select_edges: Select Edges with Lazy Computation

Description

A powerful edge selection function with lazy computation (only computes metrics actually referenced), multiple selection modes, and structural awareness (bridges, communities, reciprocity).

Usage

select_edges(
  x,
  ...,
  top = NULL,
  by = "weight",
  involving = NULL,
  between = NULL,
  bridges_only = FALSE,
  mutual_only = FALSE,
  community = "louvain",
  keep_isolates = TRUE,
  keep_format = FALSE,
  directed = NULL,
  .keep_isolates = NULL
)

Value

A cograph_network object with selected edges. If keep_format = TRUE, matrix, igraph, and statnet network inputs are converted back to that type. Nodes left without edges are kept and reported in a cograph_isolates_created warning, unless keep_isolates = FALSE.

Arguments

x

Network input: cograph_network, matrix, igraph, network, or tna object.

...

Filter expressions using edge columns or computed metrics. Available variables:

Edge columns: from, to, weight, plus any custom

Computed metrics

abs_weight, from_degree, to_degree, from_strength, to_strength, edge_betweenness, weight_rank

Predicates

is_bridge, is_mutual (alias is_reciprocal), is_loop, is_multiple, same_community

Endpoint labels

from_label, to_label, from_community, to_community

top

Integer. Select top N edges by a metric.

Character. Metric for top selection. Default "weight". Options: "weight", "abs_weight", "edge_betweenness", "from_degree", "to_degree", "from_strength", "to_strength", "weight_rank".

involving

Character or integer. Select edges involving these nodes (by name or index). An edge is selected if either endpoint matches.

between

List of two character/integer vectors. Select edges between two node sets. Example: between = list(c("A", "B"), c("C", "D")).

bridges_only

Logical. Select only bridge edges (edges whose removal disconnects the graph). Default FALSE.

mutual_only

Logical. For directed networks, select only mutual (reciprocated) edges. Default FALSE.

community

Character. Community detection method for same_community variable. One of "louvain", "walktrap", "fast_greedy", "label_prop", "infomap", "leiden". Default "louvain".

keep_isolates

Logical. Keep nodes that end up with no edges? Default TRUE, matching igraph::delete_edges() and tidygraph: filtering edges does not remove nodes. Set FALSE to drop them, or call remove_isolates() afterwards.

keep_format

Logical. If TRUE, matrix, igraph, and statnet network inputs are returned in that format. Default FALSE returns cograph_network.

directed

Logical or NULL. If NULL (default), auto-detect.

.keep_isolates

Deprecated. Use keep_isolates.

Details

Selection modes are combined with AND logic:

select_edges(x, top = 10, involving = "A") selects top 10 edges among those involving node A
All criteria must be satisfied for an edge to be selected

Edge metrics are computed lazily - only those actually referenced in expressions or required by selection modes are computed.

Examples

Run this code

adj <- matrix(c(0, .5, .8, 0, .5, 0, .3, .6,
                .8, .3, 0, .4, 0, .6, .4, 0), 4, 4, byrow = TRUE)
rownames(adj) <- colnames(adj) <- c("A", "B", "C", "D")

select_edges(adj, weight > 0.5)
select_edges(adj, top = 3)
select_edges(adj, involving = "A")
select_edges(adj, between = list(c("A", "B"), c("C", "D")))