Repeated "replace" results in double-word

I find that in a given scene, a character’s name is spelled in two ways: Shelly and Shelley. I use cmd-f to do a find and replace (find “Shelly” and replace it with “Shelley”). I hit the “Next” button, which highlights the first use of “Shelly.” I hit the “Replace” button, which changes that instance of “Shelly” to “Shelley,” as it should. If I then hit “Replace” again instead of “Next,” it appends another “Shelley” onto the first, resulting in “ShelleyShelley.” It will do that as many times as I hit “Replace,” resulting in “ShelleyShelleyShelleyShelley,” etc.

I’m thinking that the preferred behavior would be for it to jump to the next instance of “Shelly” and replace it. If there isn’t another instance, it should do nothing, returning the sound (don’t know what it’s called) that indicates it can’t find another instance to replace.

I don’t know if that’s technically a bug or not, but changing it would help prevent the user (OK, maybe just me) from inadvertently introducing typos.

Thanks for your attention, and thanks, as always, for an amazing product. Scrivener has allowed me to produce at a new level, and I’m a true evangelist for it.


This is standard behaviour of standard OS X Find panels (although I notice that it’s not the same in all apps these days), and not really a bug. Hitting “Replace” when nothing is selected just inserts the replacement word into the text. It sounds as though “Replace & Find” is what you want. I will, however, add this to the list to look into improving it, as it has come up before.

Thanks for the kind words!
All the best,

Sounds good, Keith. And thank you.