1 Changes between version 5.4.X and 5.5.0
2 =======================================
7 * New special functions:
8 - erfi - The imaginary error function.
9 - dawson - Compute the Dawson (scaled imaginary error) function.
11 * New functions introduced:
12 - getURL - Download an URL (HTTP, HTTPS, FTP...)
13 - splitURL - Split a URL (HTTP, HTTPS, FTP...)
14 - cov - Covariance matrix. Deprecates mvvacov. See bug #11896 fixed.
15 - ismatrix - Check if a variable is a matrix. See bug #10456 fixed.
16 - isrow - Check if a variable is a row vector. See bug #10456 fixed.
17 - iscolumn - Check if a variable is a column vector. See bug #10456 fixed.
18 - issquare - Check if a variable is a square matrix. See bug #10456 fixed.
19 - cross - Vector cross product. See bug #9941 fixed.
21 * Add a complete set of functions to read and write any HDF5 file from Scilab.
24 - daskr - differential-algebraic system solver with rootfinding 'daskr', using
25 BDF methods with direct and preconditioned Krylov linear solvers, based on ODEPACK.
27 * Based on JIMS external module, Scilab provides functions to interact
30 * erf, erfc, erfcx and calerf functions now support complex arguments.
32 * isnum has been redesigned in native code. Performance improvements up to 130x.
35 * Allow the usage of the '$' keyword in part function.
37 * The histplot command can now be used with the option polygon=%t/%f to add the
38 frequency polygon chart (thanks to Mehran Khorshidi).
40 * Multi level completion on mlist, struct, XML structures...
42 * Variable browser improvements:
43 - The variable browser also shows the size of integers and the user type of
45 See bugs #12523 and #10409
46 - It is now possible to delete variables from the variable browser.
48 - A user can now plot variables from the variable browser (this functionality was
49 already available in editvar).
51 * Added lighting effect for plot of surfaces. Lighting can be enabled
52 creating light objects or disabled by deleting them. The following function
54 - light - Creates a light graphic object
57 - Multiple domains in localization managed.
58 - Function addlocalizationdomain added to manage a new domain.
59 - Optional parameter added to gettext to manage domains.
60 - tbx_generate_pofile and tbx_build_localization added to create localization files for
63 * Windows Solution updated to Visual Studio 2012.
65 * -keepconsole option added for Scilab Windows to facilitate debugging.
66 Calling Scilab with this optin will leave the console box window opened at startup.
68 * License update: switch to the CeCILL 2.1.
74 * New calling sequence allowed for nicholschart: nicholschart(gains, phases, colors).
77 * qp_solve can now take up to 5 output arguments. The last one is an error flag,
78 if it is present, then the function will display a warning instead of an error.
81 * graypolarplot has been improved in terms of performances and rendering.
84 * nthroot is now vectorizable.
87 * New optional output argument for routh_t.
91 Differential Equations
92 ======================
94 * Netlib's Quadpack, used for definite integration, has been updated to match the upstream.
100 * Autosave feature can now use %date to append the current date to the backup filename
107 * New DAE solver: DDaskr, using BDF methods with direct Newton and preconditioned Krylov
108 linear solvers, which includes rootfinding.
109 It is available from dae function.
111 * In Modelica initialization GUI, inputs (eg. sensor) were not handled.
113 * Sundials updated to the "2.5.0" version, keeping our modifications.
115 * JGraphX updated to the "2.1.0.7" version, updating our performance hooks.
117 * API changed in the scicos_block4 interface: the uid value is not available
118 inside the simulation function.
120 * xcosAddToolsMenu added to let the user perform some actions on the graph.
122 * CBLOCK4 block added to the "User-Defined Functions"
128 * Some Xcos specific content added to xcos_toolbox_skeleton help pages.
134 * Minimal version of Flexdock (1.2.4) enforced.
137 Obsolete & Removed Functions
138 ============================
140 * Scilab 5.5.X family will be the latest family working under Windows XP/2003.
142 * Vector ^ scalar syntax declared obsolete. See bug #11524
144 * The third argument of poly function will be more strict.
145 Only the following strings are allowed: "roots", "coeff", "c", "r".
147 * Option and direction arguments of gsort will be more strict in all cases.
