Help in eliminating character patterns

[ali_hijazi]

Hello
I'm reading data from database table
one of the fields has some of the following values:

• xxx as EMP_NUM, MONTH_NUM, REGION, xxx as EMPH_KEY
• EMP_NUM, MONTH_NUM, xxx as REGION, xxx as EMPH_KEY
• PORTFOLIO, MONTH_NUM, EMPH_KEY
• xxx as EMP_NUM, REGION

what I want to accomplish is the following:
I want to eliminate substrings that start with xxx
so the output for the above sample would be:

• MONTH_NUM, REGION
• EMP_NUM, MONTH_NUM
• PORTFOLIO, MONTH_NUM, EMPH_KEY
• REGION

how can I accomplish this in a tMap component?

so in summary I want to eliminate an substring that contains xxx as YYYY 

[Vegar]

Without knowing the syntax of Talend i suggest that you do this approach. 

Splitt each string by the comma and space (', '), filtering out any items that start with 'xxx as ', and then rejoining the remaining items.

If Talend have regex support, then I suppose that is a way forward to clear out the unwanted parts as well.

[ali_hijazi]

I had this in mind
but I don't know how to use it in a tMap
you know in tMap component I'm dealing row by row 
so I need help in either RegEx or anyting else

[Shicong_Hong]

@ali_hijazi these values are in one line or there are 4 lines?

[ali_hijazi]

the above sample data is 4 rows (lines)

[Shicong_Hong · Accepted solution]

@ali_hijazi You can process the data line by line as shown below, let me know if you have any questions/concerns.

Detailed component settings:

Regards

Shicong