Mailing List Archive: 49091 messages
  • Home
  • Script library
  • AltME Archive
  • Mailing list
  • Articles Index
  • Site search
 

to pair! bug

 [1/14] from: rotenca::telvia::it at: 20-Apr-2002 14:08


This is ok:
>> to integer! "ww"
** Script Error: Invalid argument: ww ** Near: to integer! "ww" This is the bug:
>> to pair! "ww"
== none --- Ciao Romano

 [2/14] from: g:santilli:tiscalinet:it at: 20-Apr-2002 16:17


Hi Romano, On Saturday, April 20, 2002, 2:08:48 PM, you wrote: RPT> This is the bug:
>>> to pair! "ww"
RPT> == none Well, for that matter:
>> to date! "ww"
** Script Error: Invalid argument: ww ** Where: halt-view ** Near: to date! "ww"
>> to time! "ww"
== none which is rather inconsistent to me too. But maybe there's a reason for that... Carl? Are you reading? Regards, Gabriele. -- Gabriele Santilli <[g--santilli--tiscalinet--it]> -- REBOL Programmer Amigan -- AGI L'Aquila -- REB: http://web.tiscali.it/rebol/index.r

 [3/14] from: rotenca:telvia:it at: 21-Apr-2002 1:16


Hi, Gabriele have you tryied all the datatypes? Pair and time are the only one? --- Ciao Romano

 [4/14] from: g:santilli:tiscalinet:it at: 21-Apr-2002 11:59


Hi Romano, On Sunday, April 21, 2002, 1:16:41 AM, you wrote: RPT> have you tryied all the datatypes? Pair and time are the only one? I didn't. But we can try: (hmm, got a crash; seems like TO does not like some datatypes... oh, maybe it was routine! let's avoid types we don't need to check)
>> str: ""
== ""
>> for ch #"^(00)" #"^(FE)" 1 [insert tail str ch]
== ""
>> datatypes: []
== []
>> foreach value second system/words [if datatype? get/any 'value [insert tail datatypes value]]
== none
>> print datatypes
unset error datatype native action routine op function object struct library port any-type any-word any-function number series any-string any-block symbol word set-word get-word lit-word refinement none logic integer decimal money time date char pair event tuple bitset string issue binary file email url image tag block paren path set-path lit-path hash list
>> problem: reduce [routine! struct! library! port! any-type! any-word! any-function! number! series! any-string! any-block! event! string!]
== [routine! struct! library! port! any-type! any-word! any-function! number! series! any-string! any-block! event! string!]
>> foreach datatype datatypes [prin [mold datatype ": "] either error? res: try [to datatype str] [print "Error."] [print ["Gives" type? :res]]]
unset! : Error. error! : Error. datatype! : Error. native! : Error. action! : Error. op! : Error. function! : Error. object! : Error. symbol! : Gives symbol word! : Gives word set-word! : Gives set-word get-word! : Gives get-word lit-word! : Gives lit-word refinement! : Gives refinement none! : Gives none logic! : Gives logic integer! : Error. decimal! : Error. money! : Error. time! : Gives none date! : Error. char! : Error. pair! : Gives none tuple! : Error. bitset! : Gives bitset issue! : Gives issue binary! : Gives binary file! : Gives file email! : Gives email url! : Gives url image! : Gives image tag! : Gives tag block! : Gives block paren! : Gives paren path! : Gives path set-path! : Gives set-path lit-path! : Gives lit-path hash! : Gives hash list! : Error. To notice: time! : Gives none pair! : Gives none And: list! : Error. while other block types did work (but returned an empty series, maybe because confused by the null character). Regards, Gabriele. -- Gabriele Santilli <[g--santilli--tiscalinet--it]> -- REBOL Programmer Amigan -- AGI L'Aquila -- REB: http://web.tiscali.it/rebol/index.r