148 Only the following strings are allowed: option: "r" "c" "g" "lr" "lc"
151 * nfreq tagged as obsolete. Will be removed in Scilab 5.5.1.
152 Please use tabul instead.
154 * IsAScalar tagged as obsolete. Will be removed in Scilab 5.5.1.
155 Please use isscalar instead.
157 * chart tagged as obsolete. Will be removed in Scilab 5.5.1.
158 Please use nicholschart instead.
160 * mvvacov tagged as obsolete. Will be removed in Scilab 6.0.0.
161 Please use cov instead.
163 * dft removed, please use fft instead.
165 * sscanf removed, please use msscanf instead.
167 * fscanf removed, please use mfscanf for files opened with mopen or read for files opened
170 * fprintf removed, please use mfprintf for files opened with mopen or write for files
171 opened with file instead.
177 * paramfplot2d: When theta input argument was a column vector, an error occurred.
179 * Bug #2267 fixed - Wrong error message when ticks locations and labels did not have the
182 * Bug #2416 fixed - Particular case (string([]) returns []) has been added in string help
185 * Bug #3511 fixed - strindex did not return all occurrences in regexp mode.
187 * Bug #3928 fixed - An error was returned when a matrix was flipped along the third
190 * Bug #4042 fixed - squeeze returned a hypermatrix instead of a matrix when at least one
191 dimension was equal to 1.
193 * Bug #4085 fixed - num2cell help page added.
195 * Bug #4229 fixed - delip did not return an error if one element of its first input
196 argument was negative.
198 * Bug #4383 fixed - csim with "step" and "impuls" now works with direct feedthrough.
200 * Bug #4481 fixed - iir help page updated to precise that frq can be a scalar.
202 * Bug #4731 fixed - lqr failed when the time domain of an input was a number.
204 * Bug #5073 fixed - New parameter added in strtod function (decimal separator).
206 * Bug #5205 fixed - permute was slow for large hypermatrices.
208 * Bug #5207 fixed - grand can now return a hypermatrix.
210 * Bug #5365 fixed - makecell help page was in the "compatibility functions" directory
211 instead of being in "data_structures".
213 * Bug #5539 fixed - sylv help page was wrong in discrete-time cases.
215 * Bug #5616 fixed - typeof(uiCreateTree(...)) returned "Tree" instead of "uitree".
217 * Bug #5694 fixed - numdiff help page clarified.
219 * Bug #5779 fixed - xnumb number format was too small (+ unit test added).
221 * Bug #6037 fixed - macrovar help page improved.
223 * Bug #6168 fixed - zpbutt, zpch1, zpch2 and zpell help pages were unclear.
225 * Bug #6390 fixed - The "external" argument of odecd was not well documented and
228 * Bug #6427 fixed - full([%T %F]) returned an error message.
230 * Bug #6466 fixed - Example with vectorized input added in mprintf and msprintf help pages.
232 * Bug #6638 fixed - The profiler output was incorrect by a factor of 1000 under Windows.
234 * Bug #6693 fixed - modulo did not accept polynomial inputs. Help page was not updated.
236 * Bug #6752 fixed - unit test scilab.tst has been split in different tests.
238 * Bug #6840 fixed - New line_style added.
240 * Bug #6988 fixed - Error messages in modules/data_structures/src/c/hmops.c were not
243 * Bug #7026 fixed - There was no unit test for plot2d.
245 * Bug #7080 fixed - Some graphic macros did not use standard error messages.
247 * Bug #7204 fixed - geomean applied to a hypermatrix gave wrong results.
249 * Bug #7206 fixed - If the second input argument of meanf function was a hypermatrix,
250 this function returned an error.
252 * Bug #7296 fixed - %nan, %inf and -%inf enabled for the cdf* functions.
254 * Bug #7411 fixed - clf forgot to turn off the datatip mode.
256 * Bug #7486 fixed - LAPACK versions of DGELSY and ZGELSY now used.
258 * Bug #7596 fixed - A same error happening in different places now displays the same
261 * Bug #7648 fixed - CDF functions now display a warning for non integer
262 "degrees of freedom" argument.
264 * Bug #7655 fixed - An example added in type help page, for type(X)=11 and type(X)=13.
