Newline substitution.

[gskinner]

Newline substitution.

[gskinner]

Right, we currently limit the Expression and Substitution fields to single line input. I've considered treating the Substitution value as a Javascript string, so you can use "\n" and other encoded characters. Open to feedback on this.

[thorsummoner]

This is the originating request: http://stackoverflow.com/questions/22751526/regexr-com-substitution-with-new-line-feed I'm sure "peter-lee" on SO would love to talk to you more about it :)

Great tool, but newline/tab substitution is the thing I stuck with.

[gskinner]

This also causes inconsistencies with the Share>Copy>Javascript workflow, because \n in the substitution is not decoded in RegExr, but is in JS.

To resolve this, I propose leveraging JSON.parse to decode escaped characters, and adding support to the SubstLexer to include highlighting / tooltips for them as well (ideally sharing code with the RegExLexer). I might take a stab at this when I have some time.

JSON.parse('"'+subst+'"'); // should decode all escaped chars.

[gskinner]

A little more info: This seems like a better solution, but requires appropriate error handling:

try { str = eval('"'+str.replace(/"/g,'\\"')+'"')); } catch(e) { /* etc */ }

[drogo007]

The ability to clean up data (including newlines) and format text is the primary reason I tend to use regexr. I don't use regexp often enough to remember all the rules off the top of my head, so having regexr makes life very easy - except I can no longer insert special characters in the substituion. Which renders 2.0 almost completely broken for all my use cases.

[gskinner]

I just pushed a solution to this, supporting all JS escaped character formats, with appropriate errors and tooltips. This should be live in the next couple days.

[gskinner]

This is now active on the live site. I'd love any feedback you have.

[drogo007]

So a quick look this morning shows the new site now works like a charm for the kinds of things I tend to use Regexr for. Thank you!