- Subscribe to RSS Feed
- Mark Question as New
- Mark Question as Read
- Float this Question for Current User
- Bookmark
- Subscribe
- Mute
- Printer Friendly Page
NiFi processor to fetch attribute value dynamically from file or table
- Labels:
-
Apache NiFi
Created ‎11-15-2016 06:27 PM
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I want processor to fetch attribute value on run time. Example - if I am filtering twitter feeds by specific keywords, i want to maintain the list of keywords in a separate repository like file or table and not confined as a text box value. In that case, how will NiFi processor fetch that values from external file or table to attribute value.
Created ‎04-19-2017 05:59 PM
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Thanks @Timothy Spann. Looked at your comments just now after i did some modifications to existing nar file. I edited the current ScanAttribute file to iterate the dictionary words and check for its presence in twitter message. Customizing is good and look forward to make more solutions integrated into NiFi.
Created ‎11-15-2016 06:35 PM
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
It sounds like you might be able to use the ScanAttribute Processor to meet your needs.
Created ‎11-23-2016 02:10 AM
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi. Appreciate your answer. Sorry for delay in response - I was away for a while. ScanAttribute processor sounds to be the answer to my question and I did try a sample but its not picking up value and filtering the incoming tweets accordingly. Can you provide an example with snapshots.
Incoming tweets via getTwitter Processor >> ScanAttribute processor. (dictionary file : <location of text file with keywords), attribute pattern - $.text, Match criteria - atleast one value must match)
Created ‎04-17-2017 05:45 PM
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
ScanAttribute processor works fine for matching exact value and not substring of attribute value. I looked into the code {{if (dictionary.contains(entry.getValue())) }} which means matching condition if dictionary words contain any of attribute words. In my case attribute is a tweet - collection of words and would fail all as unmatched.
Need to look at a logic to see if attribute : tweet (collection of words) contain dictionary words. Customization of processor or other ways of extract text from tweet attribute.
Created ‎04-19-2017 03:47 PM
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
@Timothy Spann : Greetings. I need your assistance in customizing processor - ScanAttribute.
Flow file data considering it as tweet_msg attribute needs to be checked if it contains dictionary file words.
Created ‎04-19-2017 04:31 PM
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Scan attribute should work just put in a regular expression that does what you need and just use the one attribute for . maybe it's updated in the new one.
You can create your own processor, starting with the code in ScanAttribute if it doesn't meet your needs then deploy your NAR to the nifi/lib dir and restart nifi.
This is the formatting
https://github.com/tspannhw/nifi-nlp-processor
copy this code
Created ‎04-19-2017 05:59 PM
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Thanks @Timothy Spann. Looked at your comments just now after i did some modifications to existing nar file. I edited the current ScanAttribute file to iterate the dictionary words and check for its presence in twitter message. Customizing is good and look forward to make more solutions integrated into NiFi.
Created ‎04-19-2017 07:47 PM
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
cool, put a JIRA ticket out for that and post it to your own github. perhaps that will get mainlined. Accept your answer.
