datavolley 1.14.0
- reading a dvw file with no play-by-play data will now return a datavolley object with zero-row
plays
component, rather than throwing an error. This is consistent with the behaviour when reading a vsmm file with no play-by-play data - various improvements for reading vsm files
- small improvements to text encoding detection, calculating home_score_start_of_point and visiting_score_start_of_point
- fix issue in dv_write for matches > 5 sets
- check for entirely missing sets in validation
datavolley 1.12.0
- if the dvw file being read has been populated with official match template data (i.e. an official federation match), use the match date information there to resolve the match data in the metadata section if it’s ambiguous
datavolley 1.11.0
- the
dv_sync_video
function now does not apply offsets by default - see the function documentation for details. To restore the previous behaviour usedv_sync_video(..., offsets = dv_sync_offsets())
but it is recommended to apply offsets in your video montage software, not during the sync process (hence the change to the default behaviour) - small changes to the default sync timings
- fix an error in the line numbers returned by some validation checks
datavolley 1.10.0
- the validation step is now less strict when checking substitutions. Some files will change the players in the lineup on the line before (or in some cases, several lines before) the actual substitution code. By default these now won’t be flagged, so long as everything comes out consistent before the next rally starts. Strict checking can be re-enabled via
dv_validate(..., validation_level = 3)
ordv_read(..., extra_validation = 3)
datavolley 1.9.0
- substantial refactoring of how text encoding is detected, which should be more reliable (particularly for DV4 files)
- speed improvements to file reading
datavolley 1.7.1
- the evaluation string used for a set with a “/” evaluation code has been changed from “Error (reach over net)” to just “Error” for consistency with other skills (applies to files using VolleyMetrics conventions only)
datavolley 1.6.3
- remove dependence on
enc
package, which has been archived from CRAN - enforce the
start_coordinate
column in meta$attacks table to be integer - other small improvements
datavolley 1.6.0
- propagate player name changes (made via
remap_player_names()
) into first and last names in metadata (thanks @awosoga) - added
remap_player_info()
function (thanks @awosoga) - other small bugfixes
datavolley 1.3.3
Minor updates: - initial setter assignments (at the start of each set) are no longer marked as substitutions - better handling of corrupted position coordinates in dvw files
datavolley 1.3.0
- the raster package has been shifted from Imports to Suggests, meaning that it won’t be installed by default. This only affects plotting of raster images (heatmaps) in base graphics. If you don’t have raster installed, it will prompt
datavolley 1.2.9
- convert colour columns to hex colour strings (previously integer-encoded)
- minor speed improvements when reading
- meta$winning_symbols is now a data frame
datavolley 1.2.3
- utilities to make files scouted by ClickNScout more useful
- optional xlim, ylim to ggcourt,
- add dv_find_to_flip_coordinates function
datavolley 1.2.0
- bug fix: actions scouted with an unknown player (denoted in the dvw file with player number “$$”) were previously being incorrectly parsed
datavolley 0.16.0
- add dv_read_sq function
- improvements to match date parsing
- add columns for scores at start of point (home_score_start_of_point, visiting_score_start_of_point)
- trim whitespace from team and player names and IDs
datavolley 0.15.0
- minor bugfixes for rotation check, validation of peranavolley files
- ‘beach’ and ‘sand’ ggcourt options
- fix set number and sets_won in partially-scouted files
datavolley 0.13.0
- adjustments to
team_touch_id
andphase
to account for files in which only attacks have been scouted -
point_phase
(this is the “Point/Side out” column in the DataVolley codes list window) andattack_phase
(“Attack after reception/dig”) columns added to the play-by-play data -
dv_write
function added - additional fixes following the changes to text encoding handling in v0.12.0
datavolley 0.12.0
- substantial changes to the way that text encoding is handled, including guessing the correct encoding with
read_dv(..., text_encoding = "guess")
. Text encoding should work better now, particular on Windows - added a
date_format
parameter toread_dv
, so that the user can specify the preferred date format to use when dates are ambiguous