Alias to rel.

Coerces a char to a corresponding constant.

Converts a seqable? of chars to formDNA.

Note that nuim-code is the default ordering and is always assumed for formDNA. If a different sort-code is specified, the input sequence (expected in this code) will be reordered to match nuim-code.

Comparator for constants using the default nuim-code.

Converts a constant to a digit corresponding to an optional sort-code or the default nuim-code.

Checks if the argument is a valid constant.

Set of all 4 constants.

Converts a digit to its corresponding constant representation.

Converts a seqable? of digits (as string/char or integer) to formDNA.

Note that nuim-code is the default ordering and is always assumed for formDNA. If a different sort-code is specified, the input sequence (expected in this code) will be mapped and reordered to match nuim-code. If you just want the mapping without reordering, use something like (mapv digit->const dna).

Converts formDNA to a sequence of digits corresponding to a sort-code.

Note that nuim-code is the default ordering and is always assumed for formDNA. If a different sort-code is specified, the input dna (expected in nuim-code) will be mapped and reordered to match this code.

Generates a vdict from a given formDNA (dna).

• optional sorted? defaults to false since sorting large vspace dimensions can be expensive

FIXME: write docs

Calculates the dimension of a formDNA/dna-seq (corresponds to the number of variables in a FORM). The length of a dna-seq is 4^d for its dimension d.

• the input sequence can have any type of elements

FIXME: write docs

Extracts a single value from a dna according to a given vpoint index, which is a sequence of constants corresponding to each term→value association.

Given a formDNA (dna), generates all of its permutations and returns a map from permuted term order to the corresponding formDNA.

FIXME: write docs

Equality check for formDNA. Two formDNAs are considered equal, if they contain the same constants in the same order. Stricter than equiv-dna?, where permutations are considered equal.

Note: partial formDNA (which includes holes (:_)) cannot be compared and thus are not valid input. If you know/assume equality for holes or just want to ignore them in comparison, use equal-partial-dna?.

Equality check for partial formDNA (derived from equal-dna?), under the assumption that all holes (:_) originate from the same expression and thus their supposed value would be equal (which is the same as just ignoring them).

Equivalence check for formDNA. Two formDNAs are considered equivalent, if they belong to the same equivalence-class of dna-perspectives (i.e. if they are permutations of each other).

Note: partial formDNA (which includes holes (:_)) cannot be compared and thus are not valid input. If you know/assume equality for holes or just want to ignore them in comparison, use equiv-partial-dna?.

Equivalence check for partial formDNA (derived from equiv-dna?), under the assumption that all holes (:_) originate from the same expression and thus their supposed value would be equal (which is the same as just ignoring them).

Expands a dna-seq to a given target dimension by repeating elements.

Note: dna-seq can have any type of elements (not only constants)

Obsolete → please use expand-dna instead.

Filters a dna by selecting specific parts corresponding to a given vpoint, which acts as a coordinate vector in its value space.

• use holes :_ in vpoint to indicate a variable selection

Inverts the value of a given constant or formDNA. With multiple arguments, will relate all arguments (via rel) and then invert the result.

Note: value holes (:_) will invert to value holes.

Given a sort-code (try nuim-code or nmui-code), returns a comparator function to sort single constants, formDNA or arbitrary sequences of constants (can be mixed).

• can also compare map-entries by keys of comparable types

Creates a formDNA from arguments, which may be valid chars, keywords, integers or sequences thereof.

• valid chars are: \n \u \i \m (upper- or lowercase) and \0 \1 \2 \3
• valid integers are: 0 1 2 3
• valid keywords are: :n :u :i :m
• total argument count (including count of sequence args) must match a valid formDNA length, which is 4^d, where d is a natural number

FIXME: write docs

FIXME: write docs

FIXME: write docs

Given a formDNA (dna), generates its permutation (called a “perspective”) that matches the given perm-order.

perm-order must be a sequence of indices that correspond to each term/subdna of the dna. Think of each place in the sequence as a depth (shallowest → deepest) and each index as an identifier of the term in that depth.

For example, [0 1 2] is the default order of terms in a 3-dimensional formDNA. [1 2 0] would permute the terms, such that in the resulting formDNA term 0 is in depth 2, term 1 in depth 0 and term 2 in depth 1.

Generates a random constant. A seed (an integer) can be provided as a second argument for reproducability.

Same as rand-const, but takes a weights argument to specify the relative probability of each of the four constants to be randomly chosen.

Weights can be provided either as:

• a sequence of 4 non-negative numbers (e.g. [1 0 2 5]) in n-u-i-m order
• a map (e.g. {:i 1 :u 2}), where missing weights are 0
• a single number in the interval 0.0, 1.0 that represents the ratio of :u/:i/m against :n (whose weight is 1 - x)

Generates a random formDNA of dimension dim. A seed (an integer) can be provided as a second argument for reproducability.

Same as rand-dna, but takes a weights argument to specify the relative probability of each of the four constants to be randomly chosen.

Weights can be provided either as:

• a sequence of 4 non-negative numbers (e.g. [1 0 2 5]) in n-u-i-m order
• a map (e.g. {:i 1 :u 2}), where missing weights are 0
• a single number in the interval 0.0, 1.0 that represents the ratio of :u/:i/m against :n (whose weight is 1 - x)

Generates a random vpoint with given dimension dim (= length of the vpoint). A seed (an integer) can be provided as a second argument for reproducability.

Same as rand-vpoint, but takes a weights argument to specify the relative probability of each of the four constants to be randomly chosen.

Weights can be provided either as:

• a sequence of 4 non-negative numbers (e.g. [1 0 2 5]) in n-u-i-m order
• a map (e.g. {:i 1 :u 2}), where missing weights are 0
• a single number in the interval 0.0, 1.0 that represents the ratio of :u/:i/m against :n (whose weight is 1 - x)

Reduces a dna-seq by eliminating redundant/contingent terms.

• returns a tuple [terms dna-seq], where terms is a sequence that represents the remaining terms after reduction
• takes an optional terms sequence of any kind of items that will be used instead of the default arithmetic sequence [0 1 2 …] to represent each term (length has to match the formDNA dimension)

Note: dna-seq can have any type of elements (not only constants)

Obsolete → please use reduce-dna instead.

Relates the values of all given arguments (which must either be all constants or all formDNAs) to each other.

Note: relations with value holes (:_) can only be dominated by the mark (:m) and will otherwise result in value holes.

Reorders given formDNA/dna-seq from sort-code-from to sort-code-to.

Note:

• dna-seq can have any type of elements (not only constants)
• does NOT change the encoding of the elements, just their ordering

Obsolete → please use reorder-dna instead.

FIXME: write docs

“Value hole” – placeholder for an unknown/missing result. Denotes a constant, but like a black box, we cannot know which one.

Generates a vdict given a map vpoint->result (result is a constant).

• if the corresponding vspace is not a subset of the set of keys from vp->r, the remaining results will be filled with :n or a given default constant
• optional sorted? defaults to false since sorting large vspace dimensions can be expensive

Generates a vmap from a given vdict.

FIXME: write docs

Returns the dimension of a vmap (equivalent to dna-dimension of the corresponding formDNA).

Given a group of all perspectives from a formDNA, returns these perspectives as vmaps.

FIXME: write docs

FIXME: write docs

Generates a vspace of dimension dim, optionally with custom sort-code.

• returns a lazy-seq which may be too memory-expensive to fully realize for dimensions greater than 11 (> 200 Mio. elements in total!)

FIXME: write docs

Alias to inv.