stringregexpreplace question

[MrCheese]

argh, pulling my hair out.

considering this post: 

 

say for a string = "03a", how can I strip out the leading 0 and the a.

I have tried:

$new = StringRegExpReplace($string, '[^1-9][^0-9]', '')

 

and various combinations:

^0+[^0-9]

[^[:digit:]]

"[^0].*"

"^0*(d+)"

 

I'm going loopy!

 

 

[mikell]

This one will remove leading zero(s) and trailing non digit(s) - if exist

$new = StringRegExpReplace($string, '^0*|\D*$', "")

Edit : comments
 You want your regex to have 2 different actions, so an alternation is needed
^0*  : zero or more "0" at the beginning of the string
|       : alternation
\D*$   :  zero or more non-digit characters at the end of the string

Edited October 22, 2018 by mikell

[FrancescoDiMuro]

@mikell
Thanks for the comments! 
I'm sure that your comments help a lots who wants to learn more about SRE and SRER
Appreciate that! 

[MrCheese]

@mikell thanks for your help! Solution worked well.

[caramen]

@mikell& @FrancescoDiMuro I am sure a single file that resuming all delimiter whould be apreciate from everyone. I know the help file is already here but... even for people who want to learn this part is hard.  

 

So yeah comment like this is realy appreciated.

 

In help file it is like that : 
 

^0*  : zero or more digit at the beginning of the string

(?: ... )   Non-capturing group. Does not record the matching characters in the array and cannot be re-used as backreference.
\D

Matches any non-digit.
[^ ... ]    Matches any character not in the set: "[^0-9]" matches any non-digit. To include a caret (^) in a set, put it after the beginning of the set or escape it (\^).

But we dont got that with exemple at each delimiter :

^0*  : zero or more "0" at the beginning of the string
|       : alternation
\D*$   :  zero or more non-digit characters at the end of the string

It whould simplifi the life of all beginner. Since i am not strong at it i cannot do that  

Edited October 22, 2018 by caramen

[iamtheky]

not for nothing, but if your only criteria are stripping leading zeroes and removing trailing non-number characters

$new = number("03a")

 

Edited October 22, 2018 by iamtheky
grammar

[mikell]

iamtheky,
Very smart non-regex way 
I made a regex-oriented answer because ... of the title of the topic. Nothing personal 

caramen,
The help file provides explicit details and explanations about the syntax, and should constantly be used as reference and be read carefully to avoid confusion
Examples :

^0*  : zero or more digit at the beginning of the string

This is false. "0" means only 0 . A digit is \d, or [0-9], or [[:digit:]]  (all the same)

Ditto for ^ , which matches "start of string" but is also used inside a character class to negate its content
So to match any non-digit, you might use \D , or [^0-9] , or  [[:^digit:]]

[jchd]

On 22/10/2018 at 2:19 PM, mikell said:

A digit is \d, or [0-9], or [[:digit:]]  (all the same)

Not exactly all the same.

When/if (*UCP) is used, [0-9] does exactly what it means (ASCII 0 to ASCII 9) while both other forms include codepoins that have the Nd (Numeric Decimal) properties. And that is a number of characters: from old Unicode v5.1:

Codepoint    Glyph    CharacterName
U+00030     '0'     DIGIT ZERO
U+00660     '٠'     ARABIC-INDIC DIGIT ZERO
U+006F0     '۰'     EXTENDED ARABIC-INDIC DIGIT ZERO
U+007C0     '߀'     NKO DIGIT ZERO
U+00966     '०'     DEVANAGARI DIGIT ZERO
U+009E6     '০'     BENGALI DIGIT ZERO
U+00A66     '੦'     GURMUKHI DIGIT ZERO
U+00AE6     '૦'     GUJARATI DIGIT ZERO
U+00B66     '୦'     ORIYA DIGIT ZERO
U+00BE6     '௦'     TAMIL DIGIT ZERO
U+00C66     '౦'     TELUGU DIGIT ZERO
U+00CE6     '೦'     KANNADA DIGIT ZERO
U+00D66     '൦'     MALAYALAM DIGIT ZERO
U+00E50     '๐'     THAI DIGIT ZERO
U+00ED0     '໐'     LAO DIGIT ZERO
U+00F20     '༠'     TIBETAN DIGIT ZERO
U+01040     '၀'     MYANMAR DIGIT ZERO
U+01090     '႐'     MYANMAR SHAN DIGIT ZERO
U+017E0     '០'     KHMER DIGIT ZERO
U+01810     '᠐'     MONGOLIAN DIGIT ZERO
U+01946     '᥆'     LIMBU DIGIT ZERO
U+019D0     '᧐'     NEW TAI LUE DIGIT ZERO
U+01B50     '᭐'     BALINESE DIGIT ZERO
U+01BB0     '᮰'     SUNDANESE DIGIT ZERO
U+01C40     '᱀'     LEPCHA DIGIT ZERO
U+01C50     '᱐'     OL CHIKI DIGIT ZERO
U+024EA     '⓪'     CIRCLED DIGIT ZERO
U+024FF     '⓿'     NEGATIVE CIRCLED DIGIT ZERO
U+0A620     '꘠'     VAI DIGIT ZERO
U+0A8D0     '꣐'     SAURASHTRA DIGIT ZERO
U+0A900     '꤀'     KAYAH LI DIGIT ZERO
U+0AA50     '꩐'     CHAM DIGIT ZERO
U+0FF10     '０'     FULLWIDTH DIGIT ZERO
U+104A0     '𐒠'     OSMANYA DIGIT ZERO
U+1D7CE     '𝟎'     MATHEMATICAL BOLD DIGIT ZERO
U+1D7D8     '𝟘'     MATHEMATICAL DOUBLE-STRUCK DIGIT ZERO
U+1D7E2     '𝟢'     MATHEMATICAL SANS-SERIF DIGIT ZERO
U+1D7EC     '𝟬'     MATHEMATICAL SANS-SERIF BOLD DIGIT ZERO
U+1D7F6     '𝟶'     MATHEMATICAL MONOSPACE DIGIT ZERO
U+E0030     '󠀰'     TAG DIGIT ZERO

It's a pity that most glyphs don't show up correctly. That's the extra price you pay for being born in a non-latin script area.

[mikell]

I apologize for my too early generalization  

[jchd]

It's simply that we, occidentals, are soooo used to think only within some kind of ASCII that we too often forget that the vast majority of humans don't use any form of latin scripts, letters, digits and punctuation.

Unicode tries hard (and succeeeds) to include less known (to occidentals) scripts and hence makes the less "common" scripts a tangible reality.

[mikell]

As my previous code in post #(unknown) didn't use (*UCP) I assume that I may be forgiven (for this time)