- Fixes for compliance with quanteda v3 (thank you @kbenoit)
- All images for building the vignette now live inside the package
lnt_read_lines and lnt_read now ignore Word lock files
- Updated namespacing for planned v3 modularisation of quanteda (thank you @kbenoit)
- Fixed tests that caused problems on CRAN
- Added support for last remaining new format (zip).
- Fixed tests for compliance with dplyr 1.0.0 (dependency through tidytext).
- Added support for new formats introduced by Nexis Uni and Lexis Advance.
- Improved lnt_convert and added option for data.frame.
- Bug fixes and internal improvements.
- Added support for different download formats (.RTF, .DOC, .PDF[limited]).
- Fixed problems with changes in quanteda 1.5.
- Added function retrieve BibTeX entries from LNToutput.
- Added dim() method for LNToutput class.
- Improved lnt_lookup() by adding more word boundary options.
- Minor bug fixes in lnt_similarity().
- Improved way of telling users about missing packages (can now be installed directly).
- Improved headline cleaning in lnt_read() (Edition is now removed from headline).
- Minor bug fixes in lnt_similarity().
- Corrected inconsistency in lnt_convert().
- Some better error messages.
- Updated tests for lnt_diff() .
- Rewrote lnt_read() to be more stable (and faster) which rendered lnt_checkFiles() unnecessary (now deprecated).
- Added lnt_convert() to transform objects created by lnt_read() to formats used in popular text-as-data analysis packages.
- Enhanced lnt_similarity() which tended to crash when comparing longer texts due to memory limitations.
- Added lnt_diff() to display results from lnt_similarity() in a diff-like viewer.
- Added lnt_lookup() which can be used to check if the nexis keyword search worked properly or to apply simple dictionaries to subset the data.
- Added several methods to work with the S4 class
LNToutput.
- Added vignette with basic usage.
- Enhanced documentation of of all functions.
- Started using testthat tests