Better approach to fixing the TabVoice glissando problem.

Better approach to fixing the TabVoice glissando problem.

[MikeSol, 2011-05-03 03:56:15]

A better approach to the TabVoice glissando problem, although I'm still not
quite sure why it works, but it seems to work!

Cheers,
MS

[Neil Puttock, 2011-05-03 21:14:02]

On 2011/05/03 03:56:15, MikeSol wrote:
> A better approach to the TabVoice glissando problem, although I'm still not
> quite sure why it works, but it seems to work!

The default procedure for noteToFretFunctionm, `determine-frets', returns a
reversed list, so the following change to the helper function
`determine-frets-and-strings' should suffice:

(scm/translation-functions.scm)

(reverse string-fret-fingering-tuples)) ;; end of determine-frets-and-strings

[mike_apollinemike.com, 2011-05-03 22:27:47]

On May 3, 2011, at 2:14 PM, n.puttock@gmail.com wrote:

> On 2011/05/03 03:56:15, MikeSol wrote:
>> A better approach to the TabVoice glissando problem, although I'm
> still not
>> quite sure why it works, but it seems to work!
> 
> The default procedure for noteToFretFunctionm, `determine-frets',
> returns a reversed list, so the following change to the helper function
> `determine-frets-and-strings' should suffice:
> 
> (scm/translation-functions.scm)
> 
> (reverse string-fret-fingering-tuples)) ;; end of
> determine-frets-and-strings
> 

Well done!  That doesn't have any adverse side effects in the tab regtests, so:

Pushed c6f097f9ed1c19b9922fca890ce9f923178af738

Cheers,
MS

[Neil Puttock, 2011-05-03 22:30:11]

On 2011/05/03 21:14:02, Neil Puttock wrote:

> (reverse string-fret-fingering-tuples)) ;; end of determine-frets-and-strings

Hmm, this messes up ordering of harmonics, though they're already broken judging
by the following:

\version "2.15.0"

music = \relative c' {
  dis\2\glissando <e\2\harmonic a\1>
  <e\2 a\1\harmonic >\glissando dis\2
}

\new StaffGroup <<
  \new Staff { \clef "G_8" \music }
  \new TabStaff { \clef "moderntab" \music }
>>

[Neil Puttock, 2011-05-03 22:36:08]

On 3 May 2011 23:27, mike@apollinemike.com <mike@apollinemike.com> wrote:
> On May 3, 2011, at 2:14 PM, n.puttock@gmail.com wrote:
>
>> On 2011/05/03 03:56:15, MikeSol wrote:
>>> A better approach to the TabVoice glissando problem, although I'm
>> still not
>>> quite sure why it works, but it seems to work!
>>
>> The default procedure for noteToFretFunctionm, `determine-frets',
>> returns a reversed list, so the following change to the helper function
>> `determine-frets-and-strings' should suffice:
>>
>> (scm/translation-functions.scm)
>>
>> (reverse string-fret-fingering-tuples)) ;; end of
>> determine-frets-and-strings
>>
>
> Well done!  That doesn't have any adverse side effects in the tab regtests,
so:
>
> Pushed c6f097f9ed1c19b9922fca890ce9f923178af738

That's worrying.  I got changes in tablature-slide.ly and
tablature-harmonic.ly (both swapped harmonics).

[mike_apollinemike.com, 2011-05-03 22:39:46]

On May 3, 2011, at 3:36 PM, Neil Puttock wrote:

> On 3 May 2011 23:27, mike@apollinemike.com <mike@apollinemike.com> wrote:
>> On May 3, 2011, at 2:14 PM, n.puttock@gmail.com wrote:
>> 
>>> On 2011/05/03 03:56:15, MikeSol wrote:
>>>> A better approach to the TabVoice glissando problem, although I'm
>>> still not
>>>> quite sure why it works, but it seems to work!
>>> 
>>> The default procedure for noteToFretFunctionm, `determine-frets',
>>> returns a reversed list, so the following change to the helper function
>>> `determine-frets-and-strings' should suffice:
>>> 
>>> (scm/translation-functions.scm)
>>> 
>>> (reverse string-fret-fingering-tuples)) ;; end of
>>> determine-frets-and-strings
>>> 
>> 
>> Well done!  That doesn't have any adverse side effects in the tab regtests,
so:
>> 
>> Pushed c6f097f9ed1c19b9922fca890ce9f923178af738
> 
> That's worrying.  I got changes in tablature-slide.ly and
> tablature-harmonic.ly (both swapped harmonics).

Crap...I was so focused on the glissandi that I didn't even see that.  They came
up in my regtests, but my eyes zoomed right to the glissandi and I didn't see
any difference.

I'll revert the push - sorry!

