How and when is a Name-part Order decided?

Hi,

I’m creating a CSL(-M) style file and have a question regarding Name-part Order.

As described in the document, there are two orders: “given-name family-name” and “family-name given-name.”

I wonder where and when the order is decided. I tried both citeproc-js and citeproc-lua. It seems that citeproc-js identify an order depending on the language property of a bibliography item. However, citeproc-lua seems to distinguish an order for each name (I could mix so-called Latin names and Japanese names and the order for each item was decided correctly).

How and when is a Name-part Order decided?

Best,
Shigeru

We don’t have spec rules for that, details are up to the individual processors, though you might find something in the test-suite. I was convinced that citeproc-js handles this based on characters, not item language, but if you’ve tested it, I’m sure you’re right.

Thank you very much for your reply. I found that citeproc-js determines the Name-part order for each name instead of referring to the language setting of a bibliography item but overrides when the language is “ja” or “zh”.

https://github.com/Juris-M/citeproc-js/blob/73bc1b44bc7d54d0bfec4e070fd27f5efe024ff9/citeproc_commonjs.js#L13755-L13757

I believe this is why I got the result in the initial post. I’ll confirm and submit an issue to the developers.

I actually think that makes sense, absent a dedicated language per creator: as I understand it, Chinese/Japanese name orders should also be respected when names are transliterated, so what citeproc-js does seems like the best possible approximation to me.

Thank you very much for your reply. As you pointed out, our (or my) needs might not be standard. Making this option and passing it to a cite processor is my idea at the current moment.