NEWS.8.md

Summary of bugs fixed for version 8.5.0 (tbd):

For (bug #XXXXX) see https://savannah.gnu.org/bugs/?XXXXX

Improvements and fixes

GUI

Build system / Tests

Documentation

Summary of bugs fixed for version 8.4.0 (2023-11-05):

For (bug #XXXXX) see https://savannah.gnu.org/bugs/?XXXXX

Improvements and fixes

• mkoctfile: Skip parsing for less flags (bug #64590).
• Fix getting number of non-zero elements in SuiteSparse matrices.
• delaunayn.m: Avoid erroneous simplex removal for integer inputs (bug #64658).
• delaunayn: Add precision loss warning for large int inputs (bug #64658).
• delaunayn: Also warn for negative integers with large magnitude (bug #64658).
• Correctly load all-zero sparse matrices from text files (bug #64696).
• Allocate correct storage for interleaved complex mxArray objects (bug #64687).
• mkoctfile: Prefer parsing over passing on argument after unknown argument (bug #64725).
• fft: Avoid segmentation fault with ND-arrays (bug #64729).
• fft: Avoid overwriting input for inplace operations (bug #64733).
• intersect.m: Correct third output with "stable" flag (bug #60347).
• Avoid use-after-free issue in parser (bug #60882).
• betainc.m: Improve integer input logic and validation (bug #64726).
• webread: Apply timeout also for initial connection (bug #64826).

GUI

Build system / Tests

• mkpp.m: Remove stray text causing error with demo code.
• Ensure that new test for saving sparse matrices writes a text file.
• Search for tests in and install octave-value template sources (bug #64696).

Documentation

• Correct variable name typo in documentation of slice().
• Expand documentation for gcd() and lcm().
• datevec: Clarify that heuristics are used to detect the format of a date string (bug #64563).
• datenum: Add help text that inputs can be arrays, add examples (bug #54679).
• datevec: Add help text that inputs can be arrays (bug #54679).
• Describe various pitfalls with floating point ranges (bug #64692).
• Add range example and some copyedits.

Summary of bugs fixed for version 8.3.0 (2022-08-08):

Improvements and fixes

• undo_string_escapes: Return '\0' for NUL byte input (bug #64051).
• Stop excessive memory usage for minor grid ticks (bug #64017).
• Include needed headers in installed .cc files (bug #63922).
• Fix floating point exception when sparse array reshaped to 0x0 (bug #64080).
• unicode2native: Fix conversion to UTF-16 (bug #64139).
• memory: Fix execution on Cygwin.
• uniconv-wrappers: Avoid freeing uninitialized pointer (bug #64182).
• lo-regexp.cc: Only include PCRE2/PCRE headers if feature test succeeded (bug #64172).
• Fix segfault when saveobj() used with old-style class files (bug #64220).
• Fix restoring breakpoints from dbstatus breakpoint structure (bug #64271).
• colorbar.m: Fix undefined variable d (bug #64287).
• Reset all axes properties except 'Units' and 'Position' with cla ('reset').
• unicode2native.m: Restore support for one input parameter (bug #64331).
• native2unicode, unicode2native: Support empty second argument (bug #64331).
• fullfile.m: Return empty cell for input with empty cell (bug #64377).
• audioplayer.m: Emit error if input data Y is empty (bug #64379).
• Remove obsolete callback constructor code for @audioplayer and @audiorecorder (bug #64379).
• Avoid false detection of "non-real pivot" in ichol() (bug #63785).
• Fix building against SuiteSparse 7.1.0 when SPQR is disabled (bug #64446).
• ellipsoid.m: Avoid returning geometry data unless it is explicitly requested to avoid overwriting ans.
• subsasgn: Avoid panic on invalid field names (bug #64213).
• lex.ll: Remove $ as a valid character for identifiers.
• fileparts.m: Allow for relative path on different drive on Windows (bug #64462).
• datevec.m: Add some ISO 8601 conform formats (patch #9964).
• delaunayn.m: Avoid erroneous simplex removal for integer inputs (bug #64658).

Build system / Tests

• betaincinv.m: Slightly increase test tolerance (for macOS 13).
• colorbar.m: Add tests (bug #64287).
• sparse-qr.cc: Fix compatibility with SuiteSparse 7.1.0 / CXSparse 4.0.4.
• pcg.m: Use eps tolerance in test for ishermitian (bug #63787).
• fileparts.m: Overhaul BISTs.

