summaryrefslogtreecommitdiff
path: root/math
AgeCommit message (Collapse)AuthorFilesLines
2020-05-18libixion: clean some pkglintwiz2-3/+5
2020-05-18libixion: use .CURDIRwiz1-3/+3
distinfo and patches paths must be absolute
2020-05-17gnumeric: Update to 1.12.47nia2-8/+7
Gnumeric 1.12.47 Andreas: * Fix chart and image name roundtrip through ODF. [#477] * Fix print flag roundtrip for sheet objects through ODF. [#486] Jean: * Fix crash when the embedded spreadsheet is invalid. [#481] Morten: * Fix dialog size problem. * Add "--set CELL=CONTENTS" option to ssconvert. * Improve tests. * Improve speed on exit with lots of conditional formatting. * Improve error message [#472] * Impose ordering of ssconvert --export-graphs * Doc fixes. * Fix CELL crash. [#479] * Handle export options for ssconvert --export-graphs. * Fix deps crash. [#480] * Fix chart name roundtrip through XLSX. [#478] * Fix sheet object print flag roundtrip through XLSX. [#486]
2020-05-17pytest from versioned dependsadam4-27/+26
2020-05-16math/bc-gh: Update to 2.7.2gdt3-8/+32
Upstream NEWS highlights: (translations and bugfixes omitted) ## 2.7.2 This is a production release with one major bug fix. The `length()` built-in function can take either a number or an array. If it takes an array, it returns the length of the array. Arrays can be passed by reference. The bug is that the `length()` function would not properly dereference arrays that were references. This is a bug that affects all users. **ALL USERS SHOULD UPDATE `bc`**. ## 2.7.0 There is only one new feature: **`bc` now has a built-in pseudo-random number generator** (PRNG). The PRNG is seeded, making it useful for applications where `/dev/urandom` does not work because output needs to be reproducible. However, it also uses `/dev/urandom` to seed itself by default, so it will start with a good seed by default. On top of that, four functions were added to `bc`'s [extended math library][16] to make using the PRNG easier: * `frand(p)`: Generates a number between `[0,1)` to `p` decimal places. * `ifrand(i, p)`: Generates an integer with bound `i` and adds it to `frand(p)`. * `srand(x)`: Randomizes the sign of `x`. In other words, it flips the sign of `x` with probability `0.5`. * `brand()`: Returns a random boolean value (either `0` or `1`). ## 2.4.0 * The `bc` `&&` and `||` operators were made available to `dc` through the `M` and `m` commands, respectively. * `dc` macros were changed to be tail call-optimized. The last item, tail call optimization, means that if the last thing in a macro is a call to another macro, then the old macro is popped before executing the new macro. This change was made to stop `dc` from consuming more and more memory as macros are executed in a loop. The `q` and `Q` commands still respect the "hidden" macros by way of recording how many macros were removed by tail call optimization.
2020-05-16libnumbertext: Update to 1.0.5nia2-8/+10
2018-10-09: 1.0.5 release: * et.sor: new Estonian module * Java: fix path separator for Windows by Rens Toonen * spellout: fix DEFPATH in spellout by Rene Engelhard * README: conversion to MarkDown by Miklós Vajna * da.sor: fix "en million", bug report by Hsonesson1 * de.sor: add function year by c-kuehl * eo.sor: fix ordinal-number, ie. "1-a" by Adolfo Jayme Barrientos * fi.sor: fixed and more currencies by Tuomas Hietala * lt.sor: fix ordinal numbers and currencies by Aurimas Fišeras * pl.sor: fix ordinal numbers 4x, 7x, bug report by tengwar * sq.sor: add missing 1000-9999
2020-05-14Doesn't support Python 2.7joerg1-1/+3
2020-05-13py-lmfit: updated to 1.0.1adam3-18/+8
Version 1.0.1 Release Notes ============================ **Version 1.0.1 is the last release that supports Python 3.5**. All newer version will require 3.6+ so that we can use formatting-strings and rely on dictionaries being ordered. New features: - added thermal distribution model and lineshape - introduced a new argument ``max_nfev`` to uniformly specify the maximum number of function evalutions **Please note: all other arguments (e.g., ``maxfev``, ``maxiter``, ...) will no longer be passed to the underlying solver. A warning will be emitted stating that one should use ``max_nfev``.** - the attribute ``call_kws`` was added to the ``MinimizerResult`` class and contains the keyword arguments that are supplied to the solver in SciPy. Bug fixes: - fixes to the ``load`` and ``__setstate__`` methods of the Parameter class - fixed failure of ModelResult.dump() due to missing attributes - ``guess_from_peak`` function now also works correctly with decreasing x-values or when using pandas - the ``Parameter.set()`` method now correctly first updates the boundaries and then the value Various: - fixed typo for the use of expressions in the documentation - removal of PY2-compatibility and unused code and improved test coverage - removed deprecated ``isParameter`` function and automatic conversion of an ``uncertainties`` object - inaccurate FWHM calculations were removed from built-in models, others labeled as estimates - corrected spelling mistake for the Doniach lineshape and model - removed unsupported/untested code for IPython notebooks in lmfit/ui/*
2020-05-13py-ephem: updated to 3.7.7.1adam2-7/+7
3.7.7.1: Unknown changes
2020-05-13hs-*: add PLIST files for a few more Haskell packagesrillig2-0/+35
2020-05-13math/pari23: remove redundant SUBST blockrillig1-7/+1
At least on NetBSD 8.0 x86_64, this is not necessary.
2020-05-13math/boolstuff: suppress USE_TOOLS+=perl warningrillig1-1/+2
2020-05-12py-numba: updated to 0.49.1adam3-8/+11
Version 0.49.1: This is a bugfix release for 0.49.0, it fixes some residual issues with SSA form, a critical bug in the branch pruning logic and a number of other smaller issues: * Fixed Threading Implementation Typos * Fixes Remove references to cffi_support from docs and examples * Fix invalid type in resolve for comparison expr in parfors. * Fix erroneous rewrite of predicate to bit const on prune. * Fixes SSA local def scan based on invalid equality assumption. * Fixes naming error in array_exprs * Fix. Incorrect race variable detection due to SSA naming. * Make literal_unroll function work as a freevar. * Unset the memory manager after EMM Plugin tests * Fix some SSA issues * Pin to sphinx=2.4.4 to avoid problem with C declaration * Fix unifying undefined first class function types issue * Update example in 5m guide WRT SSA type stability. * Restore numba.types as public API
2020-05-11hs-*: add PLIST filesrillig11-0/+560
These PLIST files have been autogenerated by mk/haskell.mk using HS_UPDATE_PLIST=yes during a bulk build. They will help to track changes to the packages. The Haskell packages didn't have PLIST files because their paths contained package hashes. These hashes are now determined by mk/haskell.mk, which makes it easy to generate easy to read PLIST files.
2020-05-09Stop using boost/detail/endian.hppjoerg4-5/+50
2020-05-06revbump after boost updateadam17-31/+34
2020-05-04py-numpy: updated to 1.18.4adam2-10/+8
1.18.4: BLD: add i686 for 1.18 builds BUG: random: ``Generator.integers(2**32)`` always returned 0. BLD: fix path to libgfortran on macOS REV: Reverts side-effect changes to casting BLD: put openblas library in local directory on windows DOC: Change import error "howto" to link to new troubleshooting...
2020-05-03math/py-statsmodels: Update to 0.11.1minskim3-30/+202
Major Features: - Allow fixing parameters in state space models - Add new version of ARIMA-type estimators (AR, ARIMA, SARIMAX) - Add STL decomposition for time series - Functional SIR - Zivot Andrews test - Added Oaxaca-Blinder Decomposition - Add rolling WLS and OLS - Replacement for AR Performance Improvements: - Cythonize innovations algo and filter - Only perform required predict iterations in state space models - State space: Improve low memory usability; allow in fit, loglike
2020-05-03math/R-nloptr: remove no-op SUBST blockrillig1-7/+1
2020-05-02math/pari: remove no-op SUBST blockrillig1-7/+1
When building on NetBSD-8.0-x86_64, the file pari.cfg does not contain any references to TOOLS_DIR.
2020-05-02R: dependency tex-ifluatex,tex-xetex -> tex-iftex (commented out)markd1-3/+2
2020-05-02tex-mathspec: dependency tex-ifxetex -> tex-iftexmarkd1-2/+3
2020-04-30py-numba: remove patch that was removed from distinfowiz1-15/+0
2020-04-29(math/py-numpy16) Add missing distinfo (right ?)mef1-0/+11
2020-04-28py-lmfit: updated to 1.0.0adam2-12/+12
Version 1.0.0 Release Notes New features: - no new features are introduced in 1.0.0. Improvements: - support for Python 2 and use of the ``six`` package are removed. Various: - documentation updates to clarify the use of ``emcee``.
2020-04-28py-numexpr: updated to 2.7.1adam2-7/+7
Changes from 2.7.0 to 2.7.1 - Python 3.8 support has been added. - Python 3.4 support is discontinued. - The tests are now compatible with NumPy 1.18. - `site.cfg.example` was updated to use the `libraries` tag instead of `mkl_libs`, which is recommended for newer version of NumPy.
2020-04-28py-scipy: updated to 1.4.1adam3-23/+131
SciPy 1.4.1 is a bug-fix release with no new features compared to 1.4.0. Importantly, it aims to fix a problem where an older version of pybind11 may cause a segmentation fault when imported alongside incompatible libraries. SciPy 1.4.0 is the culmination of 6 months of hard work. It contains many new features, numerous bug-fixes, improved test coverage and better documentation. There have been a number of deprecations and API changes in this release, which are documented below. All users are encouraged to upgrade to this release, as there are a large number of bug-fixes and optimizations. Before upgrading, we recommend that users check that their own code does not use deprecated SciPy functionality (to do so, run your code with ``python -Wd`` and check for ``DeprecationWarning`` s). Our development attention will now shift to bug-fix releases on the 1.4.x branch, and on adding new features on the master branch.
