Populate information from a csv file into a LaTeX document (specifically into the fields/arguments of commands therein)

Question

I have some personal information stored in a plain CSV file personal_info.csv in the following format:

personal_info.csv

83 Fancy Avenue,Nowheresville,Gotham City 24061
123 456 7890
someone@xyz.com

I need to populate the relevant fields of the below main.tex (non-working/mockup) using class moderncv with this personal information as shown below.

Mock-up (non-working concept) of main.tex

\documentclass[11pt,a4paper,sans]{moderncv}
\moderncvstyle{classic}

\firstname{John}
\familyname{Doe}

%% NOT WORKING MOCKUP! (Relevant content to be read from csv file `personal_info.tex`)
\address{\firstfieldofline1}{\secondfieldofline1}{\thirdfieldofline2} 
\phone[mobile]{\firstfieldofline2}
\email{\firstfieldofline3}

\begin{document}
   \makecvtitle
\end{document}

to produce the desired output main.pdf (compiled with pdflatex).

How can this be achieved?

Answer 1

With readarray (and listofitems included).

First, I employ readarray macros

\readarraysepchar{\\}
\readdef{personalinfo.csv}\myrawdata

to take the contents of personalinfo.csv and place it into the macro \myrawdata, all the while converting end-of-record tokens into \\. Then, with listofitems macros,

\setsepchar{\\/,}
\readlist*{\mydata}{\myrawdata}

I set up a 2-tier parsing of \myrawdata and place the result into the array named \mydata. The \\ token separates the 1st tier items, and the , separates the 2nd-tier items within each 1st-tier item.

The syntax of listofitems allows the array to be recalled by the indices of the parsing. Thus \mydata[1,2] will extract the tokens of \myrawdata that occurred before the 1st \\ and, within that set of tokens, those that are between the 1st and 2nd ,.

The MWE:

\RequirePackage{filecontents}
\begin{filecontents*}{personalinfo.csv}
83 Fancy Avenue,Nowheresville,Gotham City 24061
123 456 7890
someone@xyz.com
\end{filecontents*}
\documentclass[11pt,a4paper]{moderncv}
\usepackage{readarray}

\moderncvstyle{classic}
\moderncvcolor{green}
\renewcommand{\familydefault}{\rmdefault}

\firstname{John}
\familyname{Doe}
\title{Title}

\readarraysepchar{\\}
\readdef{personalinfo.csv}\myrawdata
\setsepchar{\\/,}
\readlist*{\mydata}{\myrawdata}

\address{\mydata[1,1]}{\mydata[1,2]}{\mydata[1,3]} 
\phone[mobile]{\mydata[2,1]}
\email{\mydata[3,1]}

\begin{document}
\makecvtitle
\end{document}

Answer 2

Quick and dirty expl3 implementation:

First you load the CSV file with \LoadCSV{<label>}{<file>}. This command will open the file and load the contents of the CSV into <label>. The contents of the file are read verbatim (all catcodes are other (12), except for spaces (10)).

Then you get a field with \GetCSV{<label>}{<line>}{<field>}.

Replacing your mockup code with:

\LoadCSV {mycsv} {\jobname.csv}
\address{\GetCSV{mycsv}{1}{1}}{\GetCSV{mycsv}{1}{2}}{\GetCSV{mycsv}{1}{3}} 
\phone[mobile]{\GetCSV{mycsv}{2}{1}}
\email{\GetCSV{mycsv}{3}{1}}

it prints:

Full code:

\RequirePackage{filecontents}
\begin{filecontents*}{\jobname.csv}
83 Fancy Avenue,Nowheresville,Gotham City 24061
123 456 7890
someone@xyz.com
\end{filecontents*}

