Find/any doesn't seem to find strings in blocks..
[1/5] from: bhandley::zip::com::au at: 20-Aug-2000 11:35
I would have expected this to work. Am I missing something?
>> find/any ["my-string"] "my*"
== none Brett. --
[2/5] from: etcha:gsat:au at: 20-Aug-2000 12:00
whats wrong with:
>> blah: "my-string"
>> blah: parse blah "-"
== ["my" "string"]
>> if find blah "my" [print ["yes"]]
its not what you want, but you cant search the string for something thats not there, eg "my*" :) etcha
[3/5] from: rebol:techscribe at: 19-Aug-2000 21:44
>> help find
[...] /any -- Enables the * and ? wildcards. [...] Case 1:
>> string: "xdefgjkljyjkljweruiouz" >> find/any string "y*z"
== "yjkljweruiouz" Case 2:
>> block-with-string: reduce [string]
>> find/any block-with-string "y*z"
== none Should find's behavior be the same in both cases? At 12:00 PM 8/20/00 +1000, you wrote:
>whats wrong with: >>> blah: "my-string"
<<quoted lines omitted: 24>>>>== http://www.zipworld.com.au/~bhandley/rebol >>
;- Elan [ : - ) ] author of REBOL: THE OFFICIAL GUIDE REBOL Press: The Official Source for REBOL Books http://www.REBOLpress.com visit me at http://www.TechScribe.com
[4/5] from: bhandley:zip:au at: 20-Aug-2000 23:45
Hi Elan, Good question. I assumed one sort of behaviour but it is not necessarily obvious is it? Having thought on it a little I think the /any refinement should be activated for the block situation or the help/documentation changed to indicate it will not work. My preference is for the former since we should get those other great refinements as part of the bargin. I suspect that the behaviour should not be the same in both cases as the cases are of different types. Since Find is defined as "Finds a value in a series..." - its behaviour should depend on the values of the series it is given. The values of a string are characters so Find returns us the string with the current index of the string set appropriately - resulting in a part string. So, I suggest that it makes sense that the result of the find on a block should be the block with the current index set appropriately, as it does now, but with the /any refinement enabled to influence the matching. Any comments? Brett.
[5/5] from: bhandley:zip:au at: 21-Aug-2000 0:19
Hm. Which means I want a consistent behaviour (setting the index appropriately according to the value found) even if it looks different (one returns what looks like a part string - the other returns what looks like a part block, not a part string). Brett.
- Quoted lines have been omitted from some messages.
View the message alone to see the lines that have been omitted