- Change definition and processing code for BDS 71 (birth land parents) to character (#9)
- Adds two demo files:
examples/Laura_S.json and examples/Kevin_S.json
- Changes schema
inst/schema/bds_v3.0.json to point to master branch
- Replace non-functional
inst/examples/maria.json by an older version
Non-breaking change in the V3.0 schema definition
- Changed the type definition of BDS field 71 (parental birth country) from
integer to number
Added support in V3.0 schema definition
- Adds support in
read_bds() and write_bds() for
- parental country of birth (BDS 71, uses government Table 34 for codes)
- parental level of eduction (BDS 62/66)
- parity (BDS 471)
- 4-digit postal code (BDS 16)
Other changes
- Removes the empty array writing from
write_bds()
- Adds a developer vignette outlining the steps needed to add a BDS number
- Renames source file names for improved consistency
- Access
dplyr::case_match() by requiring dplyr 1.1.0
- Correct documentation errors
- Major overhaul of
bdsreader package
Major changes
- Reorganises calculations in function
read_bds() into 13 steps
- The new argument
validate to read_bds() can bypass the jsonvalidate::json_validate() to speed up data reading. The default is FALSE. Use validate = TRUE to obtain diagnostic information.
- Function
read_bds() now writes intermediate JSON file to the working directory by setting intermediate = TRUE.
- Much faster reading of DDI BDS fields by smarter data mungling
- Prepares for deprecation of JSON schema v1.0 and v2.0 by splitting processing in
read_bds() according to major version number
- Translates communication into the English language
Minor changes
- Updates all tests according to new functionality
- Retires the
verify() function and its tests
- Cleans up code by
lintr
- Updates example code, tests and documentation to the JSON V3.0 schema
- Solves problem with
write_bds() that failed to save gestational age for lollypop data
- Tweaks reading logic so that lollypop data are read
- Redefines ‘reference’ instead of ‘Reference’ in V3.0 schema
- Splits
write_bds() in subfunctions per major version
- Extends tests for
read_bds() to cover three major versions
- Updates
write_bds() so that it conforms to JSON BDS schema V3.0
- Prevent upstairs assignments within the
switch() statement
- Makes the reading of the data more lenient so that
name and src are properly read for V1.0-V3.0 data
- Repairs V3.0 examples so that these conform to V3.0 schemas
Bug fixes
- Repairs errors in selected bds fields in V2.0 and V3.0 schemas:
- 66 (added code “09”)
- 911 –> 912
- 944 –> 945
- 960 –> 961
- 965 –> 966
- 942 –> 943
- 944 –> 945
- 973 (added)
- 1001 (added)
- Makes
parse_valid() robust against BDS sequence
- Changes default format in
write_bds() to JSON schema V3.0
- Updates dependency versions
Major changes
- Defines BDS JSON schema V3.0 for exchanging child-level data between electronic systems. V3.0 uses an English data definition and redefines the nesting structure to a more logical structure. The new schema mostly follows alterations developed by Eaglescience BV for the SRM made in the period 2022-2023.
- From now on, the advice is to adopt V3.0 for data exchange. Versions V1.0, V1.1 and V2.0 will be retired in the future.
- Adds support for version V3.0 to functions
read_bds() and write_bds()
- Changes the key for calculating the D-score to
dscore(..., key = "gsed2212")
- Tests new functionality on 74 test files stored in package jamesdemodata
Minor changes
- Removes deprecated tidyselect
.data$var syntax
- Refreshes GH action scripts
- Adds script to simulate user bugs and requests
- Adds example data
maria and examples/maria.json
- Repairs an error with the coding of “smoking during pregnancy”
- Update documentation
- Removes class attribute
target, so now the target data is just a list with elements named "psn" and "xyz".
- Breaking change: Replace return value of
read_bds() from “tibble with attribute” to an object of class target. This will provide a complete export of the data when converted into JSON format.
- Do not use dot parameters anymore for
jsonlite::FROMjson(), which errors on alien argument names
- Make the
Format field a required element in schema bds_2.0.json
- Solves a bug that prevented
read_bds(txt) to read data from a URL
- Adds a small dataset
minidata for demo purposes
- Adds function
export_as_bds() that converts donordata into JSON files
- Repairs a problem with the validation of JSON string input (#4)
- Updates the test files to more recent
jsonvalidate output
- Hacks away parts of
parse_valid() that does not work under format 2.0
- Specify dependency on
jsonvalidate 1.3.2
- Sets return value of
set_schema()$schema to the long file name including path
- Solves a bug that prevented reading DDI under format = “2.0”
- Solves various runtime bugs
- Solves a problem with empty names in
bds_write()
- Implements
auto_format that stores the format in the data file (#2)
- Use
jamesdemodata package for testing and demo’s
- Removes some example files from
inst/examples
- Adds examples
maria1.json and maria2.json
- Updates README
This is a major update that introduces schema versioning for JAMES. Schema versioning allows input data to be formatted according to one of multiple JSON-schemas.
- Introduces schema versioning;
- Adds multi-version support to
read_bds(), write_bds() and related functions;
- Makes
format the most prominent user-facing argument. Alternatively, the user can also specify the schema file directly using the schema argument;
- Shortens names of schema files to
bds_v{x.y}.json;
- Renamee existing schemas as follows:
-
bds_schema_str.json –> bds_v1.0.json (format = 1)
-
bds_schema.json –> bds_v1.1.json
-
bds_schema_V2.json – bds_v2.0.json (format = 2)
- For clarity, renames the path from
inst/json to inst/schemas;
- Updates the
$id field in the schema’s to their permanent locations.
The default is format = 2L, which is incompatible with early users who code their data with bds_schema_str.json. To read/write with the older format, call read_bds(..., format = 1L) or write_bds(..., format = 1L). The schema argument is primarily meant for development and testing.
- Adds GHA R-CMD-check for Ubuntu 18.04
- Tweaks package description
- Switches on continuous integration
- Adds Github action
pkgdown
- Adds Github action
R-CMD-check
- Replaces
docs folder by gh-pages branch
- Set
bds_schema_v1.1.json as default!! (not all testfile yet work)
- Replaces the
jamestest package by jamesdemodata
- Supports (testing of)
bds_schema_v1.0.json (old) and bds_schema_v1.1.json (new)
- Adds a
file argument to write_bds()
- Updates scripts to account for double schema
- Removes the
appendLF = FALSE argument from message
- Adds new
bds_write() function
- Sets tighter ranges in
bds_schema_v1.1.json
- Changes the default JSON validation schema to
bds_schema_v1.1.json
- Adds example json from
bdsreader package
- Adds item splitting code for two BDS fields (head lag, walking)
- Adds a check on the existing of proper attribute in
persondata()
- Adds a
NEWS.md file to track changes to the package