266 * Bug #7684 fixed - Introduction demonstration splitted into subsections.
268 * Bug #7781 fixed - The second parameter of iqr function had not effect.
270 * Bug #7782 fixed - lcm and gcd help pages improved to tell the user how to use both
273 * Bug #7824 fixed - title function properties did not support an indexed color.
275 * Bug #7826 fixed - chart tagged as obsolete.
277 * Bug #7828 fixed - Slight improvements in nicholschart.
279 * Bug #7848 fixed - The third argument of correl function is now optional.
281 * Bug #7877 fixed - iirgroup function fixed.
283 * Bug #7927 fixed - Output "flag" in qmr function was not well documented.
285 * Bug #7960 fixed - plzr could not produce pole zero plot for a simple transfer function.
287 * Bug #8058 fixed - The user can now set the tolerances of intc function.
289 * Bug #8098 fixed - cumsum could not be applied to rational matrices.
291 * Bug #8162 fixed - Area of stability of plzr was wrong for continuous systems
294 * Bug #8211 fixed - 'parameters' module demonstration finalized.
296 * Bug #8234 fixed - strtod did not return an empty matrix when the input argument was an
299 * Bug #8264 fixed - Matlab to Scilab dictionary help page updated for atan2.
301 * Bug #8319 fixed - dbphi(hypermat) and phasemag(hypermat) returned a matrix instead of
304 * Bug #8373 fixed - clear can now handle a matrix of strings argument.
306 * Bug #8415 fixed - optim_moga, optim_nsga, optim_nsga2 can now take list as input
307 arguments, as explained in their help pages.
309 * Bug #8462 fixed - bvodeS could make Scilab unstable.
311 * Bug #8470 fixed - bvode displayed some output in terminal window and not in
314 * Bug #8479 fixed - The latest Saxon version was not supported.
316 * Bug #8511 fixed - sprand now uses grand instead of rand and grand functions. Internal
317 state of the random generator is no more changed.
319 * Bug #8561 fixed - ddassl, ddasrt, ddaskr: abs and rel tolerance sizes are now checked.
321 * Bug #8597 fixed - grand/clcg4 could display uncontrolled messages as warning.
323 * Bug #8607 fixed - Some error messages in modules/overloading/macros were not standard
326 * Bug #8614 fixed - Unit test for barhomogenize added.
328 * Bug #8636 fixed - roots help page updated (default algorithm value was wrong).
330 * Bug #8667 fixed - Handling of %nan in min, max, median functions was not properly
333 * Bug #8680 fixed - "end" output argument of regexp function has been changed.
335 * Bug #8687 fixed - typeof function failed on uint8, depending on the format
338 * Bug #8695 fixed - optim_ga used old (initial) values instead of newly-computed ones.
340 * Bug #8778 fixed - Call_ScilabOpen, TerminateScilab could not be called more
341 than 80 times in a loop.
343 * Bug #8779 fixed - gsort did not preserve the order of equal elements, in
346 * Bug #8820 fixed - squeeze did not return a matrix when the number of dimensions
347 of the result was less or equal to 2.
349 * Bug #8824 fixed - taucs_chfact returned a segfault (not the case in nwni mode).
351 * Bug #8840 fixed - fileparts did not manage matrix of strings.
353 * Bug #8856 fixed - Non regression test added for [k,l,m,...]=find(a==5).
354 Non regression test of bug #476 updated.
356 * Bug #8857 fixed - Non regression test of bug #477 updated.
358 * Bug #8858 fixed - Non regression test of bug #480 updated.
360 * Bug #8862 fixed - mget and mput could not read and write 64 bit data from
363 * Bug #9004 fixed - bitcmp function called with one input argument returned an error.
365 * Bug #9020 fixed - exists function did not accept matrix as first input argument.
367 * Bug #9059 fixed - tbx_build_macros and genlib did not stop even if an error occurred.
369 * Bug #9109 fixed - nfreq tagged as obsolete.
371 * Bug #9158 fixed - zeros called with a big number returns an empty matrix.
373 * Bug #9208 fixed - Added three optional output arguments to optim,
374 to retrieve #iterations, #evaluations and a termination indicator.
376 * Bug #9385 fixed - The type checking in trigonometric functions has been added.