Documentation

• Return documentation for help properties (bug #64067).
• Add note to isprime.m on inputs over 2^64.

Summary of bugs fixed for version 8.2.0 (2023-04-13):

For (bug #XXXXX) see https://savannah.gnu.org/bugs/?XXXXX

Improvements and fixes

• pr-output.cc: Fix output for format native-bit (bug #63940).
• Fix evaluation of & and | expressions in conditional contexts (bug #63935).
• Avoid clang warning about very unlikely buffer overflow.
• mpoles.m: Overhaul function and use absolute tolerance for zero poles (bug #63937).
• perms.m: Change "unique" output order to reverse lexicographic to match non-unique order (bug #63962).
• Remove trailing '\r' from curl dir list (bug #63851).
• fopen: Use "UTF-8" as default encoding for fopen (bug #63930).
• Don't use encoding facet when writing bytes to stream (bug #63931).
• fopen, unicode2native: Fix converting the encoding of short char arrays with invalid UTF-8 (bug #63930).
• fopen: Try to gather complete UTF-8 surrogates when converting encoding (bug #63930).
• Fix display of scalar complex variables with mixed Inf/NaN and floating point values (bug #63961).
• fopen: Do not convert encoding for file streams with libc++ (bug #63930).

GUI

• Speedup loading and saving preferences dialog (bug #63909).

Build system / Tests

• inpolygon.m: Fix demo code (bug #63865).
• if.tst: New test for bug #63935.
• acinclude.m4: Correct typo in #define PCRE2_CODE_UNIT_WIDTH.
• lu: Add self-test with complex valued input.
• Disable visibility flags by default (bug #61855, bug #63916).
• Check whether using STL from LLVM or Apple (bug #63930).

Documentation

• Improve documentation for linspace and logspace functions.
• Correct and improve documentation for sparse() function.

Summary of important user-visible changes for version 8 (2023-03-07):

General improvements

• Octave's libraries are now built using symbol visibility by default. That means that fewer symbols are exported from these libraries. Configure with --disable-lib-visibility-flags to export all symbols (as in previous versions).

• filter is now 5X faster, which also speeds up deconv, fftfilt and arma_rnd by the same amount.

• integral can now return a second argument containing the error estimate from the underlying integrator.

• perms now accepts a second argument "unique" to return only unique permutations for inputs with repeated elements. It is faster and takes less memory to call perms ('aaaabbbbcccc', "unique") than to call unique (perms ('aaaabbbbcccc'), "rows").

• quadgk can now accept the ArrayValued input parameter to handle array-valued input functions.

• delaunayn now has consistent trivial simplex checking and removal for all input dimensions, simplex checking 3D inputs is now vectorized, and >3D simplex checking performance has been improved. Simplexes points are now ordered so they will all have positive outward normal vectors. Input type checking has also been added for improved error handling.

• factor now factorizes all 64-bit integers in roughly the same time. Previously, the product of two large primes took much longer to factorize than highly composite inputs.

• Refine option is now implemented in functions ode45, ode23, and ode23s.

• Octave is now compatible with PCRE2 (UTF-8). PCRE2 is preferred over PCRE if both are installed. Configure with --without-pcre2 if you prefer Octave to use PCRE in this case.

• mean now internally processes data as type double to reduce likelihood of hitting overflow or precision limits with other types (bug #63848).

Graphical User Interface

• The GUI has a dark style and several new icons in toolbars for better visibility and higher contrast.

• A new terminal widget has been added for the GUI. It is still experimental and is disabled by default. It can be enabled with the command line parameter --experimental-terminal-widget.

• More fonts for the documentation browser are included.

Graphics backend

• When printing, the option -svgconvert is now the default. If you want to use the more limited EPS-based tool chain (the former default)—e.g., if you encounter inaccurate image rendering—you can pass the -nosvgconvert option to the print function.

• Additional properties have been added to the figure graphics object:

  • "innerposition" (equivalent to "position")
  • "windowstate" (not yet implemented)

• Legend now features a new property "itemhitfcn", allowing the execution of a callback function when a legend item is clicked.

Matlab compatibility

• inline functions now support all Matlab methods. The performance of inline functions has also been improved.

• sub2ind now supports index values outside of the size specified by the subscripts.

• cylinder now accepts a scalar for the radius argument.

• clock now has an optional second output ISDST which indicates if Daylight Savings Time is in effect for the system's time zone.

• print now accepts option -image to specify the "opengl" renderer and -vector to specify the "painters" renderer.

