[JimmXinu]

'in' is special to Python, but not to regexp, so it doesn't need to be escaped in regular expressions. Nor do commas. Parans--'()'--however, are and should be escaped with '\' when not being used for expression grouping.

This worked for me with this story:

replace_metadata:
 category=>Alice in Wonderland, 2010=>Alice in Wonderland (2010)
 characters=>Alice K\.=>Other.Alice Kingsleigh&&category=>Alice in Wonderland \(2010\)
 characters=>Mad Hatter/Tarrant Hightopp=>Other.Tarrant Hightop (Mad Hatter)&&category=>Alice in Wonderland \(2010\)

This pattern:

characters=>Cheshire Cat=>Other.\1&&category=>Alice in Wonderland

...doesn't work because \1 is replaced by the first matched paran group--and this pattern doesn't have one.

This pattern instead:

characters=>(Cheshire Cat)=>Other.\1&&category=>Alice in Wonderland \(2010\)

...will change 'Cheshire Cat/Chessur' to 'Other.Cheshire Cat/Chessur'. The '/Chessur' part is kept because the pattern only matches 'Cheshire Cat', leaving the rest. To remove it, use:

characters=>(Cheshire Cat).*=>Other.\1&&category=>Alice in Wonderland \(2010\)

Then the pattern matches the entire string and changes 'Cheshire Cat/Chessur' to 'Other.Cheshire Cat'.