Btukfyl

I have been searching on here for information about comma separated lists, but the only info I have found is how to sort, or iterate over a single list, which isn't quite what I need. If there is a relevant question somewhere I can't find, please feel free to link to it and close this.

What I am after is a way of merging several comma separated lists into a single tabular-type table to generate an answer key to several forms of a multiple choice section of an exam.

So, I would have something like the following:

begin{document}

magicFunction{ 

    {A, b, b, a, a, b, b, c}% Test Form A, form is first entry.

    }{

    {B, a, b, c, c, e, a, a}% Test form B

    }{

    {C, c, c, c, a, a, b, c}% Test form C

    }

end{document}

Which I would want to expand into the following:

begin{document}

begin{tabular}{cccc}

textbf{Problem Number} & textbf{Form A} & textbf{Form B} & textbf{Form C} \hlinehline

1  & b & a & c \

2  & b & b & c\hline

3  & a & c & c \

4  & a & c & a \hline

5  & b & e & a \

6  & b & a & b \hline

7  & c & a & c \

end{tabular}

end{document}

The header info isn't hard to just program in, especially since I don't mind having different functions for each number of forms (ie a "2 form" function vs "3 forms" vs "4 forms"; 4 would be the most). The problem I'm having is figuring out how to place the correct answer forms in the correct places. If it were transposed so that I could just inject the entire line I could write the & symbol in the list instead of using a comma-separated list, but for typesetting reasons it's better to have the answers listed vertically since most exams are more like 12-30 multiple choice problems. Moreover I would prefer the magicFunction to be length-independent... meaning that each form will have the same number of answers (so the table will always be full) but any given test might have a different number of answers in it, so that can't be hard coded. I could pass the number of answers in as an argument to the function, but I don't know how to iterate over multiple comma-separated lists simultaneously rather than concatenating the process.

Any help would be hugely appreciated. I don't mind using LaTeX3, which is what I imagine would be the most succinct way to do this, but I'm still clumsy and heavy handed trying to get through parsing all the syntax and naming schemes in LaTeX3, so when I try anything with it myself and get errors I can't figure out what the errors are actually being generated by.

Thanks!

documentclass{article}

usepackage{xparse}



ExplSyntaxOn



NewDocumentCommand{magicFunction}{m}

 {

  jason_magic:n { #1 }

 }



int_new:N l__jason_magic_cols_int

int_new:N l__jason_magic_rows_int

tl_new:N l__jason_magic_table_tl



cs_new_protected:Nn jason_magic:n

 {

  int_zero:N l__jason_magic_cols_int

  int_zero:N l__jason_magic_rows_int

  tl_map_inline:nn { #1 }

   {

    int_incr:N l__jason_magic_cols_int

    __jason_magic_makeclist:n { ##1 }

   }

  __jason_magic_maketable:

  % print the table

  begin{tabular}{ c *{l__jason_magic_cols_int}{c} }

  l__jason_magic_table_tl

  end{tabular}

 }



% syntactic sugar for avoiding long strings

cs_new:Nn __jason_magic_clist:n

 {

  l__jason_magic_ int_eval:n { #1 } _clist

 }



% store the parts in clists and get the number of rows

cs_new_protected:Nn __jason_magic_makeclist:n

 {

  clist_clear_new:c { __jason_magic_clist:n { l__jason_magic_cols_int } }

  clist_set:cn { __jason_magic_clist:n { l__jason_magic_cols_int } } { #1 }

  int_set:Nn l__jason_magic_rows_int

   {

    int_max:nn

     { l__jason_magic_rows_int }

     { clist_count:c { __jason_magic_clist:n { l__jason_magic_cols_int } } }

   }

 }



cs_new_protected:Nn __jason_magic_maketable:

 {

  % make the first row

  tl_set:Nn l__jason_magic_table_tl { Problem~number }

  int_step_inline:nn { l__jason_magic_cols_int }

   {

    tl_put_right:Nx l__jason_magic_table_tl

     {

      & Form~int_to_Alph:n { ##1 }

     }

   }

  tl_put_right:Nn l__jason_magic_table_tl { \ hline }

  % make the following rows

  int_step_inline:nn { l__jason_magic_rows_int }

   {

    tl_put_right:Nx l__jason_magic_table_tl { ##1 }

    int_step_inline:nn { l__jason_magic_cols_int }

     {

      tl_put_right:Nx l__jason_magic_table_tl

       {

        & clist_item:cn { __jason_magic_clist:n { ####1 } } { ##1 }

       }

     }

    tl_put_right:Nn l__jason_magic_table_tl { \ }

    int_if_odd:nF { ##1 } { tl_put_right:Nn l__jason_magic_table_tl { hline } }

   }

 }



ExplSyntaxOff



begin{document}



magicFunction{ 

    {b, b, a, a, b, b, c}% Test Form A, form is first entry.

    {a, b, c, c, e, a, a}% Test form B

    {c, c, c, a, a, b, c}% Test form C

}



end{document}

The idea is to allocate a clist for every column. Then the table is built by extracting the relevant item from each clist, by stepping the row index.

Not a complete answer but something for you to think about. The memoir class provides for what it calls automated tabulars where the input essentially consists of a comma-seperated list of entries to be tabulated in either a row-ordered (autorows) fashion or column-ordered (autocols) fashion.

For instance:

documentclass{memoir}

begin{document}

autorows{c}{3}{l}{one, two, three, four, five, six, seven}

end{document}

will produce a tabular looking like (apologies, I'm a GOM and haven't yet managed to upload a graphic)

one   two  three

four  five six

seven

See section 11.8.2 Automatic tabulars in the manual for more information (> texdoc memoir).