v1.3.2

• Use the new delayed S3 registration mechanism in 3.6.0 for dplyr method implementations.
• Bump minimum dplyr version required to 0.7.0.

v1.3.1

• Presto now might return data in POST response (#89)
• Presto now might not always return column information in each response (#93)
• Better error message for unknown column types (#86, #87)
• Adapt to presto changes for type casting translation (#88)
• Add CHAR data type (#91)

v1.3.0 as it appears on CRAN

• Fix Rcpp compilation under Windows (#79)
• SET/RESET SESSION queries are now correctly respected when used under dbGetQuery.
  PrestoConnection no longer has parameters slot but dbConnect remains backward
  compatible. Manual change to parameter is still possible via conn@session$setParameter() (#77)
• Adapt to changes in dplyr version 0.7.0, mainly around the remote
  backend support being split to dbplyr. This should be backwards compatible
  back to dplyr 0.4.3 (#76)
• Add support for the REAL data type (#70)
• Allow specifying the connection source (#68)
• Drop RCurl dependency (#67)
• Return DECIMAL data types as characters as opposed to numeric's
  previously (#64, fixes #60)
• Add support for new integer data types (INTEGER, SMALLINT, TINYINT) (#59, fixes #56)
• Migrate the json to data.frame construction from pure R to Rcpp for 10x
  speed gains! (thanks @saurfang) (#57, #58)
• Fix dbListFields to use the nextUri instead of infoUri (#55)
• Don't drop data for duplicate column names (#53)

v1.2.1

• Handle responses with no column information (fixes #49)
• Add retries for GET and POST responses with error status codes
• Skip test cases for ones that need locale modification if we cannot set the locale for the OS.
• Adapt to changes in the upcoming dplyr and testthat versions.

Release v1.2.0 as it appears on CRAN

• Add a session.timezone parameter to dbConnect and src_presto which
  defaults to UTC. This affects the timestamps returned for Presto data types
  "TIMESTAMP". We handle the ambiguity by assigning a time zone to every
  POSIXct column returned. Note that if you are doing as.character() directly
  on these columns, the values you obtain will be different from what happened
  before.
• Fix the way we handle zero row multiple column query results. This will
  affect LIMIT 0 queries specifically.