v0.8.1

What's Changed

• Add DOI badge to README by @jalvesz in #1097
• Simplified code to support modularized features + update README by @jvdp1 in #1098
• Fix: Race Condition in Parallel Fypp Preprocessing by @JAi-SATHVIK in #1103
• Addition of a prefix to the names of the libraries by @jvdp1 in #1102
• fix (linalg_iterative): remove pure attribute from stdlib_linop inner_product by @jalvesz in #1106
• Modularization of remaining modules by @jvdp1 in #1100
• Updated version and changelog.md towards v0.8.1 by @jvdp1 in #1107

Full Changelog: v0.8.0...v0.8.1

v0.8.0

What's Changed

• linalg: enable 80-bit extended precision for whole library, xdp by @perazz in #839
• Update author picture in API documentation webpage by @jvdp1 in #842
• linalg: Matrix Inverse by @perazz in #828
• Rewrite fypp directive to avoid the use of removesuffix in stdlib_math_meshgrip by @jvdp1 in #847
• Conversion functions from degrees to radians and vice-versa. by @MilanSkocic in #845
• sort_index: use of only int_index iterators inside sort_index by @jvdp1 in #848
• linalg: Cholesky factorization by @perazz in #840
• Remove msys from CI and install fypp via msys2 package by @minhqdao in #854
• CI: upgrade to upload-artifact@v4 by @perazz in #872
• API: make LAPACK-backed APIs since Release <=0.7.0 stable by @perazz in #873
• linalg: QR factorization by @perazz in #832
• Swap by @jalvesz in #869
• Fix missing argument to linalg_invert_inplace_ call in linalg_invert_split by @cyrilgandon in #880
• fix(factorial): result should be of the default kind, not double default kind by @cyrilgandon in #876
• linalg: vector norms by @perazz in #871
• loadtxt/savetxt: do not require space after last entry by @perazz in #877
• svd - check the size of storage space by @jvdp1 in #881
• fix fypp_deployment build and set maxrank default to 4 as in cmake by @jalvesz in #882
• Fix warnings [-Wunused-xxx] from compilation by @cyrilgandon in #879
• Sparse algebra support with OOP API by @jalvesz in #760
• Fix Gaussian distribution formula in documentation by @moriglia in #889
• loadtxt: increase length of internal buffers by @perazz in #893
• Revert "call error_stop -> error stop" by @perazz in #894
• promote ascii functions to elemental by @jalvesz in #886
• linalg: Matrix norms by @perazz in #885
• CI: deprecate macos-12 by @perazz in #905
• Add [install] section to fpm.toml by @moriglia in #906
• linalg eye: allow generalized return type and kind by @perazz in #902
• linalg: hermitian by @perazz in #896
• linalg: Moore-Penrose pseudo-inverse (pinv) by @perazz in #899
• linalg: introduce 64-bit integer size BLAS/LAPACK implementation (ilp64) by @perazz in #888
• linalg: Schur decomposition by @perazz in #892
• sparse linalg: simplify SELLC spmv kernel by @jalvesz in #912
• linalg eig: generalized eigenvalue problem by @perazz in #909
• Remove mingw32 by @chuckyvt in #914
• Remove unused by @jalvesz in #917
• linalg: refactor blas/lapack modules by @jalvesz in #913
• Proposition for fixing CMake procedure by @jvdp1 in #923
• limit Intel Classic CI to Ubuntu 22.04 by @perazz in #925
• Fix output format in sparse example by @jalvesz in #928
• Proposal to upgrade number_of_columns by @perazz in #926
