Remove long strings

[sal88]

Hi all

Is there a calculation to remove all long text strings (say anything over 20 characters) from text fields?

Thanks

[comment]

How about:

Left ( YourField ; 20 )

 

[sal88]

The string(s) could appear anywhere within the field. I'm trying to clean up emails where there's lots of strings from hyperlinks that are in the signatures (usually from the email signature).

[comment]

Then I am afraid I don't understand what you mean by "string". Wy don't you provide an example or two of what's in the field and what you would like to see as the result.

[sal88]

Emails are retrieved as records. And the body of the email might contain something like this:

Click this link to unsubscribe from this email: https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmypcsecurity.webrootanywhere.com%2Funsubscribe.aspx%3FT%3Dasdffsdf34228e3b-77d9-4eb2-8132-778fd4dea454%26UEM%3Dsupport%40madso.uk&data=02%7C01%[email protected]%7Casd29c4c0901c808d673d792d0%7C7a80fada51274aa9aaedca692e314d1d%7C1%xxx68237670672dddata=0YhXOxZRHF6%2Fls6fY%2BA4za%2BvsGo6dPBtWMQhVQZo%2ByE%3D&reserved=0

It's these long strings of text that I would like to remove as they make reading emails difficult (especially when there are multiple emails contacenated). But there are so many different ones that I can only think to do a search by length, rather than a specific substitute calculation.

Edited May 18, 2019 by sal88

[comment]

This is not going to be simple, because Filemaker can divide text only to paragraphs (values) and words. In your example, the substring:

https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmypcsecurity.webrootanywhere.com%2Funsubscribe.aspx%3FT%3Dasdffsdf34228e3b-77d9-4eb2-8132-778fd4dea454%26UEM%3Dsupport%40madso.uk&data=02%7C01%[email protected]%7Casd29c4c0901c808d673d792d0%7C7a80fada51274aa9aaedca692e314d1d%7C1%xxx68237670672dddata=0YhXOxZRHF6%2Fls6fY%2BA4za%2BvsGo6dPBtWMQhVQZo%2ByE%3D&reserved=0

is actually 32 separate words, the way Filemaker sees it. So you need to make up your own division - probably by looking for spaces and returns as the only separators to consider.

And of course, this must be a recursive process, taking each "token" in turn and examining it for length. 

In the attached demo, I used a looping script (with 2 loops, one for paragraphs and one for tokens separated by spaces). To make this a calculation, you would need to replicate the same logic using a custom function (or functions).

ShortenLongStrings.fmp12

[OlgerDiekstra]

14 hours ago, sal88 said:

Emails are retrieved as records. And the body of the email might contain something like this:

Click this link to unsubscribe from this email: https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmypcsecurity.webrootanywhere.com%2Funsubscribe.aspx%3FT%3Dasdffsdf34228e3b-77d9-4eb2-8132-778fd4dea454%26UEM%3Dsupport%40madso.uk&data=02%7C01%[email protected]%7Casd29c4c0901c808d673d792d0%7C7a80fada51274aa9aaedca692e314d1d%7C1%xxx68237670672dddata=0YhXOxZRHF6%2Fls6fY%2BA4za%2BvsGo6dPBtWMQhVQZo%2ByE%3D&reserved=0

It's these long strings of text that I would like to remove as they make reading emails difficult (especially when there are multiple emails contacenated). But there are so many different ones that I can only think to do a search by length, rather than a specific substitute calculation.

Maybe display these emails in a webviewer so that the html code gets displayed properly? I did a google on removing urls form emails and got a lot of responses that had negative results from this. Apparently not as easy to do as it has a tendency to break a lot in the emails.

[sal88]

Thanks guys! I'll be exploring both options. I guess the new While function will be handy for comment's trimming option too.