Changes in version 1.1.0 (2026-10-01)                  

This release fixes a number of bugs, rewrites the blockmodeling
algorithms and speeds up several functions considerably. Results of
signed_blockmodel() and signed_blockmodel_general() differ from earlier
versions (see below).

Bug fixes

  - as_adj_complex() now respects the attr argument and works with named
    vertices. This also fixes laplacian_matrix_complex(),
    as_incidence_complex() and complex_walks() for attributes other than
    "type".
  - laplacian_matrix_signed(sparse = TRUE) no longer errors. Normalized
    Laplacians (signed and complex) no longer return NaN for isolated
    vertices.
  - sample_islands_signed() always returns islands.n * islands.size
    vertices (previously failed when the highest-numbered vertices had
    no edges).
  - balance_score(method = "walk") no longer overflows to NaN on dense
    networks. balance_score(method = "triangles") returns NA for
    networks without triangles.
  - triad_census_signed() returns the census in canonical order.
  - eigen_centrality_signed() returns a real vector for directed
    networks (or errors if the dominant eigenvalue is complex).
  - ggblock() colors ties correctly if only one sign is present.
    ggsigned(type = "complex") now uses attr for edge colors and no
    longer overwrites the type attribute.
  - as_signed_proj() handles vertex names containing "-", "pos" or
    "neg".
  - complex_walks() validates k.
  - signed_blockmodel_general() returned a membership that did not match
    the reported criterion, and a wrong (even negative) criterion for
    alpha != 0.5. The meaning of alpha now matches signed_blockmodel().
  - signed_blockmodel(annealing = FALSE) no longer returns the random
    initial partition for networks with large blocks, and it always
    takes the best improving move.
  - signed_blockmodel() and signed_blockmodel_general() validate k,
    alpha and blockmat (square, symmetric for undirected networks).

Maintenance

  - Signed adjacency and incidence matrices are built internally instead
    of using the deprecated attr argument of igraph. Values of multiple
    edges are still summed.
  - Shared input validation. Errors no longer include the internal call.
  - Triad census lookup tables moved from the R source to internal data
    (data-raw/triad_tables.R).
  - Removed unused C++ code and added src/Makevars.
  - sample_gnp_signed() and sample_bipartite_signed() validate p_neg.
  - Added examples for ggblock(), ggsigned() and frustration_exact().
  - Requires R >= 3.5.

Performance

  - count_signed_triangles(), signed_triangles() and
    count_complex_triangles() are vectorized (several hundred times
    faster on networks with many triangles).
  - as_incidence_complex() and degree_signed() are vectorized and use
    sparse matrices. complex_walks() is about twice as fast.
  - graph_circular_signed() computes arc lengths directly from the
    sampled angles, which also avoids NaNs from rounding errors.
  - Both blockmodeling functions share a new C++ implementation (greedy
    local search and simulated annealing) that no longer copies the
    adjacency matrix for every move. signed_blockmodel(annealing = TRUE)
    no longer uses stats::optim() and is about 25x faster.

                 Changes in version 1.0.6 (2025-11-06)                  

  - fixed a bug in eigenvector centrality calculation

                 Changes in version 1.0.5 (2025-02-05)                  

  - fix all deprecated igraph calls

                 Changes in version 1.0.4 (2024-01-08)                  

  - fix deprecated support of adjacency matrices with character values
    #26

                 Changes in version 1.0.3 (2023-12-15)                  

  - code refactoring
  - more tests #17
  - removed deprecated calls to aes_ #23

                 Changes in version 1.0.2 (2023-09-08)                  

  - fixed a bug in signed_triangles() that resulted in wrong vertex ids
    (#20)

                 Changes in version 1.0.1 (2023-01-27)                  

  - fixed an error which occurs with the new version of igraph
    (https://github.com/igraph/rigraph/pull/633)

                 Changes in version 1.0.0 (2022-12-22)                  

  - added code of conduct
  - added contributing guide
  - added frustration_exact() to vignette
  - added utility functions
    is_signed,graph_from_adjacency_matrix_signed, and
    graph_from_edgelist_signed()
  - added random graph models sample_gnp_signed(),
    sample_bipartite_signed()

                 Changes in version 0.8.1 (2022-10-18)                  

  - fixed existing check errors

                 Changes in version 0.8.0 (2022-02-13)                  

  - added frustration_exact() to compute the exact number of frustrated
    edges
  - fixed issue with aggregate on r-devel

                 Changes in version 0.7.1 (2021-04-28)                  

  - fixed #7
  - fixed copy paste error in as_unsigned_2mode()
  - fixed aggregate error in as_signed_proj()

                 Changes in version 0.7.0 (2020-10-21)                  

  - added triad_census_signed()

                 Changes in version 0.6.0 (2020-08-13)                  

  - added avatar dataset
  - speed up of blockmodeling for larger networks

                 Changes in version 0.5.3 (2020-06-30)                  

  - fixed issue in complex_walks()
  - fixed faulty calculation of directed pn_index()

                 Changes in version 0.5.2 (2020-03-04)                  

  - fixed stringsAsFactors issue in complex_matrices.R

                 Changes in version 0.5.1 (2020-02-04)                  

  - fixed C++ issue for circular arc graphs
  - fixed failing eigen centrality test

                 Changes in version 0.5.0 (2020-01-24)                  

  - added vignettes and tests

                        Changes in version 0.1.0                        

  - initial version