Cheers,
MS

[mike_apollinemike.com, 2011-05-03 22:44:53]

On May 3, 2011, at 3:39 PM, mike@apollinemike.com wrote:

> On May 3, 2011, at 3:36 PM, Neil Puttock wrote:
> 
>> On 3 May 2011 23:27, mike@apollinemike.com <mike@apollinemike.com> wrote:
>>> On May 3, 2011, at 2:14 PM, n.puttock@gmail.com wrote:
>>> 
>>>> On 2011/05/03 03:56:15, MikeSol wrote:
>>>>> A better approach to the TabVoice glissando problem, although I'm
>>>> still not
>>>>> quite sure why it works, but it seems to work!
>>>> 
>>>> The default procedure for noteToFretFunctionm, `determine-frets',
>>>> returns a reversed list, so the following change to the helper function
>>>> `determine-frets-and-strings' should suffice:
>>>> 
>>>> (scm/translation-functions.scm)
>>>> 
>>>> (reverse string-fret-fingering-tuples)) ;; end of
>>>> determine-frets-and-strings
>>>> 
>>> 
>>> Well done!  That doesn't have any adverse side effects in the tab regtests,
so:
>>> 
>>> Pushed c6f097f9ed1c19b9922fca890ce9f923178af738
>> 
>> That's worrying.  I got changes in tablature-slide.ly and
>> tablature-harmonic.ly (both swapped harmonics).
> 
> Crap...I was so focused on the glissandi that I didn't even see that.  They
came up in my regtests, but my eyes zoomed right to the glissandi and I didn't
see any difference.
> 
> I'll revert the push - sorry!
> 
> Cheers,
> MS

Back to http://codereview.appspot.com/4438092/, it seems as if this doesn't
screw up the harmonics problem.
Again, sorry for the errant push: I was too tunnel vision and should have done
the full comparison (I was just using side-by-side PDFs).

Cheers,
MS

[Neil Puttock, 2011-05-04 08:49:43]

On 3 May 2011 23:44, mike@apollinemike.com <mike@apollinemike.com> wrote:
> Back to http://codereview.appspot.com/4438092/, it seems as if this doesn't
> screw up the harmonics problem.

But they're already screwed (probably since the Tab_harmonic_engraver
was removed and the harmonics moved to the TabNoteHead print
function).

> Again, sorry for the errant push: I was too tunnel vision and should have
> done the full comparison (I was just using side-by-side PDFs).

No problem, I should've tested it before posting.

Cheers,
Neil

[mike_apollinemike.com, 2011-05-06 14:52:45]

On May 4, 2011, at 1:49 AM, Neil Puttock wrote:

> On 3 May 2011 23:44, mike@apollinemike.com <mike@apollinemike.com> wrote:
>> Back to http://codereview.appspot.com/4438092/, it seems as if this doesn't
>> screw up the harmonics problem.
> 
> But they're already screwed (probably since the Tab_harmonic_engraver
> was removed and the harmonics moved to the TabNoteHead print
> function).
> 

Carl - I know you're slammed w/ a paper, but do you have time to apply my patch
(as well as the failed commit that I pushed to master), look into the
TabNoteHead print function, and see if there's a more elegant way to do what I'm
doing w/o messing up the harmonics?  I can put some time into it in a couple
weeks, but I find that things go a lot quicker when the person who wrote a chunk
of code sizes up the problem.

Cheers,
MS

[c_sorensen, 2011-05-06 15:59:06]

On 5/6/11 8:52 AM, "mike@apollinemike.com" <mike@apollinemike.com> wrote:

> On May 4, 2011, at 1:49 AM, Neil Puttock wrote:
> 
>> On 3 May 2011 23:44, mike@apollinemike.com <mike@apollinemike.com> wrote:
>>> Back to http://codereview.appspot.com/4438092/, it seems as if this doesn't
>>> screw up the harmonics problem.
>> 
>> But they're already screwed (probably since the Tab_harmonic_engraver
>> was removed and the harmonics moved to the TabNoteHead print
>> function).
>> 
> 
> Carl - I know you're slammed w/ a paper, but do you have time to apply my
> patch (as well as the failed commit that I pushed to master), look into the
> TabNoteHead print function, and see if there's a more elegant way to do what
> I'm doing w/o messing up the harmonics?  I can put some time into it in a
> couple weeks, but I find that things go a lot quicker when the person who
> wrote a chunk of code sizes up the problem.

Unfortunately, I don't.  I hope to on Saturday.  But it may be next week.

It's on my list.  I'll get to it as soon as I can.

Thanks,

Carl

[c_sorensen, 2011-05-07 22:26:30]

On 5/6/11 8:52 AM, "mike@apollinemike.com" <mike@apollinemike.com> wrote:

> On May 4, 2011, at 1:49 AM, Neil Puttock wrote:
> 
>> On 3 May 2011 23:44, mike@apollinemike.com <mike@apollinemike.com> wrote:
>>> Back to http://codereview.appspot.com/4438092/, it seems as if this doesn't
>>> screw up the harmonics problem.
>> 
>> But they're already screwed (probably since the Tab_harmonic_engraver
>> was removed and the harmonics moved to the TabNoteHead print
>> function).
>> 
> 
> Carl - I know you're slammed w/ a paper, but do you have time to apply my
> patch (as well as the failed commit that I pushed to master), look into the
> TabNoteHead print function, and see if there's a more elegant way to do what
> I'm doing w/o messing up the harmonics?  I can put some time into it in a
> couple weeks, but I find that things go a lot quicker when the person who
> wrote a chunk of code sizes up the problem.


OK, I'm on this now.  The problem is that the procedure called to determine
frets and strings doesn't respect and maintain note ordering.  And the
ordering varies with whether or not strings are specified.  I've modified
the tablature-harmonics regtest to show the problem, and a fix will be
coming soon.

I think that will solve your problem.  Once I have tab harmonics
bulletproof, I'll check the glissando stuff out.

Thanks,

Carl