问题
Not sure if this is best posted here or on the JQ dataTables forum. I'll try here.
Doing exact regex matching on columns in a dataTable and have two cases one works great but the second not so good!
CASE 1 Column data
Joe
Bob
Sue
Bobby
Joey
So then I want all the rows that have Joe, Bob, Sue and not Bobby and Joey.
MY regex looks like this;
\b(Joe|Bob|Sue)\b
And it works great. I see filtered rows for the three names and not Bobby and Joey. I use the pipe separated list because I also store the filter list in URL parameter so the filtered version can be book marked.
Now my problem case: CASE 2 Column data
Joe
Bob
Sue
Bobby
Joey
Ann
Jo-Ann
In this case I want Bob Sue and Ann rows.
Using this regex
\b(Bob|Sue|Ann)\b
Gets the rows I want but also gets the Jo-Ann Rows, I assume because the - is being treated as a boundary between words.
Play around with [\w-] but can't seem to get it work on a pipe separated list. Below is the fiddle.
http://refiddle.com/1v1j
Thanks,
Dave
回答1:
You could create a DIY boundary.
Using either a capture or non-capturing group we can assert that the start of the string or a space character precedes while asserting that a space character or the end of the string follows your grouped pattern.
(?:^| )(Bob|Sue|Ann)(?: |$)
You could also use a negated character class if you prefer. On the right side of the alternation operator, instead of matching a space character in the above regular expression, we match a character that is not a hyphen or a word character, which is closer to the intent of expanding the word boundary and do the same for the group that follows.
(?:^|[^-\w])(Bob|Sue|Ann)(?:[^-\w]|$)
回答2:
line feed could be applied in your case
/\s*(Bob|Joe|Ann)\s*\n/g
回答3:
^(?:Bob|Sue|Ann)$
You can apply anchors as well.See demo.
https://www.regex101.com/r/fG5pZ8/3
来源：https://stackoverflow.com/questions/27732348/js-regex-matching-whole-words-with-hyphen-in-datatables