378 * Bug #9394 fixed - is_params recognized "plist" as an existing field.
380 * Bug #9395 fixed - add_params did not check its input arguments.
382 * Bug #9396 fixed - add_params accepted duplicate keys.
384 * Bug #9459 fixed - Default values of the optional plot3d arguments were not documented.
386 * Bug #9537 fixed - optimbase_configure only allowed row vectors as initial value.
387 Column vectors now allowed by transposing them.
389 * Bug #9538 fixed - optimbase_checkshape only allowed row vectors as output arguments of
390 cost function. Column vectors are now allowed by transposing them.
392 * Bug #9577 fixed - Setting neldermead_configure("-numberofvariables") is now optional,
393 setting neldermead_configure("-x0") initializes -numberofvariables
396 * Bug #9601 fixed - Cylinder demonstration fixed.
398 * Bug #9688 fixed - optim could crash when "imp" option was < 0. It is now set to 0 in
401 * Bug #9690 fixed - The "imp" option for optim could crash Scilab and was not consistent
404 * Bug #9691 fixed - "imp" option in optim help page was poorly documented.
406 * Bug #9694 fixed - Example in optim help page fixed to display correct counters.
408 * Bug #9702 fixed - Contrary to what optim help page stated, the "gc" algorithm does use
409 the "epsx" parameter.
411 * Bug #9780 fixed - gmres solver did not run with complex systems.
413 * Bug #9821 fixed - getrelativefilename did not manage matrix of strings.
415 * Bug #9851 fixed - Error message occurred because of a cut-off frequency of 0.25Hz
418 * Bug #9859 fixed - It was not possible to draw arrows in 3-D using xarrows.
420 * Bug #10146 fixed - In SciNotes, 'help on keyword' moved from bottom to top in the popup
423 * Bug #10180 fixed - det was not defined for sparse matrices.
425 * Bug #10213 fixed - sci2exp help page updated to document the impact of format function.
427 * Bug #10216 fixed - Invalid syntaxes for zeros, ones, eye, rand, like zeros(2,:).
429 * Bug #10226 fixed - When a // <empty session> line was deleted, all sessions
430 histories were folded.
432 * Bug #10254 fixed - Slight improvements in ones help page.
434 * Bug #10269 fixed - qp_solve can now take up to 5 output arguments. The last one is an
435 error flag, if it is present, then the function will issue a warning
438 * Bug #10276 fixed - qp_solve segfaulted with large matrices.
440 * Bug #10287 fixed - Error message added for complex expression as input argument of
443 * Bug #10305 fixed - Comparison of lists with empty items returned an error message.
445 * Bug #10596 fixed - exit(xxx) from Scilab failed.
447 * Bug #10818 fixed - hom, %e, %t, %f, %z and %s help pages added.
449 * Bug #10823 fixed - fullpath returned different results under Windows and Linux for
452 * Bug #10862 fixed - Add a without Internet connection installation
453 global configuration in the installer.
455 * Bug #10866 fixed - det was not equivalent to detr for rational matrices.
457 * Bug #10906 fixed - Typo fixed cls2dls help page.
459 * Bug #10930 fixed - The comments in armax function were in French.
461 * Bug #10995 fixed - Typo fixed in grand help page for Gamma law argument.
463 * Bug #11065 fixed - The second output argument of unique function contained a wrong result.
465 * Bug #11139 fixed - conj was not defined for sparse matrices (+ unit test added).
467 * Bug #11308 fixed - Calling sequences in dsearch help page were wrong.
469 * Bug #11343 fixed - The "isoview" figure property did not work when the axes
470 margins had been modified.
472 * Bug #11648 fixed - Copying graphic via the clipboard did not work.
474 * Bug #11766 fixed - nthroot has been added to m2sci help page.
476 * Bug #11792 fixed - Lists can be accessed with non integer indexes (list help page
479 * Bug #11869 fixed - "Environment" was not localized in preferences.
481 * Bug #11885 fixed - Each rand has been changed to grand in genetic algorithms and
482 simulated annealing functions.
484 * Bug #11891 fixed - Fisher ratio could be inaccurate for one-way ANOVA.
486 * Bug #11997 fixed - In case of invalid variable name, save function now returns an error
487 instead of a compatibility warning.
