Comparing strings char by char

[firace]

I want to compare 2 strings of equal length and find which characters differ and which are equal.
I managed to do this with the below code (note: must be compiled as ASCII), but one little detail that I don't understand is why I need to subtract 1 from @r0+i in the PokeC. Could anyone with a sharper brain answer this one?

Code: Select all

s1.s = "24cdef5e8db4afca39a4198d4df5d6e4"
s2.s = "14cd9f5e8db2afca39a41c8d4dfcd6eb"
r0.s = "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"

For i = 0 To Len (s1) 
  If Mid(s1,i,1) =  Mid(s2,i,1) : PokeC (@r0+i-1,'=') : EndIf
Next
    
Debug r0

Thanks!

[wilbert]

PB string positions start with 1, not 0 so your loop should be from 1 to Len(s1)

[firace]

PB string positions start with 1, not 0 so your loop should be from 1 to Len(s1)

Ouch, of course!! Thanks!

[Lord]

Why not using PeekC()?

Code: Select all

s1.s = "24cdef5e8db4afca39a4198d4df5d6e4"
s2.s = "14cd9f5e8db2afca39a41c8d4dfcd6eb"
r0.s = "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"

AdrS1.i=@s1
AdrS2.i=@s2
AdrR0.i=@r0
L.i=Len(s1)-1

For i = 0 To L
  If PeekC(AdrS1+i) =  PeekC(AdrS2+i) : PokeC (AdrR0+i,'=') : EndIf
Next

Debug r0

[STARGÅTE]

This code doesn't work with unicode.

Here an other example:

Code: Select all

Structure CharacterArray
	C.C[0]
EndStructure

s1.s = "24cdef5e8db4afca39a4198d4df5d6e4"
s2.s = "14cd9f5e8db2afca39a41c8d4dfcd6eb"
r0.s = "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"

*AdrS1.CharacterArray = @s1
*AdrS2.CharacterArray = @s2
*AdrR0.CharacterArray = @r0

Define I = 0

While *AdrR0\C[I]
	If *AdrS1\C[I] = *AdrS2\C[I]
		*AdrR0\C[I] = '='
	EndIf
	I + 1
Wend

Debug r0

@IdeasVacuum: My code is unicode compatible

[IdeasVacuum]

All these solutions rely on using ASCII - but what about Unicode strings?

[said]

Hi,
An alternative that works with ascii/unicode (and faster than Poke/Peek)

Code: Select all

s1.s = "24cdef5e8db4afca39a4198d4df5d6e4"
s2.s = "14cd9f5e8db2afca39a41c8d4dfcd6eb"
r0.s = "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"


*C1.Character = @s1
*C2.Character = @s2
*C3.Character = @r0
While *C1\c
    If *C1\c = *C2\c : *C3\c = '=' : EndIf
    *C1 + SizeOf(Character)
    *C2 + SizeOf(Character)
    *C3 + SizeOf(Character)
Wend
Debug r0

edit: updated to include the *Ci + .... (lost with copy/paste)

[Lord]

This code doesn't work with unicode.
...[/code]

The premise was:

...
(note: must be compiled as ASCII)
...

And not everybody wants/needs unicode.
Leave BASIC basic.

[STARGÅTE]

Support for Ascii compilation ends after the next LTS cycle

[Lord]

Support for Ascii compilation ends after the next LTS cycle

As I already stated: BAD NEWS!
Keep BASIC basic!

[Tenaja]

Keep BASIC basic!

ASCII vs Unicode has nothing to do with BASIC being basic.

Please imagine what you are saying, from a the perspective of other people...

Keep BASIC English.

Because English is one of the few languages without unicode characters.

[Lord]

...

Keep BASIC English.

Because English is one of the few languages without unicode characters.

So what?
My native language is not English.

[marroh]

[offtopic on]

And not everybody wants/needs unicode.

+1

[offtopic off]

[Tenaja]

[offtopic on]

And not everybody wants/needs unicode.

+1

[offtopic off]

Your complaints are moot, considering the decisions of the dev team. OTOH, you can always use a "legacy" version of PB.

[netmaestro]

Just one point to consider, depending upon how many string comparisons would test as equal you might realize a considerable speed gain by first doing a hash check on the strings and if they are the same, stop there. If not, then compare char by char.