 [5/14] from: rotenca:telvia:it at: 21-Apr-2002 21:37


Hi Gabriele
> list! : Error. > > while other block types did work (but returned an empty series, > maybe because confused by the null character). >> to-block to-string #"^(00)"
== []
>> length? to-block to-string #"^(00)"
== 0 But
>> to-block to-string #"^(01)"
== [-ç] and
> to-block #"^(00)"
== [#"^@"] --- Ciao Romano

 [6/14] from: rotenca:telvia:it at: 21-Apr-2002 21:34


Thank you Gabriele for the work.
> symbol! : Gives symbol
Interesting.
> none! : Gives none
Mhhhh, everything can become none?
> image! : Gives image
But always an image sized 0x0 and I do not know ho to change image/size, do you know it? --- Ciao Romano ----- Original Message ----- From: "Gabriele Santilli" <[g--santilli--tiscalinet--it]> To: "Romano Paolo Tenca" <[rebol-list--rebol--com]> Sent: Sunday, April 21, 2002 11:59 AM Subject: [REBOL] Re: to pair! bug
> Hi Romano, > On Sunday, April 21, 2002, 1:16:41 AM, you wrote:
<<quoted lines omitted: 9>>
> == [] > >> foreach value second system/words [if datatype? get/any 'value [insert
tail datatypes value]]
> == none > >> print datatypes
<<quoted lines omitted: 4>>
> binary file email url image tag block paren path set-path lit-path hash list > >> problem: reduce [routine! struct! library! port! any-type! any-word!
any-function! number! series! any-string! any-block! event! string!]
> == [routine! struct! library! port! any-type! any-word! > any-function! number! series! any-string! any-block! event! string!] > >> foreach datatype datatypes [prin [mold datatype ": "] either error? res:
try [to datatype str] [print "Error."] [print ["Gives" type? :res]]]
> unset! : Error. > error! : Error.
<<quoted lines omitted: 43>>
> maybe because confused by the null character). >> to-block to-string #"^(00)"
== []
>> length? to-block to-string #"^(00)"
== 0 But
>> to-block to-string #"^(01)"
== [-ç] and

 [7/14] from: g:santilli:tiscalinet:it at: 22-Apr-2002 12:55


Hi Romano, On Sunday, April 21, 2002, 9:34:48 PM, you wrote:
>> symbol! : Gives symbol
RPT> Interesting. Yup. But I was never able to understand what it is, and why it is available in system/words as IIRC Holger once said is an internal thing only (like end!, context! etc.).
>> none! : Gives none
RPT> Mhhhh, everything can become none? Once Jeff wrote something like: "this is a puzzle for you: what is to-none useful for?". I was never able to solve the puzzle.
>> image! : Gives image
RPT> But always an image sized 0x0 and I do not know ho to change image/size, do RPT> you know it? Never tried. I think you can only make images from binaries with:
>> img: make image! [1x1 #{000000}]
== make image! [1x1 #{000000}] Regards, Gabriele. -- Gabriele Santilli <[g--santilli--tiscalinet--it]> -- REBOL Programmer Amigan -- AGI L'Aquila -- REB: http://web.tiscali.it/rebol/index.r

 [8/14] from: rotenca:telvia:it at: 22-Apr-2002 18:31


Hi Gabriele
> >> none! : Gives none > RPT> Mhhhh, everything can become none? > > Once Jeff wrote something like: "this is a puzzle for you: what is > to-none useful for?". I was never able to solve the puzzle.
to none :-)
> >> image! : Gives image > RPT> But always an image sized 0x0 and I do not know ho to change
image/size, do
> RPT> you know it? > > Never tried. I think you can only make images from binaries with: > > >> img: make image! [1x1 #{000000}] > == make image! [1x1 #{000000}] >> img: make image! "000000"
== make image! [0x0 #{3030303030}] But 0x0... --- Ciao Romano

