Changelog

Unreleased

Changes to the existing modules

• Preserve rectangular matrix dimensions in coo2dense and csr2dense. #1228

• Change in module stdlib_intrinsics

  • Fix inverted MASK inclusion in stdlib_sum and stdlib_sum_kahan. Update masked Kahan sum and compensation only for selected elements.

• Fixed declaration order in external BLAS/LAPACK interfaces for compilation with gfortran under -std=f2018 #1210

• Change in module stdlib_system

  • Added set_environment_variable and delete_environment_variable #1187

Changes to the existing build system

• Fixed absolute paths in generated pkg-config file for external BLAS/LAPACK #1109

Version 0.8.1

Full release notes available at v0.8.1 tag.

Changes to the existing build system

• Changes in the build system

  • Completed refactoring of stdlib into modular components using CMake #1098 #1100 #1102

  • Resolve race condition in parallel FYPP preprocessing #1103

Changes to the existing modules

• Change in module stdlib_lingal_iterative_solvers

  • Remove pure attribute from stdlib_linop inner_product #1106

Changes to the existing documentation

• Change in README

  • Addition of a DOI badge #1097

  • Update of the table with flags for modularizing stdlib #1098

Version 0.8.0

Full release notes available at v0.8.0 tag.

Summary of the most relevant pull requests (more than 100 total)

• changes in module stdlib_linalg and related

  • Matrix Inverse #828

  • Vector norms #871

  • Matrix norms #885

  • Cholesky factorization #840

  • QR factorization #832

  • Generalized eigenvalue problem #909

  • Moore-Penrose pseudo-inverse #899

  • Schur decomposition #892

  • Matrix exponential #1038

  • Hermitian matrices #896

  • Iterative solvers #994

  • Bi-conjugate gradient stabilized method #1034

  • Pivoting QR decomposition #1045

  • Equality-constrained least-squares solver #1046

  • Generalized lagtm routine with support for arbitrary alpha and beta #1068

• changes in module stdlib_system

  • delete_file #966

  • is_directory #946

  • get_cwd and set_cwd #1014

  • exists #1026

  • null_device #945

  • subprocessing interface #911

• changes in module stdlib_math

  • Conversion functions: degrees-to-radians and vice versa #845

• changes in module stdlib_intrinsics

  • Extended intrinsic matmul for additional flexibility #951

• changes in module stdlib_io

  • Changes to loadtxt/savetxt #877 #958

• changes in module stdlib_sorting

  • sort_index: use only int_index iterators #848

  • Addition of sort_adj #849

• improvements and fixes to existing modules

  • Fixes for factorial results, warnings, argument alignment, etc. #876 #880 #879 #929

  • Fix sparse algebra support and examples #760 #928

• changes in CI/CD configurations

  • Upgrade CI/CD configurations #854 #925 #1052 #1072 #1094

• build improvements

  • Changes in CMake configurations #844 #923

  • Modularization #1033 #1050 #1066 #1081

• changes in documentation

  • Added tables summarizing preprocessing macros and flags #1083

  • Updated macOS architecture in README #1059

  • Added code coverage support #1039

Version 0.7.0

Full release notes available at v0.7.0 tag.

• new module stdlib_constants #800

  • Many mathematical constants and most common physical (codata) constants

Changes to existing scripts and modules

• changes in CI

  • Use of fortran-setup for GCC, Intel LLVM and Intel Classic #834

• change in module stdlib_hashmaps

  • Support of hash map key generic interfaces #827

• changes in module stdlib_io

  • Addition of a Fortran format specifier in loadtxt #805

• changes in module stdlib_linalg

  • Support of extended and quad precision checking #821

  • Several fixes #815 #818 #826 #830 #836

  • New procedures for Eigenvalues and Eigenvectors computation: eig, eigh, eigvals, eigvalsh #816

  • New procedures for Singular Value Decomposition: svd, svdvals #808

• changes in module stdlib_sorting

  • Renamed variable from int_size to int_index #824

  • Support of int32 index array in sort_index #829

Version 0.6.1

Full release notes available at v0.6.1 tag.

