v0.4

• Interface changes
  • glimpse() obtains default width from tibble.width option (#35, #56).
  • Don't export dim_desc() (#50, #55).
  • New has_rownames() and remove_rownames() (#44).
• Minor modifications
  • frame_data() returns 0-row but n-col data frame if no data.
  • [[.tbl_df() now falls back to regular subsetting when used with anything other than a single string (#29).
  • When used in list-columns, S4 objects only print the class name rather than the full class hierarchy (#33).
  • Add test that [.tbl_df() does not change class (#41, @jennybc).
  • Improve [.tbl_df() error message.
• Documentation
  • Improve documentation and vignette.
  • Update README, with edits (#52, @bhive01) and enhancements (#54, @jennybc).
• Code quality
  • Full test coverage (#24, #53).
  • Renamed obj_type() to obj_sum(), improvements, better integration with type_sum().
  • Cleanup for column_to_rownames() and rownames_to_columns() (#45).
  • Cleanup for repair_names() (#43). Whitespace are not touched by this function (#47).
  • Cleanup for add_row() (#46).
  • Regression tests load known output from file (#49).
  • Internal cleanup.
  • Don't test R-devel on AppVeyor because of missing directory on CRAN.

v0.3

• Features
  • New as_data_frame.table() with argument n to control name of count column (#22, #23).
  • New function repair_names() fixes missing and duplicate names (#10, #15, @r2evans).
  • frame_data() now also creates a list column if one of the entries is a list (#32).
  • New rownames_to_column() and column_to_rownames() functions, replace add_rownames() (#11, @zhilongjia).
  • Use tibble prefix for options (#13, #36).
• Documentation
  • Add pre-tibble NEWS (#39, #40).
  • Include vignette (#38).
  • Expand README.
  • Fix typos in documentation.
  • Remove use of src() from examples.
• Prepare CRAN release
  • Use new-style .travis.yml
  • Use AppVeyor for testing.
  • Finer coverage analysis (#37).
  • Check with win-builder and valgrind.
  • Fix NOTE from R CMD check.

v0.2

• Functions related to tbl and src stay in dplyr (#26). Remove unused make_tbl().
• Non-scalar input to frame_data() and tibble() (including lists) creates list-valued columns (#7).
• Use C++ implementation for as_data_frame.matrix() (#14). Also add former matrixToDataFrame() tests, and fix unwanted conversion to factor.
• as_data_frame(NULL) is 0-row 0-column data frame (#17, @jennybc). frame_data() and tibble() create empty data_frame if no rows are given (#20).
• data_frame(NULL) raises error "must be a 1d atomic vector or list".
• lst(NULL) doesn't raise an error anymore (#17, @jennybc), but always uses deparsed expression as name (even for NULL).
• trunc_mat() and print() use width argument also for zero-row and zero-column data frames (#18).
• glimpse() now (invisibly) returns x, so it can be used within a chain of dplyr verbs (@edwindj).
• base::getElement() now works with tibbles (#9).
• Remove spurious usage of "dplyr" in documentation (#3).
• Almost full test coverage.

v0.1

First release. Contains functions related to table sources, the tbl class and the tbl_df subclass, as well as output and helper functions:

Table sources

• src()
• src_df()
• src_local()
• src_tbls()
• is.src()
• same_src()

The tbl class

• tbl()
• tbl_df()
• make_tbl()
• tbl_vars()
• as.tbl(), is.tbl()

The tbl_df subclass

• as_data_frame()
• data_frame(), data_frame_()
• frame_data(), tibble()

Output functions

• glimpse()
• trunc_mat(), knit_print.trunc_mat()
• dim_desc()
• type_sum()

Helper functions

• lst(), lst_()
• add_row()
• add_rownames()