Scenario: ans*int(0,x,0,0)->x then cos(x) crashed because replaceReplaceableSymbols
would not care if it was replacing parameters in a parametered
expression
real (or imaginary) part negligence should depend on the argument value
relatively to some other values (norms of the parameters) and not of its
value absolutely!
Scenario: in calculation,
1/matrix(matrix(matrix(matrix(... matrix(1) ... )))) gave kind of a
weird result. These kind of computations also broke the fuzzer.
* Added some constant
Move mass on a new submenu of Physics
Added universal translation for a future easter-egg
* Added lenght constant menu
Added radiuses and distances submenu to lenght
Added Faraday Constant
* Constant
Added 16 usual pKa couple values
* Added some constant
( stefan-boltzmann, water triple point, wien, atmospheric pressure, vacuum impedance, bohr magneton, nuclear magneton, muon mass, atomic mass unit, bohr radius )
Missing hungarian translation
* Planck update
added some plancks units ( fondamentals units and energy/power units )
* Changed the Physics constants submenu
Here's the new pattern
1. Fundamental Constants => for the 5 fundamental constants
2. Electromagnetism => 3 constants remain in this submenu
3. Nuclear and Atomic Constnats => bohr magneton & nuclear magneton, bohr radius and particle mass submenu
4. Thermodynamics => this section is mostly unchanged
5. Gravitation => same as before
6. Speed => 2 submenus one for sound speeds and one for escape velocities
7. Mass => the mass of earth/sun/moon
8. Length => 2 submenus, one for radius of earth/sun/moon and one for distances earth-moon and earth-sun
9. Planck Units => The natural system of units
* Units
Added steradian, lumen and lux to the avaliables units
Warning steradian is a new fundamental units ( definition is nullptr )
* Remove debugging and includes
* Added 15 constants
* Update apps/toolbox.fr.i18n
Co-Authored-By: Quentin <quentin.guidee@gmail.com>
* Update apps/toolbox.fr.i18n
Co-Authored-By: Quentin <quentin.guidee@gmail.com>
* Corrected french and english
* Added hu translation
Correct de translation
added combining double acute accent
* q
* Revert "q"
This reverts commit 959786ae504ddf106249a3f799929471eecb8708.
Co-authored-by: Quentin <quentin.guidee@gmail.com>
Unsigned types are never negative, so the existant check does not make
sense. Check correctly using the newly introduced
TreeNode::IsValidIdentifier method.
A Multiplication whose children are all real or ComplexCartesian are
multiplied to form a single ComplexCartesian. That step of shallowReduce
assumed that the last child is a ComplexCartesian. Since a Unit is now
considered as real, the previous assumption can be maintained be
reordering the ExpresssionNode::Type.
First reduce the right member by replacing the units to make sure there
is a unit, then reduce without replacing to keep "complex units" such as
L, N, ...
