extract_tables()getsoutdirargument for writing out CSV, TSV and JSON files.- Fixes in vignette.
make_thumbnails()andsplit_pdf()now usetempdir()as the default output directory.extract_functions getcopyargument for copying original local files to R session's temporary directory.- PDF files used do not get copied to temporary directory by default.
- General clean-up for CRAN submission.
- Upgrade to PDFBox 2/Tabula 1.0.1 (#48)
methodargument is changed tooutputinextract_tables().- New
methodargument reflects method of extraction as in Tabula command-line Java utility. extract_text()acceptsareaas argument.
- Switch to using new document loading algorithm and localize all remote URLs. (#40)
- Removed kind of annoying message about overwriting a temporary file. (#36)
- Transferred source code repository to ropensci: https://github.com/ropensci/tabulizer
- Transferred source code repository to ropenscilabs: https://github.com/ropenscilabs/tabulizer
- Exposed a new option
widgetinlocate_areas()to control which widget is used in locating areas.
- Fixed bug in internal function
try_area_full()introduced by changes in8.
- Further expand the
locate_areas()interface to use a Shiny gadget when working within RStudio, or otherwise rely on the full functionality interface (based on graphics device events) or reduced functionality interface (relying onlocator()). (#8)
- Completely rewrite the
locate_areas()interface to rely on graphics device event handling where possible. This may behave differently across platforms or in RStudio. (#8)
- Fixed a bug in
extract_tables()such that when no tables are found, an empty list is returned (formethodvalues with list response structures). (h/t Lincoln Mullen)
split_pdfs()andmake_thumbnails()gain anoutdirargument to specify where to save the output. The file numbering of output files is also now zero-padded.- An warning in
merge_pdfs()has been fixed. stop_logging()is called when the package is attached to the search path.get_page_dims()earns adocargument and argument order inget_n_pages()is reversed.
- Added better support for specifying character encoding. (#10)
- Added support for password-protected PDF files. (#11)
- Expand file paths where needed. (h/t David Gohel)
- Improve handling of URLs when using
extract_areas()by downloading PDF to temporary directory.
- Exposed new functions
split_pdf()andmerge_pdfs()to split and merge PDFs, respectively. (#9) - Exposed a
get_n_pages()to determine the page length of a PDF document. - Moved tabula .jar file to separate package, tabulizerjars. (#2)
- Added a new function
extract_metadata()to extract PDF metadata as a list. - Added a new function
extract_text()to convert PDF contents to an R character vector. - Changed the internal
localize_file()function to use PDFBox to natively read from a URL. - Removed illogical default
fileargument value inextract_tables().
- Expanded test suite to cover
areasandcolumnsarguments and utilities. (#3) - Fixed the same bug in
make_columns()as was corrected formake_areas(). (#5)
- Fixed a bug in
make_areas()internal whenareawas specified as a length 1 list for a multi-page document. (#5, h/t Tony Hirst)
- Added a function,
extract_areas(), to interactively identify and extract page areas. Another new function,locates_areas()implements the locator functionality without performing any extraction. - Added a function,
make_thumbnails(), to convert pages into individual image files. - Added a function,
get_page_dims(), to extract page dimensions.
- Fixed a bug in the repeating of the
areaargument whenlength(area) == 1 & length(pages) > 1. (#5, #6)
- Fixed a bug in the handling of the
areaargument. (#5, #6)
- Added vignette. (#4)
- Added tests of guess parameter. (#3)
- Added a
spreadsheetargument, a la Tabula itself. - Fixed bugs in parsing of
areaandcolumnsarguments.
- Added multiple table writing options beyond the default list of matrices. (#1)
- Initial release.