://www.emacswiki.org/cgi-bin/wiki/MuMaMo

Re: two-mode: make-local-variable and cpu consumption

[Peter Münster]

On Sun, Feb 24 2008, Peter Münster wrote:

> I've two problems with two-mode.el:
> 
> 1.) make-local-variable does not work with default-mode and second-modes:
> After (make-local-variable 'default-mode) and
> (make-local-variable 'second-modes) new assignments with setq don't have
> any effect.

Here I'm still looking for any hint.

> 2.) When the current buffer is in two-mode-mode, the cpu usage of emacs is
> 100%.

For this one, I've found the solution: The string of the default-mode must
be exactly the same as indicated in the status-line.

With
(setq default-mode (list "ConTeXt" 'context-mode))
I got the hight cpu-usage.

But with
(setq default-mode (list "PDFConTeXt-en" 'context-mode))
the problem is solved.

Cheers, Peter

-- 
http://pmrb.free.fr/contact/

Re: two-mode: make-local-variable and cpu consumption

[Peter Münster]

On Tue, Feb 26 2008, Lennart Borgman (gmail) wrote:

> Perhaps you want to try MuMaMo:
>
>   http://www.emacswiki.org/cgi-bin/wiki/MuMaMo

Seems to be quite complicated to set up.

With two-mode, I only need to setup default-mode and second-modes:

(setq default-mode (list "PDFConTeXt-en" 'context-mode))
(setq second-modes (list (list "Lua" "\\startlua" "\\stoplua" 'lua-mode)
                         (list "MP" "\\startMP" "\\stopMP" 'metapost-mode)))

Cheers, Peter

-- 
http://pmrb.free.fr/contact/

Re: two-mode: make-local-variable and cpu consumption

[Lennart Borgman (gmail)]

Yes, it is bit complex. However that is normally nothing you see as a 
user. But I see it, I am still struggling with some problems with 
refontification after changes... ;-)

Could you perhaps send me an example of a file with those two modes?


Peter Münster wrote:
> On Tue, Feb 26 2008, Lennart Borgman (gmail) wrote:
> 
>> Perhaps you want to try MuMaMo:
>>
>>   http://www.emacswiki.org/cgi-bin/wiki/MuMaMo
> 
> Seems to be quite complicated to set up.
> 
> With two-mode, I only need to setup default-mode and second-modes:
> 
> (setq default-mode (list "PDFConTeXt-en" 'context-mode))
> (setq second-modes (list (list "Lua" "\\startlua" "\\stoplua" 'lua-mode)
>                          (list "MP" "\\startMP" "\\stopMP" 'metapost-mode)))

What markup does two-modes expect to see in the buffer then?

> Cheers, Peter
> 

Re: two-mode: make-local-variable and cpu consumption

[Peter Münster]

On Mon, Mar 03 2008, Lennart Borgman (gmail) wrote:

> Yes, it is bit complex. However that is normally nothing you see as a user. 
> But I see it, I am still struggling with some problems with refontification 
> after changes... ;-)
>
> Could you perhaps send me an example of a file with those two modes?

% engine=luatex
\startluacode
-- here is lua-mode
local function sort_table(s1, s2)
	local function cmp(a, b)
		if s2 ~= "" and a[s1] == b[s1] then
			return a[s2] < b[s2]
		else
			return a[s1] < b[s1]
		end
	end
	if s1 ~= "" then
		table.sort(thirddata.inifile.t, cmp)
	end
end
\stopluacode
\starttext
% here is context-mode
\startitemize
\item bla bla bla
\item bla bla bla
\item bla bla bla
\item bla bla bla
\stopitemize

\startMPcode
% here is metapost-mode
path p;
p := ((0,0)--(1,0)--(1,2)--(-0.5,2)--(-0.5,1)--(0,1)--cycle) scaled 1cm;
draw p;
\stopMPcode
\stoptext

>> (setq default-mode (list "PDFConTeXt-en" 'context-mode))
>> (setq second-modes (list (list "Lua" "\\startlua" "\\stoplua" 'lua-mode)
>>                          (list "MP" "\\startMP" "\\stopMP" 'metapost-mode)))
>
> What markup does two-modes expect to see in the buffer then?

The main markup is ConTeXt, inside \startlua...\stoplua it's lua and inside
\startMP...\stopMP it's metapost.

