Release Notes for 1.2 - sympy/sympy GitHub Wiki

These are the release notes for SymPy 1.2. You can also find release notes for previous versions.

SymPy 1.2 was released on Jul 9, 2018.

This version of SymPy has been tested on Python 2.7, 3.4, 3.5, 3.6, 3.7 and PyPy. See our Python version support policy for more information on when we plan to drop support for older Python versions.

Install SymPy with

pip install -U sympy

or if you use Anaconda

conda install sympy

Highlights

There are many changes in 1.2 (see below). Some of the highlights are

  • Python 3.3 is no longer supported. If you require Python 3.3 support, use SymPy 1.1.1. See our policy on dropping support for major Python versions.

  • Experimental LaTeX parsing with sympy.parsing.latex.parse_latex() has been added, based on the latex2sympy project. This requires antlr-python-runtime to be installed. #13706

  • The vector module has been improved to support orthogonal curvilinear coordinate systems (Szymon Mieszczak's GSoC project)

  • New module sympy.integrals.intpoly for integrating uni/bivariate polynomials over 2-polytopes. (Arif Ahmed's GSoC project)

  • Improvements to the code generation module. (Björn Dahlgren's GSoC project)

  • Improvements to the group theory module. See below for more information. (Valeriia Gladkova's GSoC project)

  • New module sympy.discrete for operating on discrete sequences. (Sidhant Nagpal's GSoC project)

Other GSoC projects, Abdullah Javed Nesar, Arif Ahmed, and Gaurav Dhingra are work in progress improvements to the integrate module (particularly, the new rubi submodule and the risch submodule), which are not yet visible to user-facing code.

Backwards compatibility breaks and deprecations

  • Units: the arguments dimension and scale_factor in Quantity have been deprecated (see #14319).

  • Units: Dimensional dependencies are now part of DimensionSystem. All related methods inside Dimension have been deprecated in favor of the methods in DimensionSystem (see #13336).

  • Matrix.dot now only works for row and column matrices. Use * or @ to do matrix multiplication. (see #13815).

  • The k argument for line3D.equation() is deprecated (see #13742).

  • lambdify no longer automatically wraps arguments with np.array, which improves the performance of calls to lambdified functions. Array wrapping is necessary if the input involves an operation that would raise an exception on scalar input, such lambdify(x, 1/x)(0), but these cases are relatively rare, and wrapping can always be done by the user if necessary. #14823

Major Changes

  • algebras

    • add new algebras submodule
    • add Quaternion class #13285

  • calculus

    • periodicity: added support for modulo operation #13226
    • fixed continuous_domain to remove singularities. #13335
    • Finite Set domains for function_range #13479
    • Fix function_range for periodic functions #13512

  • combinatorics

    • correct gray_to_bin and bin_to_gray #14468
    • improve the comparison test for convergence #14509

  • concrete

    • Improve summation of geometric series #13991
    • add the ratio test to Sum.is_convergent #14158
    • add the limit comparison test for is_convergent #14401
    • add bounded times summation convergence test #14464
    • add factoring to compute certain sums with negative exponents #14641

  • core

    • Derivatives by a variable a symbolic number of times, like diff(f(x), (x, n)) to represent d^nf/dx^n, where n can be symbolic. #13751 #13892
    • various improvements to Piecewise and piecewise_fold. #12951, #12920, #13074, #13282, #13309, #13204, #12587, #14071, #14682
    • Make Pow.subs not use fractional powers for noncommutative objects #13018
    • nullary undefined functions (like f = Function('f'); f()) are now allowed #12977
    • add a global option and context manager to disable automatic distribution (2*(x + y) -> 2*x + 2*y) #12440
    • make Mul.flatten aware of MatrixExpr #13279
    • Fix __rfloordiv__ not being reversed #13368
    • Expr now supports the 3-argument version of the builtin pow #13364
    • Allow undefined functions to be defined with assumptions (like Function('f', real=True)). #12945
    • Allow Floats to be created from NumPy floats #13426
    • Fix some incorrect comparisons between numbers #13429
    • removed x**1.0 == x hack from the core (they are now unequal, remember that == in SymPy means symbolic, not mathematical equality) #13518
    • Subs substitution now supports derivative by indexed objects in various cases #13452
    • Add more simplifications for Mod #13581
    • make f(1).is_number give False, where f = Function('f') #13619
    • allow differentiating with respect to tuples, returning an Array type #13655
    • correct sympify for numpy arrays of size 1 #13926
    • Do not over-simplify rational powers of negative integers #13895
    • Fix the negative rational powers of negative integers and rationals #14024
    • improve the performance of modular exponentiation for very large powers #14249
    • improve mod_inverse for negative numbers #14333
    • prevent infinite recursion in as_real_imag #14404
    • remove automatic distribution of infinite factors #14383
    • fix floor and ceiling of infinite quantities #14328
    • fix evaluation of negative terms with sympify(evaluate=False) #11095
    • SymPy numeric expressions now work with math.trunc #14451
    • fix Mod(x**2, x) giving 0 (it is not 0 if x is fractional) #13177
    • fix evalf with subs giving incorrect results with floor #13361
    • prevent substitutions in Derivative in cases where it changes the mathematical meaning of the expression #13803
    • allow complex powers to be rewritten as exp #14712

