Hi guys and gals
In find-and-replace, does the finding text and replacing text, take parameters allowing it to be different per row?
Hi Richard,
The “Find and replace” step can only take literal string values as an input. Unfortunately, wrapping a column in curly brackets will not perform a dynamic find and replace.
Can you tell us a little more about your use case? What are you hoping to accomplish?
I have two columns, one contains a sentence of text but it’s all lower case. The other contains a list of the properly capitalised proper nouns.
I’m attempting to replace the relevant proper nouns for that sentence, with properly capitalised versions of the same.
minnie is great and tinkerbell is thin
+
Minnie,Tinkerbell
Minnie is great and Tinkerbell is thin.
I’ve Regexed the nouns out of my list but now I need to capitalise the relevant words only.
I was able to come up with a workaround here, but it’s pretty involved and not super scalable. Still thought I would share with you while we think about other possible other solutions!
I put together this quick video for you to check out where I demo the Flow, which you can access your own copy of by pasting this ‘snippet’ anywhere on your canvas: parabola:cb:860e5e6e7dc14cd2877665d9ce49799e
By breaking up the sentence to include 1 word per column, we are able to do some dynamic matching before putting the sentence back together when we set the new column value. As I said, not the most scalable.
Hope this helps! I’ll be sure to come back to this thread if we are able to think of a better solution, and let us know if you have any additional questions.
Hi
Yeah, scalability and repeatability is certainly the issue here, with that solution I would have to overcome problems like
All my sentences are different lengths
All my sentences have different proper nouns in them, both different ones, and different numbers
Each sentence might have the noun in it more than once
If I split my sentence on spaces, I need to find a way to preserve the punctuation while being able to cope with a punctuation mark before the noun or after it (like quotes, brackets, commas, periods, colons etc)
Find and replace, if it worked with {}s was the best option that I could find because it would switch out multiple instances.
In my flow I had managed to work out how to put the lower and upper case version of the noun in a field and have it ready to find/replace/switch out. But none of the available transformations seem to be able to help me dynamically chop up the sentence into before the found noun, and after the found noun, where that noun is dynamic.
Totally understand Richard, if you’d like to send us a link to your Flow at support@parabola.io we are happy to take a deeper look at your data and see if there are any better workarounds that we can suggest!