How to get regex submatches? (using --json)

[bew]

Hello!

I have a number of files with lines like type: foo, bar or type: baz.
I'm looking to get the foo, bar & baz texts out of the JSON output.

Repro:

$ cd `mktemp -d`
$ echo 'type: foo, bar' > file1
$ echo 'type: baz' > file2

$ rg --json '^type: (.*)' | rg '"type":"match"'
2:{"type":"match","data":{"path":{"text":"file2"},"lines":{"text":"type: baz\n"},"line_number":1,"absolute_offset":0,"submatches":[{"match":{"text":"type: baz"},"start":0,"end":9}]}}
5:{"type":"match","data":{"path":{"text":"file1"},"lines":{"text":"type: foo, bar\n"},"line_number":1,"absolute_offset":0,"submatches":[{"match":{"text":"type: foo, bar"},"start":0,"end":14}]}}

We can see the submatches:

[{"match":{"text":"type: baz"},"start":0,"end":9}]

[{"match":{"text":"type: foo, bar"},"start":0,"end":14}]

But I'd expect to get (for each):

• submatch 0, for the whole match
• submatch 1, with only the text after type: so foo, bar for the file1 & baz for file2.

I also tried with --regexp without change.
Versions tried: 13.0.0 & 14.1.0

[BurntSushi]

Could you please provide a full reproduction? Give me a command that I can run as-is. You've described your input, but you need to show your input. It's okay to make something up that models your real data, just so long as it exhibits the same problem.

[bew]

You're totally right, sorry about that 😬. I updated my post with setup commands and updated outputs.

[BurntSushi]

OK, so the issue here is that submatches is not the same as capture group matches. submatches in this context refers to the list of matches within a line. For example:

$ echo 'foo bar baz' | rg --json '\w+' | rg '"match"'
{"type":"match","data":{"path":{"text":"<stdin>"},"lines":{"text":"foo bar baz\n"},"line_number":1,"absolute_offset":0,"submatches":[{"match":{"text":"foo"},"start":0,"end":3},{"match":{"text":"bar"},"start":4,"end":7},{"match":{"text":"baz"},"start":8,"end":11}]}}

Otherwise, I think you're probably looking for #2325. The main issue is that adding capture groups will result in a perf hit. So it means that you have to opt into it. Which is kinda unfortunate.

[bew]

Oh I see the confusion, thank you I'll subscribe to that issue in case it's ever going to be implemented.
I can match the groups on my side in nushell so it's not a priority for me.

[Aleem28]

hello mate .. iam new in github and i have no idea of how to use these files. I want to download some files in rapidgator but the time and speed restriction prevent me to do so
can you do a video on youtube explaing step by step how to install rapidgator downloder. How to use python script. I found few links for rapidgator but all files are in .py format I don't know how to use or run. Can anyone explain please.

[BurntSushi]

Sorry, but no, I'm not going to make a YouTube video for you. I think you'll have better luck if you seek out a newbie forum. Not sure where though. Good luck.