Forum Discussion

RoutingLoop_179's avatar
RoutingLoop_179
Icon for Cirrus rankCirrus
Mar 28, 2017

Scan or split to find last instance of character.

Hi is there an efficient way to find the last occurrence of a character in a string and split into 2 variables.

 

Currently I use "scan [TCP::payload] {%[^@]@%s} garbage sitename"

 

This splits "USER [email protected]" into 2 variables - $garbage = "USER username" and $siteman = "sitename.com"

 

This works fine until the username proportion contains a "@".

 

So "USER username@[email protected]" now splits into 2 variables $garbage = "USER username" and $sitename = "[email protected]".

 

What I'm trying to achieve is an efficient way of getting $garbage = "USER username@something" and $sitename = "sitename.com". I've looked at TCL split and index but I'll end up with multiple variables which I have to join back up and it won't be consistent as username proportion won't always contain an @, so sometimes i'll have 2 variables others three or (maybe even more if user has used multiple @ in their username).

 

If anyone has any pointers that'd be great.

 

application delivery
irules

1 Reply

  • Kai_Wilke's avatar
    Kai_Wilke
    Icon for MVP rankMVP
    Mar 29, 2017

    Hi RoutingLoop,

    you may try a combination of 

    [string last]
    and 
    [string range]
    to find the position of the last occurrence and to cut the leading/trailing values.

    set input        "test@[email protected]"
set last_at       [string last "@" $input]
set leading_part  [string range $input 0 [expr { $last_at - 1 }]]
set trailing_part [string range $input [expr { $last_at + 1 }] end]

    ... or you could try a combination of 

    [split]
    , 
    [lrange]
    , 
    [lindex]
    and 
    [join]
    to create and select the required 
    [list]
    entries and to format the resulting string. 

    set input         "test@[email protected]"
set input_list    [split $input "@" ]
set leading_part  [join [lrange $input_list 0 end-1] "@"]
set trailing_part [lindex $input_list end]

    Note: Performance wise both example are almost equal. So it will depend only on your personal preferences.

    Cheers, Kai