2020-04-27py-numpy: fix linker options on Darwinadam2-5/+5
2020-04-27py-autograd: updated to 1.3adam3-29/+71
1.3: Unknown changes
2020-04-27py-uncertainties: use py-numpy/buildink to auto-select based on Python versionadam1-3/+2
2020-04-27py-patsy: use py-numpy/buildink to auto-select based on Python versionadam1-2/+2
2020-04-27math/yacas: remove no-op SUBST blockrillig1-6/+1
2020-04-27py-numpy: updated to 1.18.3adam6-87/+103
1.18.3: BUG: Fix eigh and cholesky methods of numpy.random.multivariate_normalBUG,MAINT: Remove incorrect special case in string to number... BUG: Guarantee array is in valid state after memory error occurs... BUG: Check that `pvals` is 1D in `_generator.multinomial`. BUG: Alpha parameter must be 1D in `generator.dirichlet` 1.18.2: TST: move _no_tracing to testing._private MAINT: Large overhead in some random functions TST: Do not create gfortran link in azure Mac testing. BUG: Added missing error check in `ndarray.__contains__` MAINT: use list-based APIs to call subprocesses REL: Prepare for 1.18.2 release. BUG: fix logic error when nm fails on 32-bit 1.18.1: MAINT: Update pavement.py for towncrier. DOC: add moved modules to 1.18 release note MAINT, DOC: Minor backports and updates for 1.18.x TST: Add assert_array_equal test for big integer arrays BUG: use tmp dir and check version for cython test BUG: distutils: fix msvc+gfortran openblas handling corner case BUG: remove -std=c99 for c++ compilation MAINT: unskip test on win32 TST: add BLAS ILP64 run in Travis & Azure MAINT: only add --std=c99 where needed BUG: lib: Fix handling of integer arrays by gradient. MAINT: Do not use private Python function in testing REL: Prepare for the NumPy 1.18.1 release. 1.18.0: DOC: added note to docstring of numpy.savez BUG: Numpy scalar types sometimes have the same name DOC: Improve axes shift description and example in np.tensordot MAINT: avoid relying on `np.generic.__name__` in `np.dtype.name` ENH: supply our version of numpy.pxd, requires cython>=0.29 BUG: General fixes to f2py reference counts (dereferencing) BUG: NaT now sorts to ends of arrays DOC: Updates to nditer usage instructions BUG: Do not crash on recursive `.dtype` attribute lookup. ENH: Use AVX for float32 implementation of np.sin & np.cos DEP: Deprecate silent ignoring of bad data in fromfile/fromstring ENH: Always produce a consistent shape in the result of `argwhere` DOC: array(obj, dtype=dt) can downcast DOC: Document ma.filled behavior with non-scalar fill_value DOC: Add note to irfft-like functions about the default sizes BUG: Don't produce undefined behavior for a << b if b >= bitsof(a) MAINT: Update NEP template. ENH: random: Add the multivariate hypergeometric distribution. DOC: Fix unrendered links MAINT: Rewrite Floyd algorithm DOC: Add missing macros to C-API documentation ENH: Add axis argument to random.permutation and random.shuffle DOC: Adds documentation of functions exposed in numpy namespace BUG: Refcount fixes MAINT: Ensure array_dealloc does not modify refcount of self MAINT: Prepare master for 1.18.0 development. MAINT,BUG,DOC: Fix errors in _add_newdocs MAINT: Remove an unnessary backslash between two string literals MAINT: Update pavement to use python3 in shell commands. MAINT: Remove unnecessary backslashes (and replace others by... MAINT: Replace integers in places where booleans are expected DOC: Add missing parameter description for keepdims in MaskedArray ENH: use AVX for float32 and float64 implementation of sqrt,... DOC: reformat top-level release index DOC : Refactor Array API documentation -- Array Structure and... DOC: Fix typo in "make_mask" documentation MAINT: Delete unused _aliased_types.py BLD: Remove Trusty dist in Travis CI build BUG: Handle weird bytestrings in dtype() ENH: use towncrier to build the release note ENH: improve error message for ragged-array creation failure DOC: Update the description of byteswap BUG: i0 Bessel function regression on array-likes supporting... ENH, BUILD: refactor all OpenBLAS downloads into a single, testable... MAINT: Remove unnecessary parenthesis in numpy.ma.core MAINT: Fix wrong spelling of ufunc DOC: Remove explicit .next method calls with built-in next function... DOC: Don't override MaskedArray.view documentation with the one... BUG: Fix incorrect GIL release in array.nonzero MAINT: Warn if `_add_newdocs.py` is used to add docstrings to... MAINT: Revert 13876, "MAINT,BUG,DOC: Fix errors in _add_newdocs" MAINT,BUG,DOC: Fix errors in _add_newdocs DOC, MAINT: emphasize random API changes, remove Generator.randint DOC: Add a numpy-doc docstring to add_newdoc DOC: Fix rst rendering in data types DOC:Update the description of set_printoptions in quickstart... Fixing failure on Python 2.7 on Windows 7 Fix a typo related to the range of indices DOC: add space between words across lines BUG, DOC: add new recfunctions to `__all__` DOC: Change (old) range() to np.arange() DOC: improve np.sort docstring DOC: spellcheck numpy/doc/broadcasting.py MAINT, TST: remove test-installed-numpy.py DOC: Document array_function at a higher level. DOC: show workaround for backward compatibility DOC: Add a call for contribution paragraph to the readme BUG: Missing warnings import in polyutils BUILD: adapt "make version-check" to "make dist" DOC: emphasize need for matching numpy, git versions TST, MAINT, BUG: expand OpenBLAS version checking ENH: Chain exception for typed item assignment MAINT: Fix spelling error in npy_tempita kwarg DOC: Array API : Directory restructure and code cleanup [DOC] Remove unused/deprecated functions Update system_info.py DOC:Link between the two indexing documentation pages DOC: Update NumFOCUS subcommittee replacing Nathaniel with Sebastian DOC: update "Contributing to NumPy" with more activities/roles DOC: Improve quickstart documentation of new random Generator DEP: Speed up WarnOnWrite deprecation in buffer interface NEP: numpy.org website redesign DOC: Fix docstring of numpy.allclose regarding NaNs DEP: Raise warnings for deprecated functions PyArray_As1D, PyArray_As2D DEP: Remove np.rank which has been deprecated for more than 5... BUG, TEST: Adding validation test suite to validate float32 exp ENH,DEP: Allow multiple axes in expand_dims ENH: add pyproject.toml DOC: Update cversions.py links and wording DOC, BUILD: cleanups and fix (again) 'make dist' BUG: Fix file-like object check when saving arrays DOC: Resolve bad references in Sphinx warnings MAINT: bump ARMv8 / POWER8 OpenBLAS in CI DOC: Emphasize the need to run tests when building from source DOC:Add example to clarify "numpy.save" behavior on already open... DEP: Deprecate full and economic modes for linalg.qr DOC: Doc release BUG: fix build issue on icc 2016 TST: Add 3.8-dev to travisCI testing. DOC: Add blank line above doctest for intersect1d ENH: Propose standard policy for dropping support of old Python... DOC: Use `pip install .` where possible instead of calling setup.py MAINT: adjustments to test_ufunc_noncontigous MAINT: Improve NEP template DOC: fix documentation of i and j for tri. MAINT: Lazy import testing on python >=3.7 DEP: Deprecate PyArray_FromDimsAndDataAndDescr, PyArray_FromDims MAINT: Clearer error message while padding with stat_length=0 MAINT: remove duplicate variable assignments BUG: initialize variable that is passed by pointer DOC: fix typo in c-api/array.rst doc BUG: Add gcd/lcm definitions to npy_math.h MAINT: Mark umath accuracy test xfail. MAINT: Use equality instead of identity check with literal MAINT: Fix small typo in quickstart docs DOC, MAINT: Update master after 1.17.0 release. ENH: add c-imported modules for freeze analysis in np.random BUG: Fix DeprecationWarning in python 3.8 BUG: Remove stray print that causes a SystemError on python 3.7... BUG: Remove the broken clip wrapper BUG: avx2_scalef_ps must be static TST: Allow fuss in testing strided/non-strided exp/log loops NEP: Proposal for __duckarray__ protocol BUG: Make advanced indexing result on read-only subclass writeable TST: Clean up of test_pocketfft.py DEP: Deprecate np.alen MAINT: Workaround for Intel compiler bug leading to failing test DOC: Fix hermitian argument docs in `svd` MAINT: Fix a docstring typo. DOC: Fix links in `/.github/CONTRIBUTING.md`. ENH: Multivariate normal speedups MAINT: Improve mismatch message of np.testing.assert_array_equal DOC,MAINT: Move towncrier files and fixup categories BUG: Fixed default BitGenerator name BUG: Fix uint-overflow if padding with linear_ramp and negative... ENH: Enable huge pages in all Linux builds BUG: Fix leak in the f2py-generated module init and `PyMem_Del`... DOC: new nan_to_num keywords are from 1.17 onwards TST: Add tests for deprecated C functions (PyArray_As1D, PyArray_As1D) DOC: mention `take_along_axis` in `choose` ENH: Parse complex number from string DOC: update or remove outdated sourceforge links MAINT: Better error message for norm DOC: add backlinks to numpy.org BUG: Don't fail when lexsorting some empty arrays. BUG: Fix segfault in `random.permutation(x)` when x is a string. Doc: fix a typo in NEP21 DOC: set status of NEP 28 (website redesign) to Accepted BLD: MAINT: change default behavior of build flag appending. BUG: Fixes StopIteration error from 'np.genfromtext' for empty... BUG: fix inconsistent axes ordering for axis in function `unique` DEP: Deprecate load/dump functions in favour of pickle methods MAINT: Update NEP-30 DEP: Deprecate arrayprint formatting functions DOC: remove scipy.org from the breadcrumb formattiong BUG: Fix formatting error in exception message DOC: Address typos in dispatch docs BUG: Fix ZeroDivisionError for zero length arrays in pocketfft. BUG: Fix misuse of .names and .fields in various places TST, BUG: Use python3.6-dbg. BUG: core: Handle large negative np.int64 args in binary_repr. BUG: Fix numpy.random bug in platform detection MAINT: random: Match type of SeedSequence.pool_size to DEFAULT_POOL_SIZE Bug: Fix behavior of structured_to_unstructured on non-trivial... DOC: add two commas, move one word DOC: Clarify rules about broadcasting when empty arrays are involved. TST, MAINT: bump to OpenBLAS 0.3.7 stable DEP: numpy.testing.rand DEP: Deprecate class `SafeEval` BUG: revert detecting and raising error on ragged arrays DOC: Improve documentation of `isscalar`. MAINT: Fix bloated mismatch error percentage in array comparisons. DOC: Fix a minor typo in dispatch documentation. MAINT: Remove redundant deprecation checks MAINT: polynomial: Add an N-d vander implementation used under... DOC: clarify that PytestTester is non-public DOC: support and require sphinx>=2.2 DOC: random: fix doc linking, was referencing private submodules. MAINT: Fixes for prospective Python 3.10 and 4.0 DOC: lib: Add more explanation of the weighted average calculation. MAINT: Avoid BytesWarning in PyArray_DescrConverter() MAINT: Post NumPy 1.17.1 update. DOC: Fixed dtype docs for var, nanvar. DOC: Document project as Python 3 only with a trove classifier BUILD: move all test