• Fix generalized eig rwork size by @perazz in #929
• ieeeck: use ieee_arithmetic intrinsics by @perazz in #933
• Removed unused variables by @jvdp1 in #935
• strings: join, to_c_char by @perazz in #936
• error: implement general state_type handler by @perazz in #938
• system: subprocessing interface by @perazz in #911
• Remove other derived type from hashmap by @chuckyvt in #843
• intrinsics module with alternative implementations by @jalvesz in #915
• io: getfile by @perazz in #939
• fix: gamma functions by @wassup05 in #943
• system: OS type query by @perazz in #942
• system: null_device by @perazz in #945
• small fix for loadtxt by @wassup05 in #949
• fix: hbar value by @wassup05 in #950
• replaced make_directory with file(MAKE_DIRECTORY ...) by @jfdev001 in #956
• remove unnecessary merge(.true.,.false.,) by @perazz in #959
• system: is_directory by @perazz in #946
• Support csv file reading and writing in loadtxt and savetxt. by @zoziha in #958
• system: delete_file by @perazz in #966
• specialfunctions: generalize gamma precision by @perazz in #969
• hashmap-init-update by @chuckyvt in #974
• stdlib_linalg: declare the diag function as pure by @zoziha in #980
• linalg: slightly increase test tolerance by @perazz in #985
• Add findBLAS support to CMakeLists.txt by @zoziha in #844
• feat: Add github collaboration script and workflow documentation by @jalvesz in #987
• feat: activation intrinsics for neural networks by @jalvesz in #860
• modified stats exponential distribution procedures to use loc and scale by @sebastian-mutz in #991
• fix: collab diagram layout fix by @jalvesz in #996
• fix: make ansi examples from documentation executable by @jalvesz in #997
• fix: do-concurrent syntax by @wassup05 in #998
• fix: to_num_from_stream pointer position correction by @jalvesz in #1000
• GCC 15.1 fixes by @perazz in #1008
• docs: correct header level in rad2deg by @stigh in #1010
• Regrouped lapack handling functions by @loiseaujc in #1013
• fix: index handling in chomp by @wassup05 in #1012
• fix: windows compilation of hash maps tests and increase stack size avoiding segfaults by @jalvesz in #988
• Tridagonal matrices and associated spmv kernel by @loiseaujc in #957
• stdlib_system: essential path functionality by @wassup05 in #999
• helper functions for Filesystem error handling by @wassup05 in #1015
• fix: complex dot_product formulation by @jalvesz in #1017
• Make collaboration script executable by @wassup05 in #1018
• promote other ascii functions to elemental by @wassup05 in #977
• feat: creating and removing empty directories by @wassup05 in #1011
• feat: get_cwd and set_cwd by @wassup05 in #1014
• Addition of sort_adj to sort an rank 1 array in the same order as an input array by @jvdp1 in #849
• feat: exists to return the type of a path by @wassup05 in #1026
• feat: [linalg] add iterative solvers by @jalvesz in #994
• Modularize BLAS and LAPACK modules in CMake to make it easier to use in external projects by @eduardz1 in #1033
• feat : [linalg] add bi-conjugate gradient stabilized method by @jalvesz in #1034
• submodule procedures declaration made compatible with older cmake version by @BhoomishGupta in #1041
• Matrix exponential by @loiseaujc in #1038...

