• EN
Solved

One attribute to multiple using regEx

  • 16 October 2017
  • 7 replies
  • 29 views
F

Hello,

I am having difficulty finding something similar to this in the Knowledge Center so I thought I would ask.

I have an attribute whose value looks like this (it is a full line from a badly formatted text file).

Ticket No: 4653 Nearest Intersection: Chinton Street Seq No: 34

I need them split to an attribute list and subsequent value list

Eg:

Attribute_1 = Ticket No

Value_1 = 4653

Attribute_2: Nearest Intersection

Value_2: Chinton Street

Attribute_3: Seq No

Value_3: 34

It doesn't have to be a "list" as long as I can separate those part of the texts. I am guessing there should be a RegEx way of doing this.

The attribute names "Ticket No", "Nearest Intersection" and "Seq No" will always remain the same with their values changing. I am trying to build a script to always separate them.

Any suggestions? Thank you!

Addition:

For the first one, if there was a way to extract between 'Ticket No:' and 'Nearest' that would be fine. I can work with the formatting of the value afterwards.

I have 19 lines with the same formatting problems. With 2-3 attributes per line. I am trying to avoid too many transformers per line. If I could possibly use an attribute creator where I create the new attribute and the value would be a the substring between two known attributes on both sides, that would be great.

icon

Best answer by fariyafarhad 16 October 2017, 21:21

View original

7 replies

dustin
Rank
Userlevel 3
Badge +26

Here is my first thought. On the StringConcatenator, I put in 4 elements from the list to account for 4 separate 'words' in the Intersection, i.e. Huntsville Browns Ferry Road. You can add more list elements there if you think you would have that scenario.

F

Here is my first thought. On the StringConcatenator, I put in 4 elements from the list to account for 4 separate 'words' in the Intersection, i.e. Huntsville Browns Ferry Road. You can add more list elements there if you think you would have that scenario.

Hello @cartoscro

 

I appreciate your answer. That seems like a good way to do it. What I have though are many different lines with the same formatting issue. Using so many transformers per line may be a bit too much. I am trying to see if there is an easier Regex way of doing it.
dustin
Rank
Userlevel 3
Badge +26
Hello @cartoscro

 

I appreciate your answer. That seems like a good way to do it. What I have though are many different lines with the same formatting issue. Using so many transformers per line may be a bit too much. I am trying to see if there is an easier Regex way of doing it.
@fariyafarhad There likely is a cleaner way to do it with Regex. FME is very efficient within it's text string processing, so I have always defaulted to the line of thinking with multiple transformers.

 

dustin
Rank
Userlevel 3
Badge +26
Hello @cartoscro

 

I appreciate your answer. That seems like a good way to do it. What I have though are many different lines with the same formatting issue. Using so many transformers per line may be a bit too much. I am trying to see if there is an easier Regex way of doing it.
Slightly smaller workflow:

 

F
Hello @cartoscro

 

I appreciate your answer. That seems like a good way to do it. What I have though are many different lines with the same formatting issue. Using so many transformers per line may be a bit too much. I am trying to see if there is an easier Regex way of doing it.
@cartoscroI like the shorter workflow. Though I think I stumbled upon an answer myself. Thanks!

 

F

I decided to use a StringSearcher.

So for that one line, I set up 3 string searchers in series. If the string is matched, this transformer automatically puts the matching text in an attribute and I specified the attribute name. The first StringSearcher is set up this way.

This means one transformer per

attribute. If any one has a better suggestion, still welcome :)

takashi
Rank
Userlevel 3
Badge +17

Hi @fariyafarhad, why not use the StringSearcher with this regex?

^Ticket No\s*:\s*(.*)\s+Nearest Intersection\s*:\s*(.*)\s+Seq No\s*:\s*(.*)$
And set the Subexpression Matches List Name (e.g. _sub). You can then rename "_sub{0}.part" to "Ticket No", and so on.

Reply


Community Stats

28,818
Posts
109,853
Replies
36,633
Members
Terms & ConditionsCookie settings
A product of
Safe Software respectfully acknowledges that we live, learn and work on the traditional and unceded territories of the Kwantlen, Katzie, and Semiahmoo First Nations.