dependencies to ./test_requirements.txt BUG: lib: Fix histogram problem with signed integer arrays. REL: Update master after NumPy 1.16.5 release. BUG: test, fix regression in converting to ctypes NEP: Add initial draft of NEP-31: Context-local and global overrides.. DOC: document numpy/doc update process DOC: update np.around docstring with note about floating-point... BUG: view with fieldless dtype should raise if itemsize != 0 DOC: fix issue with __new__ usage in subclassing doc. DOC: Fix release notes table of contents NEP 32: Remove the financial functions from NumPy BLD: Update RELEASE_WALKTHROUGH and cythonize. Bump pytest from 5.1.1 to 5.1.2 TST: Remove build job since we now use Dependabot BLD: Only allow using Cython module when cythonizing. TST: Add dependabot config file. BUG: Fix format statement associated with AttributeError. BUG: Fix aradixsort indirect indexing. DOC: add the reference to 'printoptions' BUG: Do not show Override module in private error classes. DOC: Make implementation bullet points consistent in NEP 29 MAINT: Clarify policy language in NEP-29. REL: Update master after 1.17.2 release. MAINT: clean up pocketfft modules inside numpy.fft namespace BLD: remove generated Cython files from sdist MAINT: add test to prevent new public-looking modules being added BUG: random.hypergeometic assumes npy_long is npy_int64, hangs... ENH: Print the amount of memory that would be used by a failed... MAINT: use test_requirements.txt in tox and shippable, ship it... BUG: add a specialized loop for boolean matmul BUG: Fix _ctypes class circular reference. BUG: core: Fix the str function of the rational dtype. DOC: add timedelta64 signature MAINT: Extract raising of MemoryError to a helper function BUG,MAINT: Some fixes and minor cleanup based on clang analysis MAINT: Add `NPY_UNUSED` and `const` qualified suggested by clang MAINT: Silence integer comparison build warnings in assert statements MAINT: distutils: Add newline at the end of printed warnings. BUG: random: Revert gh-14458 and refix gh-14557. DOC: Fix reference NPY_ARRAY_OWNDATA instead of NPY_OWNDATA. ENH: Allow NPY_PKG_CONFIG_PATH environment variable override MAINT: remove the entropy c-extension module DOC: Add backslashes so PyUFunc_FromFuncAndDataAndSignatureAndIdentity DOC: Fix a minor typo in changelog readme BUG: Fix randint when range is 2**32 DOC: tweak np.round docstring to clarify floating-point error DOC: Add warning to NPV function API: Do not return None from recfunctions.drop_fields BUG: Fix flatten_dtype so that nested 0-field structs are flattened... DOC: Build release notes during CircleCI step BUILD: Hide platform configuration probe behind --debug-configure Mention that split() returns views into the original array MAINT: Simplify lookfor function MAINT: random: Remove a few duplicated C function prototypes. BUILD, MAINT: run tests with verbose for PyPY, also do not leak... BUG: fix release snippet failures caught only after merging BLD: add warn-error option, adds -Werror to compiler BUG: random: Create a legacy implementation of random.binomial. MAINT: remove unused functions, rearrange headers (from CC=clang) DOC: Fix a bit of code in 'Beyond the Basics' C API user guide. MAINT: Cleanup old_defines in DOC DOC: Added missing versionadded to diff(prepend) BUG: Avoid ctypes in Generators Changing ImportWarning to DeprecationWarning MAINT: handle case where GIT_VERSION is empty string MAINT: core: Remove duplicated inner loop ee->e from log, exp,... DOC: clarify input types in basics.io.genfromtxt.rst DOC: remove note about Pocketfft license file (non-existing here). DOC: Fix code that generates the table in the 'Casting Rules'... MAINT: don't install partial numpy.random C/Cython API. TST: ensure coercion tables aren't printed on failing public... DEP: remove deprecated (and private) numpy.testing submodules. BLD, DOC: fix gh-14518, add release note BUG: importing build_src breaks setuptools monkeypatch for msvc14 DOC: Note runtests.py `-- -s` method to use pytests `-s` DOC: update submodule docstrings, remove info.py files DOC: Document the NPY_SCALARKIND values as C variables. MAINT: Bump pytest from 5.1.2 to 5.1.3 DEP: remove deprecated select behaviour BUG: Add missing check for 0-sized array in ravel_multi_index BUG: dtype refcount cleanups DOC: Fix a minor typo in changelog entry MAINT: Fix typo: remoge → remove DOC: Change the promotion table checkmark to 'Y'. DEP: Complete deprecation of invalid array/memory order DOC: Add to doc that interp cannot contain NaN NEP: Accept NEP 32. NEP: Fix discrepancies in NEPs NEP: Only list "Active" NEPs under "Meta-NEPs" API: restructure and document numpy.random C-API BUG: properly define PyArray_DescrCheck{,Exact} MAINT: Remove duplicate files from .gitignore API: rearrange the cython files in numpy.random MAINT: Bump pytest from 5.1.3 to 5.2.0 MAINT: Add "MAINT" tag to dependabot commit msg DOC: Updated sphinx directive formatting DEP: Finish deprecation of non-integer `num` in linspace DOC: s/OR/AND/ in np.logical_and docstring DOC: misleading np.sinc() documentation DOC: clarify residual in np.polyfit BUILD: change to build_src --verbose-cfg, runtests.py --debug-info BUG: always free clean_sep DOC: Create `class Extension` docstring and add it to documentation. DOC: add `printoptions` as a context manager to `set_printoptions` DOC: Fix typo in NEP 29 MAINT: Use scalar math power function directly DOC: Add IPython to dependencies needed to build docs. MAINT: Bump pytest-cov from 2.7.1 to 2.8.1 MAINT: Bump pytest from 5.2.0 to 5.2.1 MAINT: Bump pytz from 2019.2 to 2019.3 MAINT: Use `extract_unit` throughout datetime BUG: fix fromfile behavior when reading sub-array dtypes BUG: random: Use correct length when axis is given to shuffle. BUG: Do not rely on undefined behaviour to cast from float to... NEP: add default-dtype-object-deprecation nep 34 MAINT: Remove unused boolean negative/subtract loops DEP: ufunc `out` argument must be a tuple for multiple outputs BUG: Fix `np.einsum` errors on Power9 Linux and z/Linux DOC: Note release notes process changes on devdocs start page Doc warnings DOC: Switch Markdown link to RST in NEP 29 TST: Divide Azure CI Pipelines into stages. DEP: Finish the out kwarg deprecation for ufunc calls DOC: Removing mentions of appveyor BUG: Default start to 0 for timedelta arange API: NaT (arg)min/max behavior API: Forbid Q<->m safe casting DEP: deprecate financial functions. DOC: Move newsfragment to correct folder DOC: cleaning up examples in maskedarray.generic MAINT: umath: Change error message for unsupported bool subtraction. ENH: Add complex number support for fromfile TST: run refguide-check on rst files in doc/* DOC: Edit NEP procedure for better discussion DOC: Post 1.17.3 release update. NEP: Accept NEP 29 as final BUG: Don't narrow intp to int when producing error messages DOC: lib: Fix deprecation markup in financial function docstrings. DOC: Change from HTTP to HTTPS BUG: clear only attribute errors in get_attr_string.h::maybe_get_attr MAINT: doc: Remove doc/newdtype_example/ Reword cautionary note about dtype.descr BUG: fix integer size confusion in handling array's ndmin argument TST, BUILD: add a gcc 4.8 run on ubuntu 18.04 Update CLASSIFIERS with python 3.8 support BUG: random: biased samples from integers() with 8 or 16 bit... DOC: Add release note about changed random variate stream from... DOC: Make changes to NEP procedure DOC: random: Remove redundant 'See Also' entry in 'uniform' docstring. MAINT: Minor typo fix MAINT: Bump pytest from 5.2.1 to 5.2.2 DOC: Adjust NEP-31 to new template. DEP: issue deprecation warning when creating ragged array (NEP... NEP: move 'NEP 29 random' from Accepted to Final DOC: Add take_along_axis to the see also section in argmin, argmax... ENH: change object-array comparisons to prefer OO->O unfuncs TST: Don't construct Fraction instances from numpy scalars Rename helper functions to not use the word rank MAINT: Use templating to merge float loops BUILD: ignore more build.log warnings BLD: Prevent -flto from optimising long double representation... BUG: raise ValueError for empty arrays passed to _pyarray_correlate MAINT: move buffer.h -> npy_buffer.h to avoid conflicts MAINT: Bump cython from 0.29.13 to 0.29.14 ENH: add isinf, isnan, fmin, fmax loops for datetime64, timedelta64 BLD: add 'apt update' to shippable MAINT: revert gh-14800, which gave precedence to OO->O over OO->? REL: Update master after 1.17.4 release. BUILD: remove SSE2 flag from numpy.random builds DOC: Update NEP29 with Python3.8 informations. BUG: Remove builtins from __all__ MAINT: Delete and ignore generated files Update FUNDING.yml MAINT: Remove uses of scalar aliases NEP: move nep 34 to accepted TST: Add s390x to the TravisCI test matrix. DOC: Note FFT type promotion TST: Test with Python3.8 on Windows. TST: Update travis.yml TST: add no_tracing decorator to refcount-sensitive codepath... MAINT: Bump pytest from 5.2.2 to 5.2.4 BUG: Fix step returned by linspace when num=1 and endpoint=False DOC: Compare 'tolist' function to 'list' in example DOC: Clarify return type for default_rng MAINT: move numpy/random/examples -> numpy/random/_examples DOC: testing: Note handling of scalars in assert_array_equal... DOC, API: add random.__init__.pxd and document random.* functions DOC: Clean up examples of low-level random access TST. API: test using distributions.h via cffi TST: skip if cython is not available MAINT: Cleaned up mintypecode for Py3 DOC: fix docstring of np.linalg.norm MAINT: Added Python3.8 branch to dll lib discovery on Windows DEV: update asv.conf.json MAINT: Bump pytest from 5.2.4 to 5.3.0 MAINT: Fix typos REV: "ENH: Improved performance of PyArray_FromAny for sequences... BUG: warn when saving dtype with metadata DEP: Deprecate the axis argument to masked_rows and masked_cols MAINT: Fix long name of PCG64 DOC, API: improve the C-API/Cython documentation and interfaces... DOC: Fix typo in numpy.loadtxt and numpy.genfromtxt documentation ENH: allow using symbol-suffixed 64-bit BLAS/LAPACK for numpy.dot... DOC: add a more useful comment to compat.py3k.py DOC: lib: Use a clearer example of ddof in the notes of the cov... TST: machinery for tests requiring large memory + lapack64 smoketest MAINT: Only copy input array in _replace_nan() if there are nans... MAINT: Bump pytest from 5.3.0 to 5.3.1 REV: "ENH: Improved performance of PyArray_FromAny for sequences... REL: Prepare for 1.18 branch MAINT: Cleaned up mintypecode for Py3 (pt. 2) BUG: Fix refcounting in ufunc object loops BUG: Exceptions tracebacks are dropped REV: Revert "Merge pull request 14794 from mattip/nep-0034-impl" API, DOC: change names to multivariate_hypergeometric, improve docs REL: Prepare for NumPy 1.18.0 release. TST: Check requires_memory immediately before the test ENH: Add support to sort timedelta64 `NaT` to end of the array MAINT: follow-up cleanup for blas64 PR ENH: f2py: add --f2cmap option for specifying the name of .f2py_f2cmap ENH: add support for ILP64 OpenBLAS (without symbol suffix)