v0.7.0

What's Changed

• Loadtxt real format update to list directed by @chuckyvt in #805
• Fix: copy_a deallocation in determinant by @perazz in #815
• lstsq: return correct array size by @perazz in #818
• linalg: remove pure attr from procedures using openmp runtime by @perazz in #826
• Rename int size to int_index by @jvdp1 in #824
• Codata constants. by @MilanSkocic in #800
• linalg: fix typo in least-squares complex space allocation by @perazz in #830
• Support of int32 index array in sort_index by @jvdp1 in #829
• linalg: Singular Value Decomposition by @perazz in #808
• ci: use fortran-setup for GCC, Intel LLVM and Intel Classic by @gnikit in #834
• Fixes for extended and quad precision checking. Add CI cheks with fpm by @jalvesz in #821
• stdlib_*laset add array bound checks in presence of hardcoded input address by @perazz in #836
• Hashap key generic interfaces by @chuckyvt in #827
• linalg: Eigenvalues and Eigenvectors by @perazz in #816
• Updated version and changelog.md towards v0.7.0 by @jvdp1 in #841

New Contributors

• @MilanSkocic made their first contribution in #800
• @gnikit made their first contribution in #834

Full Changelog: v0.6.1...v0.7.0

v0.6.1

What's Changed

• Fix example starts with by @jvdp1 in #813
• lk -> llk in *sytrd_sy2sb by @perazz in #812
• Update version and Changelog.md towards v0.6.1 by @jvdp1 in #814

Full Changelog: v0.6.0...v0.6.1

v0.6.0

What's Changed

• rotm, rotmg documentation by @perazz in #795
• Update linalg.md and example_blas_gemv.f90 by @zoziha in #797
• Update to include int32 hashmap keytype by @chuckyvt in #778
• linalg: determinant by @perazz in #798
• macos-latest -> macos-12 by @perazz in #807
• fpm deployment enhancement by @jalvesz in #802
• linalg: least squares by @perazz in #801
• least squares: relax test tolerance by @perazz in #809
• linalg: solve by @perazz in #806
• Update for the new release v0.6.0 by @jvdp1 in #811

New Contributors

• @chuckyvt made their first contribution in #778

Full Changelog: v0.5.0...v0.6.0

v0.5.0

What's Changed

• Upgrade Intel-classic compiler in macOS CI by @perazz in #777
• Provide BLAS, LAPACK backends and interfaces by @perazz in #772
• BLAS, LAPACK: Refactor comments for FORD compatibility by @perazz in #786
• add implicit none to example programs that do not have it by @gareth-nx in #780
• fpm branch: .f90 -> .F90 if preprocessed by @perazz in #787
• Ford documentation: improvements by @jvdp1 in #784
• Fix in the procedure remove of chaining hashmaps by @jvdp1 in #788
• Add brief explanation on how to include stdlib in a Makefile by @jannisteunissen in #781
• Module str2num and related tests by @jvdp1 in #789
• fpm deployment: workaround for case-insensitive fs by @perazz in #790
• Provide state/error handling for linear algebra by @perazz in #774
• Preparation for 0.5.0 release by @jvdp1 in #779

New Contributors

• @jannisteunissen made their first contribution in #781

Full Changelog: v0.4.0...v0.5.0

v0.4.0

What's Changed

• PR related to #729, gamma function of pure imaginary returns correct value by @banana-bred in #730
• Addition of GCC 13 to CI by @jvdp1 in #737
• Proposal to fix the procedure move for string_type by @jvdp1 in #736
• Fixed incorrect intent of the variable "conflict" on hashmaps in document by @Daisuke897 in #739
• faster to_lower and to_upper by @jalvesz in #733
• Added a procedure for getting all the keys in a hashmap by @degawa in #741
• Fix ci for windows and Intel ifx (#7) by @jvdp1 in #752
• Fix undefined use of types by @harmenwierenga in #747
• Fix bits() in bitset_type before initialization by @jvdp1 in #753
• Minor changes related to fypp prepocessing by @jvdp1 in #758
• Replaced btest() by popcnt() in bit_count_large() by @PierUgit in #756
• Add CLI option for building with fpm by @jalvesz in #757
• Chmod stdlib_hashmap F90 files by @jvdp1 in #762
• Proposal for a reference string to number conversion facility in stdlib by @jalvesz in #743
• Add .dat and .stream file extensions to .gitignore on master branch by @minhqdao in #768
• Add .gitignore to stdlib-fpm by @minhqdao in #769
• Add meshgrid subroutine in stdlib_math by @Ivanou34 in #764
• Fix aliased string move interface by @perazz in #773
• Preparation for 0.4.0 release by @jvdp1 in #775

New Contributors

• @banana-bred made their first contribution in #730
• @Daisuke897 made their first contribution in #739
• @jalvesz made their first contribution in #733
• @harmenwierenga made their first contribution in #747
• @PierUgit made their first contribution in #756
• @minhqdao made their first contribution in #768
• @Ivanou34 made their first contribution in #764

To use this release in fpm depend on

[dependencies]
stdlib.git = "https://github.com/fortran-lang/stdlib"
stdlib.commit = "8853ecdea0de04db9bcb73ab285ff6ec8153e909"

Full Changelog: v0.3.0...v0.4.0

stdlib-0.3.0

What's Changed

• Add hint for building error with make by @Carltoffel in #656
• fixed 32-bit integer overflow in stdlib_io_npy by @degawa in #655
• Remove support for manual make builds by @milancurcic in #657
• Fix erroneous gaussian quadrature points in gauss_legendre by @hsnyder in #660
• Gamma special function by @Jim-215-Fisher in #625
• Feature: loadtxt skiprows and max_rows by @MuellerSeb in #652
• Hash maps by @wclodius2 in #611
• Readme update by @gareth-nx in #659
• Some corrections in the specs by @jvdp1 in #663
• Fixes issue with rvs_normal where it lacks the elemental property by @14NGiestas in #665
• Avoid overflow in to_string for -huge(1)-1 by @awvwgk in #667
• Mv the directory src/tests to test by @jvdp1 in #669
• Extraction of the demo programs from the specs by @jvdp1 in #662
• Fixes #673: update docs CI by @zoziha in #681
• Add terminal and color escape sequences by @awvwgk in #580
• cross product of two vectors by @St-Maxwell in #687
• Update GCC versions in OS matrix by @milancurcic in #698
• New zfill function to left-pad a string with zeros by @ecasglez in #689
• convert pdf_norm and cdf_norm to pure while improving scale check by @HugoMVale in #679
• Check for equality between Fortran and C logicals by @jvdp1 in #692
• Kronecker Product addition to stdlib_linalg by @adenchfi in #700
• Fixed access violation in a type-bound procedure of open_hashmap_type by @degawa in #707
• Fixes #706 - failure due unassigned value to a intent(out) variable. by @14NGiestas in #708
• fixed unassigned value to a intent(out) variable in parse_header by @degawa in #711
• fixed a problem with dat and npy files in example dir not being deployed by @degawa in #713
• change stdlib-random function rtol to ishftc by @Euler-37 in #714
• Remove unused module stdlib_error from module stdlib_stats_distribution_normal by @HugoMVale in #716
• Add radix_sort by @0382 in #712
• Extension of PR#679 to stdlib_stats_distribution_exponential by @HugoMVale in #717
• Improve doc + examples of stdlib_stats_distribution_normal by @HugoMVale in #718
• fix doc stats-distribution-normal and stats-distribution-exponential by @HugoMVale in #721
• Add warning for breaking change in fpm by @Carltoffel in #715
• Remove reference to F202X by @jvdp1 in #722
• Fix broken link to fpm in README.md by @perazz in #725
• PR related to #726 and #723 by @jvdp1 in #727
• Support sorting arrays of bitsets by @degawa in #723
• Preparation for 0.3.0 release by @jvdp1 in #728

New Contributors

• @degawa made their first contribution in #655
• @MuellerSeb made their first contribution in #652
• @ecasglez made their first contribution in #689
• @HugoMVale made their first contribution in #679
• @adenchfi made their first contribution in #700
• @Euler-37 made their first contribution in #714
• @0382 made their first contribution in #712
• @perazz made their first contribution in #725

To use this release in fpm depend on

[dependencies]
stdlib.git = "https://github.com/fortran-lang/stdlib"
stdlib.commit = "df1e2f0ed0cbe2fbd9c1f20dcb5bd1a4bba95bb2"

