regex question - Printable Version

regex question - Printable Version

+- Sinisterly (https://sinister.li)
+-- Forum: Coding (https://sinister.li/Forum-Coding)
+--- Forum: Coding (https://sinister.li/Forum-Coding--71)
+--- Thread: regex question (/Thread-regex-question)

regex question - Vi-Sion - 07-07-2017

what does this regex do b(^\")b

RE: regex question - Blink - 07-07-2017

Quote:/ b(^\")b / g
b matches the character b literally (case sensitive)
1st Capturing Group (^\")
^ asserts position at start of the string

\" matches the character " literally (case sensitive)

b matches the character b literally (case sensitive)

Global pattern flags
g modifier: global. All matches (don't return after first match)

https://regex101.com/

RE: regex question - Vi-Sion - 07-07-2017

i know this but i cant figure what's a valid input in this case.

RE: regex question - saegenulled - 07-16-2017

where you've found this pattern? should it do something special? i havent seen a pattern i this structure.
is it maybe malformed? the only pattern which would look similar would be this one => \b[^\"]+\b
with this one you can look for word boundaries that haven't included double quotes

RE: regex question - Inori - 07-16-2017

As @saegenulled mentioned, it might be malformed. Actually, I'm surprised the majority of regex engines don't throw errors when compiling it.

A caret (^) matches the beginning of a string (or line when the multiline flag is enabled, which it isn't here), so "b" then the beginning of a string always fails.

It could be that whoever wrote this attempted to use it as a negative look ahead (which actually looks like "(?!pattern)") because caret also acts as a negation operator in character classes (e.g. [^xy] matches a character that isn't x or y).

On that note, it could be that the capture group was intended, but the target therein was one non-quote character, meaning the pattern was intended to be the following instead:

Code:
b([^"])b

Which is perfectly valid (no need to escape most things inside character classes), but I don't know what it would be used for without context.

RE: regex question - mothered - 07-16-2017

It doesn't look right, I haven't come across anything like that.

As an example, In MySQL, the following returns all values that "start" with the chars B & T:

Code:
'^[BT]'

And If It's hyphenated, It will return all values that "start between" the chars B & T (Inclusive of B & T):

Code:
'^[B-T]'

There's a few other variants, but these are the basics. @Inori has explained your question well.