2020-04-27Added py-numpy16 - the last version to support Python 2.7adam10-1/+1104
2020-04-27py-bottleneck: updated to 1.3.2adam3-22/+30
Bottleneck 1.3.2 Bug Fixes - Explicitly declare numpy version dependency in ``pyproject.toml`` for Python 3.8, fixing certain cases where ``pip install`` would fail. and ``@0xb0b`` for reporting. Bottleneck 1.3.1 Bug Fixes - Fix memory leak in :func:`bottleneck.nanmedian` with the default argument of ``axis=None``. - Add regression test for memory leak case Bottleneck 1.3.0 Project Updates - Bottleneck has a new maintainer, Christopher Whelan (``@qwhelan`` on GitHub). - Documentation now hosted at https://bottleneck.readthedocs.io - 1.3.x will be the last release to support Python 2.7 - Bottleneck now supports and is tested against Python 3.7 and 3.8. - The ``LICENSE`` file has been restructured to only include the license for the Bottleneck project to aid license audit tools. There has been no change to the licensing of Bottleneck. - Licenses for other projects incorporated by Bottleneck are now reproduced in full in separate files in the ``LICENSES/`` directory (eg, ``LICENSES/NUMPY_LICENSE``) - All licenses have been updated. Notably, setuptools is now MIT licensed and no longer under the ambiguous dual PSF/Zope license. - Bottleneck now uses :pep:`518` for specifying build dependencies, with per Python version specifications Enhancements - Remove ``numpydoc`` package from Bottleneck source distribution - :func:`bottleneck.slow.reduce.nansum` and :func:`bottleneck.slow.reduce.ss` now longer coerce output to have the same dtype as input - Test (tox, travis, appveyor) against latest ``numpy`` (in conda) - Performance benchmarking also available via ``asv`` - ``versioneer`` now used for versioning - Test suite now uses ``pytest`` as ``nose`` is deprecated - ``python setup.py build_ext --inplace`` is now incremental - ``python setup.py clean`` now cleans all artifacts - Compiler feature support now identified by testing rather than hardcoding - The ``BN_OPT_3`` macro allows selective use of ``-O3`` at the function level - Contributors are now automatically cited in the release notes Performance - Speed up :func:`bottleneck.reduce.anynan` and :func:`bottleneck.reduce.allnan` by 2x via ``BN_OPT_3`` - All functions covered by ``asv`` benchmarks - :func:`bottleneck.nonreduce.replace` speedup of 4x via more explicit typing - :func:`bottleneck.reduce.median` up to 2x faster for Fortran-ordered arrays Bug Fixes - Documentation fails to build on Python 3 - :func:`bottleneck.benchmark.bench` crashes on python 3.6.3, numpy 1.13.3 - :func:`bottleneck.nonreduce_axis.push` raises when :code:`n=None` is explicitly passed - :func:`bottleneck.reduce.nansum` wrong output when :code:`a = np.ones((2, 2))[..., np.newaxis]` same issue of other reduce functions - Silenced FutureWarning from NumPy in the slow version of move functions - Installing bottleneck onto a system that does not already have Numpy - Memory leaked when input was not a NumPy array - Tautological comparison in :func:`bottleneck.move.move_rank` removed Cleanup - The ``ez_setup.py`` module is no longer packaged - Building documentation is now self-contained in ``make doc`` - Codebase now ``flake8`` compliant and run on every commit - Codebase now uses ``black`` for autoformatting
2020-04-27py-asteval: updated to 0.9.18adam2-8/+9
0.9.18: clarify that 0.9.17 was the final version to support Py2.7 0.9.17: add 3.8 to travis config 0.9.16: auto-run codecov only if numpy is installed
2020-04-27tex-unicode-math{,-doc}: update to 0.8q texlive 2020 versionmarkd4-18/+16
2020-04-20Consistently use the interface type of the library.joerg4-1/+66
2020-04-18Add missing includes. Bump revision.joerg4-4/+31
2020-04-18py-numba: updated to 0.49.0adam3-368/+661
Version 0.49.0: This release is very large in terms of code changes. Large scale removal of unsupported Python and NumPy versions has taken place along with a significant amount of refactoring to simplify the Numba code base to make it easier for contributors. Numba’s intermediate representation has also undergone some important changes to solve a number of long standing issues. In addition some new features have been added and a large number of bugs have been fixed! IMPORTANT: In this release Numba’s internals have moved about a lot. A backwards compatibility “shim” is provided for this release so as to not immediately break projects using Numba’s internals. If a module is imported from a moved location the shim will issue a deprecation warning and suggest how to update the import statement for the new location. The shim will be removed in 0.50.0! Highlights of core feature changes include: Removal of all Python 2 related code and also updating the minimum supported Python version to 3.6, the minimum supported NumPy version to 1.15 and the minimum supported SciPy version to 1.0. Refactoring of the Numba code base. The code is now organised into submodules by functionality. This cleans up Numba’s top level namespace. Introduction of an ir.Del free static single assignment form for Numba’s intermediate representation An OpenMP-like thread masking API has been added for use with code using the parallel CPU backends For the CUDA target, all kernel launches now require a configuration, this preventing accidental launches of kernels with the old default of a single thread in a single block. The hard-coded autotuner is also now removed, such tuning is deferred to CUDA API calls that provide the same functionality The CUDA target also gained an External Memory Management plugin interface to allow Numba to use another CUDA-aware library for all memory allocations and deallocations The Numba Typed List container gained support for construction from iterables Experimental support was added for first-class function types
2020-04-15Needs py-pip and py-setuptools_scmjoerg1-1/+3
2020-04-12octave: fix a patchadam1-6/+6
2020-04-12octave: updated to 5.2.0adam7-52/+63
Summary of bugs fixed for version 5.2.0: Improvements - Lock amd and symbfact to avoid segmentation fault with SuiteSparse - Fix segfault when calculating patch vertex normals - Fix segfault in constructing patch normals for lighting - mat2str.m: Fix for logical matrix (add `transpose`). - improve file equality check for MS Windows - ismember.m: Fix second return argument when values are complex - findobj.m: Fix typo and correctly handle numeric properties - Correctly display integer types with format hex - Cast `winqueryreg` `REG_DWORD` types to `DWORD` - Fix numerous errors in `audiowrite` / `audioread` - print.m: Pass quoted path to `octave-svgconvert` - dir.m: Fix occasionally returning incorrect folder field - Stop segfault when calling 3-input form of `diag` with cell arrays - improve message for setting breakpoint in nonexistent function - Use framebuffer object for printing invisible Qt figures on mac - Escape backslash characters in `EPS` output - Use replacement characters to display non UTF-8 strings in figures - inputrc: Add warning about modification - documentation.cc (global_search): just return if query string is empty - Default (c)transpose for old style class arrays - Always reserve at least 1 element of storage for sparse matrices - Don't segfault at exit after reading malformed HDF5 file - Fix pause() with no arguments called on Windows - refactor minimum eigenvalue index search in `qp` - qp: fix obscure corner case when calculating `qp` caused by a typo - use `std::streampos` instead of `std::ios::streampos`. - fix `pause` and `kbhit` with glibc 2.28 end-of-file state behavior - Fix return of left-handed vectors when inputs are complex - waitbar.m: Fix hang when using `createcancelbtn` property - shading.m: Fix unexpected error when multiple `hggroups` present - dec2base.m: Correctly handle zero matrix input - eliminate duplicate graphics callback object stack - Improve performance when closing figures - Reduce time to process `pkg -forge list` by 30X - Feed `fig2dev` with PDF files converted from svg - Improve error checking for `iconv_open`. - poly.m: fix the fix for input of complex conjugate pairs - mkoctfile: use the `TMPDIR` environment variable if set - line.m: Fix creation of unwanted axes - Accept 4-input form for quiver3 - pkg.m: restore installing packages from zip archives - axis.m: Fix issues with `equal` argument - ordeig.m: Do not fail on 1x1 matrices GUI - fix editors search and replace in selection - undo a complete replace all action in the editor - fix missing unlock of mutex when gui editor is not present - prevent unnamed editor tab from being closed by `rmdir` - prevent editor from closing files that are not affected by `rmdir` - fix creating keyword files for editor autocompletion - fix dock widget position when being dragged out of the main window Build system / Tests - configure: prevent overlinking when SUNDIALS is disabled - use QtCore and Qt5Gui modules instead of Qt5OpenGL - make building with Qt4 QGLWidget work again - build: adapt mk-qthelp.pl to changes in Texinfo 6.7. - BIST should not rely on the current directory to be writable. - bp-table.cc: Fix test. - maint: Make old style class tests Matlab compatible. - Add BIST tests for `unlink` Documentation - fix many spelling errors - help: improve documentation for the `startup.m` user script. - avoid conflict with @seealso macro in Texinfo 6.6 - mk-doc-cache.pk: Also translate `@seealso` here. - __makeinfo__.m: Also translate `@seealso` here - polar.m: Document that input is expected to be in radians - gallery.m: Fix typo in docstring for poisson - Add 1024x1024 Octave logo icon - add `.editorconfig` for default file encoding and indentation. - octave.doap: Copy localized (short) description over from appdata - org.octave.Octave.appdata.xml: Add German and French description - languages/*.ts: updated language files - quiver3.m: Fix texinfo typo in docstring from cset f7b10bd40045. - doc: fix `.` Operator Index entry to build with Texinfo 6.7. - doc: improve example of using global variables from oct-files. - doc: Create en-dashes and em-dashes correctly in documentation. - doc: expand `page_output_immediately` doc string, mention `page_screen_output`. - doc: Clarify documentation of `history_file` - doc: Improve documentation for `get_help_text`, `get_help_text_from_file`. - doc: Add documentation for `GNUTERM` variable used with gnuplot - doc: Improve `contour` docstring example - doc: Add example to show limitations of the given BLAS integer size. - doc: Improve documentation of sparse functions. - doc: Improve spelling of `bicgstab` - doc: Remove `Map` functions `keys`, `values`, `remove` from unimplemented list - doc: Add function index entry for alias `inverse` - add content rating declaration to AppStream metadata - doc: Remove stray semicolons from `pie`, `pie3` calling forms. - doc: Redo documentation for `rats` function. - doc: Specify `position` property for text objects is a three-element vector - doc: state that Octave can only load HDF5 files created by itself - doc: Fix names of options in `issorted` - doc: Small fixes to `mat2cell` and `polyeig` docstrings - doc: Improve example code for using `-pdflatexstandalone` - doc: Refer to "root object" rather than "root figure object". - doc: Place cross-reference from list of graphics object properties back to object - doc: document the newline character in warning messages - doc: pkg versions can be more general than "x.y.z"
