Ruby .split（各种分隔符）将分隔符保留在匹配数组中

Question

I'm dealing with an export file that contains a note field in an exported .csv file. The notes exported look like this.

"AC) Dianne - # or code dialed is incorrect. AC) dianne - # or code dialed is incorrect. WDB) Dianne - Wrong Number. AC) Dianne - # or code dialed is incorrect."

Within this field we have a user who left the note and the actual note itself.

The delimiters in the files are "AC)" or "WDB)"

I need to write these as:

AC) Dianne - # or code dialed is incorrect.
AC) dianne - # or code dialed is incorrect.
WDB) Dianne - Wrong Number.
AC) Dianne - # or code dialed is incorrect.

Using a regular expression and the ruby function .split I can output the text following the delimiter but I lose the user who captured the notes.

Ruby

notes.split( /AC\)|WDB\)/ ).each do |n|
    puts n  
end     

Output

Dianne - # or code dialed is incorrect.
dianne - # or code dialed is incorrect.
Dianne - Wrong Number.
Dianne - # or code dialed is incorrect.

In the code above I have no idea who the user (AC, WDB) who left the individual note.

I'm not sure if I need to switch to a .scan, alter the regex (ie. include a lookbehind), etc. Does anyone have any idea how I can capture the user and text to look like this?

Output

AC) Dianne - # or code dialed is incorrect.
AC) dianne - # or code dialed is incorrect.
WDB) Dianne - Wrong Number.
AC) Dianne - # or code dialed is incorrect.

Answer 1

Just split the input according to the below lookahead,

(?=AC\)|WDB\))

Lookarounds are zero width assertions. It won't match any character but it was used for condition checking purposes.

Code:

> "AC) Dianne - # or code dialed is incorrect. AC) dianne - # or code dialed is incorrect. WDB) Dianne - Wrong Number. AC) Dianne - # or code dialed is incorrect.".split(/(?=AC\)|WDB\))/)
=> ["AC) Dianne - # or code dialed is incorrect. ", "AC) dianne - # or code dialed is incorrect. ", "WDB) Dianne - Wrong Number. ", "AC) Dianne - # or code dialed is incorrect."]

Answer 2

You can try this -

notes.gsub!(/\s+((?:AC\))|(?:WDB\)))/, '\n\1')

Test out the regex here