Search for unicode character (ranges)

[Karellen]

You could also use the Characters list in the Reports module. Double clicking on the character list will jump you to the next instance.

[kovidgoyal]

This is a python based regex engine, you need \U followed by 8 digits so insert three leading zeros. For example for pouting cat face (😾 U+1f63e)

\U0001f63e

[Karellen]

Originally Posted by kovidgoyal

This is a python based regex engine, you need \U followed by 8 digits so insert three leading zeros. For example for pouting cat face (😾 U+1f63e)

\U0001f63e

Thanks.

Why does the Search find the character, then not recognize it when trying to replace it?

[kovidgoyal]

That's a limitation of Qt with non-BMP unicode characters find will select only the first utf-16 codepoint making up the character. I could possibly workaround it, but its a lot of effort for a niche use case. Just use replace all for this case.

[Karellen]

Originally Posted by kovidgoyal

That's a limitation of Qt with non-BMP unicode characters find will select only the first utf-16 codepoint making up the character. I could possibly workaround it, but its a lot of effort for a niche use case. Just use replace all for this case.

Ok, thanks.

[kovidgoyal]

Actually the workaround turns out to be quite easy

https://github.com/kovidgoyal/calibr...f210a1129f00ab

[Comfy.n]

Originally Posted by Karellen

Thanks.

Why does the Search find the character, then not recognize it when trying to replace it?

Today I had a similar issue in Notepad++ while trying to remove the first character of every line in a text file with regex.

Using ^(.), it would mark the first character occurrences in the search panel, but would not actually replace them ('\1').

Then I found two alternative ways to do it: one is by using ^.?(.*) instead, and the other is surprisingly simple - Alt-selecting vertically the text "column" and deleting it.

[Karellen]

Originally Posted by kovidgoyal

Actually the workaround turns out to be quite easy

https://github.com/kovidgoyal/calibr...f210a1129f00ab

Oh, nice!!.
Yep, looks like a simple fix... but probably took a bit to figure out that's what was needed.

Originally Posted by Comfy.n

and the other is surprisingly simple - Alt-selecting vertically the text "column" and deleting it.

Oh, geez. I didn't even know you could do that.
Thanks for the tip!!

[Azraelo]

Thank you for the feedback. With the syntax "\U0001D4B7" it finds some characters.

If i have the text "𝐧𝒐𝗏𝑬𝔩𝓤𝗌𝒷.𝓬𝑶𝐦" and search for this, it shows me 4 occurences (also the wrong ones).
But when replacing, then it properly only replaces just the one character it should match.

I assume this also corresponds to your bugfix.

In some cases it doesn't seem to work consistently, but I'm looking forward for this bugfix to get a better visual information about the matches. Then i'll continue testing.