Package 'owdata'

List the datasets shipped by published packages

Description

Dataset detail is harvested for published packages only; an unpublished package appears in owd_packages() but not here.

Usage

owd_datasets(pkg = NULL)

Arguments

pkg	Optional character vector of package names to keep. NULL (the default) returns all rows.

Value

A tibble with one row per dataset, with columns pkg_name, dataset_name, title, description, n_rows, n_vars, n_vars_dictionary, csv_url and xlsx_url.

Examples

owd_datasets()
owd_datasets(pkg = "washmalawi")

---

Install openwashdata packages from the openwashdata R-universe

Description

A thin wrapper around utils::install.packages() that points at https://openwashdata.r-universe.dev (with CRAN as fallback for dependencies).

Usage

owd_install(pkg = NULL, ...)

Arguments

pkg	Character vector of package names to install. NULL (the default) installs every published package after an interactive confirmation showing the count; in non-interactive sessions NULL is an error, so scripts must name packages explicitly.
...	Passed on to utils::install.packages().

Value

Invisibly, the character vector of package names requested.

Examples

## Not run: 
owd_install("washmalawi")
owd_install()

## End(Not run)

---

List the openwashdata packages in the catalog

Description

Returns the harvested package table. Every detected package is listed, whether or not it has been published; use published to restrict the result.

Usage

owd_packages(published = NULL)

Arguments

published

Optional logical. TRUE keeps published packages, FALSE keeps the rest, NULL (the default) returns all rows.

Details

A package counts as published when its CITATION.cff carries a Zenodo DOI. There are no version tiers, so a version number says nothing about whether a package is published.

Value

A tibble with one row per package and these columns: pkg_name, title, description, version (leading v stripped), published, reviewed (NA until the pkgreview check is wired), maintainer, maintainer_orcid, authors, license, date, keywords, spatial_coverage, temporal_coverage, doi, cff_version, url_github, url_docs, n_datasets, latest_release, default_branch, created_at, last_commit, topics, has_dictionary, has_extdata, and the tooling provenance columns washr_version, brand_ref, pkgreview_standard, has_r_cmd_check, has_pkgdown_workflow and has_jsonld.

Examples

owd_packages()
owd_packages(published = TRUE)

---

Search the openwashdata catalog

Description

Runs a case-insensitive regular expression over package, dataset and variable metadata, so owd_search("groundwater") answers "which package has data about groundwater". The search runs against a prebuilt flat index, so it is a single vectorized match regardless of catalog size.

Usage

owd_search(
  query,
  fields = c("title", "description", "variable_name", "dataset_name")
)

Arguments

query	A single regular expression (case-insensitive). Plain words work as-is.
fields	Character vector of fields to search. Any of "title", "description", "variable_name", "dataset_name", "pkg_name". The default covers titles, descriptions, variable names and dataset names.

Value

A tibble of hits with columns pkg_name, match_type ("package", "dataset" or "variable"), dataset_name, field and text (the matched text). Zero rows when nothing matches.

Examples

owd_search("sanitation")
owd_search("water", fields = "variable_name")

---

List the documented variables across the catalog

Description

The cross-package variable dictionary, harvested from each package's data-raw/dictionary.csv. Published packages only, as in owd_datasets().

Usage

owd_variables(pkg = NULL, dataset = NULL)

Arguments

pkg	Optional character vector of package names to keep.
dataset	Optional character vector of dataset names to keep.

Value

A tibble with one row per documented variable, with columns pkg_name, dataset_name, variable_name, variable_type (first class only) and description.

Examples

owd_variables(pkg = "washmalawi")
head(owd_variables())

---