[1.7.0.305] GREP & two following conjunctions

[Wojciech Krakowiak 22]

Here's the case: I want to place a non-breaking space before all conjunctions in document.

I use this GREP rule:
find: (\b)(a|i|o|u|w|z|A|I|O|I|W|Z)(\b\s)
replace: $2$3

Everything works fine until it meets two following conjunctions, like "i w". Then it changes this expression into "I", and the second conjunction "w" gets moved to the place of next conjunction in document and all following conjunctions in document get moved to the place of next conjunctions.

To clarify it, download Publisher document that I attach, use the mentioned grep rule and see what happens with all conjunctions.

I hope this is understandable

Without this fix, this GREP rule is useless in polish language, because we have a lot combinations like " i w ", " a w "

 

test.afpub

[Seneca 397]

2 hours ago, Wojciech Krakowiak said:

I want to place a non-breaking space before all conjunctions in document.

Hello @Wojciech Krakowiak.

Your find Grep adds space (not the non-breaking space) after the conjunction and not before.

I modified your grep slightly. Please test it now and let me know whether this works for you now.

FIND: \b(a|i|o|u|w|z|A|I|O|I|W|Z)\b\s

REPLACE: $1(add the non breaking space from the pop up menu)

 

Edit: This ads the non-breaking space after the conjunction. I think this is what you want.

[walt.farrell 2,932]

3 hours ago, Wojciech Krakowiak said:

(a|i|o|u|w|z|A|I|O|I|W|Z)

By the way, shouldn't that be (a|i|o|u|w|z|A||O|U|W|Z) instead?

[Seneca 397]

10 minutes ago, walt.farrell said:

(a|i|o|u|w|z|A||O|U|W|Z)

Well spotted.

However, even this one has a small typo :-):

(a|i|o|u|w|z|A||O|U|W|Z)

[Old Bruce 900]

49 minutes ago, Seneca said:

Well spotted.

However, even this one has a small typo :-):

(a|i|o|u|w|z|A||O|U|W|Z)

I think there is a missing EYE in there.

[walt.farrell 2,932]

1 hour ago, Seneca said:

Well spotted.

However, even this one has a small typo :-):

(a|i|o|u|w|z|A||O|U|W|Z)

Almost. (a|i|o|u|w|z|A|I|O|U|W|Z)

Thanks. I knew I had messed it up, but thought I had fixed it before.

[Wojciech Krakowiak 22]

@Seneca @Old Bruce @walt.farrell – wow, thanks for your interest and discussion.

However, publisher still changes position of the conjunctions. See screenshots: before, after and what happened

 

[Wosven 600]

You can use simpler expressions, and run it twice (for the second one in  "i w") :

_space_([aiouwzAIOUWZ])_space_

replace with:

_nobreakSpace_$1_space_

 

 

Comment: [abc]  expression will select a (ONE) character depending of the ones listed between the brackets, no need to use "or" (the "|").

Edited April 27, 2019 by Wosven
Added comment

[Seneca 397]

9 hours ago, Wojciech Krakowiak said:

wow, thanks for your interest and discussion.

The GREP below works for me.

I made the assumption that any 1 letter word needs to be connected to the next word. That pretty much covers all the cases:

FIND: \b(\w)\b\s

REPLACE: $1(non-breaking space)

Edit: Wait. This still moves the one letter words around. I definitely think that this is a bug.

[Sean P 1,603]

Hi guys! Thank you very much for your help with this! I've reproduced the issue - it seems to be gobbling up the first 'w' after the i and then replacing all the other single letter conjuctions shifted 1 letter along.

I'll get this passed on to development.

[Patrick Connor 4,117]

We believe this has been addressed in the latest (#371) beta build. Please retest this behaviour.