I think there still is no way to specify cautionary accidentals should look back more than one bar. Is this likely to be added to a future update (soon)?
This would be great to specify per layout.
There are still a few problems with cautionaries in general — as in this 4-year-old post
It would be great to address some of these — especially on repeated notes. I have tweaked settings but it still throws up a lot of oddities, espcially in keyboard instruments.
normally, by default, you shouldn’t need a cautionary accidental if it’s more than one measure away.
you can certainly go, on a case-by-case basis, and add them in manually.
I’ve done this in the very rare cases where there might have been confusion.
But I know that speaking for myself, I would NOT want cautionary accidentals appearing in my score when separated by more than a full measure.
There are plenty of instances where a cautionary would be expected more than one measure away, especially in atonal music. Of course I realise I can add them in manually – for the 90-minute long orchestral score I’m working on now, that is a tedious job., hence my question.
At least in Dorico Pro there are settings to apply accidentals differently for different styles of music. Beyond that, applying them manually is always available.
Thanks — yes, I’m aware of and have tried all the options. There isn’t an option for changing the number of bars of context for cautionary accidentals, and this feature has been asked for by a number of people on the forums… including in that post I linked from five years ago.
In addition the two examples of unidiomatic behaviour in that previous thread are still problems in the latest Dorico version… That’s why I asked the question.
It’s true that courtesy accidentals are sometimes needed after more than one barline – even in music of the common practice period. But IMO this is a job for a human editor rather than a rule. There are way too many different circumstances and edge cases.
I appreciate everyone’s thoughts about typesetting practice or their preferred workflow. I was however asking about the possibility of a software feature about which a number of people have been asking on this forum over some years, and following up on a number of the problems with Dorico’s handling of cautionaries, which haven’t yet been resolved, particularly in keyboard parts. I guess I should have addressed this directly to @dspreadbury.
@dspreadbury and other members of the Development Team read all posts; so you have effectively asked already. Of course, as you say, people have been asking related questions for years. No one knows when it might bubble up to the top of the feature queue, if it does.
Yeah, fingers crossed. 
Completely agree. It’s come up a handful of times before, but I’d love settings for Dorico to ignore intervening bars of rests, whether it’s a single bar …
… or a multibar rest:
A setting to specify the number of bars of music to look, like Finale’s Cautionary Accidentals plug-in, would be useful too:
I had an editor 20+ years ago that insisted all accidentals that deviated from the key signatures required a cautionary, no matter the distance from the alteration. That’s not something I endorse, but seems like it should be an option too as some definitely use that system.
Agreed! And the repetition of accidentals in chords in multi-staff instruments, or the placement of them in the bar, can be quite frustrating… It means that everything requires a lot of proof-reading and manual fiddling, especially in chromatic music where there are quite a lot of chords with simulataneously different adjustments of the same note in different octaves.
I can agree with a cautionary after a bar/multibar rest, but not in your 3rd example.
All other situations I think are context-specific, hence probably done manually.
Is that a problem?
I find the opposite of that - scores that haven’t been proof-read - is more of an issue…
Of course it is a problem. The objective, surely, is to have as much automatically correct as possible, so that there is as little as possible manual correction when proofing.
You’ve just inadvertently agreed with me.
Thank you 
