olegz0/acramalpha

Mathematical equations' parser & differentiator

AcramAlpha

Mathematical equations' parser & differentiator

Compilation

• cmake CMakeLists.txt ; make.

If ADVANCED_MODE is defined in "options.hpp", equation simplifications are better.

WARNING: ADVANCED_MODE may result in unexpected simplification if such operations like "* * " and " - *" were redacted via scripts. Turing it off leads to only those simplifications defined in "simplify.dat".

Running

• example: ./diff files/*.in

The executable "diff" expects for a list of files containing mathematical equations as an argument. For each file the output is stored in the same directory as the file.

The output is a LaTeX file containing 2 formulas: the original equation f(x) and its differential f'(x). Also the result of all equations are stored in the working directory in "output.tex".

Under Linux where packages texlive and texlive-lang-cyrillic are installed the LaTeX files will be automatically converted to PDF with removing the mediators.

Preconditions

• Mathematically correct (no 2/4 or *).
• Contains only "* x *" as a variable.
• Using only operations specified in "lib/scripts/".

Any parsing difficulties will be reported though.

The examples are stored in "files" directory.

Permissions

• Input file is not to have ".in" extension.
• Using spaces.
• Using unary "* - *" even in sequences.
• Using double numbersnot starting from point (1, 1.2, 12e-1 - OK, .2 - NO).

Description

Library (./lib)

• Equation: storage for mathematical formulas as binary tree.
• scriptparser: some weird functions based on parsing scripts using macro'.
• TexFile: creating a texfile with appropriate header and converting it.
• system: operating with OS environment (available only for Linux).

Scripts (./lib/scripts)

• functions.dat: the most important one; specifies operations names for parsing and programm, priority, differential calculation and LaTeX formula generation.
• simplify.dat: specification upon degenerated cases for simple calculations (x0=0, sqrt(0)=0, etc.*). For unary functions the inverse one is also specified.
• calculate.dat: specifies how to calculate an operations when numeric constants provided as operands ("/" not included as story fractions look better).
• duplex.dat: specifies how to calculate an operations when operands provided are equal.
• relation.dat: specification upon operations' characteristics and co-relations (like commutativity and associativity).

Issues

• Operations' relations are not used by their full potential, sketches are to be found in Equation::simplify.
• Convertion doesn't work not under Linux.
• Binary suffix operations like log(a, b) not supported - only prefic ones.
• Advanced mode implies certain prescription (+ ~ add, * ~ multiplicate, etc.) and therefore not common.