RegExReplace can't process (unicode comma)

[Jem]

Code: Select all

“宁肯我负天下人，不可天下人负我”……

Do this: Select above text, copy (Ctrl-C) and press ctrl-alt-shift-c, the above (unicode double-quote) didn't change to (unicode single-quote) even when there's no (unicode comma) in it. i.e. the RegEx can't process this somehow? Bug?

Code: Select all

^!+c::  ; Control+Alt+Shift+C hotkey.
  s := clipboard
  s := RegExReplace(s, "“([^，]+)”", "‘$1’") ; replaces (unicode double-quote) to (unicode single-quote)
  clipboard := s
  Send ^v
Return

[toralf]

You may add (*UCP) at the start of the needle. That might help, not tested though.

[Jem]

deleted

[toralf]

Put it before the brace

[toralf]

Code: Select all

s := RegExReplace(s, "(*UCP)“([^，]+)”", "‘$1’")

not tested

[toralf]

Looking at your needle, I do not think it has something to do with Unicode. Your needle seems not to find the string in quotes, because the string contains a comma and space

[Delta Pythagorean]

Wouldn't it be better to use the escaped code of the quote?

[swagfag]

look at the regex u wrote:

• find me a substring that contains a pair of funny looking opening quotes
• onemany chars that arent a funny looking comma
• immediately followed by a pair of funny looking closing quotes

what do u think happens when the regex read head expects a set of funny looking closing quotes but is looking at a funny comma? exactly - it fails. this has nothing to do with unicode

---

string contains a comma and space

wrong

---

use the escaped code of the quote

u can, but ultimately it doesnt matter

[Jem]

I realised this is my fault so I would like to delete this post, how do I do that?

[gregster]

Please don't empty posts that have so many answers already. That doesn't help anyone, is bad for forum integrity and against forum rules.
No one will hold your question against you, but might help others.
Thank you.

[Jem]

Thanks gregster for pointing this out, but I posted this when I was having a great headache; well now is getting slightly worse; so there are things I didn't 'see' at that time, and now after I realised it, this is a mistake on my part and nothing can be gained by anyone to see it in all honesty. So it is best not to pollute the forum, imho, and delete this post entirely.

You can find out why I say this by just looking at the codes of the 'recovered' 1st post.

Let me know if this post can be deleted after you've done the research. Cheers!
Jem

[Frosti]

I agree with gregster. I read this post and learned a new regex option (*UCP). Nice!

[toralf]

Yes, leave it. It is also now in the ask for help section.
Btw: I and and I assume most of all users have had these kind of posts. I had several of them, were I thought I was on the right track. Had long arguments with Chris to find out at the end that I was wrong or had made a mistake. So, don”t worry, you are in good company...

[gregster]

Thanks guys, for showing support and encouragement for this new user!

@Jem: Please have a look at your PMs!