Re: new bar-lines / issue 1320

[Thomas Morley]

2012/6/4 Werner LEMBERG <address@hidden>:
>>> Uh, oh, please not a letter.  I almost always use seriffed fonts, and
>>> it looks rather strange.  What slightly longer but easy to remember
>>> symbolics?
>>>
>>>   |
>>>
>>>   ||
>>
>> How about ! then?  It actually has both | and . in it, and it _is_ a
>> sentence ending punctuation.
>>
>>>   || |:
>>
>> !: or !|: (I don't actually know what you want here)
>
> Well, || alone should be a pair of thin bars, while || in combination
> with other elements means a thick bar.  To use `!' or `.', I don't
> care (but I have a conservative preference for `.').
>
> My idea was to use the vertical bar symbol for everything, thus coming
> nearer to the visible result.
>
>>>   | ||
>>
>> Uh, |! or does this even exist?
>
> Yes, of course.  A thin bar followed by a thick bar ends a piece.
> This looks like a misunderstanding.
>
>
>    Werner


My basic idea to design (new) barlines was to asign _every_ _single_
substring of a barline-glyph to a _single_ print-procedure.
With the code as it is, a glyph like "|| |:" will be converted into a
list of single substrings and " "-strings will be filtered out.
i.e. "|| |:" -> '( "|" "|" "|" ":")
In the end it will produce three thin barlines and a colon.

Well, I could add a procedure, preprocessing "|| |:", converting this
into sth else.
But I have already some doubts about increased compiling time.
So I'd prefer to alter "." to a new sign, if wanted.
In our private discussions Marc already suggested the uppercase i, but
it seems there is no sympathy for it.
Apart from having the same conservative preference for "." as Werner,
I'd vote for "!".
Currently I don't see any reason why it shouldn't work.
Will test it this evening.

If this approach is suitable for replacing bar-line.cc and span-bar.cc
(and some parts of output-lib.scm) the question whether something even
exists is obsolet.

-Harm