Changelog

fuzzy_join now supports formula notation for match_fun, multi_match_fun, and index_match_fun.
Many changes to internals to make compatible with newest versions of dplyr and tidyr (#58 and #60, @jennybc)
difference, distance, geo and stringdist joins now add a distance column even if there are no overlapping rows (fixes #57)
distance joins now support the case where there’s only one column in common (fixes #43)
Fixed typos in documentation (#40, @ChrisMuir and #53, @brooke-watson)

Added interval_join, which joins tables on cases where (start, end) intervals overlap between the two columns. This adds IRanges from Bioconductor to SUGGESTS.
Added genome_join, which is a more specific case of interval_join that joins tables on based on (chromosome, start, end), where the chromosome must agree and (start, end) must overlap.
Added index_match_fun argument to fuzzy_join, which handles functions (such as interval_join and genome_join) that operate on the original columns rather than all pairs of columns
Fixed bug when matching multiple columns to the same column (#28)
Fixed bug in which rows were sometimes duplicated when no distance column was specified (#21)
Added more unit tests

Fixed bug that failed when single column data frames (not tbl_dfs) were given (#13)
Updated README for newest versions of dplyr and janeaustenr

Features

Added option not only to join based on a maximum distance but also to append a distance column, using the distance_col argument. This is available in difference_join, distance_join, stringdist_join, and geo_join (#10)

Fixed to ignore groups while preserving the groups of x in the output (#11)
Fixed to append .x and .y to all common columns, not just those in by (#12)