Replace "and" in a List of Authors with Comma Except the Last One

[Mohammad]

I have to format the bibliography data in many ways! One case is APA format as below!

Q1

A bibtex entry has a bibtex-author field, normally authors name are separated with and, I have to format the output

to mimic the real situation I give below an example.

\define results() $(first)$, and $(last)$
<$vars authorlist="Maarten R. Dobbelaere and Pieter P. Plehiers and Ruben Van de Vijver and Christian V. Stevens and Kevin M. Van Geem">
<$vars first={{{ [<authorlist>split[ and ]butlast[]join[, ]] }}} last={{{ [<authorlist>split[ and ]last[]] }}}>
-- <<results>>

</$vars>
</$vars>

Do you propose a simpler solution?

[pmario]

Do you have a link to the APA style guide?

[pmario]

I did find this one: Reference List: Author/Authors // Purdue Writing Lab

It says the last Author should be separated with &

[pmario]

Since the last element has to be treated different “search-replace” isn’t possible. … So may be there is no better way.

[Mark_S]

Maybe not simpler, but different …

<$vars authorlist="Maarten R. Dobbelaere and Pieter P. Plehiers and Ruben Van de Vijver and Christian V. Stevens and Kevin M. Van Geem"
replace1=" and "
replace2=",([^,]*)$"  
>
<$vars first={{{ [<authorlist>search-replace:g:regexp<replace1>,[, ]search-replace:g:regexp<replace2>,[, and $1]]}}} 
>
<<first>>
</$vars>
</$vars>

[Mohammad]

@pmario
I use this one https://guides.library.uq.edu.au/referencing/apa6 but there are others like
https://apastyle.apa.org/

[Mohammad]

Hi Mark,
Your solution is shorter! while uses sophisticated regexp
Small and smart!

Thank you!

[Mohammad]

@pmario

It says the last Author should be separated with &

That’s true, the hard part is to show the author name as surname Initials , e.g. Nguyen, T., Carnevale, J. J., Scholer, A. A., Miele, D. B., & Fujita, K.

Assume a name like Ruben Van de Vijver where the last three words are surname!
Of course some bibtex generator uses Ruben {Van de Vijver}

So I gave up!

[Mark_S]

Your solution will not look right if there is only one author, I think.

[Mohammad]

I check the number of authors! if greater than 1, then I use the solution!

\define show-authors-in-references()
<!-- display authors as comma separated list, removes "and" -->
\whitespace trim
<$vars number-authors={{{[{!!bibtex-author}split[ and ]count[]]}}}>
<$list filter="[<number-authors>compare:number:gt[1]]" emptyMessage="""<$view  field="bibtex-author"/>""" variable=null>
<$vars authorlist={{!!bibtex-author}}>
<$vars first={{{ [<authorlist>split[ and ]butlast[]join[, ]] }}} last={{{ [<authorlist>split[ and ]last[]] }}}>
<$text text={{{ [<first>addsuffix[, and ]addsuffix<last>] }}}/>
</$vars>
</$vars>
</$list>
</$vars>
\end

[Mark_S]

The regexp solution takes care of that case automatically.

[Mohammad]

Ah, I did not know this!
So, your solution is shortest in this context!

Mark what do you propose for the hard part below?

@Mohammad

That’s true, the hard part is to show the author name as surname Initials , e.g. Nguyen, T., Carnevale, J. J., Scholer, A. A., Miele, D. B., & Fujita, K.

Assume a name like Ruben Van de Vijver where the last three words are surname!
Of course some bibtex generator uses Ruben {Van de Vijver}

[Mark_S]

I don’t suppose we can assume no middle names?

[Mark_S]

This requires human knowledge. You would need a pre-process that analyzes standard known special forms (e.g. Van de xxxxxxx} and maybe converts them all into something like Van#de#xxxxxxx. After that you can run your process to convert first and middle names into initials. At the end you would convert “#” into spaces.

I chose “#” because I know of no name with that symbol. I did know someone with an apostrophe in her name, so punctuation can sometimes be part of names.

[Mohammad]

There are some references with no middle name!
Normally these bibtex entries are downloaded from online databases like
https://www.sciencedirect.com/science/article/abs/pii/S0197458021002955

many authors have no middle name!

[Mark_S]

I heard there were people stimulating their brains electrically with home-made equipment.

[Mohammad]

@Mark_S

The regexp solution takes care of that case automatically.

Tested and worked great!

I replaced my solution with yours and acknowledged your invaluable help in Refnotes plugin!

[Mohammad]

@Mark_S

This requires human knowledge. You would need a pre-process that analyzes standard known special forms (e.g. Van de xxxxxxx} and maybe converts them all into something like Van#de#xxxxxxx. After that you can run your process to convert first and middle names into initials. At the end you would convert “#” into spaces.

What do you think, if we process the {surname with space}? It seems bibtex uses {...} to indicate this is a multi words surname or a name which contains spaces!

[Mark_S]

If the source author list is already formatted for you, that should make things easier. But you’ll have to use some extra escaping in regexp because { } are used in regexp.

[Mohammad]

Thank you!
Yes, the multi words surname are wrapped into {}.
I will go through and will let you know the result!