regex url next slash mark?

youtuber · March 2, 2018

I would like to get only this part of this url https://autoit.com/test1 but all my tests are failing thanks for this . thanks

$url = "https://autoit.com/test1/test2/"
$Reg = StringRegExp($url, 'https?://*.*[^/]+', 3)

ConsoleWrite($Reg[0] & @CRLF)

ripdad · March 3, 2018

Much simpler than straining with a regex. Took me 2 minutes and I know it's accurate.

Local $url = 'https://autoit.com/test1/test2/'
Local $str = StringLeft($url, StringInStr($url, '/', 0, 4) - 1)
MsgBox(0, '', $str)

Bilgus · March 3, 2018

Took me 5 and its more flexible

$url = "https://autoit.com/test1/test2/test3/test4, http://autoet.com/test5/test6/, ftp://autovt.com/test7/test8/ "

$Reg = StringRegExp($url, '(?i).*?(..?tp.?://[^/]*.?[^/]*)', 3)
; (?i) caseless matching
; .*? 0 or more characters but doesn't keep them
; ( start a capturing group
; . any character 'f or h'
; .? 0 or 1 char 't or `none`'
; tp literally tp
; .? 0 or 1 char 'http>S<'
; : literal colon
; // literal slashes
;[^/]* all characters till a slash
;.? or or 1 character
;[^/]* all characters till a slash (again)
;) end capturing group
ConsoleWrite($Reg[0] & @CRLF & $Reg[1] & @CRLF& $Reg[2] & @CRLF)

Bilgus · March 3, 2018

oh outputs:

https://autoit.com/test1
http://autoet.com/test5
 ftp://autovt.com/test7

youtuber · March 3, 2018

@ripdad is also not suitable for me because I will do url verification at the same time.

@Bilgus your friend regex has worked well for me thank you.

(?i).*?(..?tp.?://[^/]*.?[^/]*)

maybe there is another alternative @mikell

mikell · March 3, 2018

For the fun

$sReg = StringRegExpReplace($url, '(.*?/){3}[^/]+\K.*', "")

But the usual way is more understandable/accurate/secure

$aReg = StringRegExp($url, 'https?://[^/]+/[^/]*', 1)

iamtheky · March 3, 2018

2 hours ago, youtuber said:

is also not suitable for me because I will do url verification at the same time.

Are you doing an inetget on the return to verify the url or are you just validating that the first 8 characters are https://? either way it's plenty suitable within the existing parameters.

mikell · March 3, 2018

2 hours ago, iamtheky said:

either way it's plenty suitable within the existing parameters

I totally agree. The code from ripdad in post #2 works the same way than my srer (find the 4th occurence of "/" and grab all chars on the left)
A less classy look maybe (though this could be discussed, for sure) but equal efficiency

Jury · March 4, 2018

$url = "https://autoit.com/test1/test2/"
$Reg = StringRegExpReplace($url, '(https://.*?/\w+).*?$', '$1')

ConsoleWrite($Reg & @CRLF)

Bilgus · March 4, 2018

OP said he wanted to validate the URL at the same time so how about we start getting crazy

$url = "https://autoit.com/test1/test2/test3/test4, http://autoet.com/test5/test6/, http://blank.com, http://blank.com/Nope/ ftp://autovt.com/test7/test8/ http://autoet.com/test9/test10/ "

$aReg = StringRegExp($url, '(?i)(?x) (?(DEFINE) (?<scheme>..?tps?://)(?<host>\w+\.\w{3}/?)(?<path>[^/\s,]+/))(?&scheme)(?&host)(?&path)(?&path)', 3)

For $i = 0 to UBound($aReg) - 1
    ConsoleWrite($aReg[$i] & @CRLF)
Next

https://autoit.com/test1/test2/
http://autoet.com/test5/test6/
 ftp://autovt.com/test7/test8/
http://autoet.com/test9/test10/

Edited March 4, 2018 by Bilgus
Forgot the outputs

youtuber · March 4, 2018

@Bilgus Thank you for your help

@mikell and @Bilgus I would like to ask you another question, what should be the pattern to cut the http:// or https:// and www. at the beginning?

https://autoit.com/test1/test2/
http://aut-oet.com/test5/test6/
https://www.autovt.com/test7/test8/
http://aut.oet.com/test9/test10/

