REBOL "issues"
[1/16] from: dukeofperl:ml1 at: 5-Nov-2010 21:49
I'm reading:
http://www.rebol.com/docs/core23/rebolcore-3.html
Am I to understand that _all_ issues have the _#_ prefix? If so, then
are binary values "issues" as well? From the above document:
#{42652061205245424F4C}
--
Duke
[2/16] from: izkata:gma:il at: 5-Nov-2010 23:57
All issue!s start with #, but not all that start with # are issue!s:
>> type? #"r"
== char!
>> type? #foo
== issue!
>> type? #{44}
== binary!
>> issue? #"r"
== false
>> issue? #{44}
== false
On Fri, Nov 5, 2010 at 10:49 PM, Duke Normandin <dukeofperl-ml1.net> wrote:
> I'm reading:
> http://www.rebol.com/docs/core23/rebolcore-3.html
<<quoted lines omitted: 6>>
> To unsubscribe from the list, just send an email to
> lists at rebol.com with unsubscribe as the subject.
--
<*>
[3/16] from: dhsunanda:g:mail at: 6-Nov-2010 10:24
> Am I to understand that _all_ issues have the _#_ prefix? If so, then
> are binary values "issues" as well? From the above document:
>
> #{42652061205245424F4C}
The documentation is a little unclear there.
Or, from another perspective, # is somewhat overloaded:
#{....} is binary!
#"." is character!
#.... is issue!
An issue! cannot contain a opening or closing brace, nor can it contain
a double quote, so the three uses of # are quite separate.
Sunanda
[4/16] from: dukeofperl:ml1 at: 6-Nov-2010 5:51
On Sat, 6 Nov 2010, Sunanda wrote:
> > Am I to understand that _all_ issues have the _#_ prefix? If so, then
> > are binary values "issues" as well? From the above document:
<<quoted lines omitted: 7>>
> An issue! cannot contain a opening or closing brace, nor can it contain
> a double quote, so the three uses of # are quite separate.
So _exactly_ what is the significance of _#_?
BTW, is "overloading" a REBOL practice that I'll be discovering to be
the rule and not the exception? :)
--
Duke
[5/16] from: gregg:pointillistic at: 6-Nov-2010 10:27
Hi Duke,
DN> So _exactly_ what is the significance of _#_?
It's simple a lexical element in the REBOL grammar. Imagine the parser
going along, breaking a string of characters up and identifying the
types and values. When it comes to a space, the next character guides
it. Is it a digit? Then we need to go further and see if it's a
decimal or an integer. If we see a dollar sign, it's a money! value.
If we see a # followed by an alphanum, it's an issue. If it's a #
followed by a quote, bracket, or curly brace, it's a char, serialized
value, or binary, etc.
More importantly for humans, issue! was intended for serial numbers,
IDs, and the like, which makes the # sigil a good fit.
DN> BTW, is "overloading" a REBOL practice that I'll be discovering to
DN> be the rule and not the exception? :)
No. This is not considered overloading. REBOL is still an ASCII
language, and the lexical space is very tight. # has been pressed into
service for a few notations, which makes it a bit of an exception.
-- Gregg
[6/16] from: dukeofperl:ml1 at: 6-Nov-2010 11:55
On Sat, 6 Nov 2010, Gregg Irwin wrote:
> Hi Duke,
> DN> So _exactly_ what is the significance of _#_?
<<quoted lines omitted: 6>>
> followed by a quote, bracket, or curly brace, it's a char, serialized
> value, or binary, etc.
I see the rationale! Thanks...
--
Duke
[7/16] from: tim-johnsons:web at: 6-Nov-2010 11:06
* Izkata <izkata-gmail.com> [101105 22:03]:
> All issue!s start with #, but not all that start with # are issue!s:
> >> type? #"r"
<<quoted lines omitted: 7>>
> >> issue? #{44}
> == false
;; :) Another console session
>> series? #foo
== true
>> type? #foo
== issue!
>> type? #{44}
== binary!
>> series? #{44}
== true
--
Tim
tim at johnsons-web.com or akwebsoft.com
http://www.akwebsoft.com
[8/16] from: dukeofperl:ml1 at: 6-Nov-2010 17:10
On Sat, 6 Nov 2010, Tim Johnson wrote:
> * Izkata <izkata-gmail.com> [101105 22:03]:
> > All issue!s start with #, but not all that start with # are issue!s:
<<quoted lines omitted: 19>>
> >> series? #{44}
> == true
So an "issue" is also a "series" which in turn is also what? etc
etc. How far up does it goes? I can see why a parser may need this,
but programmers? Let me get my feet a little wetter with REBOL before
I make the mistake of voicing an _uninformed opinion. :)
BTW, maybe leave an empty line before and after your words of wisdom -
they'll stand out better. :)
--
Duke
[9/16] from: dhsunanda::gmail at: 7-Nov-2010 0:17
> So an "issue" is also a "series" which in turn is also what? etc
> etc. How far up does it goes? I can see why a parser may need this,
> but programmers? Let me get my feet a little wetter with REBOL before
> I make the mistake of voicing an _uninformed opinion. :)
>
> There is a slightly out-of-date datatype hierarchy here:
http://www.mail-archive.com/list-rebol.com/msg02743.html
It looks like a fun execise to produce a complete hierarchy by finding all
datatypes and checking which contain the others. Easy to do in an R3 script.
