Match strings between specific tags and convert to wikilinks

Question

This question is similar to another one I asked here: Match strings between delimiting characters but I could not modify in order to perform the new task. (Solution should work with EmEditor or Notepad++)

I need to match text between specific tags, I.e. <b class="b2">I have a lot of text, more text, some more text, text</b> and then

1. Convert first character only after opening tag to lowercase (with the exception of the pronoun "I")
2. Convert content between commas to wikilinks (and eliminate the tag).

I have tried running a number of regexes to get close to this with multiple steps, i.e.

(<b class="b2">)(.)
[[\L\2

</b>
]]

(\[\[)(\w+), (\w+)(\]\])
\1\2]], [[\3\4

Input text:

Any text <b class="b2">I make laugh</b>: Ar. and P. γέλωτα. Some more text <b class="b2">Delight</b>: P. and V. [[τέρπω]].
Any text <b class="b2">I amuse oneself, pass the time</b>: P. διάγειν.
Any text <b class="b2">It amuses oneself with, pass the time over, amuse</b>: Ar. and P.

Expected output:

Any text [[I make laugh]]: Ar. and P. γέλωτα. Some more text [[delight]]: P. and V. [[τέρπω]].
Any text [[I amuse oneself]], [[pass the time]]: P. διάγειν.
Any text [[it amuses oneself with]], [[pass the time over]], [[amuse]]: Ar. and P.

Answer 1

This a one-step solution:

• Ctrl+H
• Find what: (?:<b class="b2">|\G(, (?=.*</b>)))(I )?([^,<]+)(?:</b>)?
• Replace with: $1[[$2\l$3]]
• check Wrap around
• check Regular expression
• UNCHECK . matches newline
• Replace all

Explanation:

(?:                 # non capture group
    <b class="b2">  # literally
  |                 # OR
    \G              # restart from last match position
    (               # group 1, a comma and a space
      ,             # a comma and a space
    (?=.*</b>)      # positive look ahead, make sure we have a closing tag after
    )               # end group 1
)                   # end group
(I )?               # group 2, UPPER I and a space, optional
([^,<]+)            # group 3, 1 or more any character that is not comma or less than
(?:</b>)?           # optional end tag

Replacement:

$1          # content og group 1 (i.e. comma & space)
[[          # double openning square bracket
$2          # content of group 2, (i.e. "I ")
\l$3        # lowercase the first letter of group 3 (i.e. all character until comma or end tag)
]]          # double closing square bracket

Result for given example:

Any text [[I make laugh]]: Ar. and P. γέλωτα. Some more text [[delight]]: P. and V. [[τέρπω]].
Any text [[I amuse oneself]], [[pass the time]]: P. διάγειν.
Any text [[it amuses oneself with]], [[pass the time over]], [[amuse]]: Ar. and P.
[[be at ease]], v.: P. and V. ἡσυχάζειν, V. ἡσύχως ἔχειν.

Screen capture: