How to apply a GREP style to a name at the end of a para in a caption.

[fidgetyfishPLATFORM]

I can't figure out how to apply a GREP style to my Paragraph style to make the name after the end of the paragraph bold. The name could have a dot in it, or two dots and it could be anything from 1 to 100 characters long. Can you help, here's an example "The imposing 1904 NER.... to colour lights. M.Rhodes

[winterm]

So, that 'name after the end of the paragraph' may or may not have in it: upper and lower case letters, dots, spaces (I assuming it since it may be up to 100 chars long)... Combination dot+space inside 'the name' also possible? Is there any formal difference from the rest of the text?

It seems to me like "Heather, Peter, Jennifer" - can you tell boys from girls with a GREP?

I'd be happy to be corrected, though

[fidgetyfishPLATFORM]

Here are some examples from the captions I have

F.F.P. Media

M.Jamieson

P.Shannon & N.Allsopp

Both D.Allen

D. Allen (2)

So as you can see there are quite a few variations, dots, spaces, ampersands, special characters perhaps maybe even the odd glyph character like ü or é.  I was hoping to be able to use something like this

\Z(?<=).~.

but I'm not getting any results, I've already got a paragraph style and a character style all set up. My GREP for the beginning of the paragraph works, that apply the same character style 'bold green italic' to this string

.+(-~>)

That basically is used if I need to indicate an image Above-, Left-, Right- or Below-

[winterm]

Well, this pretty simple one seems to work on my side, exactly on your given samples:

\.\s\K(\u\.)+.+

Let’s give it a try and see what happens.

The code is not bullet proof at all, it highly depends on your actual text.

There still may be false positives and negatives.

Try and come back with comments!

[fidgetyfishPLATFORM]

This is working for most of the names, some it doesn't but if it captures say, 80% that cut's down my work and I only have to manually apply the style to the ones that it misses.

An example of 1 that didn't work is:

. both FFP Media

Thanks for your help, it's saved me a lot of time, the book will be 288pages!

S

[winterm]

This updated regex will catch cases like . both FFP Media, too:

\.\s\K(both\s|(\u\.)+).+

But only with both, not any other word! Is it sufficient?

As I said earlier, without knowing how actual text looks... etc.

[vinny38]

This might sound like a daft question, but... is there any chance that the text before the "names" never contains any full stop, except the one at the end?

If that so, you might think the other way round and applying a nested style would just be sufficient...

It's probably to easy to be true, but it's worth asking...

[fidgetyfishPLATFORM]

This is a specific example where it didn't work.

The 1961 box at Colwich, built at the time of electrification and resignalling of the WCML closed and the work of it’s 40 levers was added to the Stoke signalling centre. FFP Media

[vinny38]

Really? It does work for me...

By the way, can you be more specific about your workflow to create those captions? Do you use Metadata variables, dynamic or static captions?

[Obi-wan Kenobi]

Aha!  Once again: Variable!? 

By Security, Grep style:

Bold

~v$

(^/)

[vinny38]

Aha!  Once again: Variable!? 

?

Pas compris là...

[fidgetyfishPLATFORM]

No The caption aren't in the metadata, the author has just supplied these in word and I have to copy paste them across, so they're Static captions. I've applied an object style that applies the para style to the box and the grep hopefully catches most of the styling in there for Right- Text in Caption. Author. of Image

At the moment with you're help it's capturing 90% of the image authors which is fab, the others I'm manually styling.

Sorry I don't understand the Variable!?  By Security, Grep style: Bold ~v$ (^/) bit, I wish I did.

S

[vinny38]

OK no variables then...

But I don't understand why your last example doesn't work.

Can you make sure no character style is applied?

Can you share the result?

[Srishti Bali]

Hi fidgetyfishPLATFORM,

I would like to know if the steps suggested above worked for you, or the issue still persists.

Kindly update the discussion if you need further assistance with it.

Thanks,

Srishti

[fidgetyfishPLATFORM]

Only this worked

\.\s\K(Both\s|(\u\.)+).+

nothing else did, I have another book, similar issue. this time around, the above didn't work and neither did this

\.\s\K(\u\.)+.+

This time the captions are not as similar, here's a page that gives you an idea, I've manually applied the character style to the highlighted areas but really I want a GREP solution for this as I know the book will be around 300pages at the end.

Can anyone help?

[winterm · Accepted solution]

Basically, you need to catch the last sentence? Well, there’s no universal definition for *sentence* in GREP.

These expressions will catch the longest possible run at the end of a paragraph that does not contain a period, or question, or exclamation mark (one of a possible *sentence* definitions).

If there’s no punctuation at the very end, this one is enough:

[^.?!]+$

This one will take care about .?! at the end of the last sentence, too:

[^.?!]+[.?!]?$

However, both fail with samples like this:

Taxes raised by 30.5%!

You’ll catch only 5%! using the second regex, and nothing at all using the first.

Not a good solution for exercise book, but may work well enough for actual job.

[fidgetyfishPLATFORM]

The 1st option [^.?!]+$ does what I need it to do so thanks, that was very helpful. I wish I understood GREP expressions better, they're so useful.

Sarah

[grudgemom]

What about manually putting in a non printing character like a non-breaking space in between the caption and the name, then formatting your grep to say find the nbsp and change everything to the end of the paragraph to this format? (Sorry no help with the actual grep code for this)