Mercurial > hg > octave-lojdl
changeset 13318:f3b665972bb5
strjust.m: Tweak code for performance. Add more input validation and tests.
* strjust.m: Use repmat for faster array construction. Simplify code by using
default values to function. Validate input is 2-D character string.
author | Rik <octave@nomad.inbox5.com> |
---|---|
date | Tue, 11 Oct 2011 12:46:24 -0700 |
parents | d2b799e3178a |
children | eb0ce6ffefb0 |
files | scripts/strings/strjust.m |
diffstat | 1 files changed, 54 insertions(+), 51 deletions(-) [+] |
line wrap: on
line diff
--- a/scripts/strings/strjust.m +++ b/scripts/strings/strjust.m @@ -18,10 +18,11 @@ ## <http://www.gnu.org/licenses/>. ## -*- texinfo -*- -## @deftypefn {Function File} {} strjust (@var{s}, @var{pos}) +## @deftypefn {Function File} {} strjust (@var{s}) +## @deftypefnx {Function File} {} strjust (@var{s}, @var{pos}) ## Return the text, @var{s}, justified according to @var{pos}, which may ## be @samp{"left"}, @samp{"center"}, or @samp{"right"}. If @var{pos} -## is omitted, @samp{"right"} is assumed. +## is omitted it defaults to @samp{"right"}. ## ## Null characters are replaced by spaces. All other character ## data are treated as non-white space. @@ -41,69 +42,71 @@ ## @seealso{deblank, strrep, strtrim, untabify} ## @end deftypefn -function y = strjust (s, pos) +function y = strjust (s, pos = "right") if (nargin < 1 || nargin > 2) print_usage (); - endif - - if (nargin == 1) - pos = "right"; - else - pos = tolower (pos); - endif - - if (ndims (s) != 2) - error ("strjust: input must be a string or character matrix"); + elseif (! ischar (s) || ndims (s) > 2) + error ("strjust: S must be a string or 2-D character matrix"); endif if (isempty (s)) y = s; - else - ## Apparently, Matlab considers nulls to be blanks as well; however, does - ## not preserve the nulls, but rather converts them to blanks. That's a - ## bit unexpected, but it allows simpler processing, because we can move - ## just the nonblank characters. So we'll do the same here. + return; + endif - [nr, nc] = size (s); - ## Find the indices of all nonblanks. - nonbl = s != " " & s != "\0"; - [idx, jdx] = find (nonbl); + ## Apparently, Matlab considers nulls to be blanks as well; however, does + ## not preserve the nulls, but rather converts them to blanks. That's a + ## bit unexpected, but it allows simpler processing, because we can move + ## just the nonblank characters. So we'll do the same here. + + [nr, nc] = size (s); + ## Find the indices of all nonblanks. + nonbl = s != " " & s != "\0"; + [idx, jdx] = find (nonbl); - if (strcmp (pos, "right")) - ## We wish to find the maximum column index for each row. Because jdx is - ## sorted, we can take advantage of the fact that assignment is processed - ## sequentially and for duplicate indices the last value will remain. - maxs = nc * ones (nr, 1); - maxs(idx) = jdx; - shift = nc - maxs; - elseif (strcmp (pos, "left")) - ## See above for explanation. - mins = ones (nr, 1); - mins(flipud (idx(:))) = flipud (jdx(:)); - shift = 1 - mins; - else - ## Use both of the above. - mins = ones (nr, 1); - mins(flipud (idx(:))) = flipud (jdx(:)); - maxs = nc * ones (nr, 1); - maxs(idx) = jdx; - shift = floor ((nc + 1 - maxs - mins) / 2); - endif + if (strcmpi (pos, "right")) + ## We wish to find the maximum column index for each row. Because jdx is + ## sorted, we can take advantage of the fact that assignment is processed + ## sequentially and for duplicate indices the last value will remain. + maxs = repmat (nc, [nr, 1]); + maxs(idx) = jdx; + shift = nc - maxs; + elseif (strcmpi (pos, "left")) + ## See above for explanation. + mins = ones (nr, 1); + mins(flipud (idx(:))) = flipud (jdx(:)); + shift = 1 - mins; + else + ## Use both of the above to achieve centering. + mins = ones (nr, 1); + mins(flipud (idx(:))) = flipud (jdx(:)); + maxs = repmat (nc, [nr, 1]); + maxs(idx) = jdx; + shift = floor ((nc + 1 - maxs - mins) / 2); + endif - ## Adjust the column indices. - jdx += shift (idx); + ## Adjust the column indices. + jdx += shift(idx); - ## Create a blank matrix and position the nonblank characters. - y = " "(ones (1, nr), ones (1, nc)); - y(sub2ind ([nr, nc], idx, jdx)) = s(nonbl); - endif + ## Create a blank matrix and position the nonblank characters. + y = repmat (" ", nr, nc); + y(sub2ind ([nr, nc], idx, jdx)) = s(nonbl); endfunction -%!error <Invalid call to strjust> strjust(); -%!error <Invalid call to strjust> strjust(["a";"ab"], "center", 1); + %!assert (strjust (["a"; "ab"; "abc"; "abcd"]), %! [" a";" ab"; " abc"; "abcd"]); -%!assert (strjust (["a"; "ab"; "abc"; "abcd"], "center"), +%!assert (strjust ([" a"; " ab"; "abc"; "abcd"], "left"), +%! ["a "; "ab "; "abc "; "abcd"]); +%!assert (strjust (["a"; "ab"; "abc"; "abcd"], "CENTER"), %! [" a "; " ab"; "abc "; "abcd"]); +%!assert (strjust (["";""]), ""); + +%% Test input validation +%!error <Invalid call to strjust> strjust () +%!error <Invalid call to strjust> strjust (["a";"ab"], "center", 1) +%!error <S must be a string> strjust (ones(3,3)) +%!error <S must be a string> strjust (char (ones(3,3,3))) +