 [9/14] from: rotenca:telvia:it at: 23-Apr-2002 10:00


Hi Gabriele,
> Once Jeff wrote something like: "this is a puzzle for you: what is > to-none useful for?". I was never able to solve the puzzle.
What do you think about this: probe context [none: 1 a: none b: to-none ""] --- Ciao Romano

 [10/14] from: g:santilli:tiscalinet:it at: 23-Apr-2002 23:19


Hi Romano, On Tuesday, April 23, 2002, 10:00:54 AM, you wrote:
>> Once Jeff wrote something like: "this is a puzzle for you: what is >> to-none useful for?". I was never able to solve the puzzle.
RPT> What do you think about this: RPT> probe context [none: 1 a: none b: to-none ""] Hmm, that does not seem much more useful than defining: none2: none and then doing: probe context [none: 1 a: none b: none2] so, I don't know if that really solves the puzzle. Jeeeeeeeff! Where are you? Regards, Gabriele. -- Gabriele Santilli <[g--santilli--tiscalinet--it]> -- REBOL Programmer Amigan -- AGI L'Aquila -- REB: http://web.tiscali.it/rebol/index.r

 [11/14] from: rebol:optushome:au at: 25-Apr-2002 0:44


----- Original Message ----- From: "Gabriele Santilli" <[g--santilli--tiscalinet--it]> To: "Romano Paolo Tenca" <[rebol-list--rebol--com]> Sent: Wednesday, April 24, 2002 7:19 AM Subject: [REBOL] Re: to pair! bug
> Hi Romano, > > On Tuesday, April 23, 2002, 10:00:54 AM, you wrote: > > >> Once Jeff wrote something like: "this is a puzzle for you: what is > >> to-none useful for?". I was never able to solve the puzzle.
For converting none when a file is loaded??
>> d: load "none"
== none
>> type? d
== word!
>> none? d
== false
>> d: to-none d
== none
>> none? d
== true Cheers, Allen K

 [12/14] from: holger:rebol at: 24-Apr-2002 15:05


On Mon, Apr 22, 2002 at 12:55:56PM +0200, Gabriele Santilli wrote:
> Once Jeff wrote something like: "this is a puzzle for you: what is > to-none useful for?". I was never able to solve the puzzle.
I am not aware of that puzzle, or what Jeff had in mind, but here is my guess: it lets you add code to an 'any clause without breaking the 'any, e.g. if you want to call a sequence of functions until one of them returns non-none/false, but also need to do some calculations in between, you could use something like: any [ func1 to-none count: count + 1 func2 to-none count: count + 1 func3 ... ] -- Holger Kruse [kruse--nordicglobal--com]

 [13/14] from: g:santilli:tiscalinet:it at: 25-Apr-2002 11:53


Hi Holger, On Thursday, April 25, 2002, 12:05:17 AM, you wrote: HK> any [ HK> func1 HK> to-none count: count + 1 HK> func2 HK> to-none count: count + 1 HK> func3 HK> ... HK> ] That seems useful, indeed! BTW, since we are at it, would that be more efficient than: any [ func1 (count: count + 1 none) func2 (count: count + 1 none) func3 ... ] Regards, Gabriele. -- Gabriele Santilli <[g--santilli--tiscalinet--it]> -- REBOL Programmer Amigan -- AGI L'Aquila -- REB: http://web.tiscali.it/rebol/index.r

 [14/14] from: holger:rebol at: 25-Apr-2002 11:21


On Thu, Apr 25, 2002 at 11:53:59AM +0200, Gabriele Santilli wrote:
> Hi Holger, > On Thursday, April 25, 2002, 12:05:17 AM, you wrote:
<<quoted lines omitted: 16>>
> ... > ]
I am not sure about execution time, but in terms of memory use it is more efficient, yes, because it avoids additional block series. -- Holger Kruse [kruse--nordicglobal--com]

Notes
  • Quoted lines have been omitted from some messages.
    View the message alone to see the lines that have been omitted