Skip to contents

Changelog

Source: NEWS.md

hydroloom 1.2.0

This release introduces an S3 class hierarchy (hy_topo, hy_leveled, hy_node, hy_flownetwork) that lets hydroloom functions validate input at dispatch time and provide guided error messages when the wrong network representation is passed. Existing code that passes data.frame or hy objects continues to work without changes – the new classes are assigned automatically and are transparent to downstream consumers. Package developers who depend on hydroloom should note that returned objects now carry subclass attributes (e.g. hy_topo) which are stripped by hy_reverse() and by standard dplyr operations.

  • Outlet detection is explicit (#85): a row is an outlet when its toid is not in id. Canonical reserved values (0 / ""), NA, implicit absence, foreign reserved values, and unique-per-outlet ids are all accepted.
  • hy() still normalizes NA toid to the canonical reserved outlet value, so x$toid == 0 after hy() keeps working.
  • make_node_topology() and is.hy() accept NA and orphan toid; the related errors are removed.
  • check_hy_outlets() warns only on id/toid type mismatch.
  • sort_network() warns only when a network has zero outlets.
  • Improve performance of add_levelpaths() by converting to data.table
  • Add S3 class hierarchy: hy_topo, hy_leveled, hy_node, hy_flownetwork – #73
  • hy() gains add_topo parameter to auto-build toid from fromnode/tonode
  • New exported helpers: hy_network_type(), is_dendritic(), hy_capabilities()
  • Print methods for hy_topo, hy_node, hy_flownetwork
  • Producer functions now stamp output classes: add_toids() -> hy_topo, sort_network() -> hy_topo, add_levelpaths() -> hy_leveled, make_node_topology() -> hy_node, to_flownetwork() -> hy_flownetwork
  • add_divergence() sets attr(x, "dendritic") <- FALSE on output
  • add_toids(return_dendritic = FALSE) is deprecated; use to_flownetwork()
  • S3 method dispatch: functions now dispatch on subclass (e.g. .hy_topo, .hy_leveled) with guided error messages for wrong input class
  • Functions that require hy_leveled (e.g. add_pfafstetter(), navigate_hydro_network(), to_flownetwork()) fall through from hy_topo/hy_node when required columns are already present
  • Fix pre-existing bug in make_to_dt() where dendritic branch failed on tibble input (data.table with = FALSE syntax on plain data.frame)
  • Fix get_bridge_flowlines() correctness on networks with independent terminals: make_nondendritic_topology() collapsed all rows carrying the reserved outlet value into one synthetic node, misclassifying bridges and exhausting memory on continental networks
  • to_flownetwork() and get_bridge_flowlines() now accept hy_node input without divergence/levelpath, auto-converting to a non-dendritic edge list with a warning that main-path info is dropped
  • hy_flownetwork no longer inherits from hy – it is a separate junction table where id is not guaranteed to be a primary key, and it does not pass is.hy(). Documented behavior already; the implementation now matches
  • Documentation:
    • Add class-level roxygen pages for hy_topo, hy_leveled, hy_node, and hy_flownetwork describing representation pattern, required columns, supported functions, and conversion paths
    • Add divergence case study to vignette("non-dendritic") showing how a secondary path is dropped in hy_topo form and preserved in hy_flownetwork form
    • Add hy_capabilities() pipeline walkthrough to vignette("network_navigation") demonstrating the hy -> hy_node -> hy_topo -> hy_leveled -> hy_flownetwork progression
    • Style and clarity pass across all vignettes
  • Domain decomposition gains a containment relationship that lets callers declare, post-decomposition, that one domain (a catchment aggregate, in HY_Features terms) is enclosed by another – endorheic basins, drainage-divide remnants, or any isolated component the caller wants treated as contained. New set_containment() records the declaration; get_containing_domain() reads it back; get_domain_graph() returns one row per declaration (with nexus_id = NA and relation_type = "contained"); recompose() gains a containment argument that defaults to "ignore" (each basin’s accumulated value stops at its own outlet) and optionally "accumulate" (the contained basin’s accumulated value is added at the containing domain’s outlet and routed downstream through the containing basin’s extensive network). Containment does not appear in nexus_registry because no flow crosses a hydro nexus between the two basins.
  • Deprecation notice: A future release will require that hy_topo objects have unique id values (one row per catchment). Non-dendritic connectivity with duplicated ids in a toid-based edge list will need to be represented as hy_flownetwork (via to_flownetwork()). Developers who currently pass non-dendritic toid tables through hydroloom functions should migrate to to_flownetwork() or make_node_topology() for non-dendritic workflows.

hydroloom 1.1.3

CRAN release: 2026-02-20

Update test tolerances for failing Fedora CRAN tests

hydroloom 1.1.2

CRAN release: 2026-02-20

Hydroloom 1.1.2 introduces new functionality in accumulate_downstream(), reworks the make_index_ids(), and deprecates make_fromids() and format_index_ids() for clarity of package function. Deprecated functions will be removed in the next major version release.

hydroloom 1.1.1

CRAN release: 2025-10-01

  • fix bug with sort_network when duplicate entries are in the extended attributes – #52
  • add specific id search to index_points_to_lines for #24
  • error handling #49

hydroloom 1.1.0

CRAN release: 2024-08-26

hydroloom 1.0.1 and 1.0.2

CRAN release: 2024-01-09

  • Add checks if nhdplusTools is available.

hydroloom 1.0.0

CRAN release: 2023-09-29