Changes to existing scripts and modules

• changes in module stdlib_linalg_lapack

  • Renamed variable for compiler compliance #812

• change of the format in some example programs #813

Version 0.6.0

Full release notes available at v0.6.0 tag.

• new script fypp_deployment.py to support fpm in combination with fypp files #802

Changes to existing scripts and modules

• change in module stdlib_hashmap_wrappers

  • addition of int32 hashmap key type #778

• changes in module stdlib_linalg

  • addition of the procedure det to compute determinants #798

  • addition of the procedures lstsq and lstsq_space #801 #809

  • addition of the procedures solve and solve_lu #806

• change in module stdlib_linalg_blas

  • addition of the documentation for rotm and rotmg #795

• use of macOS 12 in macOS CI #807

Changes to existing documentation

• Improvement of the documentation linalg #797

Version 0.5.0

Full release notes available at v0.5.0 tag.

• new module stdlib_linalg_state #774

  • new derived type: linalg_state_type

  • new procedure: linalg_error_handling

Changes to existing scripts and modules

• addition of implicit none to all example programs #780

• change in module stdlib_hashmaps

  • fix the procedure remove_chaining_entry #788

• change in module stdlib_linalg

  • addition of the BLAS/LAPACK backends and interfaces #772

• change in module stdlib_str2num

  • fix the procedure to_${k1}$_from_stream #789

• upgrade of the Intel-classic compiler in macOS CI #777

Changes to existing documentation

• Improvement of the documentation #784 #781 #786

• Improvement of the support of fpm #787 #790

Version 0.4.0

Full release notes available at v0.4.0 tag.

• new module stdlib_str2num #743

  • new procedures: to_num, to_num_from_stream

Changes to existing scripts and modules

• change in .gitignore

  • addition of the file extensions .dat and .stream #768

  • addition of .gitignore to stdlib-fpm #769

• change in CI/CD

  • support of GCC 13 #737

  • support of Intel compiler ifx #752

• change in script fpm-deployment.sh

  • changes to facilitate fypp preprocessing for the fpm deployment #758

• change in module stdlib_ascii

  • Improved procedures to_lower and to_upper #733

• change in module stdlib_bitsets

  • initialization in bitset_type #753

  • improved procedure bit_count_large #756

• change in module stdlib_hashmaps

  • new procedure get_all_keys #741

  • new file permissions #762

• change in module stdlib_math

  • new procedure meshgrid #764

• change in module stdlib_specialfunctions_gamma

  • fix procedure gamma #730

• change in module stdlib_string_type

  • fix procedure move #736 #773

• change in SpookyV2Test.cpp

  • Fix undefined use of types #747

Changes to the existing documentation

• change in the specs stdlib_hashmaps

  • Correction of an intent of the variable “conflict” #739

• change in README.md

  • instructions to build stdlib with fpm through the fpm-deployment.sh script #757

Version 0.3.0

Full release notes available at v0.3.0 tag.

• new modules stdlib_hashmap_wrappers and stdlib_hashmap #611

  • new procedures in stdlib_hashmap_wrappers: copy_key, copy_other, fibonacci_hash, fnv_1_hasher, fnv_1a_hasher, free_key, free_other, get, hasher_fun, operator(==), seeded_nmhash32_hasher, seeded_nmhash32x_hasher, seeded_water_hasher, set, key_type, other_type

  • new procedures in stdlib_hashmaps: chaining_hashmap_type, hashmap_type, `open_hashmap_type

Changes to existing scripts and modules

• change in script doc-deployment.yml

  • update of the script #681

• change in script fpm-deployment.sh

  • fixed a problem with dat and npy files in example dir not being deployed #713

• change in module stdlib_bitsets

  • remove define assignment for bitset_64 and bitset_large #727

• change in module stdlib_hashmap_open

  • fix access violation in a type-bound procedure of open_hashmap_type #707

• change in module stdlib_io_npy_load

  • fix various bugs #708 #711

• change in module stdlib_linalg

  • addition of kronecker_product #700

• change in module stdlib_quadrature_gauss

  • fix erroneous gaussian quadrature points in gauss_legendre #660

• change in module stdlib_sorting

  • addition of radix sort #712

  • support for sorting arrays of bitset_64 and of bitset_large #723

• change in module stdlib_stats_distribution_exponential

  • convert pdf_exp and cdf_exp to pure functions #717

• change in module stdlib_stats_distribution_normal

  • convert rvs_norm to an impure elemental function #665

  • remove unused module stdlib_error from module stdlib_stats_distribution_normal #716

