Mercurial > octave
view scripts/io/textscan.m @ 17338:1c89599167a6
maint: End m-files with 1 blank line.
Simplifies automated grammarchecking script.
* scripts/@ftp/ascii.m, scripts/@ftp/binary.m, scripts/@ftp/cd.m,
scripts/@ftp/close.m, scripts/@ftp/delete.m, scripts/@ftp/dir.m,
scripts/@ftp/display.m, scripts/@ftp/ftp.m, scripts/@ftp/loadobj.m,
scripts/@ftp/mget.m, scripts/@ftp/mkdir.m, scripts/@ftp/mput.m,
scripts/@ftp/rename.m, scripts/@ftp/rmdir.m, scripts/@ftp/saveobj.m,
scripts/audio/lin2mu.m, scripts/audio/loadaudio.m, scripts/audio/mu2lin.m,
scripts/audio/record.m, scripts/audio/saveaudio.m, scripts/audio/setaudio.m,
scripts/deprecated/__error_text__.m, scripts/deprecated/cut.m,
scripts/deprecated/error_text.m, scripts/deprecated/isstr.m,
scripts/deprecated/polyderiv.m, scripts/deprecated/studentize.m,
scripts/deprecated/sylvester_matrix.m, scripts/general/bicubic.m,
scripts/general/celldisp.m, scripts/general/colon.m,
scripts/general/cplxpair.m, scripts/general/del2.m, scripts/general/display.m,
scripts/general/isdir.m, scripts/general/isequaln.m, scripts/general/loadobj.m,
scripts/general/private/__isequal__.m, scripts/general/private/__splinen__.m,
scripts/general/profexplore.m, scripts/general/quadgk.m,
scripts/general/randi.m, scripts/general/repmat.m, scripts/general/saveobj.m,
scripts/geometry/delaunay.m, scripts/help/__unimplemented__.m,
scripts/help/doc_cache_create.m, scripts/help/get_first_help_sentence.m,
scripts/help/help.m, scripts/help/print_usage.m,
scripts/help/private/__additional_help_message__.m,
scripts/help/private/__strip_html_tags__.m, scripts/help/type.m,
scripts/image/imfinfo.m, scripts/image/imformats.m, scripts/image/imread.m,
scripts/image/imwrite.m, scripts/image/private/__imfinfo__.m,
scripts/image/private/__imread__.m, scripts/image/private/__imwrite__.m,
scripts/image/private/imageIO.m, scripts/image/private/imwrite_filename.m,
scripts/image/private/ind2x.m, scripts/io/beep.m, scripts/io/strread.m,
scripts/io/textread.m, scripts/io/textscan.m, scripts/linear-algebra/krylov.m,
scripts/linear-algebra/subspace.m, scripts/miscellaneous/bug_report.m,
scripts/miscellaneous/bunzip2.m, scripts/miscellaneous/cast.m,
scripts/miscellaneous/copyfile.m, scripts/miscellaneous/debug.m,
scripts/miscellaneous/dir.m, scripts/miscellaneous/dump_prefs.m,
scripts/miscellaneous/error_ids.m, scripts/miscellaneous/fileattrib.m,
scripts/miscellaneous/gunzip.m, scripts/miscellaneous/isdeployed.m,
scripts/miscellaneous/ismac.m, scripts/miscellaneous/mex.m,
scripts/miscellaneous/mexext.m, scripts/miscellaneous/mkoctfile.m,
scripts/miscellaneous/movefile.m, scripts/miscellaneous/namelengthmax.m,
scripts/miscellaneous/news.m, scripts/miscellaneous/pack.m,
scripts/miscellaneous/perl.m,
scripts/miscellaneous/private/display_info_file.m,
scripts/miscellaneous/python.m, scripts/miscellaneous/rmappdata.m,
scripts/miscellaneous/run.m, scripts/miscellaneous/tar.m,
scripts/miscellaneous/tempname.m, scripts/miscellaneous/untar.m,
scripts/miscellaneous/unzip.m, scripts/miscellaneous/what.m,
scripts/miscellaneous/zip.m, scripts/optimization/fminunc.m,
scripts/optimization/fsolve.m, scripts/optimization/fzero.m,
scripts/optimization/glpk.m, scripts/optimization/optimget.m,
scripts/optimization/optimset.m, scripts/optimization/qp.m,
scripts/optimization/sqp.m, scripts/path/pathdef.m, scripts/pkg/pkg.m,
scripts/pkg/private/build.m, scripts/pkg/private/describe.m,
scripts/pkg/private/dirempty.m, scripts/pkg/private/get_forge_download.m,
scripts/pkg/private/get_forge_pkg.m,
scripts/pkg/private/get_unsatisfied_deps.m, scripts/pkg/private/install.m,