Full Changelog: v0.2.1...v0.3.0

stdlib-0.2.1

What's Changed

• Only set Fortran arguments for Fortran compiler by @awvwgk in #637
• Fix linking issue with shared libraries by @awvwgk in #642
• Export symbols on Windows and set PIC flag for Unix by @awvwgk in #646
• Bugfix release version 0.2.1 by @awvwgk in #651

To use this release in fpm depend on

[dependencies]
stdlib.git = "https://github.com/fortran-lang/stdlib"
stdlib.commit = "d01ce7c358625182891ad7fb84777cc36e32ce19"

---

Discourse thread: https://fortran-lang.discourse.group/t/3297
Twitter: https://twitter.com/fortranlang/status/1517565551345811457

Full Changelog: v0.2.0...v0.2.1

stdlib-0.2.0

What's Changed

• Make linspace and logspace pure functions by @milancurcic in #549
• Probability Distribution and Statistical Functions -- Uniform Distribution Module by @Jim-215-Fisher in #272
• Pin specific fpm version by @awvwgk in #558
• fix some FORD links by @jvdp1 in #556
• Add testing module to allow better structuring of test suites by @awvwgk in #494
• Minor update pure/elemental in string_type module by @zoziha in #562
• Make support for quadruple precision optional by @awvwgk in #565
• Create a call for reviewers pull request template by @awvwgk in #566
• Update error in case fypp preprocessor is not found by @awvwgk in #578
• Add module for handling version information of stdlib by @awvwgk in #579
• Selection algorithms by @gareth-nx in #500
• Update of stdlib_stats.md by @jvdp1 in #586
• Add routines for saving/loading arrays in npy format by @awvwgk in #581
• Update changelog by @awvwgk in #590
• Error on no tests in CTest by @awvwgk in #588
• stdlib_selection: correction of typos and addition of some checks by @jvdp1 in #585
• Fix compilation errors with makefiles due to command-line variable assignments by @fiolj in #591
• Probability Distribution and Statistical Functions -- Normal Distribution Module by @Jim-215-Fisher in #273
• Replace the call to sort by select in stdlib_stats_median by @jvdp1 in #584
• Probability Distribution and Statistical Functions -- Uniform Distribution Module by @Jim-215-Fisher in #593
• Minor update to makefile installation instructions by @gareth-nx in #594
• Rename references to default branch by @awvwgk in #596
• Fix iomsg allocation in save_npy by @ivan-pi in #600
• [stdlib_math] add is_close routines. by @zoziha in #488
• Add getline to read whole line from formatted unit by @awvwgk in #597
• [stdlib_math] add arg/argd/argpi by @zoziha in #498
• Implement trueloc/falseloc by @awvwgk in #603
• Revised Hash functions incorporating changes in the main Stdlib repository. by @wclodius2 in #573
• Consistent spec titles by @milancurcic in #609
• Fixed tables in stdlib_hash_procedures.md by @jvdp1 in #610
• [stdlib_linalg] matrix property checks by @ghbrown in #499
• Ignore hash testing binaries and logs by @ghbrown in #613
• Probability Distribution and Statistical Functions -- Exponential Distribution Module by @Jim-215-Fisher in #276
• Made format constants public by @lewisfish in #617
• [stdlib_math] Add function diff by @zoziha in #605
• [stdlib_math] Minor update to stdlib_math module and document by @zoziha in #624
• doc: fix some typos by @cuishuang in #632
• option to disable testing by setting BUILD_TESTING to OFF by @stigh in #629
• Preparation for 0.2.0 release by @awvwgk in #631

New Contributors

• @lewisfish made their first contribution in #617
• @cuishuang made their first contribution in #632
• @stigh made their first contribution in #629

To use this release in fpm depend on

[dependencies]
stdlib.git = "https://github.com/fortran-lang/stdlib"
stdlib.commit = "ea68aae040ac1e9488d20982d3bd04ec150ae49c"

Twitter: https://twitter.com/fortranlang/status/1505632594716774408

Full Changelog: v0.1.0...v0.2.0