[Zajora]

I have created a fair number of regex fixes. I make changes to them every so often, so I'll probably edit this post if I do. I'd use code tags, but they take up too much room. If a regex should be replaced with a space, it will say "space". If there is nothing, then (logically) it should be replaced with nothing.

Of course, there is no guarantee any of these will work properly. I always check them a number of times before doing replace all, since there are a TON of ways eBooks can have formatting that wrecks these regexes.

Scenario: Apostrophes have been replaced with double quotes
Match: (?<=\w)(“|”)(?=\w)
Replace: ’

Scenario: There is a linebreak in the middle of a character's dialogue
Match: (?<=“[^”]*)</p>\s*<p[^>]*>(?!“)
Replace: space

Scenario: A tag closes, is followed by 0+ spaces or newlines, is then reopened and is then followed by a lowercase letter
Match: </(?P<tag>\w+)>\s*<(?P=tag) [^/>]+>(?=[a-z])
Match: (?<![".!?>*”“…~’])</(?P<tag>\w+)>\s*<(?P=tag) [^/>]+>
Replace: space
Notes: The second one is an alternate, which I think is better, but I'm not 100% sure it covers all the cases of the former.

Scenario: "LL" Ligatures have been replaced with a single "L".
Match: (l (?=(y|s|ed|ey|ion|en|ar|ars|er|ow|et|owed|enge|age |enging|ected|egal|ections|ect|apse|ular|op|owing| ocks|ied|ier|ies|ing|ingly|ered|icit|est)(\W)))|(l (?![(–<-])(?=\W))|(?<=’)l(?=\W)|(?<= (wi|du|a|we|te|sma|ca|sti|fu|fa|chi|sha|wa|pha|se| bi|ha|ki|pu|ce|ba|ski|hi|fi|fe|he|ro|ta|i|sme|bri| sta|we))l(?=\W)
Replace: ll
Notes: This regex doesn't really work that well, but it's faster than doing it manually. I would recommend using the spellcheck afterwards and catching the most common ones. This regex is actually a bunch of individual ones chained together by ORs (|) so it's easier to see what's doing what.

Scenario: More than 1 space in a row
Match: (?<=\S) {2,}(?=\S)
Replace: space

Scenario: There are tags (which may be nested) that are either empty or just have a number in them
Match: (<[^/>]*>)+\s*\d*\s*(</[^>]*>)+
Replace:
Notes: This may remove things you'd like to keep, such as scenebreaks/whitespace, or the chapter links.

Scenario: There's a linebreak or spaces before a closing tag
Match: (?<![".!?>*”“…~’])</(?P<tag>\w+)>\s*<(?P=tag) [^/>]+>
Replace:

For future use:

Scenario:
Match:
Replace: