If a quotation mark or an apostrophe (including characters entered by Alt 0145 / 0146 / 0147 / 0148) appears at the BEGINNING of an Index entry (in the "Insert index entry" dialogue), LibreOffice reads the mark IN ERROR as a character to be alphabetised and opens a new alphabetical category for it. SOLUTION: Any quotation mark or apostrophe, in the "Insert Index entry" dialogue, should be ignored and the following character used instead for alphabetising.
There are other characters which are problematic in alphabetisation: e.g. the Period/Full Stop in acronyms. Take this from a real alphabetical index: "S.E. Metropolitan Tramways Co. 393 Sackville, Lord John, and cricket 462" One would expect "S.E." to come after "Sackville": but it has been placed before. Perhaps there should be a default option, to ignore non-alphanumeric characters in indexes.
Have closed and reported in more detail at Bug 131315 - Index: Implement letter by letter alphabetising.
On second thoughts, this issue is still open. Bug 131315 is related and contains a request to standardise the indexing method to "letter by letter" mode.
Thank you for reporting the bug. Unfortunately without clear steps to reproduce it, we cannot track down the origin of the problem. Please provide a clearer set of step-by-step instructions on how to reproduce the problem. I have set the bug's status to 'NEEDINFO'. Please change it back to 'UNCONFIRMED' once the steps are provided
Created attachment 158764 [details] Writer doc. showing index problem with non-alphanumeric characters Open the attached writer doc. Look at the index at the end. You can see the various problems involved with indexing entries with non-alphanumeric characters. 1. Notice under "S" that an entry is out of order because it contains a full stop (period). 2. Notice that the entries containging either an apostrophe or quotation mark at the beginning generate spurious entries under meaningless headings.
(In reply to R. Green from comment #3) > On second thoughts, this issue is still open. Bug 131315 is related and > contains a request to standardise the indexing method to "letter by letter" > mode. Please explain, why bug 129500 is not covered by the more general bug 131315 => NEEDINFO
Moving to a standardised system of letter by letter indexing would fix this bug – BUT is it likely to be implemented in the near future? If not, it would be better, IMV, for now, to just "patch" the existing system by improving the way it alphabetises non-alphabetical characters.
(In reply to R. Green from comment #7) > Moving to a standardised system of letter by letter indexing would fix this > bug – BUT is it likely to be implemented in the near future? If not, it > would be better, IMV, for now, to just "patch" the existing system by > improving the way it alphabetises non-alphabetical characters. If you are unable to contract a developer to do this patching immediately, I think it is better to keep your bug 131315 as it went through UX review. *** This bug has been marked as a duplicate of bug 131315 ***