[ Thread Index |
Date Index
| More lilynet.net/tablatures Archives
]
On 9/16/09 12:41 AM, "Marc Hohl" <marc@xxxxxxxxxx> wrote:
> Carl Sorensen schrieb:
>>
>> On 9/14/09 1:30 PM, "Marc Hohl" <marc@xxxxxxxxxx> wrote:
>>
>>
>>> Carl Sorensen schrieb:
>>>
>>>> [...]
>>>>
>>>> No question it's the right way to do it for 2.13.4.
>>>>
>>>> But the whole thin (including the alists) could be copied to tablature..ly
>>>> for immediate use with 2.12, couldn't it?
>>>>
>>>> Carl
>>>>
>>>>
>>> Hm, after a bit of investigating I can only produce error messages.
>>> What I did: I included the definition of c0-pitch-alist together with
>>> the definition for add-new-clef in tablature.ly, which didn't work:
>>>
>>> In procedure cdr in expression (cdr (assoc # c0-pitch-alist)):
>>> /home/marc/lilypond/usr/share/lilypond/current/scm/parser-clef.scm:119:45:
>>> Wrong type (expecting pair): #f
>>> (by the way, I don't understand this error, because when I #(display
>>> c0-pitch-alist), I looks
>>> as it should be)
>>>
>>
>> (assoc # c0-pitch-alist) looks fishy to me.
>>
>> I think that the error comes from the cdr call, which needs a pair, but the
>> assoc call returns #f.
>>
> Yes, and my first thought was that there were error building up the alist,
> but the list seems to be ok...
Have you tested by displaying the list from within scm/parser-clef.scm?
You define c0-alist in scm/parser-clef.scm, and then you define-public
c0-alist in tablature.ly. I'm thinking that that may lead to two different
copies of c0-alist, and scm/parser-clef.scm is using the local one, not the
public one.
>>
>>
>>> Then I added the supported-clefs list to the file with the result that
>>> the error message went away, but lilypond still claims that the clef
>>> "moderntab" can't be found.
>>>
>>> Then again, I tried to hardcode the values for the moderntab clef,
>>> and this resulted in the attached file, but it still doesn't work. I
>>> also tried to
>>> include the definition for make-clef-set from scm/parser-clef.scm
>>> in tablature.ly, but the function string-match is not known at this stage.
>>>
I think this is the way to go. But I'm surprised that string-match isn't
known. It's part of the core guile implementation.
http://www.gnu.org/software/guile/manual/html_node/Regexp-Functions.html#Reg
exp-Functions
What error did you get when you did this?
>>
>> I didn't get an attachment with this file, so I don't know how to
>> troubleshoot it for you.
>>
> Oh, yes, I wanted to attach the file ... I hope in some years from now the
> email clients are clever enough to warn you if you mention "attached" or
> "send ... to you" that there is propably some attachment missing :-)
I'm really surprised that nobody makes a client that checks that. I would
use it multiple times a month. I hate it when I say "Here is ...,." and
then forget to do the attaching before I hit send.
HTH,
Carl