\documentclass[11pt,a4paper,sans]{moderncv}

\usepackage{expl3}
\ExplSyntaxOn
\ior_new:N \l__krishna_ior_tmpa
\cs_new:Npn \krishna_csv_load:nn #1 #2
  {
    \cs_if_free:cTF { l__krishna_ #1 _seq }
      { \__krishna_csv_load:nn { l__krishna_ #1 } {#2} }
      { \msg_error:nnn { krishna / csv } { name-used } {#1} }
  }
\cs_new:Npn \__krishna_csv_load:nn #1 #2
  {
    \ior_open:Nn \l__krishna_ior_tmpa {#2}
    \seq_new:c { #1 _seq }
    \exp_args:Nc
      \__krishna_csv_read:N { #1 _seq }
    \ior_close:N \l__krishna_ior_tmpa
  }
\cs_new:Npn \__krishna_csv_read:N #1
  {
    \ior_str_map_inline:Nn \l__krishna_ior_tmpa
      { \seq_put_right:Nn #1 {##1} }
  }
\cs_new:Npn \krishna_csv_get:nnn #1 #2 #3
  {
    \exp_args:Nf
    \clist_item:nn
      { \seq_item:cn { l__krishna_ #1 _seq } {#2} }
      {#3}
  }
\msg_new:nnn { krishna / csv } { name-used }
  { The~CSV~name~`#1'~is~already~taken }
\cs_new_eq:NN \LoadCSV \krishna_csv_load:nn
\cs_new_eq:NN \GetCSV \krishna_csv_get:nnn
\ExplSyntaxOff

\moderncvstyle{classic}

\firstname{John}
\familyname{Doe}

\LoadCSV {mycsv} {\jobname.csv}
\address{\GetCSV{mycsv}{1}{1}}{\GetCSV{mycsv}{1}{2}}{\GetCSV{mycsv}{1}{3}} 
\phone[mobile]{\GetCSV{mycsv}{2}{1}}
\email{\GetCSV{mycsv}{3}{1}}

\begin{document}
   \makecvtitle
\end{document}

But @Mensch's idea of a .tex file with definitions is far simpler :)

Answer 3

Let us say we have the following code:

\documentclass[11pt,a4paper,sans]{moderncv}

% moderncv themes
\moderncvstyle{classic} % casual, classic, banking, oldstyle and fancy
\moderncvcolor{blue} 

\usepackage[utf8]{inputenc}

\usepackage[scale=0.75]{geometry}

% personal data
\name{John}{Doe}
\title{Resumé title}
\address{street and number}{postcode city}{country}
\phone[mobile]{+1~(234)~567~890}
\phone[fixed]{+2~(345)~678~901}
\phone[fax]{+3~(456)~789~012}
\email{john@doe.org}
\homepage{www.johndoe.com}
\social[linkedin]{john.doe}
\social[twitter]{jdoe}
\social[github]{jdoe}
\extrainfo{additional information}
\photo[64pt][0.4pt]{example-image-a}
\quote{Some quote}

\setlength{\footskip}{66pt}


\begin{document}

\makecvtitle

\section{Education}
\cventry{year--year}{Degree}{Institution--3}{City--4}{\textit{Grade}--5}{Description--6}  % arguments 3 to 6 can be left empty
\cventry{year--year}{Degree}{Institution}{City}{\textit{Grade}}{Description}

\section{Master thesis}
\cvitem{title}{\emph{Title}}
\cvitem{supervisors}{Supervisors}
\cvitem{description}{Short thesis abstract}

\end{document}

Now move the personal data into a new file mydata.tex:

\name{John}{Doe}
\title{Resumé title}
\address{street and number}{postcode city}{country}
\phone[mobile]{+1~(234)~567~890}
\phone[fixed]{+2~(345)~678~901}
\phone[fax]{+3~(456)~789~012}
\email{john@doe.org}
\homepage{www.johndoe.com}
\social[linkedin]{john.doe}
\social[twitter]{jdoe}
\social[github]{jdoe}
\extrainfo{additional information}
\photo[64pt][0.4pt]{example-image-a}
\quote{Some quote}

After that you can use the new code (simply used \input instead the original personal datas):

\documentclass[11pt,a4paper,sans]{moderncv}

% moderncv themes
\moderncvstyle{classic} % casual, classic, banking, oldstyle and fancy
\moderncvcolor{blue} 

\usepackage[utf8]{inputenc}

\usepackage[scale=0.75]{geometry}

% personal data
\input{mydata.tex} % <===========================================

\setlength{\footskip}{66pt}


\begin{document}

\makecvtitle

\section{Education}
\cventry{year--year}{Degree}{Institution--3}{City--4}{\textit{Grade}--5}{Description--6}  % arguments 3 to 6 can be left empty
\cventry{year--year}{Degree}{Institution}{City}{\textit{Grade}}{Description}

\section{Master thesis}
\cvitem{title}{\emph{Title}}
\cvitem{supervisors}{Supervisors}
\cvitem{description}{Short thesis abstract}

\end{document}

There are no differences in the both resulting pdfs ...

Post the shown mydata.tex with the code for the cv on github and keep a copy of mydata.tex for your own purpose, of course filled with your real informations for a real cv ...