Parse surprise

[1/11] from: lmecir:mbox:vol:cz at: 14-Jan-2003 17:14

Hi all, accidentally I tried this:

>> parse "abcdefgh" [thru 2 a: (print a) to 1 a: (print a)]

cdefgh abcdefgh == false It was quite a surprise for me. This isn't as useful as it may complicate block parsing IMO. What do you think? Regards -L

[2/11] from: cyphre:seznam:cz at: 14-Jan-2003 18:24

Hi Ladislav, interesting, just tried

>> parse "abcdefgh" [thru 3 a: (print a) copy rslt to 1 a: (print a print

rslt)] defgh abcdefgh none == false why is the content of rslt set to none ? Shouldn't it be the rslt "b"? regards, Cyphre

[3/11] from: petr:krenzelok:trz:cz at: 14-Jan-2003 18:34

Ladislav Mecir wrote:

> Hi all, > > accidentally I tried this: > >>>parse "abcdefgh" [thru 2 a: (print a) to 1 a: (print a)] > > cdefgh > abcdefgh > == false

what? :-) I did not know it is possible ... is that needed? Isn't it the same as using skip 2 skip 1? Well, as for parse ... still waiting for favorite to/thru [a | b | c], which can't be so easily done imo ... -pekr-

[4/11] from: g:santilli:tiscalinet:it at: 14-Jan-2003 19:30

Hi Ladislav, On Tuesday, January 14, 2003, 5:14:18 PM, you wrote: LM> What do you think? It's a surprise for me too, and could be a problem in block parsing indeed (on the other hand, I wonder what kind of dialect would be doing something like "to 5" or "thru 44"...). I have to think if this behavior can be useful for something... Regards, Gabriele. -- Gabriele Santilli <[g--santilli--tiscalinet--it]> -- REBOL Programmer Amigan -- AGI L'Aquila -- REB: http://web.tiscali.it/rebol/index.r

[5/11] from: rotenca:telvia:it at: 14-Jan-2003 19:19

Hi,

> why is the content of rslt set to none ? Shouldn't it be the rslt "b"? >> parse "abcdefgh" [copy x thru 5 copy y thru 4] print [x y]

abcde none

>> parse "abcdefgh" [copy x thru 5 copy y thru 5] print [x y]

abcde none

>> parse "abcdefgh" [copy x thru 5 copy y thru 6] print [x y]

abcde f It seems that copy works if the second to/thru value is after the first. Other strange thing:

>> parse "abcdefgh" [copy x to 0] print [x]

abcdefgh

>> parse "abcdefgh" [copy x to 1] print [x]

none

>> parse "abcdefgh" [copy x to 2] print [x]

a Perhaps is undocumented because it is a unfinished or buggy feature. --- Ciao Romano

[6/11] from: rotenca:telvia:it at: 14-Jan-2003 19:22

Hi,

> Well, as for parse ... still waiting for favorite to/thru [a | b | c], > which can't be so easily done imo ...

This can be done with parse-compile. Obviously a native parse command should be better. --- Ciao Romano

[7/11] from: otherchaz::mindspring::com at: 15-Jan-2003 3:05

> It's a surprise for me too, and could be a problem in block > parsing indeed (on the other hand, I wonder what kind of dialect > would be doing something like "to 5" or "thru 44"...). > > I have to think if this behavior can be useful for something... >

this behavior can be useful for an RPG IV dialect http://publib.boulder.ibm.com/cgi-bin/bookmgr/BOOKS/QB3AGZ03/3.3.1

[8/11] from: sqlab:gmx at: 16-Jan-2003 10:18

I find it more annoying, that it uses the absolute position starting from the head and not from the actual position.

>> parse next "abcdefgh" [thru 2 a: (print a) to 1 a: (print a)]

cdefgh abcdefgh AR Ladislav Mecir wrote:

> Hi all, > accidentally I tried this:

<<quoted lines omitted: 11>>

> [rebol-request--rebol--com] with "unsubscribe" in the > subject, without the quotes.

-- +++ GMX - Mail, Messaging & more http://www.gmx.net +++ NEU: Mit GMX ins Internet. Rund um die Uhr f�r 1 ct/ Min. surfen!

[9/11] from: lmecir:mbox:vol:cz at: 17-Jan-2003 16:19

Hi all, how about this?

>> a: does [parse "" [(return "done")]] >> a

== true Regards -L

[10/11] from: rotenca:telvia:it at: 17-Jan-2003 19:05

Hi Ladislav,

> how about this? > > >> a: does [parse "" [(return "done")]] > >> a > == true

Break, throw and return do not stop a parse paren (like object spec), only make error! is able to do it. I should like to have a method to stop a parse rule inside paren code and reurn a value like true or false (or something else). --- Ciao Romano

[11/11] from: lmecir:mbox:vol:cz at: 18-Jan-2003 11:18

Hi Romano, would you like to have something like PE? pe/pe "" [any (pe/ret "done")] ; == "done" ----- Original Message ----- From: "Romano Paolo Tenca"

> I should like to have a method to stop a parse rule inside paren code and > reurn a value like true or false (or something else).

The source of PE: pe: make object! [ pe: function [ "Parses a series according to rules." input [series!] "Input series to parse" rules [block! string! none!] "Rules to parse by" /all "Parses all chars including spaces." /case "Uses case-sensitive comparison." ] [value] [ res?: false either rebol/words/all [ error? value: try [ apply/r/a :parse compose [ (either all ['all] []) (either case ['case] []) ] reduce [input rules] ] res? ] [ res?: false return get/any 'res ] [value] ] res?: false res: none ret: func [ value [any-type!] ] [ res?: true error? set/any 'res get/any 'value make error! "" ] ]

Notes

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