Cheers, Peter

-- 
http://pmrb.free.fr/contact/

Re: two-mode: make-local-variable and cpu consumption

[Lennart Borgman (gmail)]

Peter Münster wrote:
> On Mon, Mar 03 2008, Lennart Borgman (gmail) wrote:
> 
>> Yes, it is bit complex. However that is normally nothing you see as a user. 
>> But I see it, I am still struggling with some problems with refontification 
>> after changes... ;-)
>>
>> Could you perhaps send me an example of a file with those two modes?

Ok, thanks Peter, that finally gave me a reason to fix some code in 
MuMaMo to make these things simple. I have just uploaded a new beta of 
nXhtml here:

   http://ourcomments.org/Emacs/DL/elisp/nxhtml/beta/

If you use that beta with the following little code you will get what 
you want I believe:


(mumamo-easy-make-chunk-fun mumamo-test-lua-chunk
   "\\startluacode" "\\stopluacode" 'lua-mode)

(mumamo-easy-make-chunk-fun mumamo-test-MP-chunk
   "\\startMPcode" "\\stopMPcode" 'metapost-mode)

(define-mumamo-turn-on test-lua-and-MP
   "Just a quick test with lua and MP."
   ("lua MP test" context-mode
    (mumamo-test-lua-chunk
     mumamo-test-MP-chunk)))

Though I don't have lua-mode and context-mode here so I have not tested. 
Do you want to test?
	
> % engine=luatex
> \startluacode
> -- here is lua-mode
> local function sort_table(s1, s2)
> 	local function cmp(a, b)
> 		if s2 ~= "" and a[s1] == b[s1] then
> 			return a[s2] < b[s2]
> 		else
> 			return a[s1] < b[s1]
> 		end
> 	end
> 	if s1 ~= "" then
> 		table.sort(thirddata.inifile.t, cmp)
> 	end
> end
> \stopluacode
> \starttext
> % here is context-mode
> \startitemize
> \item bla bla bla
> \item bla bla bla
> \item bla bla bla
> \item bla bla bla
> \stopitemize
> 
> \startMPcode
> % here is metapost-mode
> path p;
> p := ((0,0)--(1,0)--(1,2)--(-0.5,2)--(-0.5,1)--(0,1)--cycle) scaled 1cm;
> draw p;
> \stopMPcode
> \stoptext
> 
>>> (setq default-mode (list "PDFConTeXt-en" 'context-mode))
>>> (setq second-modes (list (list "Lua" "\\startlua" "\\stoplua" 'lua-mode)
>>>                          (list "MP" "\\startMP" "\\stopMP" 'metapost-mode)))
>> What markup does two-modes expect to see in the buffer then?
> 
> The main markup is ConTeXt, inside \startlua...\stoplua it's lua and inside
> \startMP...\stopMP it's metapost.
> 
> Cheers, Peter
> 

Re: two-mode: make-local-variable and cpu consumption

[Peter Münster]

On Tue, Mar 04 2008, Lennart Borgman (gmail) wrote:

> (define-mumamo-turn-on test-lua-and-MP
>   "Just a quick test with lua and MP."
>   ("lua MP test" context-mode
>    (mumamo-test-lua-chunk
>     mumamo-test-MP-chunk)))
>
> Though I don't have lua-mode and context-mode here so I have not tested. Do 
> you want to test?

Ok, I've tested it quickly. Seems to work all right. The ConTeXt part has
another background color, but I suppose that this is configurable.
So thank you for your mode, I'll keep it on my disk, even if it's quite
heavy... ;)

Thanks for your help!
Peter

-- 
http://pmrb.free.fr/contact/

Re: two-mode: make-local-variable and cpu consumption

[Lennart Borgman (gmail)]

Peter Münster wrote:
> On Tue, Mar 04 2008, Lennart Borgman (gmail) wrote:
> 
>> (define-mumamo-turn-on test-lua-and-MP
>>   "Just a quick test with lua and MP."
>>   ("lua MP test" context-mode
>>    (mumamo-test-lua-chunk
>>     mumamo-test-MP-chunk)))
>>
>> Though I don't have lua-mode and context-mode here so I have not tested. Do 
>> you want to test?
> 
> Ok, I've tested it quickly. Seems to work all right. The ConTeXt part has
> another background color, but I suppose that this is configurable.

Yes.

> So thank you for your mode, I'll keep it on my disk, even if it's quite
> heavy... ;)

Thanks for testing.

> Thanks for your help!
> Peter
>