Publisher: one-character prepositions don't work in V2

[pgraficzny]

For me this regex works in version 1 and version 2, in Polish.

[pgraficzny]

And this regular expression and replacement text are the best. And I think that's what it was all about.

[Stepaan]

On 2/15/2023 at 10:49 AM, pgraficzny said:

And this regular expression and replacement text are the best. And I think that's what it was all about.

May I please know where you got the book (Děti to chtěj vědět taky)? Because I was involved in its creation, but we made it in InDesign. Thanks. 😄

[pgraficzny]

10 hours ago, Stepaan said:

May I please know where you got the book (Děti to chtěj vědět taky)? Because I was involved in its creation, but we made it in InDesign. Thanks. 😄

From this topic, a few posts earlier.

[pgraficzny]

On 11/15/2022 at 10:12 PM, Stepaan said:

🙂

Thank you. The loukash's way works, mine doesn't. In V2... Look for all single characters at the end of a row.

One_char_prep.afpub

From here 🙂

[bajqo]

Thank you @pgraficzny your solution works but you must use this only if you are done with whole text. Because the syntax (\s[iuawzoIUAWZO])(\s) find both spaces. Non breaking and basic space. So if you use it for search and then replace all then it works. But If you edit or add some text. Then you use syntax (\s[iuawzoIUAWZO])(\s) for search again it will find every space again. Then after using replace all it will add one more space before single letters. After another search another space before letters.

[pgraficzny]

2 hours ago, bajqo said:

Thank you @pgraficzny your solution works but you must use this only if you are done with whole text. Because the syntax (\s[iuawzoIUAWZO])(\s) find both spaces. Non breaking and basic space. So if you use it for search and then replace all then it works. But If you edit or add some text. Then you use syntax (\s[iuawzoIUAWZO])(\s) for search again it will find every space again. Then after using replace all it will add one more space before single letters. After another search another space before letters.

This is why I usually use (\s\b.{1,2}\b)\s - also previously described. You can repeat many times and spaces are not duplicated.

[bajqo]

2 hours ago, pgraficzny said:

This is why I usually use (\s\b.{1,2}\b)\s - also previously described. You can repeat many times and spaces are not duplicated.

Thanky.

If i change the expresion to (\s\b.{1,2,3}\b)\s expr it will find three letter words too?

For replace you use \1<NBSP> ?

[Seneca]

1 hour ago, bajqo said:

If i change the expresion to (\s\b.{1,2,3}\b)\s expr it will find three letter words too?

Hello @bajqo,

No you need this (\s\b.{1,3}\b)\s.

[walt.farrell]

1 hour ago, bajqo said:

If i change the expresion to (\s\b.{1,2,3}\b)\s expr it will find three letter words too?

The syntax of {} is {min,max} so no, that won't work. But {1,3} should.

[bajqo]

34 minutes ago, Seneca said:

Hello @bajqo,

No you need this (\s\b.{1,3}\b)\s.

Great. Thank you.

[pgraficzny]

8 hours ago, bajqo said:

Thanky.

If i change the expresion to (\s\b.{1,2,3}\b)\s expr it will find three letter words too?

For replace you use \1<NBSP> ?

Look at the screen, after \1 I choose Non-Breaking Space from the context menu.

[fabula_rasa]

This bug is still around! With expressions where you give a set of specific letters/numbers/clumps of letters, Affinity just decides that it's fine to replace some of them with another random one from a set. So with an expression formulated like this:

Find: (\b)(a|i|o|u|w|z|A|I|O|U|W|Z|ze|Ze|nr|Nr|Za|za|We|we|Dr|dr)(\b )
Replace: \2[NBSP]

we get randomly "o" becoming "w" and so on.

The query posted by pgraficzny works fine (thanks for that, by the way :)). But it's very general. And if I only want to find/replace some specific set of things (e.g. specific list of various scientific titles or not all two-letter words), it is no longer valid.

The bug only happens when you "replace all". If you go one by one, it works as intended. Type of document does not matter (tried newly-created, imported, old, new etc.).

Any chance it will be fixed? As I read in this thread that it worked fine in v. 1.

[walt.farrell]

16 minutes ago, fabula_rasa said:

This bug is still around! With expressions where you give a set of specific letters/numbers/clumps of letters, Affinity just decides that it's fine to replace some of them with another random one from a set. So with an expression formulated like this:

Find: (\b)(a|i|o|u|w|z|A|I|O|U|W|Z|ze|Ze|nr|Nr|Za|za|We|we|Dr|dr)(\b )
Replace: \2[NBSP]

we get randomly "o" becoming "w" and so on.

I don't think that "this bug" has been reported. If you have a sample document that can be used to demonstrate it, that would be useful.

 

[fabula_rasa]

Just now, walt.farrell said:

I don't think that "this bug" has been reported. If you have a sample document that can be used to demonstrate it, that would be useful.

 

Sure. Occurs in any and all documents for me. Here, have a test one I made today (the one I made a screenshot from).

If it wasn't reported, I will do it now. I assumed (silly, I admit) that the original poster reported it.

grep_test.afpub

[Klababa]

Tohle mi funguje:

najít

(\<[szkvaiouSZKVAIOU])[ ]+

zaměnit

$1<NBSP>

bez mezery před "$1"

[bajqo]

On 2/17/2024 at 12:14 AM, Klababa said:

Tohle mi funguje:

najít

(\<[szkvaiouSZKVAIOU])[ ]+

zaměnit

$1<NBSP>

bez mezery před "$1"

Funguje aj s predložkami ako sú na, sa, do, alebo aby, asi, ale ...