scripts/pkg/private/is_architecture_dependent.m,
scripts/pkg/private/list_forge_packages.m, scripts/pkg/private/rebuild.m,
scripts/pkg/private/shell.m, scripts/pkg/private/uninstall.m,
scripts/plot/axes.m, scripts/plot/box.m, scripts/plot/closereq.m,
scripts/plot/diffuse.m, scripts/plot/ezpolar.m, scripts/plot/findfigs.m,
scripts/plot/gco.m, scripts/plot/guidata.m, scripts/plot/guihandles.m,
scripts/plot/hdl2struct.m, scripts/plot/linkprop.m, scripts/plot/peaks.m,
scripts/plot/print.m, scripts/plot/private/__add_datasource__.m,
scripts/plot/private/__axis_label__.m, scripts/plot/private/__clabel__.m,
scripts/plot/private/__color_str_rgb__.m, scripts/plot/private/__contour__.m,
scripts/plot/private/__default_plot_options__.m,
scripts/plot/private/__errcomm__.m, scripts/plot/private/__file_filter__.m,
scripts/plot/private/__fltk_file_filter__.m,
scripts/plot/private/__getlegenddata__.m,
scripts/plot/private/__gnuplot_open_stream__.m,
scripts/plot/private/__gnuplot_print__.m,
scripts/plot/private/__go_draw_axes__.m,
scripts/plot/private/__interp_cube__.m, scripts/plot/private/__is_function__.m,
scripts/plot/private/__line__.m, scripts/plot/private/__marching_cube__.m,
scripts/plot/private/__next_line_style__.m, scripts/plot/private/__patch__.m,
scripts/plot/private/__pie__.m, scripts/plot/private/__pltopt__.m,
scripts/plot/private/__quiver__.m, scripts/plot/private/__scatter__.m,
scripts/plot/private/__stem__.m, scripts/plot/private/__uigetdir_fltk__.m,
scripts/plot/private/__uigetfile_fltk__.m,
scripts/plot/private/__uiobject_split_args__.m,
scripts/plot/private/__uiputfile_fltk__.m, scripts/plot/refresh.m,
scripts/plot/saveas.m, scripts/plot/shg.m, scripts/plot/specular.m,
scripts/plot/sphere.m, scripts/plot/struct2hdl.m, scripts/plot/subplot.m,
scripts/plot/uicontextmenu.m, scripts/plot/uicontrol.m, scripts/plot/uipanel.m,
scripts/plot/uipushtool.m, scripts/plot/uiresume.m,
scripts/plot/uitoggletool.m, scripts/plot/uitoolbar.m, scripts/plot/uiwait.m,
scripts/plot/waitforbuttonpress.m, scripts/polynomial/pchip.m,
scripts/polynomial/polyeig.m, scripts/polynomial/ppval.m,
scripts/prefs/addpref.m, scripts/prefs/getpref.m, scripts/prefs/ispref.m,
scripts/prefs/private/loadprefs.m, scripts/prefs/private/prefsfile.m,
scripts/prefs/private/saveprefs.m, scripts/prefs/setpref.m,
scripts/set/private/validargs.m, scripts/set/unique.m,
scripts/signal/arch_fit.m, scripts/signal/arch_rnd.m,
scripts/signal/arch_test.m, scripts/signal/arma_rnd.m,
scripts/signal/durbinlevinson.m, scripts/signal/fractdiff.m,
scripts/signal/freqz.m, scripts/signal/freqz_plot.m, scripts/signal/hurst.m,
scripts/signal/periodogram.m, scripts/signal/private/rectangle_lw.m,
scripts/signal/private/rectangle_sw.m, scripts/signal/private/triangle_sw.m,
scripts/signal/spectral_adf.m, scripts/signal/spectral_xdf.m,
scripts/signal/stft.m, scripts/signal/synthesis.m, scripts/signal/yulewalker.m,
scripts/sparse/colperm.m, scripts/sparse/eigs.m, scripts/sparse/etreeplot.m,
scripts/sparse/gmres.m, scripts/sparse/private/__sprand_impl__.m,
scripts/sparse/spdiags.m, scripts/sparse/sprandn.m, scripts/specfun/bessel.m,
scripts/specfun/betaln.m, scripts/specfun/expint.m,
scripts/special-matrix/gallery.m, scripts/startup/__finish__.m,
scripts/statistics/base/qqplot.m, scripts/statistics/distributions/tcdf.m,
scripts/statistics/distributions/wienrnd.m,
scripts/statistics/models/logistic_regression.m,
scripts/statistics/models/private/logistic_regression_derivatives.m,
scripts/statistics/models/private/logistic_regression_likelihood.m,