  • codegen

    • Improvements to the code generation module. (Björn Dahlgren's GSoC project), including:
      • many additions to the sympy.codegen.ast module
      • new sympy.codegen rewriting module with codegen related expression optimizations
      • new sympy.codegen.algorithms module with code generatable algorithms using codegen.ast
      • new sympy.codegen.approximations module with code generatable approximations
    • add a code printer for GLSL (glsl_code) #12713
    • add support for tensorflow 1.0+ #13413
    • added printer flag so that printer can be passed to codegen() #13587
    • Add support for fmod in ccode #13692
    • add more functions to rcode #13840
    • add more functions to jscode #13832
    • add support for Max and Min in fcode and octave #13903
    • fix pycode() and 'lambdify' for Piecewise functions #13969
    • fix imaginary number printing in octave #13978
    • add re, im, arg functions to octave codegen #14013
    • Fix printing of Fortran code with variables that are the same when compared case insensitively #14003
    • Add support for sympy.sign in .printing.pycode #14010
    • support zoo and -oo in lambdify #14306
    • lambdify: Improved speed of generated code when unpacking arguments #14691
    • lambdify now generates actual Python functions rather than lambdas #14713
    • the dummify argument to lambdify now by default only converts the arguments to Dummy when necessary #14713
    • Add lambdified function source to the linecache. This makes inspect.getsource, IPython's ??, IPython tracebacks, and most debuggers show the source code for lambdified functions #14739

  • crypto

    • Add Goldwasser Micali Encryption #13666

  • discrete

    • new submodule sympy.discrete for operating on discrete sequences with the following functions:
      • fft, ifft (fast discrete Fourier transforms), ntt, intt (number theoretic transform) #14725
      • convolution (convolutions on discrete sequences) #14745, #14783
      • fwht and ifwht (fast Walsh-Hadamard transform) #14765
      • Added the ability to evaluate linear recurrences without obtaining closed form expressions #14816

  • functions

    • Improve trigonometric function evaluation and logarithmic rewrite #13109
    • euler can now compute Euler polynomials #13228
    • add some rewrite methods for sinc #11870
    • use gmpy to compute binomial if it is installed #13394
    • fix for B-Splines of higher than 1 degree with repeated knots #12214
    • allow Min and Max to be rewritten in terms of Abs #13614
    • add some assumptions for Catalan numbers #13628
    • Added an interpolating_spline function for symbolic interpolation of splines #13829
    • Implement more special values of polylogarithm, without exp_polar for s=1 #13852
    • Improved legendre polynomial for negative n #13920
    • Add evaluation of polygamma(0, z) for all rational z #14045
    • More general unpolarification for polylog argument #14076
    • fix floor sometimes giving the wrong answer #14167
    • return zeta(2*n) and zeta(-n) as expressions #14178
    • automatic log expansion of numbers removed #14134
    • improve absolute value of complex numbers raised to the power of complex numbers #14278
    • improve evaluation of inverse trig functions #14321
    • allow exp to be rewritten as sqrt #14358
    • evaluate transcendental functions at complex infinity #14406
    • improve automatic evaluation of reciprocal trig functions #14544
    • Support added for computing binomial(n, k) where k can be noninteger and n can be any number. #14019 by Yathartha Joshi
    • Fix binomial(n, k) for negative integer k #14575
    • Optimize binomial evaluation for integers #14576
    • implemented partition numbers (partition()) #14617
    • automatically reduce nested floor and ceiling expressions #14631
    • improve performance Mod of factorial and binomial expressions #14636
    • improve floor and ceiling rewriting and equality testing #14647

  • geometry:

