Skip to main content
Alexandros
Alexandros
Gold ⭐️
January 18, 2020
Solved

NIFI - Extract data from flowfile to write a Query

Hi, I have a flowfile with this format:

<td>Item1</td>
<td class="dest">50.3421</td>
<td class="dest">20.5547</td>

I need to write a query with the parameters, so I need to extract the numbers, with this for example:

(\d{2}.\d{4})

And put the 2 resultos in:

Insert into table(a,b) VALUES($1,$2)

How can I do this?

Thanks!

 

I tryed with replaceText but it's not what im looking for

MattWho
MattWhoAnswer
Titanium ⭐️⭐️⭐️
January 22, 2020

@Alexandros 

 

You can accomplish this use ReplaceText with a more complex Java regular expression.

The Replace Text is designed to replace every occurrence of the string matched by your java regular expression with the replacement value.  So you are probably seeing your replacement value inserted into your existing content twice.

Try using the following java regular expression which will match your entire 3 lines of content:

.*\n.*?(\d{2}.\d{4}).*?\n.*?(\d{2}.\d{4}).*


Leave your replacement value as you already have it and make sure you have Evaluation mode still set to Entire text.

 

Hope this helps,

Matt

View in thread

2 replies

DennisJaheruddi
Platinum ⭐️
DennisJaheruddi
Platinum ⭐️
January 20, 2020

The first thing that comes to mind is the Extract Text processor. It allows you to get (multiple) parts from the text and put it into attributes.

 

org.apache.nifi.processors.standard.ExtractText

Translate
- Dennis Jaheruddin If this answer helped, please mark it as 'solved' and/or if it is valuable for future readers please apply 'kudos'.
MattWho
Titanium ⭐️⭐️⭐️
MattWhoAnswer
Titanium ⭐️⭐️⭐️
January 22, 2020

@Alexandros 

 

You can accomplish this use ReplaceText with a more complex Java regular expression.

The Replace Text is designed to replace every occurrence of the string matched by your java regular expression with the replacement value.  So you are probably seeing your replacement value inserted into your existing content twice.

Try using the following java regular expression which will match your entire 3 lines of content:

.*\n.*?(\d{2}.\d{4}).*?\n.*?(\d{2}.\d{4}).*


Leave your replacement value as you already have it and make sure you have Evaluation mode still set to Entire text.

 

Hope this helps,

Matt

Translate
Terms & ConditionsAccessibility statement