The several responses to my floundering with the register have been very helpful, though I would have to confess that I have ended up 'messing with' things (suggested by Hans re the [key]) and getting a satisfactory result in almost every case without always understanding why. But I can say that I have the 'sorting' issue resolved if it is a main entry, including if that entry is surrounded by quote marks or has one part of that entry formatted differently (e.g. italics).

But I don't seem able to apply this to subentries! I cannot solve the sorting of subentries that have special features (e.g. I might have needed italics for part of a subentry, or the subentry is surrounded by quote marks).

Here are my two situations (and in each case they appear out of alphabetical order in the subentry list):

1. \index{animals+‘special kinds’}: in this case ‘special kinds’ appears in the subentry list at the bottom of the list, after one that starts with 'v'. I 'messed with' this by adding keys, e.g., \index[Animals] etc. but  the item disappeared from the index altogether.

2. \index{Plenary Council+{\it periti} (experts)}: in this case it is the italicised periti that appears out of place, after the letter 'i' rather than after 'p'. Again I tried putting various keys but this did not help.

I guess my confusion is this: I assumed that the [key] establishes the literal string which determines sort order. That seems to be the case for a main entry. How do I get it to work for a subentry?

Julian

On 1/29/2022 11:02 AM, jbf via ntg-context wrote:

Thanks for this response. I'll have to work on this (but tomorrow... it's late at night for me at the moment). I can see part of what you mean: I can use, for example \index[myindex]{\it Book title} (Book Author) and get the correct result, but not sorted properly, so I have to understand how, as you say, to 'set the sort entry to the unformatted version' which is not clear to me at the moment. I'll tackle it on the morrow when I'm thinking more clearly!

there is key and entry with key between []

when sorting, the key wins but because there can be duplicates the entry itself is also part of the final sort key

the accumulates sort key is sanitized and after that sorting happens in several stages (these can be defined / adapted) according to language, taking numbers into account and finally using the unicode ordering ...

you can fool the system by messing with the [key]

it's not the easiest subsystem (but it has a long history ... as with many subsystems the principles are not much different than mkii and the code seldom changes but of course evolved)

Hans


-----------------------------------------------------------------
                                          Hans Hagen | PRAGMA ADE
              Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
       tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl
-----------------------------------------------------------------