scripts/statistics/tests/anova.m, scripts/statistics/tests/bartlett_test.m,
scripts/statistics/tests/chisquare_test_homogeneity.m,
scripts/statistics/tests/chisquare_test_independence.m,
scripts/statistics/tests/cor_test.m,
scripts/statistics/tests/f_test_regression.m,
scripts/statistics/tests/hotelling_test.m,
scripts/statistics/tests/hotelling_test_2.m,
scripts/statistics/tests/kolmogorov_smirnov_test_2.m,
scripts/statistics/tests/kruskal_wallis_test.m,
scripts/statistics/tests/manova.m, scripts/statistics/tests/mcnemar_test.m,
scripts/statistics/tests/prop_test_2.m, scripts/statistics/tests/run_test.m,
scripts/statistics/tests/sign_test.m, scripts/statistics/tests/t_test.m,
scripts/statistics/tests/t_test_2.m,
scripts/statistics/tests/t_test_regression.m,
scripts/statistics/tests/u_test.m, scripts/statistics/tests/var_test.m,
scripts/statistics/tests/welch_test.m,
scripts/statistics/tests/wilcoxon_test.m, scripts/statistics/tests/z_test.m,
scripts/statistics/tests/z_test_2.m, scripts/strings/strcat.m,
scripts/strings/strjoin.m, scripts/strings/strsplit.m,
scripts/testfun/__have_feature__.m, scripts/testfun/__printf_assert__.m,
scripts/testfun/__prog_output_assert__.m, scripts/testfun/__run_test_suite__.m,
scripts/time/clock.m, scripts/time/datenum.m, scripts/ui/errordlg.m,
scripts/ui/private/message_dialog.m: End m-files with 1 blank line.
author | Rik <rik@octave.org> |
---|---|
date | Wed, 28 Aug 2013 08:33:02 -0700 |
parents | b81b9d079515 |
children | 2ce1f1966fc3 |
line wrap: on
line source
## Copyright (C) 2010-2013 Ben Abbott ## ## This file is part of Octave. ## ## Octave is free software; you can redistribute it and/or modify it ## under the terms of the GNU General Public License as published by ## the Free Software Foundation; either version 3 of the License, or (at ## your option) any later version. ## ## Octave is distributed in the hope that it will be useful, but ## WITHOUT ANY WARRANTY; without even the implied warranty of ## MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU ## General Public License for more details. ## ## You should have received a copy of the GNU General Public License ## along with Octave; see the file COPYING. If not, see ## <http://www.gnu.org/licenses/>. ## -*- texinfo -*- ## @deftypefn {Function File} {@var{C} =} textscan (@var{fid}, @var{format}) ## @deftypefnx {Function File} {@var{C} =} textscan (@var{fid}, @var{format}, @var{n}) ## @deftypefnx {Function File} {@var{C} =} textscan (@var{fid}, @var{format}, @var{param}, @var{value}, @dots{}) ## @deftypefnx {Function File} {@var{C} =} textscan (@var{fid}, @var{format}, @var{n}, @var{param}, @var{value}, @dots{}) ## @deftypefnx {Function File} {@var{C} =} textscan (@var{str}, @dots{}) ## @deftypefnx {Function File} {[@var{C}, @var{position}] =} textscan (@var{fid}, @dots{}) ## Read data from a text file or string. ## ## The string @var{str} or file associated with @var{fid} is read from and ## parsed according to @var{format}. The function behaves like @code{strread} ## except it can also read from file instead of a string. See the documentation ## of @code{strread} for details. ## ## In addition to the options supported by @code{strread}, this function ## supports a few more: ## ## @itemize ## @item @qcode{"collectoutput"}: ## A value of 1 or true instructs textscan to concatenate consecutive columns ## of the same class in the output cell array. A value of 0 or false (default) ## leaves output in distinct columns. ## ## @item @qcode{"endofline"}: ## Specify @qcode{"\r"}, @qcode{"\n"} or @qcode{"\r\n"} (for CR, LF, or ## CRLF). If