2020-04-12qhull: updated to 2019.1adam3-13/+19
Qhull 2019.1 (7.3.2): bugs fixed - Fixed broken HTML links identified by Firefox Link Analyzer. Replace old links with web.archive.org (archive.is) - Moved live code out of qh_NOtrace regions - qh-faq.htm#learn: fixed the Delaunay examples (rbox generates a square, but does not generate two squares) - geom2_r.c/qh_maxmin: qh_scalelast for 'd Qbb' to qh.MAXabs_coord instead of qh.MAXwidth [J.R. Roussel, D. Sterratt] - geom2_r.c/qh_maxsimplex: estimate targetdet as prevdet*qh.MAXwidth, set mindet to 10*qh_RATIOmaxsimplex [J.R. Roussel, D. Sterratt] - global_r.c/qh_freebuild: in high-D, simplicial facets may use long memory - global_r.c/qh_initqhull_globals: set qh.DOcheckmax and qh.KEEPnearinside if qh.APPROXhull ('Wn', rbox 100 D3 W1e-1 | qhull QJ W1e-1 Tv) - libqhull_r.c/qh_addpoint: add qh.lastmerges to qh_addpoint trace - libqhull_r.c/qh_buildcone_mergepinched: called qh_partitioncoplanar with uninitialized 'dist' - libqhull_r.c/qh_build_withrestart: if qh_joggle_restart, set v.partitioned for qh.del_vertices. Avoid error in qh_delvertex from qh_freebuild - libqhull_r.c/qh_partitionall: if qh.KEEPnearinside, qh_partitionpoint for non-outside pointset - libqhull_r.c/qh_partitioncoplanar: define qh.coplanar_apex to prevent infinite loop of repartition point - merge_r.c/qh_checkzero: check opposite vertex against -2*qh.DISTround to account for qh_distround and 'Rn' - merge_r.c/qh_comparevisit: fix compare of 'unsigned int' - poly2_r.c/qh_findbestfacet: for Delaunay triangulations, may be an adjacent triangle [F. Drielsma] Updated discussion at qh-code.htm/findfacet and examples at user_eg_r.c and user_eg2_r.c - poly2_r.c/qh_findfacet_all: do not search visible facets, and if qh_NOupper, f.upperdelaunay facets [P. Virtanen, S. Dominguez, J.Arkin] - poly2_r.c/qh_initialhull: reverse orientation if clearly flipped. qh_ERRsingular if maybe flipped [S. Caron, P. Virtanen, D. Sterratt, others] - poly2_r.c/qh_initialhull: test for coplanar qh.interior_point with 'allerror' (dist >= -qh.DISTround) [S. Caron, P. Virtanen, D. Sterratt, others] - poly2_r.c/qh_initialhull: call qh_joggle_restart for qh.DELAUNAY if cospherical input sites [P. Celba] - poly2_r.c/qh_initbuild: qh_scalelast for 'd Qbb' to qh.MAXabs_coord instead of qh.MAXwidth [J.R. Roussel, D. Sterratt] - random_r.c/qh_argv_to_command_size: returned size may be < 1. If so, rbox.c/main reports an error - rboxlib_r.c/qh_rboxpoints: report error for coincident, regular points 'rbox r C1,1e-13' - stat_r.c/qh_initstatistics: missing call to qh_allstatistics for qh_newstats. qh_checkoutput ignored - userprintf.c/qh_fprintf: fixed 2019.0.1 bug if 'qh_qh' is NULL for qh_QHpointer - Qhull.cpp/outputQhull: fixed missed conditional if qh.GOODvertex == -qh.GOODpoint documentation and web site - Add notes that Qhull 64-bit code uses 50% more memory than 32-bit code. This slows down Qhull. A future version of Qhull will include memory and performance improvements for 64-bit code. - Add the Qhull identity pipeline to html/qhalf.htm - Enabled PDF download from portal.acm.org - Error exit from qhull,qconvex,etc. returns a 6xxx error code if available - Rename md5sum files back to *-tgz.md5sum and *-zip.md5sum Both Firefox and IE convert *.tgz.md5sum and *.zip.md5sum as .tgz/.zip files - html/qh-code.htm: add sections on debugging Qhull - html/qh-faq.html: reviewed startup questions for Qhull - html/qh-opto.htm: the centrum indices for option 'i' are one more than the centrum indices for option 'Ft' build and test - CMakeLists.txt: export QhullTargets for config.cmake.in and find_package [tamasmeszaros] - eg/q_benchmark: add 'uname -a' for host identification - eg/q_test.bat: update for changes to q_test removed redirects on every line (invoke with redirection) - eg/q_test: add some tests for option warnings and errors - Makefile: add dependencies for 'install', 'test', 'qtest', and 'qconvex-prompt' [V. Delecroix] - Makefile: set -m32 as default for gcc builds. Uses less memory on 64-bit hosts. disable with 'make M32=' - Remove 'unused parameter' warnings if qh_NOmerge - Tested qh_QHpointer, qh_NOmem, qh_NOmerge, qh_NOtrace, qh_KEEPstatistics 0 - Tested qh_NOmerge with qh_NOtrace and qh_KEEPstatistics 0. Simpler code, fewer errors, 35% smaller - Tested random qh_distplane and qh_setplane qh_RANDOMdist 1e-5, qh_RANDOMdist 1e-13 - user_r.h: define qh_RANDOMdist to perturb all calls to qh_distplane, qh_newfacetplane by 'Rn' other changes - Replace 'unsigned' with 'unsigned int', for consistency - Replace warning 6263 with 7079 in qconvex_r.c, etc. - Resolved C warnings for -Wsign-conversion, mostly type casts Continue to use -Wno-sign-conversion for C++ due to countT, boolT, int vs. size_t mismatches - Reviewed gcc 8.1.0 warnings. See Makefile CC_WARNINGS and CXX_WARNINGS Type cast (int)sizeof(facetT),etc. as needed - Reviewed error and warning codes for multiple instances of the same code - Reviewed file references such as libqhull_r.c vs libqhull.c - Reviewed multi-line, error message width for 80 column output - global.c/qh_initflags: Add 'qh_QHpointer' to qh_option for option list - geom2_r.c/qh_joggleinput: report QH6010, current joggle is too large, as qh_ERRinput - geom2_r.c/qh_maxsimplex: report error if qh.MAXwidth not defined, for estimating the determinates - geom2_r.c/qh_minmax: if tracing, log min and max coordinates [J.R. Roussel, D. Sterratt] - io_r.c/qh_printbegin: rewrote warning QH7053, use 'p' or 'o' for Voronoi centers - io_r.c/qh_produce_output2: fflush to maintain execution order of output - libqhull_r.c/qh_buildtracing: change qh.lastfacets, qh.lastplanes, qh.lastdist to 'int' - libqhull_r.c/qh_build_withrestart: change error QH6229, 101 attempts to joggle input, to qh_ERRinput and 51 attempts - merge_r.c/qh_renameridgevertex: report error if old vertex not found - poly_r.c/qh_update_vertexneighbors: use qh_setappend instead of qh_setunique for newly created facets (restores 2015.2 code) - poly2_r.c/qh_check_point: limit 'Tv' errors to qh_MAXcheckpoint per facet - poly2_r.c/qh_initialhull: rewrite QH6329 for cospherical input sites and add qh_printvertexlist [P. Celba] - qset_r.c/qh_setlarger_quick: optimize qh_setlarger for quick memory - rboxlib_r.c/qh_rboxpoints2: fix strncat to resolve -Wstringop-truncation warning [D. Sterratt] increased command[] to 250 and decreased seedbuf[] to 50 - user_r.h/qh_JOGGLEmaxretry: reduced from 100 iterations to 50 iterations - user_r.c/qh_new_qhull: allow "qhull" for the qhull_cmd parameter - user_eg_r.c/findDelaunay: qh_findbestfacet returns the closest Delaunay triangle or an adjacent triangle - user_eg2_r.c/findDelaunay: qh_findbestfacet returns the closest Delaunay triangle or an adjacent triangle - userprintf_r.c/qh_fprintf: set qh.last_errcode for MSG_ERROR (6000) codes - RboxPoints::qh_fprintf_rbox: print error message to stderr if cpp_object undefined Qhull 2019.0.1 (7.3.1) Qhull options - Disallow unknown options and other option errors, allow option warnings with 'Qw' - New option 'Qallow-short': allow input with fewer or more points than coordinates - New option 'Qwarn-allow': allow option warnings - New option 'TAn': stop after adding n vertices - New option 'Tf': flush after qh_fprintf for debugging segfaults - New experimental option 'Q14-merge-pinched-vertices': merge pinched vertices due to dupridge Excessive merging leads to additional topological and wide facet errors - New experimental option 'Q15-check-duplicates': check for duplicate ridges with the same vertices - Option 'Q1' merges by mergetype/angle instead of mergetype/distance, coplanar merges first In 2015.2 by default, Qhull merged by mergetype/angle with lower angles first - Option 'TP-1' turns on tracing after qh_buildhull and qh_postmerge It traces qh_check_maxout, qh_prepare_output, qh_triangulate, qh_voronoi_center - Options '--help', '-?', and '--' display a short help message Qhull does not support '--' options. Git for Windows v2.21.0 does not support isatty ('winpty qhull' is OK) - Allow Qhull extra options Q12 and Q14 for qconvex/qdelaunay/qhalf/qvoronoi - Allow option 'QRn' (random rotate) for qdelaunay/qhalf/qvoronoi - Alphabetize short prompts (qhull .) - Alphabetize long prompts w/ normal options first (qhull -) - Alphabetize option help by option code - Add 'Number of merged pinched vertices' to summary ('s') - Add 'TI file' to help message for qconvex/qdelaunay/qhalf/qvoronoi - Dropped experimental option 'Q16-simplex-merge'. Centrums are required for convexity - Error if 'd', 'v', or 'Hn,n,n' are used together - Ignore missing space after 'TI' or 'TO'. Allow 'TIinputfile' and 'TOoutputfile' - Prefix option errors and warnings with 'qhull option error:' or 'qhull option warning:' - Remove hidden option 'normals' from help prompts for qdelaunay, qhalf, qvoronoi - Reword short prompts to 'Pdrop-d0:0D0', 'QbBound-0:0.5', and 'off_format' - Updated qh-quick.htm, 'qhull -', and 'qhull .' for 'T...' and 'Q...' options. Separate help prompts for 'T options' and 'Trace options' - Use '-' instead of '_' for word delimiter in option names Qhull builds, scripts, and debugging - Add qh.traceridge, qh.traceridge_id, and qh_delridge for debugging by ridge in qh_fprintf - Check for qhull with 'which qhull' instead of 'qhull', avoids hangs in git bash - Makefile/CC_WARNINGS: Add instructions for collecting and testing gcc warnings - Makefile/benchmark: Add q_benchmark, 10 runs of each - Makefile/clean: Delete linked files from src/libqhull and src/libqhull_r, as in their Makefiles - Makefile/help: Add 'make all' to 'make help' prompts - Makefile/qhullx: Add test of testqset and testqset_r - Makefile/qhullx: Create bin/ directory if missing - Makefile/test: Add user_eg3 help prompt - Makefile/testall: Add q_benchmark, 1 run of each - eg/make-vcproj.sh/qhullx: Build bin/qhullx from src/qhullx/ for comparison - eg/q_benchmark: new script for measuring Qhull precision and performance with qtest.sh - eg/q_test: update error messages for missing bin/ and bin/libqhull_r.dll - eg/q_test: tests of Tf and TAn - eg/qhull-zip.sh: add note that it requires road-script.sh from http://www.qhull.org/road/ - eg/qhull-zip.sh: check source depencencies for prompts - eg/qtest.sh: new script for testing Qhull and summarizing its trace files - libqhull_r/Makefile: install target creates $INCDIR/libqhull_r and $LIBDIR - libqhull_r.h/qh_ERRdebug: define error status for debugging exits Qhull fixed bugs - html/qh-optp.htm: options 'PDk' and 'Pdk' require option 'Pg' to return the closest facet - Makefile/qhullx: fixed qhullx compile of testqset and testqset_r - PointCoordinates.cpp/appendPoints: read 'in >> p' instead of 'in >> p >> ws' otherwise istream fails if file does not end with whitespace - geom_r.c/qh_findbesthorizon: fix test for Zfindjump (bestdist jumps by searchdist) - geom_r.c/qh_findbesthorizon: increased minsearch of coplanar facets if ischeckmax and more than 100 neighbors - geom_r.c/qh_sethyperplane_det: dropped precision statistic Znearlysingular (it's Zminnorm) - geom2_r.c/qh_detjoggle: use 'En' (roundoff) if specified by options - geom2_r.c/qh_distround: move adjustment for 'Rn' into qh_distround. 