Jan Nieuwenhuizen | 1 Sep 08:42 2004
Picon

Re: Proposed patch: Unroll installation loops

Matthias Neeracher writes:

> lilypond uses the $(foreach) facility of gnumake in several places in
> its installation. While I'm sure this is a little bit more efficient,
> it creates problems on MacOS, where due to the fink build setup, paths
> are often about 30 characters longer: Since foreach creates a single
> shell line for the whole installation command, the resulting line
> sometimes exceeded the maximum permissible argument size.

What shell are you using?  Have you considered sending a bug report for
the shell?

Jan.

--

-- 
Jan Nieuwenhuizen <janneke <at> gnu.org> | GNU LilyPond - The music typesetter
http://www.xs4all.nl/~jantien       | http://www.lilypond.org
Matthias Neeracher | 1 Sep 10:20 2004
Picon

Re: Proposed patch: Unroll installation loops


On Aug 31, 2004, at 11:42 PM, Jan Nieuwenhuizen wrote:

> Matthias Neeracher writes:
>
>> lilypond uses the $(foreach) facility of gnumake in several places in
>> its installation. While I'm sure this is a little bit more efficient,
>> it creates problems on MacOS, where due to the fink build setup, paths
>> are often about 30 characters longer: Since foreach creates a single
>> shell line for the whole installation command, the resulting line
>> sometimes exceeded the maximum permissible argument size.
>
> What shell are you using?

bash, although at the time I did this patch, it might have been zsh

>  Have you considered sending a bug report for the shell?

Your message actually prompted me to retry without the patch, and right 
now I can't reproduce the failure. However, I'm sure that it would 
return the second I omitted the patch :-)

If I may ask conversely, what is there to be gained from using 
$(foreach) ? It seems that the output produced by for is a little bit 
more compact and readable, and the results are identical.

Matthias
Jan Nieuwenhuizen | 1 Sep 11:34 2004
Picon

Re: Proposed patch: Unroll installation loops

Matthias Neeracher writes:

> bash, although at the time I did this patch, it might have been zsh

I'm sure.  It is a bug for any GNU software to impose arbitrary limits.

> If I may ask conversely, what is there to be gained from using
> $(foreach) ?  It seems that the output produced by for is a little
> bit more compact and readable, and the results are identical.

The patch is fine; until we all run GNU it's impractical to ignore
other broken softwares, if they can be so easily supported...

Jan.

--

-- 
Jan Nieuwenhuizen <janneke <at> gnu.org> | GNU LilyPond - The music typesetter
http://www.xs4all.nl/~jantien       | http://www.lilypond.org
Pedro Kroger | 1 Sep 19:15 2004
Picon

feature request: lilypond and stdin


Hi,

I'm wandering if the feature of reading from stdin (and maybe writing to
stdout) could be added to the upcoming 3.0.

Cheers,

Pedro
Han-Wen Nienhuys | 1 Sep 20:23 2004
Picon
Picon

Re: Proposed patch: Unroll installation loops

neeracher <at> mac.com writes:
> return the second I omitted the patch :-)
> 
> If I may ask conversely, what is there to be gained from using 
> $(foreach) ? It seems that the output produced by for is a little bit 
> more compact and readable, and the results are identical.

It was actually prompted by some incompatibility in bash long ago,
IIRC. I can't remember the details.

> 
> Matthias
> 
> 
> 
> _______________________________________________
> lilypond-devel mailing list
> lilypond-devel <at> gnu.org
> http://lists.gnu.org/mailman/listinfo/lilypond-devel

--

-- 

 Han-Wen Nienhuys   |   hanwen <at> xs4all.nl   |   http://www.xs4all.nl/~hanwen 
Han-Wen Nienhuys | 1 Sep 20:24 2004
Picon
Picon

feature request: lilypond and stdin

kroegerlistas <at> pedrokroeger.net writes:
> 
> Hi,
> 
> I'm wandering if the feature of reading from stdin (and maybe writing to
> stdout) could be added to the upcoming 3.0.

writing to stdout is slightly hairy, especially if there are multiple
output files, so I am not fond of this feature. stdin already works
over here. 

--
 Han-Wen Nienhuys   |   hanwen <at> xs4all.nl   |   http://www.xs4all.nl/~hanwen 
Jan Nieuwenhuizen | 1 Sep 20:25 2004
Picon

Re: feature request: lilypond and stdin

Pedro Kroger writes:

> I'm wandering if the feature of reading from stdin (and maybe writing to
> stdout) could be added to the upcoming 3.0.

This is an old feature: It should mostly work unless something new is broken.

Jan.

--

-- 
Jan Nieuwenhuizen <janneke <at> gnu.org> | GNU LilyPond - The music typesetter
http://www.xs4all.nl/~jantien       | http://www.lilypond.org
Werner LEMBERG | 1 Sep 15:01 2004
Picon

improving multibar rests


I suggest to add something like

  \times 10/8 R4*5*4 |

to the docs as an example how to produce a multibar rest within an odd
rhythm.

    Werner
Werner LEMBERG | 1 Sep 15:13 2004
Picon

improving the doc appearance


I ask the docmeister to do the following in the texinfo input files,
improving the appearance of the output if viewed with `info' (gee,
there are still dolts like me who do that :-)

  . Remove multiple spaces between words.  Things like

       this is  stretched     text

    appear as-is.  I suggest to have two spaces after a sentence
    ending (which makes the input file easier to edit with Emacs), but
    this isn't necessary.

  . Remove all unnecessary indentation.  For example, don't do this:

      this is a text
        which has some
        unnecessary indentation --

      note that this also is valid for

       <at> example
        example
          blocks
        like this example
       <at> end example

    but this (starting in the first column):

      this is a text
(Continue reading)

Werner LEMBERG | 1 Sep 22:20 2004
Picon

Re: entering music without \time


Sorry for the late reply

> To me it makes sense to notate bar lengths by example (i like in
> general when you can show what you mean by example; that way you
> don't have to learn a new complicated syntax).
> 
> But there is a major problem: How would you tell the difference
> between e.g.  3/4 and 6/8?

With a property you could define a default which can always be
overridden with \time.

    Werner

Gmane