- Minor fix in documentation to comply with new CRAN tests
- Re-document to fix HTML5 problems
- Added a
NEWS.md
file to track changes to the package.
This release contains very minor changes to comply with CRAN policies.
- Don't use
LazyData
, since the package doesn't contain data
Other changes
- The function
read.sss()
has a new argumentverbose = FALSE
that suppresses any messages. If you want the messages, setverbose = TRUE
- This release fixes two issues that might introduce unintended consequences:
- dealing with embedded tags (#9)
- dealing with self-closing tags (#10)
- Tests conform to
testthat_3.0.0
- Mostly small cosmetic changes to keep the package up to date.
read.sss()
will now also guess*.dat
files for the data, in addition to*.csv
and*.asc
- You can now specify only the .sss file in
read.sss()
, and the function will guess the name of the .asc - Added
label.table
attribute - Internal changes:
- Imports
xml2
instead ofXML
- Better error checking using
assert_that()
- Imports
- Added support to read csv files
- Now supports multi-byte files (UTF-8)
- Expanded unit tests with real-world data
- No functional changes
- Added
.Rbuildignore
to comply with CRAN requirements for R-3.1.0
- Minor changes to remove redundant files and folders
- Now uses
packageStartupMessage()
rather thanmessage()
to display startup message
- First experimental release to CRAN
Improvements:
- Performance gains as a result of using
plyr::quickdf()
rather thanas.data.frame()
- Imports
plyr
andXML
rather than declaring dependency
Improvements:
- Substantial gains in speed by rewriting
fast.read.fwf()
andchangeValues()
to uselapply()
rather than loops. - Documentation updates
Fixed issues:
- Fixed issues with converting logical values to TRUE/FALSE
- Convert values where appropriate
- Add
variable.lable
attribute containing question text
- Refactored all function names to use "camelCase"
- Implemented
read.fast.fwf()
to read fixed width file