    • Second moment and product moment of area of a two-dimensional polygon can now be computed. #13939 by Keshri Kumar Rushyam.
    • Second moment and product moment of area of an ellipse and a circle can now be computed. #14190 by Keshri Kumar Rushyam.
    • pairwise intersections. #12963
    • Added closing_angle which represents the angle of one linear entity relative to another. #13002
    • Added method to compute the exradius of a triangle #13318
    • Implemented a length property for the Curve class. #13328
    • make arbitrary_point of Plane run through all points #13807
    • Remove integral from ellipse circumference calculation #14435

  • groups

    • It is now possible to check if a finitely presented group is infinite with _is_infinite method (this may return None in a number of cases where the algorithm cannot establish whether a group is infinite). The order method now returns S.Infinity for some infinite groups. (#12705)
    • subgroup method was added for permutation and finitely presented groups to return the subgroup generated by given group elements (#12658). Subgroups of finitely presented groups end up having different generators from the original group but a special class FpSubgroup can be used as a substitute if having the same generators is important. (#12827)
    • The presentation of a finitely presented group can be simplified with simplify_presentation
    • Group homomorphisms have been implemented as a GroupHomomorphism class in sympy.combinatorics.homomorphisms. A homomorphism can be created using the function homomorphism and additionally one can create special types of group action homomorphisms for permutation groups with block_homomorphism and orbit_homomorphism. (#12827 and #13104)
    • Rewriting systems for finitely presented groups are implemented. If a confluent rewriting system is found, this allows to reduce elements to their normal forms. Otherwise, at least the found reduction rules cab be used to equate some group elements that otherwise would be treated as different previously. See #12893 for more information.
    • Convert between permutation and finite presentation of groups by using presentation and strong_presentation method of permutation groups (#12986 and #13698) and _to_perm_group() for finitely presented groups (#13119)
    • Compute Sylow subgroups of permutation groups with sylow_subgroup method. (#13104)
    • A number of methods and attributes previously only available for permutation groups now can be used with finite finitely presented groups. These include derived_series, lower_central_series, center, derived_subgroup, centralizer, normal_closure, is_abelian, is_nilpotent, is_solvable and elements. (#13119)

  • holonomic

    • Generalize Holonomic to work on symbolic powers #12989

  • integrals

    • Rewrite Abs and sign as Piecewise for integration#13930
    • Integrate Max and Min by rewriting them as Piecewise #13919
    • Add new trigonometric rule in manual integration #12651
    • Reorder Piecewise output of integration, so that generic answer is first #13998
    • Add symbolic Riemann sums for definite integrals #13988
    • In manualintegrate, when integrating by parts, do not use u = non-poly algebraic #14015
    • fix wrong result integration of rational functions in some cases when variables have assumptions #14082
    • simplify some convergence conditions for integrals #14092
    • add a manualintegrate rule for expanding trig functions #14408
    • improve manualintegrate for substitutions u=(a*x+b)**(1/n) #14480
    • better handling of manual, meijerg, risch flags in integrate #14475
    • add floor to certain discontinuous integrals #13808
    • add integration of orthogonal polynomials of general degree #14521
    • allow integrals with inequalities, such as Integral(x, x>2) #14586

  • interpolation

    • A spline of given degree that interpolates given data points can be constructed with interpolating_spline. #13829 by Jashan.

  • liealgebras

    • fix instantiate of CartanType with a string like CartanType('A34') #14568

  • matrices

    • Computing the Smith Normal Form of a matrix over a ring is now implemented as the function smith_normal_form from sympy.matrices.normalforms (#12705)
    • Derivatives for expressions of matrix symbols now experimentally supported.
    • Matrices can be used as deriving variable.
    • Matrix norm for order 1 has been added #12616
    • add MatrixExpr.from_index_summation, which parses expression of matrices with explicitly summed indices into a matrix expression without indices, if possible #13542
    • Add functionality for infinity norm of matrices #13986
    • add numeric check for pivot being zero in Bareiss determinant method #13877
    • Implement Kronecker product of Matrices #14264
    • give a simpler representation of exp(M) when M is real #14331
    • avoid loss of precision in jordan_form #13982
    • made Mod elementwise for matrices #14498
    • implement Cholesky and LDL decompositions for Hermitian matrices #14474

  • ntheory

    • Added totient and mobius range methods to Sieve #14628

  • physics.continuum_mechanics

    • Added applied_loads and remove_load methods #14751
    • added method to find point of contraflexure #14753
    • Added support for composite beams #14773
    • added apply_support and max_deflection methods #14786

