Mercurial > octave
view scripts/general/num2str.m @ 29949:f254c302bb9c
remove JIT compiler from Octave sources
As stated in the NEWS file entry added with this changeset, no one
has ever seriously taken on further development of the JIT compiler in
Octave since it was first added as part of a Google Summer of Code
project in 2012 and it still does nothing significant. It is out of
date with the default interpreter that walks the parse tree. Even
though we have fixed the configure script to disable it by default,
people still ask questions about how to build it, but it doesn’t seem
that they are doing that to work on it but because they think it will
make Octave code run faster (it never did, except for some extremely
simple bits of code as examples for demonstration purposes only).
* NEWS: Note change.
* configure.ac, acinclude.m4: Eliminate checks and macros related to
the JIT compiler and LLVM.
* basics.txi, install.txi, octave.texi, vectorize.txi: Remove mention
of JIT compiler and LLVM.
* jit-ir.cc, jit-ir.h, jit-typeinfo.cc, jit-typeinfo.h, jit-util.cc,
jit-util.h, pt-jit.cc, pt-jit.h: Delete.
* libinterp/parse-tree/module.mk: Update.
* Array-jit.cc: Delete.
* libinterp/template-inst/module.mk: Update.
* test/jit.tst: Delete.
* test/module.mk: Update.
* interpreter.cc (interpreter::interpreter): Don't check options for
debug_jit or jit_compiler.
* toplev.cc (F__octave_config_info__): Remove JIT compiler and LLVM
info from struct.
* ov-base.h (octave_base_value::grab, octave_base_value::release):
Delete.
* ov-builtin.h, ov-builtin.cc (octave_builtin::to_jit,
octave_builtin::stash_jit): Delete.
(octave_builtin::m_jtype): Delete data member and all uses.
* ov-usr-fcn.h, ov-usr-fcn.cc (octave_user_function::m_jit_info):
Delete data member and all uses.
(octave_user_function::get_info, octave_user_function::stash_info): Delete.
* options.h (DEBUG_JIT_OPTION, JIT_COMPILER_OPTION): Delete macro
definitions and all uses.
* octave.h, octave.cc (cmdline_options::cmdline_options): Don't handle
DEBUG_JIT_OPTION, JIT_COMPILER_OPTION): Delete.
(cmdline_options::debug_jit, cmdline_options::jit_compiler): Delete
functions and all uses.
(cmdline_options::m_debug_jit, cmdline_options::m_jit_compiler): Delete
data members and all uses.
(octave_getopt_options long_opts): Remove "debug-jit" and
"jit-compiler" from the list.
* pt-eval.cc (tree_evaluator::visit_simple_for_command,
tree_evaluator::visit_complex_for_command,
tree_evaluator::visit_while_command,
tree_evaluator::execute_user_function): Eliminate JIT compiler code.
* pt-loop.h, pt-loop.cc (tree_while_command::get_info,
tree_while_command::stash_info, tree_simple_for_command::get_info,
tree_simple_for_command::stash_info): Delete functions and all uses.
(tree_while_command::m_compiled, tree_simple_for_command::m_compiled):
Delete member variable and all uses.
* usage.h (usage_string, octave_print_verbose_usage_and_exit): Remove
[--debug-jit] and [--jit-compiler] from the message.
* Array.h (Array<T>::Array): Remove constructor that was only intended
to be used by the JIT compiler.
(Array<T>::jit_ref_count, Array<T>::jit_slice_data,
Array<T>::jit_dimensions, Array<T>::jit_array_rep): Delete.
* Marray.h (MArray<T>::MArray): Remove constructor that was only
intended to be used by the JIT compiler.
* NDArray.h (NDArray::NDarray): Remove constructor that was only
intended to be used by the JIT compiler.
* dim-vector.h (dim_vector::to_jit): Delete.
(dim_vector::dim_vector): Remove constructor that was only intended to
be used by the JIT compiler.
* codeql-analysis.yaml, make.yaml: Don't require llvm-dev.
