Encapsulating specific lines with angle brackets

[Elnaz Kia]

Hi all,
I want to encapsulate every line that starts with @ with angle brackets (as in the following example). The files are in multiple sub-directories. I tried regex but it replaces the whole line with <@.*> instead of keeping the text. I would appreciate any help with this.

<@Begin>
<@Languages: zho>
<@Participants: C0026 IDc0026 Student>
<@ID: zho|corpus|C0026|0;00.00|male|||Student|||>
<@Media: c0026c05ri1_2, audio>
<@Transcriber: >
<@Situation: introduction>
*C0026: wo3 zai4 wo3 xue2 xiao4.
*C0026: xian4 zai4 zai4 (…) dian4 nao3 lao3 shi1 de1 ban1.
*C0026: he2 wo3 (.) zai4 NAME.
<@End>

[PeterJones]

@Elnaz-Kia ,

In the Search, you need to match something like <@(.*?)> where you put the captured text in a capture group (and make it capture as little as possible); in the Replacement, use $1 to refer to the capture group value. see official regex “capture groups” documentation in the manual

[guy038]

Hello, @elnaz-kia, @peterjones and All,

May be I’m completly wrong but there is what I understand :

For any line, beginning with the @ character, @elnaz-kia wants to surround all its contents with two angle brackets !

If so, the following regex S/R should work :

• SEARCH (?-s)^@.+

• REPLACE <$0>

• Tick preferably the Wrap around option

• Select the Regular expression search mode

• Click on the Replace All button

Voila !

Best regards,

guy038

[PeterJones]

@guy038 ,

I agree with your interpretation. I just didn’t see enough in my first reading to get that far.

Because the original post didn’t show a “before” and “after” state, I missed the “every line that starts with @” portion. Since I didn’t see enough information, I gave a brief reply that introduced the concepts required, and linked to the appropriate reading for those concepts.

But yes, doing it my way with the “every line that starts with @”, I would change my FIND to (?-s)^@(.*)$ and the full replacement would be <$1>.

For @Elnaz-Kia ,

The substitution escape sequences section of the same usermanual page briefly describes @guy038’s $0 notation and my $1 is described in the section on $ℕ in the document. Both are essentially equivalent, with @guy038 making use of the automatic “whole match” capture group and mine using an explicit capture group.