'QJn' did not account for 'Rn' - global_r.c/qh_checkflags: handle two-digit Q options in hiddenflags - global_r.c/qh_freebuild: do not delete unattached ridges, otherwise a double delete. Should not occur - global_r.c/qh_option: fix potential overflow of qh.qhull_options - global_r.c/qh_initflags: fixed incorrect indentation for option 'TO file' - global_r.c/qh_initqhull_start2: set qh.NOerrexit True - io_r.c/qh_eachvoronoi: allow fp==NULL for QhullQh::qh_fprintf to appendQhullMessage - io_r.c/qh_readpoints: error if 'd' or 'v' with 'H', otherwise memory corruption - libqhull_r.h/qh.MERGEvertices: fixed comment, 'Q3' disables this feature - libqhull_r.c/qh_buildcone_onlygood: cannot qh_delfacet inside FORALLnew_facets - libqhull_r.c/qh_partitionpoint: if qh.NARROWhull and bestdist>qh.MINoutside and isnewoutside, may need to move facet to qh.newfacet_list for qh.next_facet - libqhull_r.c/qh_printsummary: Zdistconvex is for checking convexity instead of testing convexity - libqhull_r.c/qh_findhorizon: coplanar horizon if apex >= -qh.MAXcoplanar, instead of > -qh.MAXcoplanar - mem_r.c/qh_memcheck: define null qh_memcheck if qh_NOmem - merge_r.c/qh_mergecycle_all: turn on tracing, 'TMn', for coplanar horizon merge of one facet - merge_r.c/qh_mergefacet2d: do not create ridges for neighborB. The opposite neighbor may be tested - merge_r.c/qh_test_centrum_merge: coplanar if centrum >= -qh.centrum_radius, instead of > -qh.centrum_radius - merge_r.c/qh_test_nonsimplicial_merge: coplanar if centrum >= -qh.centrum_radius, instead of > -qh.centrum_radius - merge_r.c/qh_tracemerging: fixed "qhull has merged" prompt - merge_r.c/qh_willdelete: empty f.neighbors and f.ridges, references are no longer valid, will be deleted - poly_r.c/qh_checkflipped: flipped if qh.interior_point >= -qh.DISTround, instead of > -qh.DISTround - poly2_r.c/qh_addfacetvertex: initialize vertex_i in case f.vertices is empty - poly2_r.c/qh_check_maxout: need to include qh.DISTround when checking f.maxoutside with qh.maxoutside - poly2_r.c/qh_checkconvex: fix documentation for qh_DATAfault. It is restricted to qh_initialhull with qh.ZEROcentrum - poly2_r.c/qh_checkconvex: coplanar if >= -qh.DISTround, instead of > -qh.DISTround - poly2_r.c/qh_checkconvex: check simplicial ridges if !qh.MERGING && qh.ZEROcentrum (unlikely or not-possible) - poly2_r.c/qh_findgood: restrict qh.GOODvertex ('QVn') to good facets - poly2_r.c/qh_findgood_all: unless 'Pg', ignore qh.GOODclosest for nearest facet to thresholds. 'PdD' reports 0 good facets - poly2_r.c/qh_replacefacetvertex: initialize vertex_n in case f.vertices is empty - qconvex.c,qdelaunay.c,etc.: remove automatic 'Qx' for 5-d and higher. Set by default in qh_initqhull_globals - qvoronoi_r.c/qh_prompt: remove disabled option 'TRn' from help prompt - random_r.c/qh_argv_to_command: fixed missing braces, same behavior as before - rbox_r.c: return 0 exit status on prompt - rboxlib.c/qh_rboxpoints: use qh_fprintf_stderr instead of qh_fprintf_rbox. Avoid qh_errexit_rbox - rboxlib.c/qh_rboxpoints2: clear rbox_inuse on exit, as in error exit - rboxlib_r.c/qh_rboxpoints2: incorrect string arg to 'rbox error: unknown flag' - rboxlib_r.c/qh_rboxpoints2: remove superfluous qh_free(*simplex) - stat_r.c/qh_initstatistics: due to NOerrors cannot use qh_fprintf, use qh_fprintf_stderr instead - stat_r.c/qh_stddev: return sqrt of absolute value - testqset_r.c: if qh_NOmem, do not call qh_memfreeshort - user_eg_r.c,user_eg2_r.c: if qh_NOmem, call qh_freeqhull(qh, qh_ALL) instead of qh_memfreeshort - user_eg3_r.cpp: return 0 exit status on prompt Qhull fixed bugs introduced by 2018.0 - libqhull_r.c/qh_addpoint: clear qh.retry_addpoint on success - libqhull_r.c/qh_buildcone_mergepinched: cannot qh_delfacet inside FORALLnew_facets - libqhull_r.c/qh_partitioncoplanar: clear qh.repart_id on exit for infinite recursion detect - geom2_4.c/qh_detmaxoutside: move qh_option call to qh_errexit - merge_r.c/mergetypes[]: fixed order by mergeT - merge_r.c/qh_all_vertexmerges: fix retryfacet for repeated calls - merge_r.c/qh_forcedmerge: f.dupridge retained after MRGdupridge hides MRGdegenerate in qh_test_degen_neighbors and qh_test_redundant_neighbors - merge_r.c/qh_postmerge: call qh_freemergesets - merge_r.c/qh_test_nonsimplicial_merge: clear iscoplanar if isconcave and any vertex is not coplanar - poly2_r.c/qh_triangulate: move reset of f.degenerate into delete facet loop Qhull C++ - QhullFacet.cpp: add hasNext() and hasPrevious() - QhullVertex.cpp: add hasNext() and hasPrevious() - qhulltest-64/QhullFacet_test.cpp,QhullVertex_test.cpp: test hasNext, hasPrevious, and hasPrevious iterator - user_eg3.cpp: use 'qhull-cout' for option 'Tz' (trace to cout) Qhull errors, warnings, and tracing - Error messages rewritten and error codes changed (see qh_fprintf) Renumbered Qhull warnings as 'QH7xxx' if needed (e.g., QH9375) Reused unused error and message codes ("def counters" in user_r.h/msgcode) Please check your code if it depends on error and message codes - Initialize variables to avoid potential uninitialized warnings - Use qh_ERRqhull for Qhull internal errors, "qhull internal error (qh_...): ..." - Use 'defined()' for qh_QHpointer_dllimport and qh_dllimport (-Wundef) - Use 'defined()' for __cplusplus, _MSC_VER, __MWERKS__, and __POWERPC__ (-Wundef) - Use 'Early exit' instead of 'At a premature exit' - libqhull_r.h/qh_ERRother: define error status for non-internal errors - libqhull_r.h/qh_ERRtopology: define error status for qhull topology errors - libqhull_r.h/qh_ERRwide: define error status for wide facets due to qhull precision error - geom_r.c/qh_setfacetplane: add qh_printfacet if qh.tracefacet for qh.tracefacet_id - global_r.c/qh_freebuild: record previousfacet/ridge/vertex in case a memory fault occurs - global_r.c/qh_init_flags: improved error message for 'cannot open file' for 'TO f' - global_r.c/qh_init_flags: improved warning message for 'unknown option' - global_r.c/qh_initbuild: add trace options to "Trace level ..." - global_r.c/qh_initqhull_globals: error if qh.VORONOI without qh.DELAUNAY - global_r.c/qh_initqhull_globals: error if numpoints is <1 or >qh_POINTSmax (INT_MAX-16) - global_r.c/qh_lib_check: use C-style comment for _CrtSetBreakAlloc - io_r.c/qh_printfacetheader: add "max" if f.merges==qh_MAXnummerge (511) - io_r.c/qh_readpoints: error if numinput is <1 or >qh_POINTSmax (INT_MAX-16) - io_r.c/qh_readpoints: moved up 'chop' of qh.rbox_command to report short input errors - io_r.c/qh_readpoints: replace warning QH7073 with error message. Override with Qallow-short - libqhull_r.c/qh_addpoint: if 'T4', check qh.visible_list before qh_partitionvisible - libqhull_r.c/qh_buildtracing: add delta cpu/facets/hyperplanes to QH1049 'qh_addpoint: add p%d...' - merge_r.c/qh_findbest_pinchedvertex: replaced assert with a 'qhull internal error' - merge_r.c/qh_check_dupridge: drop advice about reporting wide merges for error QH6271 - merge_r.c/qh_mark_dupridges: topological error if multiple dupridges for one facet, not handled - merge_r.c/qh_test_nonsimplicial_merge: if Qx and many merges, replace centrum_radius with maxoutside - poly_r.c/qh_matchneighbor: drop extra warnings for duplicate facet (QH7084) - poly_r.c/qh_matchnewfacets: check for simplicial facets on qh.newfacet_list, otherwise memory overwrite by qh_matchneighbor - poly2_r.c/qh_checkconvex: identify simplicial checking of vertices (qh.ZEROcentrum, default) - poly2_r.c/qh_checkconvex: distinguish above from coplanar for non-convex initial simplex - poly2_r.c/qh_checkconvex: "non-convex initial simplex" for qh_joggle_restart - poly2_r.c/qh_checkfacet: check f.id and f.vertex_visit - poly2_r.c/qh_checklists: check and repair facet and vertex lists for infinite loops or overwritten facets/vertices - poly2_r.c/qh_checkpolygon: if qh.NEWfacets, check visible facets for empty f.neighbors and f.ridges - poly2_r.c/qh_checkpolygon: add qh_checkvertex - poly2_r.c/qh_checkpolygon: check f.previous and v.previous links (for C++) - poly2_r.c/qh_checkvertex: add parameters for allchecks and waserror (like qh_checkfacet) - poly2_r.c/qh_checkvertex: check v.id and v.vertex_visit - qvoronoi.c/hidden_options: hide option 'FM'. Maple output is not available - rboxlib_r.c/qh_rboxpoints: use QH6... for error codes instead of warnings - stat_r.c/qh_stddev: check for division by 0 - testqset_r.c: send all messages, except 'OK' to stderr - testqset_r.c: reassign unique QH8nnn codes, 8001/etc in use by libqhull_r - user_r.c/qh_printhelp_internal: add descriptive message for a Qhull internal error - user_r.c/qh_printhelp_narrowhull: add option 'Qs' (search all points) - user_r.c/qh_printhelp_topology: add descriptive message for a Qhull topology error - user_r.c/qh_printhelp_wide: add descriptive message for a wide merge error - user_eg_r.c: add fflush and '=======' markers to interleave stdout with stderr (MSYS2) - user_eg2_r.c: add fflush and '=======' markers to interleave stdout with stderr (MSYS2) - user_eg2_r.c: add warning about calling qh_addpoint after qh_qhull with 'QJ' (joggle) Qhull documentation and statistics - Add Index to README.txt - Add links to Wiki, http://github.com/qhull/qhull/wiki - Add links to local home page (index.htm) - Add GitHub, Scholar, and Code to navigation bar on home page - Change numbered fixups to 'QH110nn FIX:' and renamed as MSG_FIX - Chrome and Internet Explorer load '.c' and '.h' files as text/html from the Web These browsers along with Firefox do not read disk files as text/html Changed 'Functions' link from a local link to "http://www.qhull.org/src/libqhull_r/index.htm" - Document implicit parameters at end of parameter list with '/* qh. ... */' - Limit maximum length of literals to 1800 characters - Move 'called by' comments to start of 'notes:' - Option 'QRn' appends the actual QRn to the summary line ('s') - Remove references to 'Qx' in 5-d and higher. It is automatically set by default. - Replace references to 'src/libqhull/*.htm' with reentrant 'src/libqhull_r/*_r.htm' - Retitle 'Qhull internals' as 'Qhull code' - Use 'nearly adjacent vertices' instead of 'nearly coincident points' - Use 'dupridge' instead