stuck with {appname:regexreplace:....}

[dutchclimber]

Situation 1: {AppName} = "TeamViewer (EN)"

Situation 2: {AppName} = "Team Viewer (DE)"

Situation 3: {AppName} = "TeamViewer"

Situation 4: {AppName} = "Team Viewer"

 

Now I would like to use {appname:regexreplace: :} To ged rid of the trailing " (xx)" with the result

"TeamViewer" (as in 1 or 3) or "Team Viewer" (as in 2 or 4).

 

I was hoping a {appname:regexreplace:(.*)(\s\(.*\)):$1} would suffice but that only works for 1 and 2

 

Any suggestions?

R

[FranciscoR]

Regex function is your friend. That's the way to go, yep.

 

{appname:regex:(.*)(\s\(\w+\))?:1}

Edited April 29, 2009 by FranciscoR

[dutchclimber]

Regex function is your friend. That's the way to go, yep.

 

{appname:regex:(.*)(\s\(\w+\))?:1}

 

Just so I understand....

(.*) grabs (just about) anything in a 'greedy' fashion; parenthesis make it available as $1

 

\s is 'space

\( is (

\w+ is one or more word characters

\) is )

last 4 are within () so available as $2

? Makes the preceding item $2 optional. Greedy, so the optional item is included in the match if possible.

 

doesn't the first (.*) grab everything and ignore the rest? i tested this in Expresso and it seems to produce

 

IN OUT

TeamViewer (EN) TeamViewer (EN)

Team Viewer (DE) Team Viewer (DE)

TeamViewer TeamViewer

Team Viewer Team Viewer

So I just get what I started with. In that case a (.*) would have sufficed I suppose ;-)

 

Sorry for turning this forum into a regex newbie school but it's just so damn promising once I get it....

 

Thanks for the patience.

R

[FranciscoR]

I know this is confusing in the beggining. But you seem to be doing well. =)

 

1. Regexreplace is for replaces w/regex. What you are trying to do is to capture a 'matchgroup', hence you have to use regex function. And the code for 'matchgroup' #1 is '1', not '$1' (regexreplace).

 

2. \w+ stands for any alphanumeric character, greedy, yeah. It will stop on'\)'. Number of ways to do it.

 

3. If you test partial matches in Expresso, '(.*)' will match everything, yep. But if you have 2 matchgroups in Ketarin, {appname:regex:(.*)(\s.*):1}, you will capture only the first matchgroup.

 

Better way to understand how this works is to do a couple of tests.

 

------------------------

 

Edit: I see what you mean, sorry. Use instead

 

(\w+[\s]?\w+)(\s\(\w+\))?

Edited April 29, 2009 by FranciscoR

[dutchclimber]

1: bunch of chars, maybe one or more whitespaces followed by a bunch of chars

2: whitespace, a '(', bunch of chars, ')' and all this on a 'maybe' basis

 

I get it !

Thanks!

R