Calibre problem with 'REPLACE WITH' RegExp

[DaveClouding]

Hello and thanks to everyone,

I am quite new to Calibre and I am trying to use it's ebook converter (last version of Calibre) to operate a string Search and Replace inside an epub ebook's text during the conversion process.

I carefully read the Calibre RegExp tutorial as well as several other articles I googled for, and played around with it for a whole couple of hours, but I still could not figure out how to reach my goal.

Let me explain, using an example, which is the problem I am facing:

Let we say I need to find and remove, inside the book's text, all the occurrences of the hyphen (-) character which are included between two alphabetical characters.

I.E the expression 'preg-nancy' should be converted into the correct word 'pregnancy'.

For this reason, according to the Calibre manual, I put, inside the SEARCH EXPRESSION field, the following Regular Expression:

[A-Za-z]-[A-Za-z]

The processor correctly finds what I mean and, jumping back to the above example, it selects the portion 'g-n' inside the word 'preg-nancy' (and all the other words with the hyphen inside).

Now, I really cannot understand which RegExP I have to put into the REPLACE WITH field to let the processor to just remove the hyphen (-) character between the two alphabetical ones, while leaving all the other character inside the word untouched, thus transforming the word 'preg-nancy' into the correct word 'pregnancy'.

I tried many combinations, but I really cannot find the correct solution.

Any help would be highly appreciated.

Thanks in advance.

Dave

[kovidgoyal]

Search: ([A-Za-z])-([A-Za-z])
Replace: \1\2

[PeterT]

Try
search: ([A-Za-z])-([A-Za-z])
replace: \1\2

The ()'s "save" the matched bit and then the \1 \2 reference the saved bits.