of 'duplicate ridge' for ridges with more than two neighbors. Reserve 'duplicate ridge' for two ridges with the same vertices. - Various edits to documentation (*.htm) - Update copyright to 2019 - eg/q_test: Add examples to q_test "front ends" - html/index.htm: Add 'Limitations of merged facets' to the Qhull manual - html/index.htm: Add links to qdelaun_f.htm and qvoroin_f.htm (furthest-site) - html/qconvex.htm,etc: update synopsis and options - html/qh-code.htm: Add 'qh_addpoint' trace to 'Performance of Qhull' - html/qh-code.htm: Add a section on 'Debugging Qhull' with subsections for error codes, infinite loops, trace options, core dumps, segfaults, qtest.sh, and memory errors. - html/qh-impre.htm: Add 'Topological error' to 'Precision problems' - html/qh-impre.htm: Add 'Topological errors' to 'Limitations of merged facets' - html/qh-impre.htm: Add "narrow range of values" to 'Uneven dimensions' - html/qh-optq.htm: Add note to 'QJ' (joggle) about calls to qh_addpoint - html/qh-optt.htm#'Tc': 'T4' makes additional checks beyond check-frequently - html/qhalf.htm: Split the first example to clarify the "feasible point" for qhalf - html/qhalf.htm: Use 'feasible point' instead of 'interior point' for halfspace intersection Same as Qhull::feasiblePoint and distinguished from qh.interior_point - html/qhull.man,qhull.txt: updated man page - libqhull_r.h: update documentation for PRINTgood, GOODpoint, GOODvertex - qhull_ra.h: include '<limits.h>'. Used for INT_MAX,etc. - stat.h: add zretryadd stats for qh_merge_pinchedvertices - stat.h: add zvertextests for distance tests of vertices for merging - libqhull_r/*: document global variables as 'qh.' - libqhull_r/*: fix libqhull/... links - libqhull_r.c/qh_partitioncoplanar: reword Zpartflip as Zpartcorner with updated messages - merge_r.c/qh_merge_degenredundant: rename statistic Zneighbor to Zredundant - stat_r.c: new or reworded statistics for Zpartcorner, Zparthidden, Zparttwisted, Zpinchduplicate, Zpinchedvertex - stat_r.c: rename 'renamed vertex statistics' to 'statistics for vertex merges' - stat_r.c: add merge statistics for MRGconcavecoplanar - stat_r.c/qh_allstat: add ' ' before parens in statistics - user_eg_r.c,user_eg2_r.c: rewrote prompt Qhull changes - For simple initializers, remove spaces (e.g., 'for (i=0; ...') - Normalize spacing for libqhull - Normalize spacing as 'for (i=0; ...' for simple initializers - Review use of '>' for comparison to negative quantities. Usually should be '>=', e.g., for coplanar tests - Reassigned 8xxx and 9xxx codes as trace codes if traced at T3 or lower - Remove unused variables - Test floating point values with 0.0 instead of 0 (qh.premerge_centrum,lower_bound,upper_bound) - Use SETsecondt_() and SETelemt_() instead of SETelem_() when assigning to a specific type - Use '(type *)' for pointer casts. It makes the cast more visible. - Use void for empty parameter lists - geom_r.c/qh_findbesthorizon: mark startfacet visited, avoids unnecessary visit - geom2_r.c/qh_detmaxoutside: determine precision target qh.MAXoutside (option _maxoutside at error exit) replaces 'maxoutside' computation - geom2_r.c/qh_maxsimplex: add ratio test for false narrow hull with search for prevdet [J. Romain] - global_r.c/qh_freebuild: free global sets first instead of last, otherwise dangling references - io_r.c/qh_compare_nummerge: renamed qh_compare_facetmerge due to conflict with merge_r.h - io_r.c/qh_compare_vertexpoint: remove this routine. Not usable in libqhull_r due to 'qh' - io_r.c/qh_printfacetridges: do not print ridge if already printed (e.g., duplicate neighbors for merge of simplicial facets) - io_r.c/qh_produce_output2: fflush of qh.fout to show statistics and verification after output (MSYS2) - libqhull_r.h/qh.ALLOWwide: rename qh.NOwide to qh.ALLOWwide for 'Q12-allow-wide' - libqhull_r.h/qh.IGNOREpinched: replaced qh.MERGEpinched with reversed meaning - libqhull_r.c/qh_addpoint: 'TVn' stops after qh_all_vertexmerges, not before qh_resetlists. Matches documentation - merge_r.h/MRGdupridge: renamed MRGridge - merge_r.h/qh_ANGLEconcave,etc.: replace mergetype indicators with qh_ANGLEnone and m.mergetype - merge_r.c/qh_all_vertexmerges: replace qh_getmergeset_initial with qh_getmergeset. No facets created - merge_r.c/qh_all_vertexmerges: move while loop into qh_all_vertexmerges - merge_r.c/qh_append_mergeset: ignore merges into a MRGdegen facet (f.degenerate) - merge_r.c/qh_buildcone_mergepinched: clean up code and design documentation - merge_r.c/qh_compare_anglemerge: renamed qh_compareangle and sort by mergetype as well as angle - merge_r.c/qh_compare_facetmerge: merge MRGcoplanar before MRGconcave, opposite order from 2015.2 sort MRGanglecoplanar by angle - merge_r.c/qh_degen_redundant_facet,qh_test_degen_neighbors: moved into qh_mergefacet due to qh_willdelete - merge_r.c/qh_getmergeset,qh_getmergeset_initial: set f.tested after testing ridges, not before - merge_r.c/qh_next_facetmerge: split out next merge from qh_all_merges and replace qsort - merge_r.c/qh_findbest_pinchedvertex,qh_findbest_ridgevertex: dropped selecting merge direction via qh_vertex_isbelow. Expensive and evidence mixed - merge_r.c/qh_freemergesets,qh_initmergesets: pair qh_initmergesets with qh_freemergesets - merge_r.c/qh_freemergesets,qh_initmergesets: drop the 'all' parameter - merge_r.c/qh_mergecycle_all: move qh_merge_degenredundant, etc. to end from qh_premerge - merge_r.c/qh_mergesimplex: delete facet1 from facet2.neighbors at end, not at ridge - merge_r.c/qh_post_merges,qh_all_merges: remove setting of v.delridge on all vertices, not needed - merge_r.c/qh_premerge: replace parameter 'apex' with 'apexpointid'. Only used for logging - merge_r.c/qh_premerge: add qh.newfacet_list as an implicit parameter - merge_r.c/qh_test_degen_neighbors: split from qh_degen_redundant_neighbors and qh_mergefacet - merge_r.c/qh_test_redundant_neighbors: split from qh_degen_redundant_neighbors and qh_mergefacet - poly_r.c/qh_attachnewfacets: clear f.neighbors and f.ridges for visible facets, will be deleted - poly_r.c/qh_matchnewfacets: remove call to qh_checkflipped_all for qh.FORCEoutput, checked by qh_setfacetplane - poly_r.c/qh_update_vertexneighbors: renamed qh_updatevertices. A future version will optimize qh_update_vertexneighbors_cone: optimization for cone of new facets - poly2_r.c/qh_checkconvex: immediate skip of f.tricoplanar facets. Tricoplanar facets not tested. Called before qh_triangulate - poly2_r.c/qh_initbuild: move 'create sentinels' before qh_createsimplex so that qh.vertex_tail is v0 like f0 - poly2_r.c/qh_initialvertices: cleaned up code for qh_INITIALmax (8-d) - poly2_r.c/qh_makenewfacets: clear f.neighbors and f.ridges for visible facets, will be deleted - poly2_r.c/qh_matchdupridge: optimize dupridge for flipped->!flipped, good merge, not !flipped->flipped - poly2_r.c/qh_matchdupridge_coplanarhorizon: dropped routine. Results same or slightly worse. Complex addition, rarely used - poly2_r.c/qh_triangulate: rename facet and vertex lists as triangulated_facet_list,etc. qh_triangulate does not maintain qh.newfacet_list or qh.visible_list - poly2_r.c/qh_triangulate: ignore qh.NEWfacets. qh_checkpolygon is not valid inside qh_triangulate - rboxlib_r.c/qh_rboxpoints2: split function from qh_rboxpoints to avoid -Wclobbered - stat_r.c/qh_printstatlevel: remove unused NULLfield Qhull C++ changes - Use NULL instead of 0 for pointers - Remove old style casts such as '(size_t)...' - Define initializers for default constructors even if explicitly assigned - Update static initializer list for s_empty_facet, s_empty_ridge, s_empty_vertex - Coordinates.h: default constructor clears coordinate_array - Coordinates.h: default iterator constructor clears i - RboxPoints.cpp/qh_fprintf_rbox: add empty default to switch statement
2020-04-12Fix build with newer automake.joerg1-3/+2
2020-04-12Revbump all Go packages after default version switch to 1.14.bsiegert2-4/+4
2020-04-12Recursive revision bump after textproc/icu updateadam11-22/+22
2020-04-11py-xarray: updated to 0.15.1adam3-15/+34
v0.15.1: This release brings many new features such as :py:meth:`Dataset.weighted` methods for weighted array reductions, a new jupyter repr by default, and the start of units integration with pint. There's also the usual batch of usability improvements, documentation additions, and bug fixes. Breaking changes - Raise an error when assigning to the ``.values`` or ``.data`` attribute of dimension coordinates i.e. ``IndexVariable`` objects. This has been broken since v0.12.0. Please use :py:meth:`DataArray.assign_coords` or :py:meth:`Dataset.assign_coords` instead. New Features - Weighted array reductions are now supported via the new :py:meth:`DataArray.weighted` and :py:meth:`Dataset.weighted` methods. See :ref:`comput.weighted`. - The new jupyter notebook repr (``Dataset._repr_html_`` and ``DataArray._repr_html_``) (introduced in 0.14.1) is now on by default. To disable, use ``xarray.set_options(display_style="text")``. - Added support for :py:class:`pandas.DatetimeIndex`-style rounding of ``cftime.datetime`` objects directly via a :py:class:`CFTimeIndex` or via the :py:class:`~core.accessor_dt.DatetimeAccessor`. - Support new h5netcdf backend keyword `phony_dims` (available from h5netcdf v0.8.0 for :py:class:`~xarray.backends.H5NetCDFStore`. - Add partial support for unit aware arrays with pint. - :py:meth:`Dataset.groupby` and :py:meth:`DataArray.groupby` now raise a `TypeError` on multiple string arguments. Receiving multiple string arguments often means a user is attempting to pass multiple dimensions as separate arguments and should instead pass a single list of dimensions. - :py:func:`map_blocks` can now apply functions that add new unindexed dimensions. - An ellipsis (``...