  • physics.mechanics

  • physics.optics

    • Added transverse_magnification #10625
    • Implemented Brewster's Angle in Optics Module #13756

  • physics.quantum

    • Adding support for simplifying powers of tensorproducts #13974

  • physics.vector

    • fix error pretty printing physics vectors #14717

  • physics.wigner

    • Make wigner_9j work for certain half-integer argument combinations #13602

  • plotting

    • Fix plotting of constant functions #14023

  • polys

    • degree now requires a generator to be specified for multivariate polynomials (use total_degree for the total degree). #13173
    • total_degree function added #13277
    • Allow itermonomials to respect non-commutative variables #13327
    • Improve srepr of polynomial rings and fraction fields #13345
    • add Poly.norm, which computes the conjugates of a polynomial over a number field. #13304
    • add srepr printer for DMP #13367
    • added Finite ring extensions to the agca submodule #13378
    • fix factor for expressions with floating point coefficients #13198
    • Changed ModularInteger to use fast exponentiation #14093
    • fix minpoly(I) over domains that contain I #14382
    • fix the quartic root solver for certain quartic polynomials with complex coefficients #14522
    • implement Dixons and Macaulay multivariate resultants #14370
    • support gcd and lcm for compatible irrational numbers #14365
    • optimized Lagrange Interpolation #14603

  • printing

    • Latex printer now supports custom printing of logarithmic functions by newly by accepting ln_notation as a boolean (default=False) keyword argument. (#14180)
    • Add mathml and cxxcode to from sympy import *. #12937
    • Add sympy_integers option to the str printer to print Rationals in a SymPy recreatable way. #13141
    • change the set pretty printer to print with {} #12087
    • fix LaTeX dagger printing #13672
    • use a better symbol for Equivalent in the pretty printers #14105
    • fix pretty printing of DiracDelta #14104
    • change LaTeX printing of O() from \mathcal{O} to \big{O} #14166
    • print Poly terms in the correct order in the latex printer #14317
    • fix latex printing of SeqFormula #13971
    • add abbreviated printing for units #13962
    • allow user defined mul_symbol in latex() #13798
    • add a presentation MathML printer #13794
    • fix an issue with some missing parentheses from pprint and latex #13673
    • fix some instances where the str printer would not reuse the same printer class #14531
    • improve printing of expressions like Pow(Mul(a,a,evaluate=False), -1, evaluate=False) #14207
    • add conjugate to Mathematica printing #14109
    • the printers no longer dispatch undefined functions like Function subclasses (makes Function('gamma') print as γ instead of Γ) #13822
    • Make LaTeX printer print full trig names for acsc and asec #14774

  • series:

    • Bidirectional limits can now be computed by passing dir=+- argument to limits.limit function. #11694 by Gaurav Dhingra.
    • Evaluate limits of sequences with alternating signs #13976

  • sets:

    • Ordinals can now be represented in Cantor normal form, and arithmetics can be performed which includes addition, multiplication, and basic exponentiation. #13682 by Ashish Kumar Gaurav
    • SetExpr and arithmetic expressions of sets now supported (via multimethods). #14106
    • Singleton sets (Reals, Complexes, etc.) can now be accessed without S #11383, #12524, #13572, #14844
    • Add set expressions #2721, #14301
    • ImageSet now supports multiple sets #14145
    • be more careful in ConditionSet.subs #14564

  • simplify

    • improve simplification of Min and Max #13054, #13599
    • New function gammasimp separated from combsimp. combsimp does not apply gamma function simplifications that may make an integer argument non-integer. Also various improvements to both. #13210
    • remove unused fu flag from simplify. #13264
    • add rational flag to simplify. #13264
    • improved simplification of hyperbolic functions #13259
    • improve the efficiency of cse #13221
    • Allow cse of unevaluated expressions #13271
    • fix logcombine for logs of numbers #14070
    • disable simplification of inverses in simplify() unless the new inverse flag is set #14422
    • simplify() now simplifies non-commutative expressions #14520

  • solvers

    • Enable initial condition solving in dsolve #11264
    • Added support for solving inequalities involving rational functions with imaginary coefficients #13296
    • the solvers now work with non-Symbols such as Indexed #13415
    • add failing_assumptions function #12147
    • Improve the solver for linear 1st order ODE systems of size 2 #13812
    • fix classify_ode(dict=True) for Eq #14663
    • improve handling of infinite solutions in solve #14749

