Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
0.8.3 * `stirling` numbers of `1st` (signed/unsigned) & `2nd` kind * lex generation of `involutions`, in progress (using tentative FXT algorithm) * `combine/complete` algebraic composition method (eg `k-Derangements = Composition of k-Combinations of Fixed Points + (n-k)-Derangement`) 0.8.4 * default or natural `mergesort` * fix a typo in `difference` utility method * `complement` utility method with sorting detection and automatic sorting if needed to get correct difference/complement * `Abacus.Partial` combinatorial iterator for `Partial` & `Explicit` (eg fixed) combinatorial data & conditions, eg can compose (w/ "complete" method) to produce conditional combinatorial templates eg conditional permutations w/ `boolean`/`algebraic` conditions (i.e "positive/negative" values by position, etc..) * `Abacus.Partial` can take (array of) array of values or (array of) array of conditional values + positions * `Abacus.CombinatorialIterator.output` method can map the combinatorial item to output "base/encoding" (& `Abacus.Tensor` support for multiple bases/encodings) 0.8.5 * `Abacus.CombinatorialIterator.compose` generic method to compose(& un-compose) combinatorial objects (& aliases per composition method eg `.connectWith`, `.completeWith`, `.projectOn`, `.concatWith`, ..) * (max) `M-compositions` (lex/lex,reflected forward/backward) * correct initial composition with M max part both forward & backward * fix wrong computation of `#compositions` with M maximum part 0.8.6 * fix still another typo in `difference` utility method * rename `Abacus.CombinatorialIterator.compose` to `Abacus.CombinatorialIterator.fuse` * optimise generation of comnditional combinatorial tensor data (ie Partial, CombinatorialTemplate) * add Abacus.CombinatorialTemplate which uses a pattern template to parse and generate conditonal partial data fused with other combinatorial objects to produce combinatorial objects matching template patterns * fix `fuse` methods for tuples, ordered permutations, etc.. when base is different than dimension (as in permutations and so on..) * make `Abacus.Partial` a sub-type of `Abacus.Tensor` ie `Abacus.Tensor.partial` 0.8.7 * fix a typo in random combination mergesort * add counting (only) of derangements with k fixed points & permutations with k cycles (stirling numbers) * `Abacus.CombinatorialIterator` support for sequences (ie sums) of other `CombinatiorialIterators` as actual combinatorial iterator instead of just abstract base class * subsets in (subset-)lex(&minimal), colex, combination & binary order (forward/backward) (partialy complete) 0.8.8 * fixed-length items (& attached effective length as extra last pos) & loopless algorithms (only subsets, todo for partitions & compositions) * adjust C/P/T methods to account for correct transformations of fixed-length objects with attached effective length * custom output functions or arrays or strings * generate partitions with k parts as partitions with m max part with conjugate output * conditional tensor data, can have custom min,max range instead of default 0,n-1 * partial support for minimal ordering (subsets, todo for tensors, permutations,combinations) * (n,k)-gray code output function for tuples * fixed-length items (& attached effective length as extra last pos) for partitions & compositions * fixed some issues with random generation of partitions/compositions and reflected/colex orderings, including with fixed-length representation * Iterator.DUAL takes into account dir of application (esp. for fixed-length representations) to construct correct reflected & complemented duals, ..
- Loading branch information