no value is given, it will be inferred from the file. If set ## to "" (empty string) EOLs are ignored as delimiters and added to ## whitespace. ## ## @item @qcode{"headerlines"}: ## The first @var{value} number of lines of @var{fid} are skipped. ## ## @item @qcode{"returnonerror"}: ## If set to numerical 1 or true (default), return normally when read errors ## have been encountered. If set to 0 or false, return an error and no data. ## As the string or file is read by columns rather than by rows, and because ## textscan is fairly forgiving as regards read errors, setting this option ## may have little or no actual effect. ## @end itemize ## ## When reading from a character string, optional input argument @var{n} ## specifies the number of times @var{format} should be used (i.e., to limit ## the amount of data read). ## When reading from file, @var{n} specifies the number of data lines to read; ## in this sense it differs slightly from the format repeat count in strread. ## ## The output @var{C} is a cell array whose second dimension is determined ## by the number of format specifiers. ## ## The second output, @var{position}, provides the position, in characters, ## from the beginning of the file. ## ## If the format string is empty (not: omitted) and the file contains only ## numeric data (excluding headerlines), textscan will return data in a number ## of columns matching the number of numeric fields on the first data line of ## the file. ## ## @seealso{dlmread, fscanf, load, strread, textread} ## @end deftypefn ## Author: Ben Abbott <bpabbott@mac.com> function [C, position] = textscan (fid, format = "%f", varargin) BUFLENGTH = 4096; # Read buffer emptfmt = 0; # Signals deliberately empty format string ## Check input if (nargin < 1) print_usage (); endif if (isempty (format)) emptfmt = 1; format = "%f"; endif if (! ischar (format)) error ("textscan: FORMAT must be a string"); endif ## Determine the number of data fields & initialize output array num_fields = numel (strfind (format, "%")) - numel (strfind (format, "%*")); C = cell (1, num_fields); if (! (isa (fid, "double") && fid > 0) && ! ischar (fid)) error ("textscan: first argument must be a file id or character string"); endif args = varargin; if (nargin > 2 && isnumeric (args{1})) nlines = args{1}; else nlines = Inf; endif if (nlines < 1) printf ("textscan: N = 0, no data read\n"); return; endif if (! any (strcmpi (args, "emptyvalue"))) ## Matlab returns NaNs for missing values args(end+1:end+2) = {'emptyvalue', NaN}; endif ## Check default parameter values that differ for strread & textread ipos = find (strcmpi (args, "whitespace")); if (isempty (ipos)) ## Matlab default whitespace = " \b\t" args(end+1:end+2) = {'whitespace', " \b\t"}; whitespace = " \b\t"; else ## Check if there's at least one string format specifier has_str_fmt = regexp (format, '%[*]?\d*s', "once"); ## If there is a string format AND whitespace value = empty, ## don't add a space (char(32)) to whitespace if (! (isempty (args{ipos+1}) && has_str_fmt)) args{ipos+1} = unique ([" ", args{ipos+1}]); endif endif if (! any (strcmpi (args, "delimiter"))) ## Matlab says default delimiter = whitespace. ## strread() will pick this up further args(end+1:end+2) = {'delimiter', ""}; delimiter = ""; else delimiter = args{find (strcmpi (args, "delimiter")) + 1}; endif collop = false; ipos = find (strcmpi (args, "collectoutput")); if (! isempty (ipos)) ## Search & concatenate consecutive columns of same class requested if (isscalar (args{ipos+1}) && (islogical (args{ipos+1}) || isnumeric (args{ipos+1}))) collop = args{ipos+1}; else warning ("textscan: illegal value for CollectOutput parameter - ignored"); endif ## Remove argument before call to strread() below args(ipos:ipos+1) = []; endif if (any (strcmpi (args, "returnonerror"))) ## Because of the way strread() reads data (columnwise) this parameter ## can't be neatly implemented. strread() will pick it up anyway warning ('textscan: ReturnOnError is not fully implemented'); else ## Set default value (=true) args(end+1:end+2) = {"returnonerror", 1}; endif ## Check if a headerlines argument is specified headerlines = find (strcmpi (args, "headerlines"), 1); if (! isempty (headerlines)) ## Yep. But it is stray when reading from strings... if (ischar (fid)) warning ("textscan: 'headerlines' ignored when reading from strings"); endif endif if (ischar (fid)) ## Read from a text string if (nargout == 2) error ("textscan: cannot provide position information for character input"); endif str = fid; else st_pos = ftell (fid); ## Skip header lines if requested if (! isempty (headerlines)) ## Beware of missing or wrong headerline value if (headerlines == numel (args) || ! isnumeric (args{headerlines + 1})) error ("Missing or illegal value for 'headerlines'" ); endif ## Avoid conveying floats to fskipl args{headerlines + 1} = round (args{headerlines + 1}); if (args{headerlines + 1} > 0) ## Beware of zero valued headerline, fskipl would skip to EOF fskipl (fid, args{headerlines + 1}); args(headerlines:headerlines+1) = []; st_pos = ftell (fid); elseif (args{headerlines + 1} < 0) warning ("textscan.m: negative headerline value ignored"); endif endif ## Read a first file chunk. Rest follows after endofline processing [str, count] = fscanf (fid, "%c", BUFLENGTH); endif ## Check for empty result if (isempty (str)) warning ("textscan: no data read"); return; endif ## Check value of 'endofline'. String or file doesn't seem to matter endofline = find (strcmpi (args, "endofline"), 1); if (! isempty (endofline)) if (ischar (args{endofline + 1})) eol_char = args{endofline + 1}; if (! any (strcmp (eol_char, {"", "\n", "\r", "\r\n"}))) error ("textscan: illegal EndOfLine character value specified"); endif else error ("textscan: character value required for EndOfLine"); endif else if (! ischar (fid)) ## Determine EOL from file. Search for EOL candidates in first BUFLENGTH chars eol_srch_len = min (length (str), BUFLENGTH); ## First try DOS (CRLF) if (! isempty (strfind (str(1 : eol_srch_len), "\r\n"))) eol_char = "\r\n"; ## Perhaps old Macintosh? (CR) elseif (! isempty (strfind (str(1 : eol_srch_len), "\r"))) eol_char = "\r"; ## Otherwise, use plain UNIX (LF) else eol_char = "\n"; endif else eol_char = "\n"; endif ## Set up the default endofline param value args(end+1:end+2) = {"endofline", eol_char}; endif if (!ischar (fid)) ## Now that we know what EOL looks like, we can process format_repeat_count. ## FIXME The below isn't ML-compatible: counts lines, not format string uses if (isfinite (nlines) && (nlines >= 0)) l_eol_char = length (eol_char); eoi = findstr (str, eol_char); n_eoi = length (eoi); nblks = 0; ## Avoid slow repeated str concatenation, first seek requested end of data while (n_eoi < nlines && count == BUFLENGTH) [nstr, count] = fscanf (fid, "%c", BUFLENGTH); if (count > 0) ## Watch out for multichar EOL being missed across buffer boundaries if (l_eol_char > 1) str = [str(end - length (eol_char) + 2 : end) nstr]; else str = nstr; endif eoi = findstr (str, eol_char); n_eoi += numel (eoi); ++nblks; endif endwhile ## OK, found EOL delimiting last requested line. Compute ptr (incl. EOL) if (isempty (eoi)) printf ("textscan: format repeat count specified but no endofline found\n"); data_size = nblks * BUFLENGTH + count; else ## Compute data size to read incl complete EOL data_size = (nblks * BUFLENGTH) + eoi(end + min (nlines, n_eoi) - n_eoi) ... + l_eol_char - 1; endif fseek (fid, st_pos, "bof"); str = fscanf (fid, "%c", data_size); else fseek (fid, st_pos, "bof"); str = fread (fid, "char=>char").'; endif endif ## Strip trailing EOL to avoid returning stray missing values (f. strread). ## However, in case of CollectOutput request, presence of EOL is required; ## also in case of deliberately entered empty format string eol_at_end = strcmp (str(end-length (eol_char) + 1 : end), eol_char); if (collop || emptfmt) if (! eol_at_end) str(end+1 : end+length (eol_char)) = eol_char; endif elseif (eol_at_end) str(end-length (eol_char) + 1 : end) = ""; ## A corner case: str may now be empty.... if (isempty (str)); return; endif endif ## Call strread to make it do the real work C = cell (1, num_fields); [C{:}] = strread (str, format, args{:}); ## I.c.o. empty format, match nr. of cols to nr. of fields on first read line if (emptfmt) ## Find end of first line eoi = index (str, eol_char); if (eoi) ## str contains an EOL, proceed with assessing nr. of columns ncols = countcols (C, str(1 : eoi-1), delimiter, whitespace); ## See if lowermost data row must be completed pad = mod (numel (C{1}), ncols); if (pad) ## Textscan returns NaNs for empty fields C(1) = [C{1}; NaN(ncols - pad, 1)]; endif ## Replace NaNs with EmptyValue, if any ipos = find (strcmpi (args, "emptyvalue")); if (ipos) C{1}(find (isnan (C{1}))) = args{ipos+1}; endif ## Compute nr. of rows nrows = floor (numel (C{1}) / ncols); ## Reshape C; watch out, transpose needed C(1) = reshape (C{1}, ncols, numel (C{1}) / ncols)'; ## Distribute columns over C and wipe cols 2:end of C{1} for ii=2:ncols C(ii) = C{1}(:, ii); endfor C{1} = C{1}(:, 1); endif endif ## If requested, collect output columns of same class if (collop) C = colloutp (C); endif if (nargout == 2) ## Remember file position (persistent var) position = ftell (fid); endif endfunction ## Assess nr of data fields on first line of data function ncols = countcols (C, str, dlm, wsp) if (isempty (dlm)) ## Field separator = whitespace. Fold multiple whitespace into one str = regexprep (str, sprintf ("[%s]", wsp), " "); str = strtrim (str); ncols = numel (strfind (str, " ")) + 1; else ncols = numel (regexp (str, sprintf ("[%s]", dlm))) + 1; endif endfunction ## Collect consecutive columns of same class into one cell column function C = colloutp (C) ## Start at rightmost column and work backwards to avoid ptr mixup ii = numel (C); while (ii > 1) clss1 = class (C{ii}); jj = ii; while (jj > 1 && strcmp (clss1, class (C{jj - 1}))) ## Column to the left is still same class; check next column to the left --jj; endwhile if (jj < ii) ## Concatenate columns into current column C{jj} = [C{jj : ii}]; ## Wipe concatenated columns to the right, resume search to the left C(jj+1 : ii) = []; ii = jj - 1; else ## No similar class in column to the left, search from there --ii; endif endwhile endfunction %!test %! str = "1, 2, 3, 4\n 5, , , 8\n 9, 10, 11, 12"; %! fmtstr = "%f %d %f %s"; %! c = textscan (str, fmtstr, 2, "delimiter", ",", "emptyvalue", -Inf); %! assert (isequal (c{1}, [1;5])); %! assert (length (c{1}), 2); %! assert (iscellstr (c{4})); %! assert (isequal (c{3}, [3; -Inf])); %!test %! b = [10:10:100]; %! b = [b; 8*b/5]; %! str = sprintf ("%g miles/hr = %g kilometers/hr\n", b); %! fmt = "%f miles/hr = %f