[L2Ork-dev] $0 weirdness

[Ivica Bukvic]

Forgot to post this (a draft was never sent for some reason):

Another update. I further refined this process (hopefully for the better).
Now you can have multiple escaped arguments in a single iemgui
label/send/receive, array name, and/or gatom label. This, however, needs to
be thoroughly tested. All, please test and report.

Thank you.

Best,

Ico

On Mon, Dec 7, 2015 at 1:07 AM, Ivica Bukvic <ico at vt.edu> wrote:

> OK, try the latest commit. Basically, it only converts the # into $ if it
> is not followed by a numerical digit that may suggest it is an arg rather
> than a string.
>
> There are still weird/illogical ways of breaking this, something that
> vanilla also exhibits. For instance, having:
>
> A#0B will expand #0 into patch's instance
>
> Whereas
>
> A$0B will truncate $
>
> OTOH, both $0AB and #0AB reflect the same behavior...
>
> Any thoughts on this?
>
> On Sun, Dec 6, 2015 at 4:16 PM, Ivica Ico Bukvic <ico at vt.edu> wrote:
>
>>
>>
>> On 12/6/2015 1:03 PM, Albert Graef wrote:
>>
>> On Sun, Dec 6, 2015 at 5:14 PM, Ivica Ico Bukvic <ico at vt.edu> wrote:
>>
>>> To get a better idea what I am doing here, is # only used at the
>>> beginning of a string (e.g. #0-something or #1), or can it be used anywhere
>>> (e.g. boo-#0 or blah#2boo)? Also, should it be escaped only if it is
>>> followed by a number? (e.g. #0 should be escaped, whereas # or #abc should
>>> not).
>>>
>>
>> AFAICT, this is only used with the built-in non-IEM number and symbol
>> boxes (floatatom and symbolatom) in stored patch files. The # must be
>> followed by a number. Apparently the #0, #1 etc.can occur anywhere in the
>> send/receive symbol (I tested with stuff like foo-#0-bar and it works as
>> expected). $0 aka #0 expands to the unique patch id, while $i aka #i
>> expands to the ith creation argument of the abstraction in its mother
>> patch, if any.
>>
>> If you can provide some clarification what are traditional uses of this,
>>> perhaps I can put in some filtering to figure out how we can hopefully
>>> address both scenarios.
>>>
>>
>> That would be great.
>>
>>
>>> Finally, why do we even have #0 when one can simply write $0?
>>>
>>
>> Max compatibility? Plain old cruft? I have no idea, you'll have to ask
>> Miller. ;-)
>>
>> The number and symbol boxes seem to be the only places where this
>> "convenience" is in use, everywhere else (including the IEM GUI objects)
>> the $i's are stored as \$i (which, incidentally, also works with number and
>> symbol boxes in reading a patch, but both vanilla and extended Pd insist on
>> storing it again as #i when writing the patch). At least that's what I
>> could deduce from playing around with a few test patches.
>>
>>
>> Wouldn't it be easier then to simply edit all such instances to use
>> exclusively # since they are interchangeable and be done with this? A
>> simple search and replace shell script would fix all your patches in a
>> single run (assuming your script is intelligent to find all of them or if
>> they are co-located in the same folder).
>>
>>
>> Albert
>>
>> --
>> Dr. Albert Gr"af
>> Computer Music Research Group, JGU Mainz, Germany
>> Email:  aggraef at gmail.com
>> WWW:    https://plus.google.com/+AlbertGraef
>>
>>
>> _______________________________________________
>> L2Ork-dev mailing listL2Ork-dev at disis.music.vt.eduhttp://disis.music.vt.edu/listinfo/l2ork-dev
>>
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://disis.music.vt.edu/pipermail/l2ork-dev/attachments/20151208/62783c6a/attachment.html>