lambda is more efficient then bind

_identifiers are reserved for use for instance in the standard library

Tomáš Pecka authored 4 years ago and Jan Trávníček committed 4 years ago

Although #pragma once is not a standard, compilers nowadays usually
support it. We don't have any tricky build system so we should be OK.

This also fixes many errors regarding using reserved identifiers (names
starting with double underscore).

removal of an unnecessary recursion, addition of forgotten move, removal of
unnecessary const qualifier, change int to -> size_t

the generator is parametrized with number of states, initial states, and final
states; additionally duplicate, redundant, unreachable states are generated

transition function is filled in transition partitions:
{state, unreachable, redundant} vs {state, unreachable, redundant}

The epsilon transitions from the new initial state were labelled with empty,
should be epsilon

const regerence and rvalue reference may be simulated by value semantics

The register class should have explicit constructors, the member variable data should
be initialized in the member initializer list, should be named m_data, and the
finish callback should accept the data by const reference

Explicit constructors, transform, and any_of

Add explicit to constructors, refactor code to use standard library algorithms
more.

Also make the test section names more clear

I redesigned the algorithm to make it similar to NPDA run implementation, most
importantly to handle cycles, also not even the very basic test that I
introduced didn't pass.

Additionaly, the graphStructuredStacks used to track the pushdown store and output were
incorrectly parametrized with InputSymbolType instead of PushdownStoreSymbolType
and OutputSymbolType.

NPDA execution through BFS-like simulation. Only exposed via Accept algo.

The variable i can be constructed from the begin and current iterator to the
input.

The ToAutomaton should be an aggregate/alias for the default algorithms.

Also add getTransition from state and transition to state to NPDA.

The algorithm generated rules of the automaton with the top of the pushdown
store symbol on the right, which is according to the algorithm, however, the
representation of PDAs and every PDA processing algorithm assumes the top of the
pushdown store is on the left.

The IsLanguageEmpty algorithm returned wrong result (the opposite boolean) because of wrong use of the set::count method. Also use set::contains to make the code clearer.

So we easily know what version of cppcheck and clang-tidy
are we running. Compiler version is announced by CMake.

Recently we moved infrastructure repositories under a new path. Docker
registries have therefore moved too.  Reflect this in our CI settings.