Two em dash is probably fine. typogrify is not expected to always do the
right thing, so sometimes it's fine to ignore the changes it suggests.
On 8/28/25 7:18 AM, David wrote:
> Question for you, Emma. SEMoS is clear on how to manage numbers obscured
> <
https://standardebooks.org/manual/1.8.4/single-page#8.7.7.9> in dates,
> but does not otherwise deal with *numbers*, so far as I can see.
>
> In three places in this book, I have an address where the number is
> "obscured":
>
> 1. c9 <
https://archive.org/details/thataffairnextd02greegoog/page/n92/
> mode/1up?q=%22Jacobs%2C+of%22>: `— Lexington Avenue.`
> 2. c21 <
https://archive.org/details/thataffairnextd02greegoog/page/
> n216/mode/1up?q=%22please+address%22>: `— Liberty Street`
> 3. c22 <
https://archive.org/details/thataffairnextd02greegoog/page/
> n227/mode/1up?q=%22and+Untrimmed%22>: `— Sixth Avenue`
>
> I thought to handle these on the analogy of "totally-obscured days of
> the month" (8.7.7.9.1), that is, just a simple em-dash. But `typogrify`
> does not like that (well, it doesn't respect it, anyway!). Should I go
> with 2-em in these cases? Or perhaps "wrap" these obscuring em-dashes
> with a hair-space, or U+200C zero-width non-joiner?
>
> Thanks!
>
>
> standardebooks/9f740fcd-d160-4798-a8f3-ae3ab5c6534bn%
40googlegroups.com
> <
https://groups.google.com/d/msgid/standardebooks/9f740fcd-d160-4798-
> a8f3-ae3ab5c6534bn%
40googlegroups.com?utm_medium=email&utm_source=footer>.