kilometers/hr"; %! c = textscan (str, fmt); %! assert (b(1,:)', c{1}, 1e-5); %! assert (b(2,:)', c{2}, 1e-5); %!test %! str = "13, 72, NA, str1, 25\r\n// Middle line\r\n36, na, 05, str3, 6"; %! a = textscan (str, "%d %n %f %s %n", "delimiter", ",","treatAsEmpty", {"NA", "na"},"commentStyle", "//"); %! assert (a{1}, int32 ([13; 36])); %! assert (a{2}, [72; NaN]); %! assert (a{3}, [NaN; 5]); %! assert (a{4}, {"str1"; "str3"}); %! assert (a{5}, [25; 6]); %!test %! str = "Km:10 = hhhBjjj miles16hour\r\n"; %! str = [str "Km:15 = hhhJjjj miles241hour\r\n"]; %! str = [str "Km:2 = hhhRjjj miles3hour\r\n"]; %! str = [str "Km:25 = hhhZ\r\n"]; %! fmt = "Km:%d = hhh%1sjjj miles%dhour"; %! a = textscan (str, fmt, "delimiter", " "); %! assert (a{1}', int32 ([10 15 2 25])); %! assert (a{2}', {'B' 'J' 'R' 'Z'}); %! assert (a{3}', int32 ([16 241 3 0])); %% Test with default endofline parameter %!test %! c = textscan ("L1\nL2", "%s"); %! assert (c{:}, {"L1"; "L2"}); %% Test with endofline parameter set to "" (empty) - newline should be in word %!test %! c = textscan ("L1\nL2", "%s", "endofline", ""); %! assert (int8 (c{:}{:}), int8 ([ 76, 49, 10, 76, 50 ])); %!test %! ## No delimiters at all besides EOL. Skip fields, even empty fields %! str = "Text1Text2Text\nTextText4Text\nText57Text"; %! c = textscan (str, "Text%*dText%dText"); %! assert (c{1}, int32 ([2; 4; 0])); %!test %% CollectOutput test %! b = [10:10:100]; %! b = [b; 8*b/5; 8*b*1000/5]; %! str = sprintf ("%g miles/hr = %g (%g) kilometers (meters)/hr\n", b); %! fmt = "%f miles%s %s %f (%f) kilometers %*s"; %! c = textscan (str, fmt, "collectoutput", 1); %! assert (size (c{3}), [10, 2]); %! assert (size (c{2}), [10, 2]); %!test %% CollectOutput test with uneven column length files %! b = [10:10:100]; %! b = [b; 8*b/5; 8*b*1000/5]; %! str = sprintf ("%g miles/hr = %g (%g) kilometers (meters)/hr\n", b); %! str = [str "110 miles/hr"]; %! fmt = "%f miles%s %s %f (%f) kilometers %*s"; %! c = textscan (str, fmt, "collectoutput", 1); %! assert (size (c{1}), [11, 1]); %! assert (size (c{3}), [11, 2]); %! assert (size (c{2}), [11, 2]); %! assert (c{3}(end), NaN); %! assert (c{2}{11, 1}, "/hr"); %! assert (isempty (c{2}{11, 2}), true); %% Test input validation %!error textscan () %!error textscan (single (4)) %!error textscan ({4}) %!error <must be a string> textscan ("Hello World", 2) %!error <cannot provide position information> [C, pos] = textscan ("Hello World") %!error <character value required> textscan ("Hello World", '%s', 'EndOfLine', 3) %! Test incomplete first data line %! R = textscan (['Empty1' char(10)], 'Empty%d %f'); %! assert (R{1}, int32 (1)); %! assert (isempty (R{2}), true); %% bug #37023 (actually a strread test) %!test %! data = textscan(" 1. 1 \n 2 3\n", '%f %f'); %! assert (data{1}, [1; 2], 1e-15); %! assert (data{2}, [1; 3], 1e-15); %% Whitespace test (bug #37333) using delimiter ";" %!test %! tc = []; %! tc{1, 1} = "C:/code;"; %! tc{1, end+1} = "C:/code/meas;"; %! tc{1, end+1} = " C:/code/sim;"; %! tc{1, end+1} = "C:/code/utils;"; %! string = [tc{:}]; %! c = textscan (string, "%s", "delimiter", ";"); %! for k = 1:numel (c{1}) %! lh = c{1}{k}; %! rh = tc{k}; %! rh(rh == ";") = ""; %! rh = strtrim (rh); %! assert (strcmp (lh, rh)); %! end %% Whitespace test (bug #37333), adding multipleDelimsAsOne true arg %!test %! tc = []; %! tc{1, 1} = "C:/code;"; %! tc{1, end+1} = " C:/code/meas;"; %! tc{1, end+1} = "C:/code/sim;;"; %! tc{1, end+1} = "C:/code/utils;"; %! string = [tc{:}]; %! c = textscan (string, "%s", "delimiter", ";", "multipleDelimsAsOne", 1); %! for k = 1:numel (c{1}) %! lh = c{1}{k}; %! rh = tc{k}; %! rh(rh == ";") = ""; %! rh = strtrim (rh); %! assert (strcmp (lh, rh)); %! end %% Whitespace test (bug #37333), adding multipleDelimsAsOne false arg %!test %! tc = []; %! tc{1, 1} = "C:/code;"; %! tc{1, end+1} = " C:/code/meas;"; %! tc{1, end+1} = "C:/code/sim;;"; %! tc{1, end+1} = ""; %! tc{1, end+1} = "C:/code/utils;"; %! string = [tc{:}]; %! c = textscan (string, "%s", "delimiter", ";", "multipleDelimsAsOne", 0); %! for k = 1:numel (c{1}) %! lh = c{1}{k}; %! rh = tc{k}; %! rh(rh == ";") = ""; %! rh = strtrim (rh); %! assert (strcmp (lh, rh)); %! end %% Whitespace test (bug #37333) whitespace "" arg %!test %! tc = []; %! tc{1, 1} = "C:/code;"; %! tc{1, end+1} = " C:/code/meas;"; %! tc{1, end+1} = "C:/code/sim;"; %! tc{1, end+1} = "C:/code/utils;"; %! string = [tc{:}]; %! c = textscan (string, "%s", "delimiter", ";", "whitespace", ""); %! for k = 1:numel (c{1}) %! lh = c{1}{k}; %! rh = tc{k}; %! rh(rh == ";") = ""; %! assert (strcmp (lh, rh)); %! end %% Whitespace test (bug #37333), whitespace " " arg %!test %! tc = []; %! tc{1, 1} = "C:/code;"; %! tc{1, end+1} = " C:/code/meas;"; %! tc{1, end+1} = "C:/code/sim;"; %! tc{1, end+1} = "C:/code/utils;"; %! string = [tc{:}]; %! c = textscan (string, "%s", "delimiter", ";", "whitespace", " "); %! for k = 1:numel (c{1}) %! lh = c{1}{k}; %! rh = tc{k}; %! rh(rh == ";") = ""; %! rh = strtrim (rh); %! assert (strcmp (lh, rh)); %! end %% Test reading from a real file %!test %! f = tmpnam (); %! fid = fopen (f, "w+"); %! d = rand (1, 4); %! fprintf (fid, " %f %f %f %f ", d); %! fseek (fid, 0, "bof"); %! A = textscan (fid, "%f %f"); %! fclose (fid); %! unlink (f); %! assert (A{1}, [d(1); d(3)], 1e-6); %! assert (A{2}, [d(2); d(4)], 1e-6); %% Tests reading with empty format, should return proper nr of columns %!test %! f = tmpnam (); %! fid = fopen (f, "w+"); %! fprintf (fid, " 1 2 3 4\n5 6 7 8"); %! fseek (fid, 0, "bof"); %! A = textscan (fid, ""); %! fclose (fid); %! unlink (f); %! assert (A{1}, [1 ; 5], 1e-6); %! assert (A{2}, [2 ; 6], 1e-6); %! assert (A{3}, [3 ; 7], 1e-6); %! assert (A{4}, [4 ; 8], 1e-6); %% Tests reading with empty format; empty fields & incomplete lower row %!test %! f = tmpnam (); %! fid = fopen (f, "w+"); %! fprintf (fid, " ,2,,4\n5,6"); %! fseek (fid, 0, "bof"); %! A = textscan (fid, "", "delimiter", ",", "EmptyValue", 999, "CollectOutput" , 1); %! fclose (fid); %! unlink (f); %! assert (A{1}, [999, 2, 999, 4; 5, 6, 999, 999], 1e-6); %% Error message tests %!test %! f = tmpnam (); %! fid = fopen (f, "w+"); %! msg1 = "Missing or illegal value for 'headerlines'"; %! try %! A = textscan (fid, "", "headerlines"); %! end_try_catch; %! fclose (fid); %! unlink (f); %! assert (msg1, lasterr); %!test %! f = tmpnam (); %! fid = fopen (f, "w+"); %! msg1 = "Missing or illegal value for 'headerlines'"; %! try %! A = textscan (fid, "", "headerlines", "hh"); %! end_try_catch; %! fclose (fid); %! unlink (f); %! assert (msg1, lasterr); %!test %! f = tmpnam (); %! fid = fopen (f, "w+"); %! fprintf (fid,"some_string"); %! fseek (fid, 0, "bof"); %! msg1 = "textscan: illegal EndOfLine character value specified"; %! try %! A = textscan (fid, "%f", "EndOfLine", "\n\r"); %! end_try_catch; %! fclose (fid); %! unlink (f); %! assert (msg1, lasterr); %!test %! f = tmpnam (); %! fid = fopen (f, "w+"); %! fprintf (fid,"some_string"); %! fseek (fid, 0, "bof"); %! msg1 = "textscan: character value required for EndOfLine"; %! try %! A = textscan (fid, "%f", "EndOfLine", 33); %! end_try_catch; %! fclose (fid); %! unlink (f); %! assert (msg1, lasterr);