``) is now supported in the ``dims`` argument of :py:meth:`Dataset.stack` and :py:meth:`DataArray.stack`, meaning all unlisted dimensions, similar to its meaning in :py:meth:`DataArray.transpose`. - :py:meth:`Dataset.where` and :py:meth:`DataArray.where` accept a lambda as a first argument, which is then called on the input; replicating pandas' behavior. - ``skipna`` is available in :py:meth:`Dataset.quantile`, :py:meth:`DataArray.quantile`, :py:meth:`core.groupby.DatasetGroupBy.quantile`, :py:meth:`core.groupby.DataArrayGroupBy.quantile` Bug fixes - Fix :py:meth:`Dataset.interp` when indexing array shares coordinates with the indexed variable - Fix recombination of groups in :py:meth:`Dataset.groupby` and :py:meth:`DataArray.groupby` when performing an operation that changes the size of the groups along the grouped dimension. By `Eric Jansen <https://github.com/ej81>`_. - Fix use of multi-index with categorical values - Fix alignment with ``join="override"`` when some dimensions are unindexed. - Fix :py:meth:`Dataset.swap_dims` and :py:meth:`DataArray.swap_dims` producing index with name reflecting the previous dimension name instead of the new one - Use ``dask_array_type`` instead of ``dask_array.Array`` for type checking. - :py:func:`concat` can now handle coordinate variables only present in one of the objects to be concatenated when ``coords="different"``. - xarray now respects the over, under and bad colors if set on a provided colormap. - :py:func:`coarsen` now respects ``xr.set_options(keep_attrs=True)`` to preserve attributes. :py:meth:`Dataset.coarsen` accepts a keyword argument ``keep_attrs`` to change this setting. :pull:`3801`) By `Andrew Thomas <https://github.com/amcnicho>`_. - Delete associated indexes when deleting coordinate variables. - Fix :py:meth:`xarray.core.dataset.Dataset.to_zarr` when using `append_dim` and `group` simultaneously. - Fix html repr on :py:class:`Dataset` with non-string keys Documentation - Fix documentation of :py:class:`DataArray` removing the deprecated mention that when omitted, `dims` are inferred from a `coords`-dict. - Improve the :py:func:`where` docstring. - Update the installation instructions: only explicitly list recommended dependencies Internal Changes - Remove the internal ``import_seaborn`` function which handled the deprecation of the ``seaborn.apionly`` entry point - Don't test pint integration in combination with datetime objects. - Change test_open_mfdataset_list_attr to only run with dask installed - Preserve the ability to index with ``method="nearest"`` with a :py:class:`CFTimeIndex` with pandas versions greater than 1.0.1 - Greater flexibility and improved test coverage of subtracting various types of objects from a :py:class:`CFTimeIndex`. By `Spencer Clark <https://github.com/spencerkclark>`_. - Update Azure CI MacOS image, given pending removal. - Remove xfails for scipy 1.0.1 for tests that append to netCDF files - Remove conversion to :py:class:`pandas.Panel`, given its removal in pandas in favor of xarray's objects. v0.15.0: This release brings many improvements to xarray's documentation: our examples are now binderized notebooks (`click here <https://mybinder.org/v2/gh/pydata/xarray/master?urlpath=lab/tree/doc/examples/weather-data.ipynb>`_) and we have new example notebooks from our SciPy 2019 sprint (many thanks to our contributors!). This release also features many API improvements such as a new :py:class:`~core.accessor_dt.TimedeltaAccessor` and support for :py:class:`CFTimeIndex` in :py:meth:`~DataArray.interpolate_na`); as well as many bug fixes. Breaking changes - Bumped minimum tested versions for dependencies: - numpy 1.15 - pandas 0.25 - dask 2.2 - distributed 2.2 - scipy 1.3 - Remove ``compat`` and ``encoding`` kwargs from ``DataArray``, which have been deprecated since 0.12. Instead, specify the ``encoding`` kwarg when writing to disk or set the :py:attr:`DataArray.encoding` attribute directly. - :py:func:`xarray.dot`, :py:meth:`DataArray.dot`, and the ``@`` operator now use ``align="inner"`` (except when ``xarray.set_options(arithmetic_join="exact")``; New Features - Implement :py:meth:`DataArray.pad` and :py:meth:`Dataset.pad`. - :py:meth:`DataArray.sel` and :py:meth:`Dataset.sel` now support :py:class:`pandas.CategoricalIndex`. - Support using an existing, opened h5netcdf ``File`` with :py:class:`~xarray.backends.H5NetCDFStore`. This permits creating an :py:class:`~xarray.Dataset` from a h5netcdf ``File`` that has been opened using other means - Implement ``median`` and ``nanmedian`` for dask arrays. This works by rechunking to a single chunk along all reduction axes. - :py:func:`~xarray.concat` now preserves attributes from the first Variable. - :py:meth:`Dataset.quantile`, :py:meth:`DataArray.quantile` and ``GroupBy.quantile`` now work with dask Variables. - Added the ``count`` reduction method to both :py:class:`~core.rolling.DatasetCoarsen` and :py:class:`~core.rolling.DataArrayCoarsen` objects. - Add ``meta`` kwarg to :py:func:`~xarray.apply_ufunc`; this is passed on to :py:func:`dask.array.blockwise`. - Add ``attrs_file`` option in :py:func:`~xarray.open_mfdataset` to choose the source file for global attributes in a multi-file dataset :pull:`3498`). By `Julien Seguinot <https://github.com/juseg>`_. - :py:meth:`Dataset.swap_dims` and :py:meth:`DataArray.swap_dims` now allow swapping to dimension names that don't exist yet. - Extend :py:class:`~core.accessor_dt.DatetimeAccessor` properties and support ``.dt`` accessor for timedeltas via :py:class:`~core.accessor_dt.TimedeltaAccessor` - Improvements to interpolating along time axes - Support :py:class:`CFTimeIndex` in :py:meth:`DataArray.interpolate_na` - define 1970-01-01 as the default offset for the interpolation index for both :py:class:`pandas.DatetimeIndex` and :py:class:`CFTimeIndex`, - use microseconds in the conversion from timedelta objects to floats to avoid overflow errors. Bug fixes - Applying a user-defined function that adds new dimensions using :py:func:`apply_ufunc` and ``vectorize=True`` now works with ``dask > 2.0``. - Fix :py:meth:`~xarray.combine_by_coords` to allow for combining incomplete hypercubes of Datasets - Fix :py:func:`~xarray.combine_by_coords` when combining cftime coordinates which span long time intervals - Fix plotting with transposed 2D non-dimensional coordinates. - :py:meth:`plot.FacetGrid.set_titles` can now replace existing row titles of a :py:class:`~xarray.plot.FacetGrid` plot. In addition :py:class:`~xarray.plot.FacetGrid` gained two new attributes: :py:attr:`~xarray.plot.FacetGrid.col_labels` and :py:attr:`~xarray.plot.FacetGrid.row_labels` contain :py:class:`matplotlib.text.Text` handles for both column and row labels. These can be used to manually change the labels. - Fix issue with Dask-backed datasets raising a ``KeyError`` on some computations involving :py:func:`map_blocks` - Ensure :py:meth:`Dataset.quantile`, :py:meth:`DataArray.quantile` issue the correct error when ``q`` is out of bounds - Fix regression in xarray 0.14.1 that prevented encoding times with certain ``dtype``, ``_FillValue``, and ``missing_value`` encodings - Raise an error when trying to use :py:meth:`Dataset.rename_dims` to rename to an existing name - :py:meth:`Dataset.rename`, :py:meth:`DataArray.rename` now check for conflicts with MultiIndex level names. - :py:meth:`Dataset.merge` no longer fails when passed a :py:class:`DataArray` instead of a :py:class:`Dataset`. - Fix a regression in :py:meth:`Dataset.drop`: allow passing any iterable when dropping variables - Fixed errors emitted by ``mypy --strict`` in modules that import xarray. - Allow plotting of binned coordinates on the y axis in :py:meth:`plot.line` and :py:meth:`plot.step` plots - setuptools is now marked as a dependency of xarray Documentation - Switch doc examples to use `nbsphinx <https://nbsphinx.readthedocs.io>`_ and replace ``sphinx_gallery`` scripts with Jupyter notebooks. - Added :doc:`example notebook <examples/ROMS_ocean_model>` demonstrating use of xarray with Regional Ocean Modeling System (ROMS) ocean hydrodynamic model output. - Added :doc:`example notebook <examples/ERA5-GRIB-example>` demonstrating the visualization of ERA5 GRIB data. `Stephan Siemen <https://github.com/StephanSiemen>`_. - Added examples for :py:meth:`DataArray.quantile`, :py:meth:`Dataset.quantile` and ``GroupBy.quantile``. - Add new :doc:`example notebook <examples/apply_ufunc_vectorize_1d>` example notebook demonstrating vectorization of a 1D function using :py:func:`apply_ufunc` , dask and numba. - Added example for :py:func:`~xarray.map_blocks`. Internal Changes - Make sure dask names change when rechunking by different chunk sizes. Conversely, make sure they stay the same when rechunking by the same chunk size. - 2x to 5x speed boost (on small arrays) for :py:meth:`Dataset.isel`, :py:meth:`DataArray.isel`, and :py:meth:`DataArray.__getitem__` when indexing by int, slice, list of int, scalar ndarray, or 1-dimensional ndarray. - Removed internal method ``Dataset._from_vars_and_coord_names``, which was dominated by ``Dataset._construct_direct``. - Replaced versioneer with setuptools-scm. Moved contents of setup.py to setup.cfg. Removed pytest-runner from setup.py, as per deprecation notice on the pytest-runner project. - Use of isort is now enforced by CI.
2020-04-11py-networkx: updated to 2.4adam3-24/+128
2.4: This release is the result of 6 months of work with over 200 commits by 67 contributors. Highlights include: - Remove deprecated code from 1.x - Support for Python 3.8 - Switched to pytest for testing - Last release to support Python 3.5 New Functions: - barycenter functions - Bethe Hessian matrix function - Eulerian Path methods - group centrality measures - subgraph monomorphisms - k-truss algorithms - onion decomposition - resistance distance - asteroidal triples - non-randomness measures - linear prufing - minimum weight bipartite matching - Incremental closeness centrality - ISMAGS subgraph isomorphism algorithm - create chordal graph of a graph New generators - Binomial tree generator - Directed joint degree generator - Random internet AS graph generator New for Layouts - spiral node layout routine - support for 3d layouts Improvements ------------ - allow average shortest path to use Floyd-Warshall method - improve read/write of GML, GEXF, GraphML - allow string or json object as input to jit_graph - attempt to allow numpy.array input in place of lists in more places - faster strongly connected components - faster Floyd-Warshall Optimization - faster global efficiency - faster transitive closure - fix unionfind; betweenness_subset; lexico-topo-sort; A*; inverse_line_graph; async label propagation; edgelist reading; Gomory-Hu flow method; label_propagation; partial_duplication; shell_layout with 1 node in shell; from_pandas_edgelist - Documentation improvement and fixes API Changes ----------- A utility function is_list_of_ints became is_bunch_of_ints and now tests int(item)==item instead of isinstance(_, int) This allows e.g. floats whose values are integer. Added utility make_list_of_ints to convert containers of integer values to lists of integers Deprecations ------------ Removed functions (marked as deprecated in NetworkX 2.1): - attracting_component_subgraphs - connected_component_subgraphs - weakly_connected_component_subgraphs - strongly_connected_component_subgraphs - biconnected_component_subgraphs - See docs for component functions for how to get subgraphs. Graph Object methods removed (marked as deprecated 2.1) - G.add_path - G.add_cycle - G.add_star - G.nodes_with_selfloops - G.number_of_selfloops - G.selfloop_edges - These are now NetworkX functions, e.g. nx.add_star(G, 5) - G.node --> use G.nodes - G.fresh_copy --> use G.__class__ Remove old names for graphview functions. - ReverseView - SubGraph - SubMultiGraph - SubMultiDiGraph - SubDiGraph - GraphView - DiGraphView - MultiGraphView - MultiDiGraphView - MultiReverseView - Use reverse_view, subgraph_view and generic_graph_view.
2020-04-03*: recursive bump for poppler-0.87wiz1-2/+2
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-08 08:37:03 +0000