How to create a macro that manipulates the author's name?

The problem that I see in providing commands that extract implicit information from the author is that there are too many corner cases to get right. Ideally, it should cope correctly with each of the following use-cases:

• author{Alexsandro Lucena Mota}
• author{Doe}
• author{Jane Doe}
• author{Jane Maria Doe}
• author{Jane Maria Ellen Doe}
• author{Jane Maria de Lucca}
• author{Mota, Alexsandro Lucena}

Handling the first five cases is fairly straightforward. Case 6 requires building in knowledge of all multi-word surnames in all languages. Coping with Case 7 would be possible, but tedious. Of course, there are further problems because some languages, such as Mandarin, write names in the reverse order, with the surname first.

Ignoring all of these issues the code below defines an Author command that sets firstname, middlename and lastname and also passes its argument along to the author command. As the image below shows, it caters for Cases 1-6, with Case 6 requiring a small amount of extra help through the addition of a ~. The Author command does not play well with Case 7.

Here is the code:

documentclass[12pt,a4paper]{memoir}
parindent0pt

usepackage{xparse}
ExplSyntaxOn
tl_new:N g_firstname_tl
tl_new:N g_middlename_tl
tl_new:N g_lastname_tl
seq_new:N l_author_names_seq
NewDocumentCommandAuthor{m}{
  author{#1}
  % clear the  first, middle and last token lists
  tl_clear:N g_firstname_tl
  tl_clear:N g_middlename_tl
  tl_clear:N g_lastname_tl
  % trim spaces and split the author name(s) on surrounding spaces
  regex_split:nxN {s+} { tl_trim_spaces:n {#1} } l_author_names_seq
  int_case:nnF {seq_count:N l_author_names_seq }
  {
  {0} {}
  {1} {seq_pop_left:NN l_author_names_seq g_lastname_tl }
  {2} {
          seq_pop_left:NN l_author_names_seq g_firstname_tl
          seq_pop_left:NN l_author_names_seq g_lastname_tl
      }
  }
  {
    seq_pop_left:NN l_author_names_seq g_firstname_tl
    seq_pop_right:NN l_author_names_seq g_lastname_tl
    tl_gset:Nx g_middlename_tl { seq_use:Nn l_author_names_seq {space} }
  }
}
cs_generate_variant:Nn regex_split:nnN { nxN }

NewDocumentCommandfirstname{}{tl_use:Ng_firstname_tl}
NewDocumentCommandmiddlename{}{tl_use:Ng_middlename_tl}
NewDocumentCommandlastname{}{tl_use:Ng_lastname_tl}
NewDocumentCommandfullauthor{}{lastname,spacefirstnamespacemiddlename}
ExplSyntaxOff

newcommandNames[1]{Author{#1}textsf{Author:} #1.newline textit{First}: firstname, textit{Middle}: middlename, textit{Last}: lastname.parmedskip}

begin{document}

Names{Alexsandro Lucena Mota}
Names{Doe}
Names{Jane Doe}
Names{Jane Maria Doe}
Names{Jane Maria Ellen Doe}
Names{Jane Maria de~Lucca}

textbf{Fails}:

Names{Mota, Alexsandro Lucena}

end{document}

The idea is simple enough: use LaTeX3 regular expressions to split the author names on the spaces and then set the last "name" to be the last name, any remaining first name to be the first name, after which anything left over becomes the middle name. The Names macro is just a helper macro to show what Author does.

The code provides a fullauthor command that prints last name, first name middle name. I did not calls this newauthorname because, to me, this sounds like the command that you would use to define the author, not to print it.

EDIT

As requested in the comments, here is a small variation that redefines the author command instead of defining a new Author command. If this is going to be used often the code in the preamble could be put into a style file, say fullauthor.sty and then included with usepackage{fullauthor}.

documentclass[12pt,a4paper]{article}
parindent0pt
letrealAuthorauthor

usepackage{xparse}
ExplSyntaxOn
tl_new:N g_firstname_tl
tl_new:N g_middlename_tl
tl_new:N g_lastname_tl
seq_new:N l_author_names_seq
renewcommandauthor[1]
{
  realAuthor{#1}
  % clear the  first, middle and last token lists
  tl_clear:N g_firstname_tl
  tl_clear:N g_middlename_tl
  tl_clear:N g_lastname_tl
  % trim spaces and split the author name(s) on surrounding spaces
  regex_split:nxN {s+} { tl_trim_spaces:n {#1} } l_author_names_seq
  int_case:nnF {seq_count:N l_author_names_seq }
  {
  {0} {}
  {1} {seq_pop_left:NN l_author_names_seq g_lastname_tl }
  {2} {
          seq_pop_left:NN l_author_names_seq g_firstname_tl
          seq_pop_left:NN l_author_names_seq g_lastname_tl
      }
  }
  {
    seq_pop_left:NN l_author_names_seq g_firstname_tl
    seq_pop_right:NN l_author_names_seq g_lastname_tl
    tl_gset:Nx g_middlename_tl { seq_use:Nn l_author_names_seq {space} }
  }
}
cs_generate_variant:Nn regex_split:nnN { nxN }

NewDocumentCommandfirstname{}{tl_use:Ng_firstname_tl}
NewDocumentCommandmiddlename{}{tl_use:Ng_middlename_tl}
NewDocumentCommandlastname{}{tl_use:Ng_lastname_tl}
NewDocumentCommandfullauthor{}{lastname,spacefirstnamespacemiddlename}
ExplSyntaxOff

newcommandNames[1]{author{#1}textsf{Author:} #1.newline textit{First}: firstname, textit{Middle}: middlename, textit{Last}: lastname.parmedskip}

author{Alexsandro Lucena Mota}     % using the new author command in the preamble

begin{document}

fullauthor

Names{Alexsandro Lucena Mota}
Names{Doe}
Names{Jane Doe}
Names{Jane Maria Doe}
Names{Jane Maria Ellen Doe}
Names{Jane Maria de~Lucca}

textbf{Fails}:

Names{Mota, Alexsandro Lucena}

end{document}