liuguangxi / pe

Basic code for solving problem on project euler

pe

baihacker's C++ library for solving problem on project euler.

Prerequirements:

C++ dev environment that supports

• C++17 or above.
• Building x86_64 targets.

Installation:

• Make sure "#include <pe.hpp>" work.

  • Put all the files in an installation folder (directory).
  • Add the installation folder to the environment variable CPLUS_INCLUDE_PATH. Other approaches are also OK.

• Configure this library

  • Execute gen_config.py in the installation folder to generate pe_config.
    • Static configuration. The values are always 1 in the generated config file. You can edit the values after generating the config file.
      • ENABLE_ASSERT whether to assert some inputs or conditions.
      • TRY_TO_USE_INT128 whether to check whether the compiler support int128 and use it. If it is 0, it int128 is disabled even if the compiler supports it.
    • Auto-deduced configuration for third party libraries. The script searches for the corresponding header file and set the value to 1 if found. See Build and use pe's dependent libraries for the pre-built third party binaries.
      • ENABLE_EIGEN whether to use Eigen library.
      • ENABLE_GMP whether to use gmp.
      • ENABLE_FLINT whether to use FLINT.
      • ENABLE_MPFR whether to use mpfr.
      • ENABLE_LIBBF wheter to use libbf.
      • ENABLE_NTL whether to use ntl.
  • Edit the pe_config by adding more configuration items manually.
    • ENABLE_OPENMP whether to use openmp. It is NOT generated by the script automatically becaused it's inferred at the compiling time. You can still define it in pe_config to prevernt the library from inferring the value from the compiling environment. If it is enabled but the environment doesn't support it, a warning message will show.

• [optional, recommended] Generate precompile header "pe.hpp.gch".

  • Run "g++ -xc++-header pe.hpp" in the installation folder.
  • Add more compiling options if necessary. Usually, they are the same as the options used to build your binary. e.g. "g++ -xc++-header pe.hpp --std=c++17 -O3 -march=native -fopenmp".

Use:

See example.c for quick start.

File list:

• pe: Including all the implementation files.
• pe.hpp: The file for generating precompile header. It only includes pe.
• pe_algo: Algorithms.
• pe_array: An array implementation with compiling time dimension length. The element count can be more than the limit of int32 and you can specify customized allocator.
• pe_base: Some pre-including headers. Some macros and typedef. Some basic inline functions.
• pe_bi32: Big integer whose base is 1 << 32.
• pe_bit: Bit operation tricks.
• pe_config: a centralized place the configure pe.
• pe_db: load and save pre-calculated result such as prime pi and prime sum.
• pe_extended_int: The extended integer.
• pe_extended_signed_int: Extended signed integer.
• pe_extended_unsigned_int: Extended unsigned integer.
• pe_fft: Fast fourier transform and polynomial multiplication.
• pe_float128: Unified float number functions of __float128.
• pe_fraction: Fraction arithmetic.
• pe_gbi: general big integer. The content corresponds to pe_nt.
• pe_geometry: Support Point2D and Point3D.
• pe_initializer: The helper class and macros used to initialize the library.
• pe_int: Basic integer utilities.
• pe_int128: Support to output int128 and the corresponding type traits.
• pe_internal: Include configurations, define necessary types/macros and include third party libraries.
• pe_io: methods and macros that simplify or fasten reading from and writing std io.
• pe_mat: Matrix operations.
• pe_memory: Memory manipulation such as allocating large memory. (windows only)
• pe_misc: misc codes.
• pe_mma: support mma: helper method or class to generate mma codes.
• pe_mod: Modular arithmetic.
• pe_nt: Basic code of number theory.
• pe_nt_base: Generate prime list, factorize integer, prime test, compute phi and mu.
• pe_parallel: A simple framework to solve problem with multi-threads. (windows only)
• pe_parallel_algo: Parallel algorithms.
• pe_persistance: KVPersistance. (it may support linux if we change the generated cmdline and check the other codes)
• pe_poly: Polynomial c++ wrapper.
• pe_poly_algo: Polynomial algorithms.
• pe_poly_base: Polynomial basic algorithms.
• pe_poly_base_flint: Flint based polynomial basic algorithms.
• pe_poly_base_libbf: Libbf based polynomial basic algorithms.
• pe_poly_base_min25: Min_25's polynomial basic algorithms. The polynomial multiplication implementation is the fastest one among mod polynomials integrated into pe.
• pe_poly_base_ntl: Ntl based polynomial basic algorithms.
• pe_rand: Random number.
• pe_range: generate an range of numbers, iterate a container with index.
• pe_sym_poly: Symbolic polynomial.
• pe_time: Support TimeDelta, TimeRecorder.
• pe_tree: Some tree based data structures.
• pe_type_traits: Type traits.