Nrthugu

As stated in the interface3 document (Part IX, 1. Creating and initializing sequences) seq_set_split: does not preserve spaces. Wishing to scan the text contained in a _tl variable into a sequence item by item for further processing (to be parsed for example) it would be advantageous to have spaces preserved. The MWE demonstrates that this can be achieved by introducing hard spaces, but is there a way to make do without the ~s?

documentclass{article}

% RN. 11 Nov 2018

%=======================

usepackage[check-declarations,log-functions]{expl3}

usepackage{xparse}

%-----------------------

ExplSyntaxOn

tl_new:N l_rn_auxOne_tl

seq_new:N l_rn_auxOne_seq



NewDocumentCommandmyScanText{m}

  {

    tl_set:Nn l_rn_auxOne_tl {#1}

    textbf{tl~variable:}~tl_use:N l_rn_auxOne_tl\

    seq_set_split:NnV l_rn_auxOne_seq {} l_rn_auxOne_tl

        textbf{seq~variable:}~seq_use:Nn l_rn_auxOne_seq {,}\

  }  %  myExperminent

ExplSyntaxOff

%-----------------------

begin{document}

  verb+myScanText{The quick brown fox jumps over the lazy dog.}+\  

  myScanText{The quick brown fox jumps over the lazy dog.}



  verb+myScanText{The~quick~brown~fox~jumps~over~the~lazy~dog.}+\

  myScanText{The~quick~brown~fox~jumps~over~the~lazy~dog.}

end{document}

Use a slower routine first splitting at spaces.

documentclass{article}

usepackage[margin=1cm]{geometry}

usepackage{xparse}



ExplSyntaxOn

tl_new:N l_rn_auxOne_tl

tl_new:N l_rn_auxTwo_tl

seq_new:N l_rn_auxOne_seq

seq_new:N l_rn_auxTwo_seq



NewDocumentCommandmyScanText{m}

 {

  tl_set:Nn l_rn_auxOne_tl {#1}

  seq_clear:N l_rn_auxOne_seq

  % split at spaces

  seq_set_split:Nnn l_rn_auxTwo_seq {~} { #1 }

  % the first item is special, pop it out and split it

  seq_pop_left:NN l_rn_auxTwo_seq l_rn_auxTwo_tl

  tl_map_function:NN l_rn_auxTwo_tl __rn_add:n

  % now do the other items, reinserting the space before them

  seq_map_inline:Nn l_rn_auxTwo_seq

   {

    __rn_add:n { ~ }

    tl_map_function:nN { ##1 } __rn_add:n

   }

  % print the data

  textbf{tl~variable:}~tl_use:N l_rn_auxOne_tl\

  textbf{seq~variable:}~seq_use:Nn l_rn_auxOne_seq {,}

 }

cs_new_protected:Nn __rn_add:n

 {

  seq_put_right:Nn l_rn_auxOne_seq { #1 }

 }

ExplSyntaxOff

%-----------------------

begin{document}



setlength{parindent}{0pt} % just for the example



myScanText{The quick brown fox jumps over the lazy dog.}



myScanText{The~quick~brown~fox~jumps~over~the~lazy~dog.}



end{document}

Alternative method: first replace spaces with a private token and then split with tl_map_function:NN

documentclass{article}

usepackage[margin=1cm]{geometry}

usepackage{xparse}



ExplSyntaxOn

tl_new:N l_rn_auxOne_tl

seq_new:N l_rn_auxOne_seq



NewDocumentCommandmyScanText{m}

 {

  seq_clear:N l_rn_auxOne_seq

  tl_set:Nn l_rn_auxOne_tl {#1}

  tl_replace_all:Nnn l_rn_auxOne_tl { ~ } { __rn_space: }

  tl_map_function:NN l_rn_auxOne_tl __rn_add:n

  % print the data

  textbf{input:}~#1\

  textbf{seq~variable:}~seq_use:Nn l_rn_auxOne_seq {,}

 }

cs_new_protected:Nn __rn_add:n

 {

  tl_if_eq:nnTF { #1 } { __rn_space: }

   {

    seq_put_right:Nn l_rn_auxOne_seq { ~ }

   }

   {

    seq_put_right:Nn l_rn_auxOne_seq { #1 }

   }

 }

cs_new_protected:Nn __rn_space: {}

ExplSyntaxOff

%-----------------------

begin{document}



setlength{parindent}{0pt} % just for the example



myScanText{The quick brown fox jumps over the lazy dog.}



myScanText{The~quick~brown~fox~jumps~over~the~lazy~dog.}



end{document}

You could use str_map_inline:Nn and append to the seq. The problem is that catcodes are all 12 then.

documentclass{article}

% RN. 11 Nov 2018

%=======================

usepackage[check-declarations,log-functions]{expl3}

usepackage{xparse}

%-----------------------

ExplSyntaxOn

tl_new:N l_rn_auxOne_tl

seq_new:N l_rn_auxOne_seq



NewDocumentCommandmyScanText{m}

  {

    tl_set:Nn l_rn_auxOne_tl {#1}

    textbf{tl~variable:}~tl_use:N l_rn_auxOne_tl\

    seq_clear:N l_rn_auxOne_seq

    str_map_inline:Nn l_rn_auxOne_tl { seq_put_right:Nn l_rn_auxOne_seq { ##1 } }

        textbf{seq~variable:}~seq_use:Nn l_rn_auxOne_seq {,}\

  }  %  myExperminent

ExplSyntaxOff

%-----------------------

begin{document}

  verb+myScanText{The quick brown fox jumps over the lazy dog.}+\  

  myScanText{The quick brown fox jumps over the lazy dog.}



  verb+myScanText{The~quick~brown~fox~jumps~over~the~lazy~dog.}+\

  myScanText{The~quick~brown~fox~jumps~over~the~lazy~dog.}

end{document}

You could pre-process the input using regex_replace_all:nnN to replace all spaces with space. (I first tried replacing spaces with ~ and ~ but this did not work so well.) Doing this your MWE produces:

Here is the code:

documentclass{article}

% RN. 11 Nov 2018

%=======================

usepackage[check-declarations,log-functions]{expl3}

usepackage{xparse}

%-----------------------

ExplSyntaxOn

tl_new:N l_rn_auxOne_tl

seq_new:N l_rn_auxOne_seq



NewDocumentCommandmyScanText{m}

  {

    tl_set:Nn l_rn_auxOne_tl {#1}

    regex_replace_all:nnN {s+} { c{space} } l_rn_auxOne_tl

    textbf{tl~variable:}~tl_use:N l_rn_auxOne_tl\

    seq_set_split:NnV l_rn_auxOne_seq {} l_rn_auxOne_tl

        textbf{seq~variable:}~seq_use:Nn l_rn_auxOne_seq {,}\

  }  %  myExperminent

ExplSyntaxOff

%-----------------------

begin{document}

  verb+myScanText{The quick brown fox jumps over the lazy dog.}+\

  myScanText{The quick brown fox jumps over the lazy dog.}



  verb+myScanText{The~quick~brown~fox~jumps~over~the~lazy~dog.}+\

  myScanText{The~quick~brown~fox~jumps~over~the~lazy~dog.}

end{document}

Note that this replaces repeated spaces with a single space, which I'm guessing is probably what you'd really want.