- better connection error handling
- fix a problem with metadata parsing does not work properly for table names
- make documentations more consistent wrt default langauge names
- add convenience functions for setting cache paths for data accessed via get_cansim_connection
- better handling of duplicated levels in metadata, ignore duplication for geography names of census tables but emit warning
- fix issue with accessing tables without footnotes
- ensure proper ordering of levels even if StatCan metadata is not ordered
- better error messages and information on how to disable peer checking when StatCan SSL certificates have problems
- automatically batch vector or coordinate data retrieval in case users request more than 300 series at a time
- enable series information by table and coordinate
- generate table template and facilitate adding vector info to aid pinpointed data download
- enable downloading of data by vector and multiple coordinates in get_cansim_data_for_table_coord_periods (breaking changes with change to parameter)
- fix problem with parsing census data tables
- fix problem with converting to factors when classification codes are attached.
- add support for local caching in parquet and feather formats
- uniform interface for sqlite, parquet, and feather caching
- principled approach to column order
- fix problem with inconsistent type parsing of notes
- better support for french language when accessing data by vector or coordinate
- tests
- fix problem with reading French tables released by the census division
- restore original column order after converting to factors
- convert geography column to factor if available
- fix problem with
add_provincial_abbreviationsthat could lead to mislabelling of provinces in some cases - improve handling of metadata, enable downloading only metadata instead of only via full table download
- fold metadata into data when accessing via vector or coordinates
- allow for cansim vectors in
view_cansim_webpageto view vector information from statcan in the browser
- improve offline handling when StatCan servers are down
- improve metadata handling when Member ID order is mixed up in metadata
- fix problem with refreshing data in get_cansim_vectors
- accommodate quirks in table 98-10-0017
- Better header parsing to avoid warning messages
- Fix problem with some semi-wide tables
- Speed up access to cached sqlite tables
- Fix problem with
get_cansim_vector_info()
- Fix bug that causes collect_and_normalize not to function on some operating systems
- Support for new semi-wide table format, e.g. Census data releases
- Improvement in offline handling of sqlite tables
- Better error handling when StatCan returns empty tables
- Add Hierarchy for Geography in sqlite tables
- Better fallback and warning messages when StatCan table categories are internally inconsistent
- Performance improvements
- deprecate
list_cansim_tablesandserach_cansim_tablesand fallback to corresponding "_cube" methods as Open Data Canada API has changed and similar functionality is available through the "_cube" methods that tie directly into StatCan APIS
- Fix issues with top level duplicate categories
- Check for expired tables in
list_cansim_sqlite_cached_tables - New auto-update feature for sqlite tables
- Exclude vignette from automatic CRAN checks to fix problem of CRAN checks failing when StatCan servers are down and lead to the package being removed from CRAN (checks are still active in local environment and when using GitHub action checks)
- add release date info to cube metadata and cube list calls
- add auto-refresh option for sqlite tables
- remove deprecated
adjust_cansim_values_by_variablefunction
- Fix problem with UTF-8 encoding on solaris
- move dbplyr dependence from Imports to Suggests
- Fold part of
normalize_cansim_valuesinto the default table and vector output, in particular always add a scaled variable column calledval_normand an imputedDatecolumn and covert categories to factors by default. - New
get_cansim_sqlitefunction that stores tables in an SQLite database and facilitates access and management of data.
- Adapt to changes in dplyr, tidyr, and tibble
- fix a bug that would not properly add hierarchies when category names are repeated
- Use system unzip if
getOption("unzip")is set to enable unzip for files larger than 4GB on unix-like systems
- Exclude all vignettes and example code from compilation as this may cause CRAN check errors when StatCan servers are down or otherwise temporarily unavailable
- Expand
get_cansim_table_notes()functionality - Add functionality to access the new cube list API
- Fix time zone problem when parsing and formatting times for the StatCan API
- Adjust package for changes in StatCan API with different metadata format
- Fixes issues arising from StatCan changing their API row limit
- Optimize vector retrieval by REF_DATE
- Fixes issues arising from StatCan changing their API
- Member Names come concatenated with Classification Code by default, this could break existing code.
- Adds option to change fields to factors
- Adds option to strip Classification Codes from fields
- Exposes timeout limit to deal with slow connections and large tables
- More robust table download functions
- Improved documentation
- Initial CRAN release
- French metadata implemented