• remove support for manual make builds #657

Changes to the existing documentation

• change in README.md #656 #659 #715 #725

• change in stdlib_stats_distribution_normal.md

  • Improvement of the documentation #718 #721

• change in stdlib_stats_distribution_exponential.md

  • Improvement of the documentation #721

• change in the structure of the project stdlib

  • extraction of the demo programs from the specs in the directory example #662

  • move the directory src/tests to test #669

• fix various docs #663

Version 0.2.1

Full release notes available at v0.2.1 tag.

• build system related bugfixes

Version 0.2.0

Full release notes available at v0.2.0 tag.

• new module stdlib_hash_32bit #573

  • new procedures: fibonacci_hash, fnv_1_hash, fnv_1a_hash, new_nmhash32_seed, new_nmhash32x_seed, new_water_hash_seed, nmhash32, nmhash32x, odd_random_integer, universal_mult_hash, and water_hash

• new module stdlib_hash_64bit #573

  • new procedures: fibonacci_hash, fnv_1_hash, fnv_1a_hash, new_pengy_hash_seed, new_spooky_hash_seed, odd_random_integer, pengy_hash, spooky_hash, spookyhash_128, and universal_mult_hash

• new module stdlib_array #603

  • new procedures trueloc, falseloc

• new module stdlib_distribution_uniform #272

• new module stdlib_selection #500

  • new procedures select, arg_select

• new module stdlib_version #579

  • new procedure get_stdlib_version

• update module stdlib_io 597

  • new procedure getline

• new module stdlib_io_npy #581

  • new procedures save_npy, load_npy

• update module stdlib_math

  • new procedures is_close and all_close #488

  • new procedures arg, argd and argpi #498

  • new procedure diff #605

Changes to existing modules

• change in module stdlib_math

  • linspace and logspace made pure #549

• change in module stdlib_string_type

  • move procedure made pure/elemental #562

• support for quadruple precision made optional #565

• change in module stdlib_io

  • Modified format constants, and made public #617

• change in module stdlib_math

  • Minor update to stdlib_math module and document #624

Version 0.1.0

Full release notes available at v0.1.0 tag.

• new module stdlib_ascii #32

• new module stdlib_bitsets #239

  • new derived types bitset_64 and bitset_large

  • new abstract base class bitset_type

• new module stdlib_error #53

• new module stdlib_io

  • new procedures loadtxt and savetxt #23 #37

  • new procedure open #71 #77

• new module stdlib_kinds #63

• new module stdlib_linalg

  • new procedures diag, eye and trace #170

  • new procedure outer_product #432

• new module stdlib_logger

  • new derived type logger_type #228 #261

• new module stdlib_math

  • new procedure clip #355

  • new procedures linspace and logspace #420

  • new procedure arange #480

  • new procedure gcd #539

• new module stdlib_optval #73 #96 #139

• new module stdlib_quadrature

  • new procedures trapz, trapz_weights, simps and simps_weights #146

  • new procedures gauss_legendre, gauss_legendre_lobatto #313

• new module stdlib_random #271

• new module stdlib_sorting

  • new procedures sort, ord_sort and sort_index #408

• new module stdlib_specialfunctions

  • new procedures legendre and dlegendre #313

• new module stdlib_stats

  • new procedure mean #124 #130 #132

  • new procedure var #144

  • new procedure moment #153

  • new procedure corr #191

  • new procedure median #426

• new module stdlib_string_type

  • new derived types string_type #320

  • new procedure move #467

• new module stdlib_stringlist_type

  • new derived types stringlist_type and stringlist_index_type #470

• new module stdlib_strings

  • new procedure to_string #444

  • new procedures strip and chomp #343

  • new procedures starts_with and ends_with #384

  • new procedure slice #414

  • new procedure find #433

  • new procedure replace_all #436

  • new procedures padl and padr #441

  • new procedure count #453

• new module stdlib_system

  • new procedure sleep #54