(?:http[s]?:\/\/)?(?:www\.)?([^\/]+\/[^\/]*)

Edited March 4, 2018 by youtuber

Bilgus · March 4, 2018

You have to turn them into non capturing groups

$url = "test.com/test0/ https://autoit.com/test1/test2/test3/test4, http://autoet.com/test5/test6/, http://blank.com, http://blank.com/Nope/ ftp://autovt.com/test7/test8/ http://autoet.com/test9/test10/ "

;This one is wrong
;$aReg = StringRegExp($url, '(?i)(?(DEFINE)(?<host>\w+\.\w{3}/?)(?<path>[^/\s,]+/))(?!..?tps?://)(?&host)(?&path)', 3)
$aReg = (?i)(?x) (?(DEFINE) (?<scheme>..?tps?://)(?<host>\w+\.\w{3}/?)(?<path>[^/\s,]+/))(?&scheme)\K(?&host)(?&path)

For $i = 0 to UBound($aReg) - 1
    ConsoleWrite($aReg[$i] & @CRLF)
Next
ConsoleWrite(@CRLF)

$bReg = StringRegExp($url, '(?i).*?..?tps?://([^/]*.?[^/]*)', 3)

For $i = 0 to UBound($bReg) - 1
    ConsoleWrite($bReg[$i] & @CRLF)
Next

autoit.com/test1/
autoet.com/test5/
blank.com/Nope/
autovt.com/test7/
autoet.com/test9/

autoit.com/test1
autoet.com/test5
blank.com, http:/
autovt.com/test7
autoet.com/test9

Edited March 4, 2018 by Bilgus
Fixed

Bilgus · March 4, 2018

So for the top one first off you can see the validation is a bit more robust

I added \K to restart the match after matching Scheme

since as far as I know you can't use a named group as a capturing group

but we want it to be there for a valid match so we match it then reset the position for matching just after

Bilgus · March 4, 2018

Really for what you want it could be shortened to

$aReg = StringRegExp($url, '(?i)..?tps?://\K\w+\.\w{3}/?[^/\s,]+/', 3)

or

$aReg = StringRegExp($url, '(?i)(?x) (?(DEFINE)(?<path>[^/\s,]+/))..?tps?://\K\w+\.\w{3}/?(?&path)(?&path)', 3)

the second being just in case you decide you want more than a single portion of the path

iamtheky · March 4, 2018

or you could just do this

;~ $str = "https://autoit.com/test1/test2/"
;~ $str = "http://aut-oet.com/test5/test6/"
$str = "https://www.autovt.com/test7/test8/"
;~ $str =  "http://aut.oet.com/test9/test10/"


$sMid = Stringmid(stringleft($str , StringInStr($str , "/"  , 0 , 4) - 1) , stringinstr($str , "aut"))

msgbox(0, '' , $sMid)

youtuber · March 4, 2018

@Bilgus unfortunately some of your regex pattern unsuccessful domains are bypassing!

$url = "https://autoitscript.com/test1/test2/" & @CRLF & _
        "http://aut-oit.com/test5/test6/" & @CRLF & _
        "https://www.autoit.com/test7/test8/" & @CRLF & _
        "http://blog.autoitscript.com/test9/test10/"

$aReg1 = StringRegExp($url, '(?i)..?tps?://\K\w+\.\w{3}/?[^/\s,]+/', 3)
$aReg2 = StringRegExp($url, '(?i).*?..?tps?://([^/]*.?[^/]*)', 3)
$aReg3 = StringRegExp($url, '(?i)(?(DEFINE)(?<host>\w+\.\w{3}/?)(?<path>[^/\s,]+/))(?!..?tps?://)(?&host)(?&path)', 3)

ConsoleWrite("----" & "$aReg1 " & "----" & @CRLF)
For $i = 0 To UBound($aReg1) - 1
    ConsoleWrite($aReg1[$i] & @CRLF)
Next
ConsoleWrite(@CRLF)

ConsoleWrite("----" & "$aReg2 " & "----" & @CRLF)
For $i = 0 To UBound($aReg2) - 1
    ConsoleWrite($aReg2[$i] & @CRLF)
Next
ConsoleWrite(@CRLF)