Probably doable in an R2 script. Have fun trying.
Sunanda.
[10/16] from: izkata:gma:il at: 6-Nov-2010 19:31
On Sat, Nov 6, 2010 at 7:17 PM, Sunanda Dh <dhsunanda-gmail.com> wrote:
> >
> > So an "issue" is also a "series" which in turn is also what? etc
> > etc. How far up does it goes? I can see why a parser may need this,
> > but programmers?
>
Path notation (array-type access) works in some way on all series! types,
math works on all number! types, an so on. It helps handle crossover.
I say "in some way" on all series! types because the file! type handles it
specially, creating a directory path instead of accessing the series in an
array-like way, and binary! doesn't seem to allow setting, just getting.
> Let me get my feet a little wetter with REBOL before
> > I make the mistake of voicing an _uninformed opinion. :)
<<quoted lines omitted: 6>>
> Probably doable in an R2 script. Have fun trying.
> Sunanda.
I can vaguely remember someone did do an R2 one years ago. It generated a
table, and saved it as an image, where <X column datatype> is a subset of <Y
column datatype>, or something like that.
--
<*>
[11/16] from: tim-johnsons:web at: 7-Nov-2010 7:47
* Duke Normandin <dukeofperl-ml1.net> [101106 16:14]:
> So an "issue" is also a "series" which in turn is also what? etc
> etc. How far up does it goes?
I believe that series! is descended from any-type! which all types
are descended from. And I believe that there is only one layer of
subtypes.
I.E. any-type!/series!/issue! etc..
> I can see why a parser may need this but programmers?
Wweellll... this programmer does not take advantage of all rebol
datatypes but I find it very handy for runtime typechecking, including
the function interface.
--
Tim
tim at johnsons-web.com or akwebsoft.com
http://www.akwebsoft.com
[12/16] from: M8R-kr0jkp:mailinator at: 8-Nov-2010 16:50
#[[Sunanda
There is a slightly out-of-date datatype hierarchy here:
http://www.mail-archive.com/list-rebol.com/msg02743.html
Sunanda]] - see also:
http://www.rebol.net/wiki/Datatypes
http://en.wikibooks.org/wiki/REBOL_Programming/Language_Features/Types
#[[Tim
I believe that series! is descended from any-type! which all types
are descended from. And I believe that there is only one layer of
subtypes.
I.E. any-type!/series!/issue! etc..
Tim]] - actually, that is not correct, it is as follows:
any-type!/series!/any-string!/issue!
(in R2)
-Ladislav
[13/16] from: tim-johnsons:web at: 8-Nov-2010 12:48
* Ladislav <M8R-kr0jkp-mailinator.com> [101108 06:53]:
> #[[Sunanda
> There is a slightly out-of-date datatype hierarchy here:
<<quoted lines omitted: 9>>
> Tim]] - actually, that is not correct, it is as follows:
> any-type!/series!/any-string!/issue!
:) Did'ya get that Duke. Don't let your head explode! It will sink
in over time. After 10 years,for me, it is *still* sinking in.
--
Tim
tim at johnsons-web.com or akwebsoft.com
http://www.akwebsoft.com
[14/16] from: dukeofperl:ml1 at: 8-Nov-2010 16:03
On Mon, 8 Nov 2010, Tim Johnson wrote:
> * Ladislav <M8R-kr0jkp-mailinator.com> [101108 06:53]:
> > #[[Sunanda
<<quoted lines omitted: 16>>
> :) Did'ya get that Duke. Don't let your head explode! It will sink
> in over time. After 10 years,for me, it is *still* sinking in.
Yep! I read it! How have you managed to resist the REBOL "layers"
tectonics for ten years. :) Must have something to do with that
Kodiak moonshine you guys have access to up there. ;D
I'm hanging in there...
--
Duke
[15/16] from: tim-johnsons:web at: 8-Nov-2010 16:04
* Duke Normandin <dukeofperl-ml1.net> [101108 15:13]:
> Yep! I read it! How have you managed to resist the REBOL "layers"
> tectonics for ten years. :) Must have something to do with that
> Kodiak moonshine you guys have access to up there. ;D
>
> I'm hanging in there...
I was a cannery rat and commercial fisherman out of Kodiak in the
70's. Now I make my moonshine (home-brewed wine and beer) in
Palmer... You can see what I see while I'm coding at
akwebsoft.com
--
Tim
tim at johnsons-web.com or akwebsoft.com
http://www.akwebsoft.com
[16/16] from: dukeofperl:ml1 at: 8-Nov-2010 18:59
On Mon, 8 Nov 2010, Tim Johnson wrote:
> * Duke Normandin <dukeofperl-ml1.net> [101108 15:13]:
> >
<<quoted lines omitted: 7>>
> Palmer... You can see what I see while I'm coding at
> akwebsoft.com
Likewise! Except I'm looking at the Eastern Slopes of the Canadian
Rockies, S.W. of Calgary, Alberta. :) and instead of getting busy
programming, I dream of fly-fishing for trout in those mountain and
foothill stream. Only about 240 days to go ...
--
Duke
Notes
- Quoted lines have been omitted from some messages.
View the message alone to see the lines that have been omitted