ipumsr 0.8.1

Minor improvements and fixes

• Small fix to vignette that was previously throwing an error in R CMD Check
  with --no-manual and --no-build-vignettes options (#80).

ipumsr 0.8.0

Breaking changes + deprecations

• Collection-specific definition functions for IPUMS microdata projects
  have been deprecated in favor of define_extract_micro(). This
  streamlines syntax across all supported microdata collections. Use the
  collection argument to specify the collection for your
  microdata extract.

  IPUMS NHGIS extracts are still created with
  define_extract_nhgis().

• Previously deprecated methods for retrieving extract history are now
  defunct. Use get_extract_history() instead.

New features

• Adds support for IPUMS API extract endpoints for IPUMS Time Use and
  IPUMS Health Surveys projects! This includes:

  • IPUMS Time Use

    • IPUMS ATUS ("atus")

    • IPUMS AHTUS ("ahtus")

    • IPUMS MTUS ("mtus")

  • IPUMS Health Surveys

    • IPUMS NHIS ("nhis")

    • IPUMS MEPS ("meps")

• Adds support for new rectangularization options for certain
  microdata collections. Use the rectagular_on argument in your
  extract definition to rectangularize on records other than person
  ("P") records.

• Time use variables can be included in IPUMS Time Use extracts. Use
  the tu_var_spec() helper with the time_use_variables argument of
  define_extract_micro() to create a time use variable
  specification.

  This includes both IPUMS-defined and user-defined time use variables. Note
  that user-defined time use variables cannot currently be created via API.
  However, if you have defined a time use variable through the IPUMS web
  interface, you can request it in an extract via API.

• Sample members can be specified in IPUMS ATUS extracts. Use the
  sample_members argument of define_extract_micro() to do so.

• "household_only" is now a supported value for data_structure in
  IPUMS USA extracts.

Minor improvements and fixes

• Fixes bug in add_to_extract.micro_extract() that caused an error if a user
  attempted to update the value of data_quality_flags in a detailed variable
  specification.

ipumsr 0.7.2

• Fix for CRAN checks

ipumsr 0.7.1

• Minor updates to README and vignettes

ipumsr 0.7.0

Function retirements

• ipumsr no longer suggests raster, rgdal, and sp (#23). Removing these
  dependencies requires the retirement of several previously deprecated
  functions:

  • All read_terra_*() functions
  • read_ipums_sp() (use read_ipums_sf() to load spatial data in sf format)
  • read_ipums_codebook() (use read_nhgis_codebook() to load an NHGIS
    codebook)
  • ipums_list_*() helper functions (instead, use ipums_list_files())
  • read_nhgis_sf() and read_nhgis_sp() (instead, use read_ipums_sf() and
    read_nhgis() to load spatial and tabular data separately, then join
    with an ipums_shape_join_*() function)

Other updates

• Fixes bug in ipums_view() where content would not display properly in
  viewer pane (#19)

• RStudio is now explicitly required to launch files in the viewer pane using
  ipums_view(). R console users can still generate stand-alone HTML files
  that can be viewed in a different browser.

• Updates the UI and fixes various bugs in ipums_website() (#54):

  • Accepted project names are now consistent with those provided in
    ipums_ddi objects. Shorthand project names as used by the IPUMS API are
    also accepted.
  • Users can now use the same syntax regardless of whether providing
    an ipums_ddi object or a project name. The project argument has been
    deprecated.
  • MacOS is now supported
  • The var argument is no longer required. Omitting var will launch the
    URL to the homepage for the specified IPUMS project.
  • The homepage_if_missing argument now defaults to FALSE.
  • The var_label argument has been deprecated

• Updates IPUMS projects listed in ipums_data_collections().

• Various documentation updates

ipumsr 0.6.3

• The ability to read an IPUMS DDI file contained within a zip archive using
  read_ipums_ddi() has been deprecated. Users must now load DDI files by
  providing the direct path to the uncompressed .xml file.

  This resolves an inconsistency between the behavior
  of read_ipums_micro_*() functions when provided a DDI file path as compared to
  an ipums_ddi object created with read_ipums_ddi().

• The ability to read IPUMS files by providing the path to their containing
  directory has been deprecated. This affects:

  • read_nhgis()
  • read_ipums_sf()
  • read_nhgis_codebook()
  • read_ipums_ddi()
  • ipums_list_files()

  These functions now require either a zip archive (with the exception of
  read_ipums_ddi()—see above) or a direct file path as input.

  This has the consequence that read_ipums_sf() with bind_multiple = TRUE
  requires a zip archive as input if multiple files are to be read and combined.

• Documentation updates for clarity

ipumsr 0.6.2

• Fixes for CRAN checks.

ipumsr 0.6.1

Fixes for CRAN checks

ipumsr 0.6.0

ipumsr 0.6.0 introduces significant updates to the client tools for the IPUMS API and revamps several IPUMS readers, particularly for IPUMS NHGIS.

Breaking Changes + Deprecations

IPUMS API

• ipumsr now supports IPUMS API version 2, and no longer supports
  either the beta version or version 1 of the IPUMS API.