  • solveset

    • Solveset now supports exponential equations with negative base. #13923 by Yathartha Joshi
    • Fix solve_univariate_inequality for finite set domains #13458
    • solve more trig equations #13941
    • improve solveset with piecewise expressions #14253
    • Improved solveset for negative base exponential equations #13844
    • allow solving equations with Abs in solveset #14449

  • stats

    • Added pre-computed cumulative distribution functions to some Random Symbols #13284, #13804, #13878
    • add trapezoidal distribution #13419
    • The characteristic function of a probability distribution can be computed with characteristic_function. #13851 and #13958 by Ethan Ward.
    • Probability of continuous random variables lying in a finite set is now zero #14254
    • add sampling for GammaInverse distribution #14250
    • Allow Geometric RVs where p is a symbol #14472
    • various improvements to sampling methods
    • fix wrong results with Or conditions #14578
    • implemented probability for discrete random variables #14119
    • rename restricted_domain to where for discrete random variables #14604
    • implement several missing classes in discrete random variables #14218

  • tensor

    • do not require assumptions to be set on Idx bounds. #12888
    • improve Indexed.free_symbols #13360

  • units

    • add SI base units to sympy.physics.units.systems #12897
    • Implemented get_dimensional_expr for Derivative #13003
    • Implemented Quantity.get_dimensional_expr() for functions #13011
    • the relation between base dimensions and derived dimensions are now part of DimensionSystem, so that they are no longer absolute. #13287
    • Add Stefan Boltzmann constant #13440
    • Add Planck charge and derived Planck units #13438
    • Add katal, gray to derived units #13658
    • Added derived unit of Radioactivity #13839
    • fix subs for Quantities #13855
    • add simplification of quantities #14286

  • vector

Minor changes

  • free_symbols and find_dynamicsymbols methods now available for vectors directly, if accompanied by a reference frame argument. #13549 by Saloni Jain

  • simplify Relational.canonical #12906

  • fix (z**r).is_algebraic where z is algebraic and r is rational #12924[<8;52;35m

  • Remove register keyword from C++17 standard in cxxcode #12964

  • Allow autowrap's tempdir to be a relative path #12944

  • Fix symbols with superscripts raised to a power in the LaTeX printer #12894

  • Fix conversion of undefined functions (like Function('f')) to Sage. #12826

  • Fix wrong result integrating derivatives of multiple variables #12971

  • Rich comparison methods now return NotImplemented when comparing against unknown types #13091

  • Add/Fix logarithmic rewrites of inverse hyperbolic functions #13099

  • Fix Mul.expand(deep=False) #13281

  • Make DiracDelta(-x) == DiracDelta(x) #13308

  • linsolve now raises error for non-linear equations #13325

  • vlatex: added brackets in multiplication #12299

  • Add "The Zen of SymPy" (import sympy.this)

  • Install isympy using setuptools, making it available on Windows. isympy can also now be run as python -m isympy #13193

  • improve evaluation performance of binomial #13484

  • Add support for more than two operands in numpy logical and/or representations in lambdify #12608

  • The Mathematica parsing function has been rewritten #13533

  • The Mathematica parsing function now has an additional_translations flag #13544

  • Fix is_real for trigonometric and hyperbolic functions #13678

  • @/matmul now only performs matrix multiplication #13773

  • Fold expressions with Piecewise before trying to integrate them #13866

  • Fix has for non-commutative Muls in some cases #14026

  • Avoid recursive calls for upper and lower gamma #14021

  • Remove the sympy_tokenize module in favor of the standard library tokenize. Fixes sympification of Python 3-only idioms, like sympify('α'). #14085

  • Check for logarithmic singularities in _eval_interval #14097

  • Simple optimization to speed up LCM #14314

  • Use extended Euclidean algorithm instead of totient calculation for modular matrix inverse #14347

Authors

The following people contributed at least one patch to this release (names are given in alphabetical order by last name). A total of 170 people contributed to this release. People with a * by their names contributed a patch for the first time for this release; 124 people contributed for the first time for this release.

Thanks to everyone who contributed to this release!

