feature request: allow basic javascript in find and replace function

[Billybobbonnet]

Hello,

My issue is quite simple, I have a list of countries in ISO 3166-1 alpha-2 format. here is how it looks:

{
  "name": "Afghanistan",
  "code": "AF",
  "country-code": "004"
}, {
  "name": "Åland Islands",
  "code": "AX",
  "country-code": "248"
}, {
  "name": "Albania",
  "code": "AL",
  "country-code": "008"
},

I want to change all the codes to their lowercase equivalent. I can select them using a simple regex find string "[A-Z]{2}"

I would love a way to apply a matchedString.toLowerCase() on every string matched by my regex. Or maybe something like this.toLowerCase();

Thanks!

[dorian-marchal]

You can use Ctrl+K U to change the selection to uppercase and Ctrl+K L to lowercase.

[Billybobbonnet]

I just did a click on find + Ctrl+K L 294 times. It was surely shorter than doing it "manually". But still, I feel ashamed 😉

I love atom and I want to stick with it. I must say that this is the kind of feature that I would like to have in my main development tool.

Keep up the good work!

[davelnewton]

A partial solution would be to allow Sublime-style replacement expressions, e.g., https://manual.macromates.com/en/regular_expressions#replacement_string_syntax_format_strings. That makes easy stuff super-easy. It's one of the few things I really miss about Sublime. That and keyboard macros :(

[benogle]

A partial solution would be to allow Sublime-style replacement expressions

Do you mean replacing ab([c]+) with de$1 style of replacement? Atom currently supports these. Or is there something in that link that I am missing? A solution to this issue without allowing js would be interesting.

[davelnewton]

No, I mean to allow Sublime-style modifiers like \u to upper-case as mentioned in the "case foldings" section:

It is possible to convert the next character to upper or lowercase by prepending it with \u or \l. This is mainly useful when the next character stems from a capture register. Example:

Find: (<a.*?>)(.*?)(</a>)
Replace: $1\u$2$3

You can also convert a longer sequence to upper or lowercase by using \U or \L and then \E to disable the case folding again. Example:

Find: (<a.*?>)(.*?)(</a>)
Replace: $1\U$2\E$3

[benogle]

Ah, neat. 👍 to that idea.

[techmag]

Google searching just led me here (rabbit hole was a tad deep however). I'm just discovering the joys of Regex within the Atom editor's search-and-destroy, ahem I mean find-and-replace features and I love it - first one that stumped me however was that yes I have a dataset with "words" needing to be upper-cased and while I can do the match I can't indicate the "replace match as UPPER".

Would love to have that as a feature as I'm sure the proper implementation of it would bring in a whole bunch of other delicious goodies as well that I've yet to discover...

[AustinDizzy]

👍 Just had to Ctrl+K L 44 times. Supporting Sublime Text style regex case manipulation would be great for this specific purpose.

[AaronCritchley]

Note that you can use Alt + Enter to select all found results so you only need to do the modification once. (Not a solution but hopefully saves some time for anyone looking for toLowerCase / toUpperCase alternatives)

[davelnewton]

That's only an alternative if you're not making use of regex replacement with case manipulation, though.

[AaronCritchley]

Absolutely, like I said, it's not a solution but should save time for the people coming here saying 'Just had to Cmd + KU 350 times!'.

[techmag]

That alt-enter trick is neat! Thank you Aaron for pointing that out!

Doubt I was the only now that didn't know it was there.

Only works on a single file at a time though right?

Even so a huge step up as you say.

Having an actual regex replacement process would of course work on either type of search (in file versus across a set of files)

[AaronCritchley]

Yeah, it would add huge value to have something fully functional.

I only learnt about Alt + Enter today, had a question sitting on SO about it for a month or so and there's a lot of others regarding similar functionality, really hope it gets implemented soon!