I cam across an issue when printing of my thesis. A Table of Contents item would link to the wrong document in the binder. As both documents have the same title, I thought it was my simple error at copying the wrong one, even though I’d copied the entire project as ToC.
After trying a few more times, it appears that since they have the same document name in the binder, only one of them, the second, is being pointed at. So I have a ToC that looks like the following:
Chapter 5: The Verb 66
5.1 Agreement 68
5.2 Incorporation 75
5.2.1 Noun incorporation 146
5.2.2 Verb incorporation 83
Chapter 7: Incorporation 140
7.1 Noun incorporation 146
When I hover my mouse over the links in Word, the underlying link name “Noun_incorporation” shows up, so maybe there’s a way I can disambiguate them by forcing one to have a different unique identifier, but how can I do that? Do I just have to change the title to be something even slightly different?
That’s strange. In fact Scrivener should be serialising duplicates so that this does not happen. I recreated your example outline in a test project, selected the entire outline and copied it as a ToC, pasting that into a “Contents” file at the top. I then compiled to RTF with titles and RTF bookmarks enabled for all Formatting rows, and opened this in a word processor. The list of bookmarks in the navigator shows:
- The Verb
- Noun incorporation
- Verb incorporation
- Noun incorporation-1
In Scrivener, have you double-checked that in your contents file, the Scrivener links are not in fact pointing to the same item in the Binder? If they are, then that would explain the problem, since multiple cross-reference links to the same bookmarked section would not generate serial-numbered versions, they would all just point back to the one bookmark (as expected).
Yes, I have indeed checked that. It was my first guess actually. I even carefully added both titles into the contents next to each other, and I get both pointing at the same (second) document. Interestingly, although I suppose this is just how the numbering works, the hierarchical numbering is accurate, by which I mean that the one that links to the wrong document has the correct section number (as if it linked to the correct document).
I’ll try again and see if there’s any difference.
I tried with another project and it appears to have worked fine. I’ll have to go to uni and see what happened with my thesis. Also I can’t seem to find my actual thesis scrivener project on my home computer right now.
Solved it. And found a bug.
The problem was, and I didn’t notice this, but the two documents were had different titles case-sensitively, but not case-insensitively. One was titles ‘Noun incorporation’ and the other ‘Noun Incorporation’, just error on my part (or a mid-thesis change of capitalisation technique). So this means that there’s a mismatch somewhere in the creation of document links and disambiguating those links when they are identical by adding a numerical suffix.
Clearly the assignment of the numerical suffix is case-sensitive, so two documents named as above doesn’t trigger the number, and then the bookmarking (might be a word issue then, I suppose) is case-insensitive and thinks of them as pointing to the same bookmark.
If I’m right, then this can be fixed by changing the way Scrivener checks for identical names and assigns numbers, and making it case-sensitive, which should then feed properly into Word’s RTF bookmarking.
In the meantime, I’ve fixed my own project by making sure all titles have the same capitalisation.