* subst-config-vals.in.sh, subst-cross-config-vals.in.sh: Don't
substitute OCTAVE_CONF_LLVM_CPPFLAGS, OCTAVE_CONF_LLVM_LDFLAGS, or
OCTAVE_CONF_LLVM_LIBS.
* Doxyfile.in: Don't define HAVE_LLVM.
* aspell-octave.en.pws: Eliminate jit, JIT, and LLVM from the list of
spelling exceptions.
* build-env.h, build-env.in.cc (LLVM_CPPFLAGS, LLVM_LDFLAGS,
LLVM_LIBS): Delete variables and all uses.
* libinterp/corefcn/module.mk (%canon_reldir%_libcorefcn_la_CPPFLAGS):
Remove $(LLVM_CPPFLAGS) from the list.
* libinterp/parse-tree/module.mk (%canon_reldir%_libparse_tree_la_CPPFLAGS):
Remove $(LLVM_CPPFLAGS) from the list.
author | John W. Eaton <jwe@octave.org> |
---|---|
date | Tue, 10 Aug 2021 16:42:29 -0400 |
parents | dbc191149841 |
children | 796f54d4ddbf |
line wrap: on
line source
######################################################################## ## ## Copyright (C) 1993-2021 The Octave Project Developers ## ## See the file COPYRIGHT.md in the top-level directory of this ## distribution or <https://octave.org/copyright/>. ## ## 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 ## <https://www.gnu.org/licenses/>. ## ######################################################################## ## -*- texinfo -*- ## @deftypefn {} {} num2str (@var{x}) ## @deftypefnx {} {} num2str (@var{x}, @var{precision}) ## @deftypefnx {} {} num2str (@var{x}, @var{format}) ## Convert a number (or array) to a string (or a character array). ## ## The optional second argument may either give the number of significant ## digits (@var{precision}) to be used in the output or a format template ## string (@var{format}) as in @code{sprintf} (@pxref{Formatted Output}). ## @code{num2str} can also process complex numbers. ## ## Examples: ## ## @example ## num2str (123.456) ## @result{} 123.456 ## ## num2str (123.456, 4) ## @result{} 123.5 ## ## s = num2str ([1, 1.34; 3, 3.56], "%5.1f") ## @result{} s = ## 1.0 1.3 ## 3.0 3.6 ## whos s ## @result{} Variables in the current scope: ## Attr Name Size Bytes Class ## ==== ==== ==== ===== ===== ## s 2x8 16 char ## Total is 16 elements using 16 bytes ## ## num2str (1.234 + 27.3i) ## @result{} 1.234+27.3i ## @end example ## ## The @code{num2str} function is not very flexible. For better control ## over the results, use @code{sprintf} (@pxref{Formatted Output}). ## ## Programming Notes: ## ## For @sc{matlab} compatibility, leading spaces are stripped before returning ## the string. ## ## Integers larger than @code{flintmax} may not be displayed correctly. ## ## For complex @var{x}, the format string may only contain one output ## conversion specification and nothing else. Otherwise, results will be ## unpredictable. ## ## Any optional @var{format} specified by the programmer is used without ## modification. This is in contrast to @sc{matlab} which tampers with the ## @var{format} based on internal heuristics. ## @seealso{sprintf, int2str, mat2str} ## @end deftypefn function retval = num2str (x, arg) if (nargin < 1) print_usage (); elseif (! (isnumeric (x) || islogical (x) || ischar (x))) error ("num2str: X must be a numeric, logical, or character array"); endif if (ischar (x)) retval = x; elseif (isempty (x)) retval = ""; elseif (isreal (x)) if (nargin == 2) if (ischar (arg)) fmt = arg; elseif (isnumeric (arg) && isscalar (arg) && arg >= 0 && arg == fix (arg)) if (isfloat (x)) fmt = sprintf ("%%%d.