• format now accepts the option "default", which is equivalent to calling format without any options to reset the default state.

• quadgk now stops iterating when error <= tolerance while the previous condition was error < tolerance.

• mean now accepts vector dimensions and options to handle NaN values. The option "a" (arithmetic mean), "g" (geometric mean), and "h" (harmonic mean) are no longer accepted, only the arithmetic mean is computed. For the geometric and harmonic mean, please use respective functions geomean and harmmean from the Octave Statistics package.

• var and std now optionally output a second argument containing the mean or weighted mean.

• integral can now accept the 'ArrayValued' option in combination with 'RelTol' and 'WayPoints'.

• File system operations that remove files (e.g., unlink or rmdir) now also remove files that have their read-only file attribute set on Windows.

• The default state for certain graphics properties has been made consistent with Matlab.

    Object      | Property         | Default State
    ------------|------------------|------------
    figure      | "dockcontrols"   | "on"

• ode45, ode23, and ode23s have improved results for options Events, OutputFcn, and Refine, along with corrected orientation of struct outputs.

• Early notice of future breaking changes: Due to many user requests that Octave should have a Matlab-compatible string class, there is work under way to implement a string class that will differ from a vector of characters. In Octave, single-quoted character arrays are currently compatible with Matlab, but double-quoted forms are not. Currently in Octave, both 'foo' and "foo" are largely interchangeable, barring certain escape sequence interpretations such as "\n" (converted to a single newline character) as opposed to '\n' (two separate characters). Matlab’s single-quoted character arrays and double-quoted strings do not process backslash escape sequences, unlike many other languages, and those escape sequences are instead processed by individual functions such as fprintf.

  Octave's behavior is likely to change in future as a consequence of implementing Matlab-style string syntax. For example, 'foo' will remain a three-element character vector, but "foo" will become a single-element string object. Some backslash escape sequences are likely to differ from their current Octave behavior, but will achieve greater Matlab compatibility. The exact implementation is a work in progress, and may or may not include methods of preserving backward compatibility.

  What this means for user code: If your code currently relies on double-quoted strings (e.g., "foo") representing character vectors as opposed to string objects, and if you intend to update to a future version of Octave incorporating the above changes, then consider replacing all double-quoted strings with single-quoted strings in your code over time (e.g., replace "foo" with 'foo'). Single-quoted strings are expected to retain current behavior. Further, if your code relies on backslash escape sequence interpretation in double-quoted strings (except for special cases like the printf family), that code may need to change as well.

Alphabetical list of new functions added in Octave 8

• clearAllMemoizedCaches
• matlab.lang.MemoizedFunction
• memoize
• normalize
• pagectranspose
• pagetranspose
• uifigure

Deprecated functions, properties, and operators

The following functions and properties have been deprecated in Octave 8 and will be removed from Octave 10 (or whatever version is the second major release after 8):

• Functions

    Function               | Replacement
    -----------------------|------------------
    shift                  | circshift
    sparse_auto_mutate     | none (see below)

• The sparse_auto_mutate function no longer has any effect on Octave's behavior. Previously, after calling sparse_auto_mutate (true), Octave would automatically convert sparse matrices to full when a sparse matrix required more memory than simply using full matrix storage. This setting was false by default for compatibility with Matlab. Now you must manually convert to full storage when desired.

• Properties

  The following property names are discouraged, but there is no fixed date for their removal.

    Object           | Property    | Replacement
    -----------------|-------------|------------
    uimenu           | label       | text
    uimenu           | callback    | menuselectedfcn

The following functions were deprecated in Octave 6 and have been removed from Octave 8.

• Functions

    Function     | Replacement
    -------------|------------
    runtests     | oruntests

• The environment variable used by mkoctfile for linker flags is now LDFLAGS rather than LFLAGS. LFLAGS was deprecated in Octave 6, and will be removed in a future version of Octave.

Summary of bugs fixed for version 8.1.0 (2023-01-22):

• Improved input validation and/or output handling for poly, pinv, patch, fill, fill3, qp, datevec, textscan, sub2ind, qr, airy, regexp, dec2bin, dec2hex, mean, and many others.
• Improved performance for complex, fftw, delaunayn, isfield, tsearch, sparse matrix exponentiation, other sparse operations, and many others.
• Overhauled @audiorecorder and @audioplayer classes.

Old release news

• Octave 7.x
• Octave 6.x
• Octave 5.x
• Octave 4.x
• Octave 3.x
• Octave 2.x
• Octave 1.x