Two or three letter initials in bibliography with Biblatex

Question

In my bibliography, I need to abreviate first names keeping digraphs and trigraphs.

• John should be abreviated as J.
• Clare should be abreviated as Cl.
• Charles should be abreviated as Ch.
• Christine should be abreviated as Chr.
• Philippe should be abreviated as Ph.
• etc.

The solution is to modify the first name in the bibliographical data from

Charles

to

{\relax Ch}arles

The following example uses the \DeclareSourcemap customization macro in Biblatex to change Charles to {\relax Ch}arles but I'm looking for a more generic solution that would automatically modify all first names starting with two or three consonants.

\documentclass{article}
\usepackage{filecontents}
 \begin{filecontents}{\jobname.bib}
  @book{Book1,
  author = {Doe, Charles},
  title  = {An Important Book},
  publisher = {Publisher},
  date = {2012},
}
@book{Book2,
  author = {Doe, Charlotte},
  title  = {Another Important Book},
  publisher = {Publisher},
  date = {2014},
}
@book{Book3,
  author = {Smith, Theodore},
  title  = {A very Important Book},
  publisher = {Publisher},
  date = {2016},
}
\end{filecontents}
\usepackage[style=verbose,firstinits=true, backend=biber]{biblatex}
\DeclareSourcemap{
 \maps[datatype=bibtex]{
    \map{
    \step[fieldsource=author, 
            match={Charles},
            replace=\regexp{\{\\relax \x20Ch\}arles}]
  }
 }
}
\addbibresource{\jobname.bib}
\begin{document}

\cite{Book1}

\cite{Book2}

\cite{Book3}

\end{document}

Answer 1

A possible starting point would be to use the code Yves de Saint-Pern wrote in class droit-fr, which I am translating below.

EDIT : The regular expressions provided here catch last names as well, which may cause issues outside Biblatex, for instance in indexes done with imakeidx. A tentative solution to this problem is provided here.

You can use it in a biber.conf file or include it in your .bbx.

Some languages may require additional regex patterns for first names, but they are pretty straightforward to add.

\DeclareStyleSourcemap{%
  \maps[datatype=bibtex]{%
    \map{%
        % Author field
      \step[fieldsource=author,%
        match={\regexp{\b(Chr|Ch|Th|Ph|[B-DF-HJ-NP-TV-XZ](l|r))(\S*,)}},%
        replace={\regexp{\{$1\}$3}}]% Protect last names (first last)
      \step[fieldsource=author,%
        match={\regexp{([^,]\s)\b(Chr|Ch|Th|Ph|[B-DF-HJ-NP-TV-XZ](l|r))}},%
        replace={\regexp{$1\{$2\}}}]% Protect last names (last, first)
      \step[fieldsource=author,%
        match={\regexp{\b(Chr|Ch|Th|Ph|[B-DF-HJ-NP-TV-XZ](l|r))([^\}])}},%
        replace={\regexp{\{\\relax\{\}$1\}$3}}]% Insert \relax after abbreviating
      % Editor field
      \step[fieldsource=editor,%
        match={\regexp{\b(Chr|Ch|Th|Ph|[B-DF-HJ-NP-TV-XZ](l|r))(\S*,)}},%
        replace={\regexp{\{$1\}$3}}]% Protect last names (first last)
      \step[fieldsource=editor,%
        match={\regexp{([^,]\s)\b(Chr|Ch|Th|Ph|[B-DF-HJ-NP-TV-XZ](l|r))}},%
        replace={\regexp{$1\{$2\}}}]% Protect last names (last, first)
      \step[fieldsource=editor,%
        match={\regexp{\b(Chr|Ch|Th|Ph|[B-DF-HJ-NP-TV-XZ](l|r))([^\}])}},%
        replace={\regexp{\{\\relax\{\}$1\}$3}}]% Insert \relax after abbreviating
}}}%

As far as the regular expressions are concerned, I am no expert but I would hazard that the expression means this:

\b(Chr|Ch|Th|Ph|[B-DF-HJ-NP-TV-XZ](l|r))(\S*,)

A word (\b) that starts with (\S* means: ends with any number of non-space characters): Chr (Chris) or Ch (Charles) or Th (Thomas) or Ph (Philippe), or any combination that starts with a consonant (the [B-DF-HJ-NP-TV-XZ part is a set of ranges: letters B to D, F to H, J to N, P to T, V to X, Z) followed by l or r (Bruno or Claire would fall within that category).