%dg", arg+7, arg); else fmt = sprintf ("%%%dd", arg); endif else error ("num2str: PRECISION must be a scalar integer >= 0"); endif else if (isnumeric (x)) ## Set up a suitable format string while ignoring Inf/NaN entries valid = isfinite (x(:)); ndgt = floor (log10 (max (abs (x(valid))))); if (isempty (ndgt) || ndgt == -Inf) ndgt = 0; # All Inf or all zero array endif if (ndgt > 15 || any (x(valid) != fix (x(valid)))) ## Floating point input ndgt = max (ndgt + 5, 5); # Keep at least 5 significant digits ndgt = min (ndgt, 16); # Cap significant digits at 16 fmt = sprintf ("%%%d.%dg", ndgt+7, ndgt); else ## Integer input ndgt += 3; if (any (! valid)) ndgt = max (ndgt, 5); # Allow space for Inf/NaN endif if (isfloat (x)) fmt = sprintf ("%%%d.0f", ndgt); else fmt = sprintf ("%%%dd", ndgt); endif endif else ## Logical input fmt = "%3d"; endif endif fmt = do_string_escapes (fmt); # required now that '\n' is interpreted. nd = ndims (x); nc = columns (x) * (nd - 1); # ND-arrays are expanded in columns x = permute (x, [2, 3:nd, 1]); if (! (sum (strrep (fmt, "%%", "") == "%") > 1 || any (strcmp (fmt, {"%s", "%c"})))) fmt = [deblank(repmat (fmt, 1, nc)), "\n"]; endif strtmp = sprintf (fmt, x); retval = strtrim (char (ostrsplit (strtmp, "\n", true))); else # Complex matrix input if (nargin == 2) if (ischar (arg)) fmt = [deblank(arg) "%-+" arg(2:end) "i"]; elseif (isnumeric (arg) && isscalar (arg) && arg >= 0 && arg == fix (arg)) fmt = sprintf ("%%%d.%dg%%-+%d.%dgi", arg+7, arg, arg+7, arg); else error ("num2str: PRECISION must be a scalar integer >= 0"); endif else ## Set up a suitable format string while ignoring Inf/NaN entries valid_real = isfinite (real (x(:))); valid_imag = isfinite (imag (x(:))); ndgt = floor (log10 (max (max (abs (real (x(valid_real)))), max (abs (imag (x(valid_imag))))))); if (isempty (ndgt) || ndgt == -Inf) ndgt = 0; # All Inf or all zero array endif if (any (x(valid_real & valid_imag) != fix (x(valid_real & valid_imag)))) ## Floating point input ndgt = max (ndgt + 5, 5); # Keep at least 5 significant digits ndgt = min (ndgt, 16); # Cap significant digits at 16 fmt = sprintf ("%%%d.%dg%%-+%d.%dgi", ndgt+7, ndgt, ndgt+7, ndgt); else ## Integer input ndgt += 3; ## FIXME: Integers must be masked to show only 16 significant digits ## See test case for bug #36133 below fmt = sprintf ("%%%d.0f%%-+%d.0fi", ndgt, ndgt); endif endif ## Manipulate the complex value to have real values in the odd ## columns and imaginary values in the even columns. nd = ndims (x); nc = columns (x); idx = repmat ({':'}, nd, 1); perm(1:2:2*nc) = 1:nc; perm(2:2:2*nc) = nc + (1:nc); idx{2} = perm; x = horzcat (real (x), imag (x)); x = x(idx{:}); fmt = [deblank(repmat(fmt, 1, nc * (nd - 1))), "\n"]; tmp = sprintf (fmt, permute (x, [2, 3:nd, 1])); ## Put the "i"'s where they are supposed to be. tmp = regexprep (tmp, " +i\n", "i\n"); tmp = regexprep (tmp, "( +)i", "i$1"); retval = strtrim (char (ostrsplit (tmp(1:end-1), "\n"))); endif endfunction ## Basic tests %!assert (num2str (123), "123") %!assert (num2str (1.23), "1.23") %!assert (num2str (123.456, 4), "123.5") %!assert (num2str ([1, 1.34; 3, 3.56], "%5.1f"), ["1.0 1.3"; "3.0 3.6"]) %!assert (num2str (1.234 + 27.3i), "1.234+27.3i") %!assert (num2str ([true false true]), "1 0 1") ## Exceptional values %!assert (num2str (19440606), "19440606") %!assert (num2str (2^33), "8589934592") %!assert (num2str (-2^33), "-8589934592") %!assert (num2str (2^33+1i), "8589934592+1i") %!assert (num2str (-2^33+1i), "-8589934592+1i") %!assert (num2str ([0 0 0]), "0 0 0") %!assert (num2str (inf), "Inf") %!assert (num2str ([inf -inf]), "Inf -Inf") %!assert (num2str ([inf NaN -inf]), "Inf NaN -Inf") %!assert (num2str ([complex(Inf,0), complex(0,-Inf)]), "Inf+0i 0-Infi") %!assert (num2str (complex (Inf,1)), "Inf+1i") %!assert (num2str (complex (1,Inf)), "1+Infi") %!assert (num2str (nan), "NaN") %!assert (num2str (complex (NaN, 1)), "NaN+1i") %!assert (num2str (complex (1, NaN)), "1+NaNi") %!assert (num2str (NA), "NA") %!assert (num2str (complex (NA, 1)), "NA+1i") %!assert (num2str (complex (1, NA)), "1+NAi") %!assert (num2str (int64 (-flintmax ()) - 1), "-9007199254740993") %!assert (num2str (int64 (-flintmax ()) - 1, 18), "-9007199254740993") ## ND-arrays are concatenated in columns %!shared m, x %! m = magic (3); %! x = cat (3, m, -m); ## real case %!test <*46770> %! y = num2str (x); %! assert (rows (y) == 3); %! assert (y, ["8 1 6 -8 -1 -6" %! "3 5 7 -3 -5 -7" %! "4 9 2 -4 -9 -2"]); ## complex case %!test <*46770> %! x(1,1,2) = -8+2i; %! y = num2str (x); %! assert (rows (y) == 3); %! assert (y, ["8+0i 1+0i 6+0i -8+2i -1+0i -6+0i" %! "3+0i 5+0i 7+0i -3+0i -5+0i -7+0i" %! "4+0i 9+0i 2+0i -4+0i -9+0i -2+0i"]); ## Clear shared variables %!shared ## Integers greater than 1e15 should switch to exponential notation %!assert <*36133> (num2str (1e15), "1000000000000000") %!assert <*36133> (num2str (1e16), "1e+16") ## Even exact integers in IEEE notation should use exponential notation %!assert <*36133> (num2str (2^512), "1.34078079299426e+154") ## Mixed integer/floating point arrays %!assert <*36133> (num2str ([2.1, 1e23, pi]), %! "2.1 9.999999999999999e+22 3.141592653589793") ## Large integers should not switch sign when printed due to overflow %!assert <*36121> (num2str (2.4e9, 15), "2400000000") ## Test for extra rows generated from newlines in format %!assert <*44864> (rows (num2str (magic (3), "%3d %3d %3d\n")), 3) ## Test that string conversion of numeric objects results in characters ## if the numbers are within range for ASCII. %!assert <*45174> (num2str ([65 66 67], "%s"), "ABC") ## Test input validation %!error <Invalid call> num2str () %!error <X must be a numeric> num2str ({1}) %!error <PRECISION must be a scalar integer .= 0> num2str (1, {1}) %!error <PRECISION must be a scalar integer .= 0> num2str (1, ones (2)) %!error <PRECISION must be a scalar integer .= 0> num2str (1, -1) %!error <PRECISION must be a scalar integer .= 0> num2str (1, 1.5) %!error <PRECISION must be a scalar integer .= 0> num2str (1+1i, {1}) %!error <PRECISION must be a scalar integer .= 0> num2str (1+1i, ones (2)) %!error <PRECISION must be a scalar integer .= 0> num2str (1+1i, -1) %!error <PRECISION must be a scalar integer .= 0> num2str (1+1i, 1.5)