  • Saurabh Agarwal*
  • Arif Ahmed
  • Jonathan Allan*
  • anca-mc*
  • Adwait Baokar*
  • Nilabja Bhattacharya*
  • Johan Blåbäck
  • Nicholas Bollweg*
  • Francesco Bonazzi
  • Matthew Brett
  • Marcin Briański
  • Bulat*
  • Ondřej Čertík
  • Arighna Chakrabarty*
  • Rishav Chakraborty*
  • Ravi charan*
  • Lev Chelyadinov*
  • Poom Chiarawongse*
  • James Cotton*
  • cym1*
  • czgdp1807*
  • Björn Dahlgren
  • David Daly*
  • der-blaue-elefant*
  • Gaurav Dhingra
  • dps7ud*
  • Rob Drynkin*
  • Seth Ebner*
  • Peter Enenkel*
  • Fredrik Eriksson*
  • Boris Ettinger*
  • eward*
  • Isuru Fernando
  • Segev Finer*
  • Caley Finn*
  • Micah Fitch*
  • Lucas Gallindo*
  • Bradley Gannon*
  • Mauro Garavello
  • Varun Garg
  • Ashish Kumar Gaurav*
  • Jithin D. George*
  • Sourav Ghosh*
  • Valeriia Gladkova
  • Jeremey Gluck*
  • Nikoleta Glynatsi*
  • Keno Goertz*
  • Nityananda Gohain*
  • Filip Gokstorp*
  • Sayan Goswami*
  • Jonathan A. Gross*
  • Theodore Han
  • Rupesh Harode*
  • James Harrop*
  • Rahil Hastu*
  • helo9*
  • Himanshu*
  • Hugo*
  • David Menéndez Hurtado*
  • Ayodeji Ige*
  • Itay4*
  • Harsh Jain
  • Samyak Jain
  • Saloni Jain*
  • Rohit Jain*
  • Shikhar Jaiswal
  • Mark Jeromin*
  • JMSS-Unknown*
  • Yathartha Joshi
  • Ishan Joshi*
  • Salil Vishnu Kapur*
  • Atharva Khare*
  • Abhigyan Khaund*
  • Sergey B Kirpichev
  • Leonid Kovalev
  • Amit Kumar
  • Jiri Kuncar
  • Akash Kundu*
  • Himanshu Ladia*
  • Rémy Léone*
  • Alex Lubbock*
  • Jared Lumpe*
  • luz.paz*
  • Colin B. Macdonald
  • Shubham Maheshwari*
  • Akshat Maheshwari*
  • Shikhar Makhija
  • Amit Manchanda*
  • Marco Mancini*
  • Cezary Marczak*
  • Yuki Matsuda*
  • Bhautik Mavani
  • Maxence Mayrand*
  • B McG*
  • Cavendish McKay*
  • Ehren Metcalfe*
  • Aaron Meurer
  • Peleg Michaeli
  • Szymon Mieszczak
  • Aaron Miller*
  • Joaquim Monserrat*
  • Jason Moore
  • Sidhant Nagpal*
  • Abdullah Javed Nesar
  • Javed Nissar*
  • Richard Otis
  • Austin Palmer*
  • Nikhil Pappu*
  • Fermi Paradox
  • Rhea Parekh*
  • Arihant Parsoya
  • Roberto Díaz Pérez*
  • Cristian Di Pietrantonio*
  • Waldir Pimenta*
  • Robert Pollak*
  • Alexander Pozdneev*
  • P. Sai Prasanth*
  • Rastislav Rabatin*
  • Zach Raines*
  • Shekhar Prasad Rajak
  • Juha Remes
  • Matthew Rocklin
  • Phil Ruffwind
  • rushyam*
  • Saketh*
  • Carl Sandrock*
  • Kshitij Saraogi
  • Nirmal Sarswat*
  • Subhash Saurabh*
  • Dmitry Savransky*
  • Nico Schlömer
  • Stan Schymanski*
  • sfoo*
  • Ayush Shridhar*
  • shruti
  • Sartaj Singh
  • Rajeev Singh*
  • Mayank Singh*
  • Arshdeep Singh*
  • Jashanpreet Singh*
  • Malkhan Singh*
  • Gleb Siroki*
  • Chris Smith
  • Ralf Stephan
  • Kalevi Suominen
  • Ian Swire*
  • symbolique*
  • Mihail Tarigradschi*
  • James Taylor*
  • Pavel Tkachenko*
  • Cédric Travelletti*
  • Tschijnmo TSCHAU
  • Wes Turner*
  • Unknown*
  • Akash Vaish*
  • vedantc98*
  • Pauli Virtanen
  • Vishal*
  • Stewart Wadsworth*
  • Ken Wakita*
  • Ethan Ward*
  • Matthew Wardrop*
  • Daniel Wennberg*
  • Lucas Wiman*
  • Kiyohito Yamazaki*
  • Yang Yang*
  • Ka Yi
  • ylemkimon*
  • zhouzq-thu*
  • Rob Zinkov*