Re: is shapeSlur broken?

[Urs Liska]

Hi David,

now I tested your new function.
OK, I didn't test more than the sources you provided, but I think they give all the necessary combinations.

So my conclusion is: This is awesome!

I won't ever live without this (as long as LilyPond is concerned) anymore - as long as it won't get broken by new versions.

One idea to make it even more comfortable and generic to use would be not to hard-code the color within the function.
If one could somehow set the color outside the function one could personalize it to ones needs.

As this is kind of a library thing, I think it isn't necessary to make this settable at runtime through the function call.
Maybe one could define a variable for the color above the function, setting #black as default.
Then anybody can easily see how to adapt it even if she/he doesn't understand the function itself.
Ah, I just realized that this way one could still set the color in the music source by redefininge the variable ...

I would be happy about this enhancement.
But I really have to admit that this is quite low priority because the function is already extremely helpful.

Best and thanks again
Urs

Am 09.05.2012 01:49, schrieb David Nalesnik:

Hi Urs,

Hi David,
as promised I tried out your updated function(s).
Well, you can't call this a complete test suite, but it seems to work perfectly. Many thanks.
Attached is a version showing that it also/still works with phrasingSlurs.

I find the warnings very useful. I assume it isn't possible to find out and display the 'real' place in the source where the problem comes from? As it is, I only know that there is a changed curve that doesn't work anymore, but don't know where it is (which can of course be difficult to pin down in larger pieces.
If it isn't possible to identify the calling line in the source code, would it be possible to mark the respective curve red? This way one could easily spot the problematic grob.

You can do either, or both.  The attached file will display a warning which includes the input location and (if you uncomment the relevant lines in shape-curve) print the curves in red.

 

That a wrong number of pairs gives strange results is OK. That way one is gently pointed towards malformed input ;-)

+1

 

So it works like a charm now :-)
If you could still add the colour or line number feature - or tell me that you won't or can't do it - I could make a useable and distributably version of the file - maybe as a package together with displayControlPoints (see other mail in the other thread).

Sure, please do!  In the meantime, I'll keep tinkering with this and I'll send along any improvements. 

Thank you very much for your comments!

Best,

David

 

_______________________________________________
lilypond-user mailing list
address@hidden
https://lists.gnu.org/mailman/listinfo/lilypond-user