489 * Bug #12034 fixed - max function did not manage empty matrices.
491 * Bug #12045 fixed - repmat returned wrong results if the values of an input matrix were
494 * Bug #12070 fixed - Removing a module can now be done in an on-line mode even if
495 it has been installed in an off-line mode.
497 * Bug #12080 fixed - lsqrsolve always printed messages.
499 * Bug #12085 fixed - Under Windows, csvWrite wrote wrong EOL.
501 * Bug #12114 fixed - libstdc++ is now compiled as static instead of using the
502 library from thirdparty.
504 * Bug #12118 fixed - ode could take complex externals.
506 * Bug #12143 fixed - "stop entity picker" (ged(11)) returned an error message.
508 * Bug #12150 fixed - datatipCreate failed with one single point.
510 * Bug #12163 fixed - unzoom did not work with a single input argument.
512 * Bug #12212 fixed - Export of a polyline in 2-D broke it into several segments.
514 * Bug #12326 fixed - There was no way to set LaTeX font size in preview.
516 * Bug #12349 fixed - In SciNotes, when the view was splitted, removing a char made the
519 * Bug #12376 fixed - Exec & edit buttons in the help pages examples were sometimes
522 * Bug #12413 fixed - ATOMS packages could not be removed if ATOMS had never been on-line.
524 * Bug #12415 fixed - PATH environment variable grew when using call_scilab in a loop.
526 * Bug #12426 fixed - By using addErrorMessage, a random string error could be displayed.
528 * Bug #12433 fixed - show_pixmap function was removed from Scilab but still used in some
531 * Bug #12443 fixed - The behavior of mopen in text file mode has been documented
534 * Bug #12463 fixed - Wrong specification for rect=[x,y,w,h] argument in xstringb
537 * Bug #12470 fixed - Variable browser was not refreshed after loading an environment.
539 * Bug #12473 fixed - Problems with "é" in mkdir French help page.
541 * Bug #12475 fixed - csvRead did not support double quoted fields.
543 * Bug #12490 fixed - plot did not call clf in its help page example producing a
546 * Bug #12506 fixed - In SciNotes, a miscolorization could occurred when returned values
549 * Bug #12508 fixed - Wrong error message in rand function fixed.
551 * Bug #12518 fixed - Polynoms were not displayed by default in the variable browser.
553 * Bug #12520 fixed - Improve the description of the size in the variable browser.
555 * Bug #12527 fixed - Scilab user functions were not listed in the variable browser.
557 * Bug #12547 fixed - In SciNotes, lines were wrongly numbered with splitted function
560 * Bug #12548 fixed - Duplicated code in xmltoformat removed.
562 * Bug #12550 fixed - optimbase and optimsimplex help pages were not standard.
564 * Bug #12551 fixed - Stack problem with diary([],"pause") and diary([],"resume") fixed.
566 * Bug #12556 fixed - A fatal error occurred when calling set with wrong instructions.
568 * Bug #12557 fixed - Valid function names were not specified in function and functions
571 * Bug #12564 fixed - Compile and run javasci help page was not clear about needed packages.
573 * Bug #12581 fixed - isfield could not support mlist or tlist.
575 * Bug #12589 fixed - Call sequence for spzeros & spones were missing in French
578 * Bug #12592 fixed - Scilab hung with plot(-0).
580 * Bug #12593 fixed - A wrong error message was returned when running genlib with an
581 error in the sci file.
583 * Bug #12600 fixed - mput did not manage unsigned integer.
585 * Bug #12606 fixed - Overloads for grand were not standard.
587 * Bug #12613 fixed - gsort did not return correct results with %nan.
589 * Bug #12614 fixed - Helpbrowser was not launched in EDT.
591 * Bug #12615 fixed - Graphics seemed to be freezed after a call to plot/bar/barh with
592 a bad LineSpec argument.
594 * Bug #12624 fixed - In case of errors in Scilab macros, "make check-TESTS" did not fail
597 * Bug #12627 fixed - At restoration, a window could be out of the screen.
599 * Bug #12629 fixed - The last example of csim help page defined a function called
600 input (overwriting the Scilab one).
