Help

Re: Extract Text From Between Two Set Characters

4876 0
cancel
Showing results for 
Search instead for 
Did you mean: 
Sal_Suissa
6 - Interface Innovator
6 - Interface Innovator

Hi!

This one seems quite easy, but having a tough time to loop LEFT and RIGHT to make it work …

I have the following examples
5-DBR_4.JPG
5-GD-WHI_1.jpg

In both instances, I want to extract the text between the “_” and the “.JPG”, so “4” and “1” respectively.
This is to populate an {Position} field.

What is the right formula to use for this?

8 Replies 8
Grant_Andrew
5 - Automation Enthusiast
5 - Automation Enthusiast

Howdy -

@Melanie_E_Magdalena and I had a crack at this for you - not sure how variable your data is, but with the samples you gave, try this:

LEFT(RIGHT(Name,5),1)
where {Name} = the field name

41

Hope that helps!!

Grant

If you can’t be sure of the relative position of the number — for instance, if your filetype suffix might be ‘jpg’ or ‘jpeg’, or if the position text might be more than a single digit — but you know the relative locations of the underscore and period stay the same, you can use MID() with more confidence:

MID(
    {Text},
    FIND(
        '_',
        {Text}
        )+1,
    FIND(
        '.',
        {Text},
        FIND(
            '_',
            {Text}
            )
        )-(FIND(
            '_',
            {Text}
            )+1)
    )

That will extract a string of any length between the first '_' in {Text} and the first '.' in {Text} that falls after the "_".

That is,

5-DBR_32.jpg will return 32
5-GD-WHI_10011.jpeg will return 10011

and

36-LT.BLUE_8.wbem will return 8.

However,

6-DRK_RED_7.GIF will return RED_7

– embedded underscores will trip up the formula.

THANK YOU!!!
This is an AMAZING find. I am slow with airtable formulas as a rule - but this got me over a hump I’d been working around for 2 weeks. Specially, how to extract the web url of the attachment field (an image in my case) that does not contain the parentheses. Testing it now but positive results so far

“parent_file (https://dl.airtable.com/.attachments/b85ad854f7d480569ea1a020c00f8792/a6ba6ba112/parent_file)”

I modified the formula above (my column is called “attachment_url” ) and it worked like a charm
My method was to

  1. add an interim formula column (sloppy I know) that pulled the full text of the attachment URL into it - “parent_file (https://dl.airtable.com/.attachments/b85ad854f7d480569ea1a020c00f8792/a6ba6ba112/parent_file)”
  2. add a second formula column that looked for stuff between the parentheses and after the first parentheses
    MID(
    {attachment_url},
    FIND(
    ‘(’,
    {attachment_url}
    )+1,
    FIND(
    ‘)’,
    {attachment_url},
    FIND(
    ‘(’,
    {attachment_url}
    )
    )-(FIND(
    ‘(’,
    {attachment_url}
    )+1)
    )

the result?
https://dl.airtable.com/.attachments/b85ad854f7d480569ea1a020c00f8792/a6ba112/parent_file

This helped me find the text between two parentheses - regardless of the length of the string between them! Sorry I repeated this because I wanted to make sure I could say this in as many ways as I could so people can find this when they google it!

Seriously, thank you so much!

The code above works with different characters ie < and >, ( and ). I wanted to use |bars| in my code/scoring system, so worked on this that my supervisor finally got working:

MID( {Text}, FIND(’|’,{Text})+1, (FIND(’|’,{Text},(FIND(’|’,{Text})+1)) - (FIND(’|’,{Text}))-1))

For some reason the one that works on two characters returned a null/blank value for the | bar| version.

Dalkat
5 - Automation Enthusiast
5 - Automation Enthusiast

This was super helpful! I have a somewhat related question, I’m hoping I could get some help on.

I’m trying to extract the value in a string query, where sometimes it’ll be the only value, and sometimes there will be others.

E.g.,
getpresently.com/hello?event=123
getpresently.com/hello?event=123&utm=newsletter&test=hi
getpresently.com/goodbye?event=12345&utm=newsletter&test=hi

I’m curious if there’s an easy way to do this. There are a few variable digits which make it hard to do a simple left/right:

  1. slug character count may vary (e.g., hello vs goodbye)
  2. event # of digits may vary (e.g., 123 vs 12345)
  3. sometimes there will be multiple strings (&utm, &test) and sometimes there will be just the one (?event). I don’t know yet if ?event will always be the first one either.

Is there any way to pull out just what’s after the “event=” but skip anything that may follow an &?

Benson_G_Munyan
4 - Data Explorer
4 - Data Explorer

So I am trying to leverage this against the following string:

New voicemail in mailbox 1 from “FIRSTNAME LASTNAME” <123456789>

Using:

MID(
Subject,
FIND(
‘"’,
Subject
)+1,
(FIND(
‘"’,
Subject
)
)-(FIND(
‘"’,
Subject
)+1)
)

But all it returns is “-1”. What have I missed? Any help is appreciated!

left(
substitute(left(Text,6+find(‘event=’,Text)),‘’),
find(‘&’,Text)-find(‘event=’,Text)-6
)