Releases: r-transit/gtfsio
Releases · r-transit/gtfsio
gtfsio v1.2.0
gtfsio v1.1.1
Bug fixes
- Now prevents
export_gtfs()
to save large round numbers in scientific notation. This was not exactly a bug, as the specification does not forbid it, but the behavior could interfere with the workflow of people using other applications (as shown in {gtfstools} #73). Also, improves the readability of the tables.
gtfsio v1.1.0
Bug fixes
- Fixed a bug in which
import_gtfs()
would append a.txt
extension to non text files found inside the GTFS feed, which would lead to an error when attempting to unzip the feed. The function now ignores non text files when reading feeds, and raises an informative warning message if any of such files is found (#23).
New features
import_gtfs()
now accepts paths and URLs without.zip
extension, as long as they still point to zip files (in other words, it accepts zip files that for some reason or another do not have.zip
extension).
Notes
- Started converting more warnings to messages when
quiet = FALSE
inimport_gtfs()
andexport_gtfs()
. - Updated standards to read and write GTFS tables and fields in R due to changes in the specification.
gtfsio v1.0.0
Bug fixes
import_gtfs()
would raise a{bit64}
-related warning when the first row of any table included a 64-bit integer. This is now fixed - when reading the first row to figure out which fields are present, all columns are read as character vectors.
New features
- Added (internal) input assertion functions, which declutter the input checking sections quite a bit.
- Added new (internal)
gtfsio_error()
function, which raises a custom-classed error condition. All errors raised in gtfsio's exported functions inherit fromgtfsio_error
and<function_name>_error
, which allows for easier and more clear error catching.
Potentially breaking changes
- Files and fields checking functions had their names changed. They now use the singular form, instead of the plural (i.e.
check_fields_exist()
was substituted bycheck_field_exists()
). This shouldn't raise reverse dependencies concern, because as of now only{gtfstools}
dev version uses such functions, which is an easy fix that doesn't concern CRAN release.
gtfsio v0.2.0
Bug fixes
import_gtfs()
would ocasionally include the full path to a table to their name in a GTFS object (#17). Fixed by Mark Padgham (@mapdge) in #18.
New features
- Added new
summary.gtfs
method. Thanks Mark Padgham (@mapdge). import_gtfs()
has a newencoding
parameter, used to handle encoded strings in their native encoding.
gtfsio v0.1.2
Bug fixes
- Fixed a bug (or perhaps, changed the behaviour) of
check_fields_exist()
andcheck_fields_types()
when a lower-level test (i.e. if the file that holds such field exists, for example) failed. Now it returnsFALSE
instead of raising an error.
gtfsio v0.1.1
Bug fixes
- Fixed a bug in
export_gtfs()
whenas_dir
was set toTRUE
andpath
was set totempdir()
in #15. The function now returns an (intentional) error iftempdir()
is passed topath
. Thanks Flavio Poletti (@polettif).
New features
import_gtfs()
has a newskip
parameter. It may be used similarly tofiles
, but you specify the files you don't want to read, instead of the ones you do. Thanks Flavio Poletti (@polettif).
gtfsio v0.1.0
- First CRAN release!