  This means that extract definitions saved in JSON format will no longer
  be compatible with ipumsr via define_extract_from_json(). To load
  extract definitions created under previous API versions, there are two
  options:

  • Rewrite the extract definition represented in the JSON file using the
    define_extract_*() function for the relevant IPUMS collection, then update
    the saved file with save_extract_to_json().

  • Update the JSON file by converting all snake_case fields to camelCase.
    For instance, "data_format" would become "dataFormat".
    The "api_version" field will also need to be changed to "version" and
    set equal to 2.

  See the IPUMS developer documentation for more details on API versioning
  and breaking changes introduced in version 2.

• The ipums_extract object structure has been updated. For IPUMS microdata
  projects, variables and samples are no longer stored as character vectors,
  but as lists. This accommodates new API version 2 features (see below). Use
  names(x$variables) instead of x$variables to access variable (and sample)
  names as a character vector.

• get_recent_extracts_info_*() functions have been deprecated. Additionally,
  tabular-formatted extract history is no longer supported, and conversion
  functions extract_tbl_to_list() and extract_list_to_tbl() have
  therefore been deprecated as well.

  Use get_extract_history() to browse previous extract definitions in list
  format.

IPUMS Readers

• read_nhgis_sf() and read_nhgis_sp() have been deprecated. Use
  read_ipums_sf() and read_nhgis() to load spatial and tabular
  data, respectively. Join data with an ipums_shape_*_join() function.

• data_layer and shape_layer arguments have been deprecated in favor of
  file_select throughout ipumsr. This provides clarity on the intended purpose
  of this argument. Deprecated functions that use the original argument names
  remain unchanged.

• Support for objects from the sp package has been deprecated because of the
  upcoming retirement of rgdal. Use read_ipums_sf() to load spatial data
  as an sf object. To convert to a Spatial* object, use sf::as_Spatial().

  For more, see r-spatial's post covering the evolution of several spatial packages.

• read_ipums_sf() no longer defaults to bind_multiple = TRUE.

• Individual ipums_list_*() functions have been moved to ipums_list_files().

• Example files in ipums_example() have been updated and include new file
  names.

IPUMS Terra

• Support for IPUMS Terra has been discontinued. This includes deprecations
  to all read_terra_*() functions, the types = "raster" option in
  ipums_list_files(), and read_ipums_codebook().

  NHGIS codebook reading was previously supported by read_ipums_codebook().
  This functionality has been moved to read_nhgis_codebook().

  For more on IPUMS Terra decommissioning, click here.

New Features

IPUMS API

• Adds API support for IPUMS NHGIS and IPUMS International!

  • Use define_extract_nhgis() to create an NHGIS extract definition.
  • Use define_extract_ipumsi() to create an IPUMS International extract
    definition.

• Adds support for IPUMS API version 2 features! This includes:

  • Detailed variable specifications for IPUMS microdata extract
    definitions, including case selections, attached characteristics,
    and data quality flags. Use var_spec() to add these specifications
    to variables in your extract definition.
  • Additional definition-wide parameters for IPUMS microdata extracts,
    including data_quality_flags and case_select_who.
  • Hierarchical extracts for IPUMS microdata extracts. Set
    data_structure = "hiearchical" to create a hierarchical extract definition.
  • Year selection for time series tables in IPUMS NHGIS extract definitions.
    Use tst_spec() to add year selections for time series tables.

• Adds API support for IPUMS NHGIS metadata

  • Use get_metadata_nhgis() to browse NHGIS data sources. Metadata
    is available in summary form for all datasets, data tables, time series
    tables, and shapefiles as well as for individual datasets, data tables, and
    time series tables.

• Allows users to set a default IPUMS collection using
  set_ipums_default_collection(). Users with a default collection do not
  need to specify the IPUMS collection in functions that require it; instead,
  the default collection is used. This is a convenience for users who rely
  primarily on a single IPUMS collection.

• wait_for_extract() wait intervals no longer double after each status
  check. Instead, intervals increase by 10 seconds with each subsequent check.

IPUMS readers

• Adds handling for fixed-width NHGIS extracts in read_nhgis().

• read_nhgis_codebook() allows reading
  of raw codebook lines (as opposed to extracting codebook information into
  an ipums_ddi object) by setting raw = TRUE. Furthermore, var_info
  generated from NHGIS codebook files has been updated to include more
  contextual information about the data variables.

• read_nhgis() now supports additional arguments to refine the data loading
  process. Users can now specify col_types manually and read a subset of
  a data file using vars and n_max.

• read_nhgis() now allows users to retain the extra header row included in
  some NHGIS files. Set remove_extra_header = FALSE to do so. In general,
  the information contained in the extra header is attached to the data from
  the NHGIS codebook file, but in some cases the extra header may differ
  slightly from the information found in the codebook.

Miscellaneous

• Various bug fixes

• Updates to documentation and vignettes for clarity

Thanks to @dtburk and @renae-r for their work on these updates!

v0.5.2

What's Changed

• Add progress bar when downloading extracts
• Removed validate argument from extract revision functions and improved warning messages for invalid extract field names when used as arguments.
• Fixed bug preventing users from providing their API key directly to submit_extract and wait_for_extract.