ConsoleWrite("----" & "$aReg3 " & "----" & @CRLF)
For $i = 0 To UBound($aReg3) - 1

    ConsoleWrite($aReg3[$i] & @CRLF)
Next
ConsoleWrite(@CRLF)

console

----$aReg1 ----
autoitscript.com/test1/
www.autoit.com/
blog.autoitscript.com/

----$aReg2 ----
autoitscript.com/test1
aut-oit.com/test5
www.autoit.com/test7;---->here www. I do not want the part
blog.autoitscript.com/test9

----$aReg3 ----
autoitscript.com/test1/
oit.com/test5/
www.autoit.com/
blog.autoitscript.com/

Looks better

$aReg4 = StringRegExp($url, '(?i).*?..?tps?://?(?:www\.)?([^/]*.?[^/]*)', 3)

Edited March 4, 2018 by youtuber

Bilgus · March 4, 2018

I updated the post you might want to recheck it Sorry I realized when I was writing a follow up post that it wasn't right

youtuber · March 4, 2018

@Bilgus @mikell or Is it better? I do not know

(?i)(?:https?://)(?:www\.)?([^/]*.?[^/]*)

Edited March 4, 2018 by youtuber

Bilgus · March 5, 2018

$url = "https://autoitscript.com/test1/test2/" & @CRLF & _
        "http://aut-oit.com/test5/test6/" & @CRLF & _
        "https://www.autoit.com/test7/test8/" & @CRLF & _
        "https://google.com/test1/test2/" & @CRLF & _
        "http://qwerty.org/test5/test?/whatabout_a_page.html" & @CRLF & _
        "https://www.tungsten.com/test7/test8/metalisbetter.htm" & @CRLF & _
        "https://falsedomains.com/test1/test2/" & @CRLF & _
        "http://who%20is%20this/test5/test6/" & @CRLF & _;NO
        "https://www.autochecker.com/test7/test8/" & @CRLF & _
        "https://autobanks.com" & @CRLF & _;NO
        "http://aut_ofitbaddomain" & @CRLF & _;NO
        "https://www.autoit.biz/test7/test8/" & @CRLF

$aReg = StringRegExp($url, '(?i)https?://(?:www\.)?\K\S+\.\w{3}/?[^/\s,]+/', 3)

For $i = 0 to UBound($aReg) - 1
    ConsoleWrite($aReg[$i] & @CRLF)
Next
ConsoleWrite(@CRLF)

$bReg = StringRegExp($url, '(?i)(?:https?://)(?:www\.)?([^/]*.?[^/]*)',3)

For $i = 0 to UBound($bReg) - 1
    ConsoleWrite($bReg[$i] & @CRLF)
Next

Its easy enough to test

;'(?i)https?://(?:www\.)?\K\S+\.\w{3}/?[^/\s,]+/'
autoitscript.com/test1/
aut-oit.com/test5/
autoit.com/test7/
google.com/test1/
qwerty.org/test5/
tungsten.com/test7/
falsedomains.com/test1/
autochecker.com/test7/
autoit.biz/test7/


;'(?i)(?:https?://)(?:www\.)?([^/]*.?[^/]*)'

autoitscript.com/test1
aut-oit.com/test5
autoit.com/test7
google.com/test1
qwerty.org/test5
tungsten.com/test7
falsedomains.com/test1
who%20is%20this/test5
autochecker.com/test7
autobanks.com
http:/
autoit.biz/test7

as you can see the first one is still more robust although I still had to change it a bit to match your data URI have a list of valid characters but I don't know that I'd want to build a regex for them lol

Edited March 5, 2018 by Bilgus

youtuber · March 5, 2018

I do not want the right / slash at the end

(?i)https?://(?:www\.)?\K\S+\.\w{3}/?[^/\s,]+/

Sign In

regex url next slash mark?

Recommended Posts

youtuber

ripdad

Bilgus

Bilgus

youtuber

mikell

iamtheky

mikell

Jury

Bilgus

youtuber

Bilgus

Bilgus

Bilgus

iamtheky

youtuber

Bilgus

youtuber

Bilgus

youtuber

Create an account or sign in to comment

Create an account

Sign in

Recently Browsing 0 members

Browse

AutoIt Resources

Release

Beta