602 * Bug #12631 fixed - A "see also" link has been added from progressionbar to waitbar
605 * Bug #12634 fixed - ATOMS modules could not be installed from an archive file
606 already existing in SCI/contrib/archives in offline mode.
608 * Bug #12637 fixed - In some help pages, some signal processing functions were not in
611 * Bug #12639 fixed - justify([], position) returned an error instead of [].
613 * Bug #12641 fixed - graypolarplot has been fully vectorized as it was too slow.
615 * Bug #12657 fixed - Computation of v1.^v2 is now done without any memory allocation,
616 when v1 and v2 are real arrays, v1 >= 0 and v2 integer.
618 * Bug #12668 fixed - Undocking SciNotes led to an exception.
620 * Bug #12678 fixed - nthroot now accepts vector/matrix as second argument.
622 * Bug #12679 fixed - Argument type check added in gcd and lcm.
624 * Bug #12686 fixed - Error returned by diff fixed.
626 * Bug #12702 fixed - When no extra parameters were needed in the cost function,
629 * Bug #12703 fixed - In SciNotes, common shortcuts 'SHIFT DELETE', 'SHIFT INSERT'
632 * Bug #12705 fixed - members function added. It allows to find the number of occurrences
633 and linear indexes for common values between two matrices of the
636 * Bug #12706 fixed - A wrong size of a matrix as input argument of cross function was
639 * Bug #12708 fixed - Incorrect display in SciNotes preferences fixed (onmouseover styles).
641 * Bug #12712 fixed - In SciNotes, autosaving can now use %date to append the current
642 date to the backup filename.
644 * Bug #12715 fixed - Variable cross in pspect and cspect has been renamed because of
645 conflicts with cross function.
647 * Bug #12736 fixed - In SciNotes, the Completion window appeared only in first tab.
649 * Bug #12747 fixed - legendre now accepts the -1 and 1 values for third argument.
651 * Bug #12749 fixed - fscanfMat help page updated.
653 * Bug #12758 fixed - Focus issue with plot3d fixed.
655 * Bug #12761 fixed - The https:// protocol was not supported on ATOMS.
657 * Bug #12772 fixed - eigs failed when trying to solve a sparse matrix eigen value problem.
659 * Bug #12774 fixed - Various typos fixed.
661 * Bug #12775 fixed - Some related functions were not listed in "See also" section of
664 * Bug #12778 fixed - Insertion of an empty matrix in an integer matrix led to a
667 * Bug #12779 fixed - savewave had a miscoding in the internal function write_wavedat.
669 * Bug #12783 fixed - There were some inconsistent error messages in dsearch.
671 * Bug #12790 fixed - Links to ZCOS files in documentation were broken.
673 * Bug #12791 fixed - More information is now given in case of failure during the ATOMS
676 * Bug #12794 fixed - calfrq.sci code did not follow Scilab standard.
678 * Bug #12795 fixed - Typos fixed in CACSD help page.
680 * Bug #12800 fixed - Typo fixed in Polynomials help page.
682 * Bug #12804 fixed - Typos fixed in routh_t help page.
684 * Bug #12807 fixed - Display of showprofile improved.
686 * Bug #12808 fixed - Add missing </td> in documentation generation (note, warning, ...).
688 * Bug #12813 fixed - flipdim function extended to any type of input data.
690 * Bug #12814 fixed - Improvements of pertrans help page.
692 * Bug #12815 fixed - levin redefined cov as a variable.
694 * Bug #12816 fixed - Numbers pasted in editvar were not parsed according to locale.
696 * Bug #12818 fixed - Segfault in set function with invalid property values dimension.
698 * Bug #12823 fixed - In help generation (toolbox) links were not correctly handled.
700 * Bug #12827 fixed - noisgen help page improved.
702 * Bug #12828 fixed - routh_t gave a wrong result if the first element of a row was zero.
704 * Bug #12829 fixed - New optional output argument added for routh_t function.
706 * Bug #12830 fixed - In SciNotes, it was not possible to execute a replace action
707 from the caret position.
709 * Bug #12831 fixed - In SciNotes toolbar, there was no button to open code navigator.
711 * Bug #12833 fixed - In SciNotes, there was no autoscroll when searching a pattern.
713 * Bug #12836 fixed - Error fixed in strcmpi help page.
