> str.match(/('[^']+'|[^,]+)/g)
["A", "B", "C", "E", "'F,G,bb'", "H", "'I9,I8'", "J", "K"]
Though you requested this, you may not accounted for corner-cases where for example:
'bob's'
is a string where '
is escaped
a,',c
a,,b
a,b,
,a,b
a,b,'
',a,b
',a,b,c,'
Some of the above are handled correctly by this; others are not. I highly recommend that people use a library that has thought this through, to avoid things such as security vulnerabilities or subtle bugs, now or in the future (if you expand your code, or if other people use it).
Explanation of the RegEx:
('[^']+'|[^,]+)
- means match either '[^']+'
or [^,]+
'[^']+'
means quote...one-or-more non-quotes...quote.
[^,]+
means one-or-more non-commas
Note: by consuming the quoted string before the unquoted string, we make the parsing of the unquoted string case easier.
与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…