715 * Bug #12839 fixed - Typo fixed in getVariablesOnStack help page.
717 * Bug #12840 fixed - Typo fixed in number_properties.xml help page.
719 * Bug #12852 fixed - Visual Studio 2012 SDK configuration was incorrect.
721 * Bug #12858 fixed - Typo fixed in debug and pause French help pages.
723 * Bug #12859 fixed - Fixed optional argument in isdef help page.
725 * Bug #12863 fixed - size(state-space, "r") returned an error.
727 * Bug #12875 fixed - phasemag returned an error for input vector containing zeros.
729 * Bug #12887 fixed - Scilab hung with auto_clear set to on and log_flags set to true.
731 * Bug #12888 fixed - sysdiag was not documented about block diagonal matrices build.
733 * Bug #12906 fixed - champ and champ1 help pages updated
735 * Bug #12909 fixed - Completion on (mt)list led to a crash.
737 * Bug #12911 fixed - Matlab to Scilab dictionary help page updated for eig.
739 * Bug #12913 fixed - linspace returned an error if the third argument was an integer type
742 * Bug #12916 fixed - power help page improved.
744 * Bug #12919 fixed - Rotation on plots was disabled after using menus.
746 * Bug #12927 fixed - ones function could not take integer type input.
748 * Bug #12931 fixed - ATOMS redefined "message" keyword.
754 * Bug #7350 fixed - The I/O ports numbering of a superblock was not updated
755 when a new port was dropped.
757 * Bug #8570 fixed - "Region to superblock" has been renamed to
758 "Selection to superblock".
760 * Bug #11518 fixed - CLR block-text was displayed out of the bounding box
763 * Bug #11776 fixed - CMSCOPE did not take into account label&Id parameter.
765 * Bug #12359 fixed - Xcos files have been converted to ZCOS to gain some space.
767 * Bug #12384 fixed - Using a Modelica part linked with an explicit link to
768 another Modelica part led to an algebraic loop error.
770 * Bug #12387 fixed - The "Modelica initialize" setup menu option did not blur
773 * Bug #12414 fixed - SWITCH_m block had different behaviors
774 for different types of inputs.
776 * Bug #12424 fixed - Calling lincos on a derivative block made Scilab crash.
778 * Bug #12460 fixed - xcosPalGenerateAllIcons sometimes crashed Scilab.
780 * Bug #12461 fixed - Cancelling zoom out by zooming in did not fully work.
782 * Bug #12561 fixed - SELECT_m and RELAY_f did not behave as expected.
784 * Bug #12568 fixed - "Recent files" menu entry is now below the "Open" one.
786 * Bug #12590 fixed - Block shape style was removed on file loading.
788 * Bug #12603 fixed - ZCOS files could contain blocks with modified (eg. invalid)
791 * Bug #12619 fixed - DLR discrete block did not display LaTeX formula
792 like CLR continuous block.
794 * Bug #12651 fixed - The 'nw' scicos_simulate option did not work while
795 passing a previous Info simulation status.
797 * Bug #12664 fixed - Inverted pendulum French localization in the Xcos demonstrations was
800 * Bug #12667 fixed - 'Recent Files' menu was not localized.
802 * Bug #12731 fixed - Code generation produced erroneous block.
804 * Bug #12732 fixed - Improper Copyright comments in the files were generated by the code
807 * Bug #12796 fixed - There was some mismatches between implicit
808 and explicit ports of superblocks.
810 * Bug #12797 fixed - I/O blocks generated by "Selection to Superblocks" had
813 * Bug #12868 fixed - There were several problems with PULSE_SC block.
815 * Bug #12869 fixed - min and max were not defined but used in Xcos generated code.
817 * Bug #12873 fixed - Scicos_flat produced an unexpected error after a diagram
820 * Bug #12874 fixed - CSCOPE marks sizes were too small.
822 * Bug #12877 fixed - Incorrect output port dimensions and types in GENSIN_f, GENSQR_f,
823 STEP_FUNCTION and STEP blocks fixed.
825 * Bug #12924 fixed - Blocks type 2004 was not handled as a valid block type.
827 * Bug #12934 fixed - Separate compilation of a superblock could fail due to under
828 determined signal sizes.